Pasar al contenido principal

House-Passed and Senate Health Bills Reduce Deficit, Slow Health Care Costs, and Include Realistic Medicare Savings

Congress Has Good Record of Implementing Medicare Savings; Medicare Savings in These Bills Are Similar to Those Implemented in the Past


Health reform legislation that has passed the House in one form and is before the Senate in another is facing a series of attacks that, taken together, suggest the legislation would do little to control health care costs and would increase budget deficits. Many of these charges are exaggerated or simply incorrect, based on the Center’s careful analysis of the legislation. In particular, a number of criticisms rest on a mistaken belief that, in recent years, Congress has repeatedly enacted provisions to achieve savings in Medicare and then generally blocked these provisions before they could take effect. Thus, critics say, no one should take seriously the provisions of the current bills that would produce Medicare savings. In fact, the Center’s analysis of major legislation affecting Medicare that Congress has enacted over the last two decades shows that Congress has permitted the vast majority of Medicare savings to take effect.

A number of commentators have leveled at least four distinct, though related, lines of attack either on the House-passed bill or on both the House bill and the pending Senate bill. As summarized below and explained in more detail in subsequent sections of this paper, these attacks do not hold up well under scrutiny:

  • First, despite charges to the contrary, the health bills approved by the House and pending in the Senate contain a wide range of measures to reform the nation’s health care system and moderate the growth of health care costs over time, particularly in Medicare. To be sure, policymakers ultimately will have to do much more to restrain health care costs. But the bills take most of the steps that we know enough about to pursue now in most of the areas that experts view as promising avenues for restraining health care spending. When Congress reconciles these bills in a conference committee, lawmakers should combine the strongest cost-control elements of each.

    In Medicare, the biggest savings in the bills stem from reducing or eliminating overpayments to private insurance companies that participate in Medicare Advantage, reducing annual payment updates for hospitals and other providers, and, in the House bill, lowering prescription drug costs. For health care as a whole, both bills include measures creating a health insurance exchange to promote competition among insurance plans, reducing the amount that health insurers spend on administrative costs, expanding research on the comparative effectiveness of different medical procedures, investing in preventive care, penalizing hospitals with excessive readmission rates, and establishing pilot projects in Medicare for bundling payments for inpatient and post-acute care. In addition, the Senate bill would impose an excise tax on high-cost insurance plans to discourage excessive health care utilization and create an independent Medicare Advisory Board to develop and submit proposals to slow the growth of Medicare spending and improve the quality of care; the board’s recommendations would take effect automatically unless Congress later modified or overturned them. (For a more in-depth discussion of these issues, see “Both Bills Contain Wide Range of Cost-Control Measures,” below.)
  • Second, the history of health legislation in recent decades demonstrates that, despite some critics’ charges, Congress has repeatedly adopted measures to produce considerable savings in Medicare and has let them take effect. For example, Congress took such action as part of major deficit-reduction packages in 1990 and 1993 and as part of more modest deficit-reduction packages in 1997 and 2005. Virtually all of the cuts that it enacted in 1990, 1993, and 2005 went into effect. After Medicare spending slowed dramatically after 1997 — in 1999, it was for the first time lower than it had been the year before — and the budget was balanced in 1998, Congress did ameliorate some of the Medicare cuts that it had enacted in 1997. But, even in those special circumstances, it allowed four-fifths of the 1997 cuts (other than those described in the next paragraph) to take effect.

    In arguing that Medicare cuts never “stick,” critics point in particular to Congress’ repeated refusal to let the reductions in physician reimbursement rates under Medicare’s so-called “sustainable growth rate” (SGR) mechanism, which it enacted in 1997, take full effect. The SGR cuts, however, represented a badly designed measure that was not intended to produce large savings (the projected SGR savings represented less than five percent of total Medicare savings in the 1997 bill), but turned into a blunt instrument that would have produced cuts far in excess of what was anticipated and would have had harsh and indefensible effects. (Moreover, even though Congress did not allow the full cuts required under the SGR formula to take effect, it has still cut the physician reimbursement rate substantially — at its current level, the reimbursement rate in 2010 will be 17 percent below the rate for 2001, adjusted for inflation.) The SGR mechanism has little in common with most of the other provisions that Congress has enacted over the years to produce savings in Medicare and that have, in fact, taken effect. This distinction is important because most of the Medicare savings provisions in the House and Senate health reform bills are similar in nature to the types of Medicare provisions that Congress has enacted in the past that have taken effect — and they differ markedly from the blunt-instrument design of the SGR cut. (For a more in-depth discussion of these issues, see “SGR and Other Medicare Cuts — What Has Worked and Why,” below.)
  • Third, critics complain that, under the House and Senate bills, total health care expenditures in the United States will rise in the near term, not fall. That’s no surprise, however, nor does it suggest something troubling about health reform. Providing insurance coverage for tens of millions of uninsured Americans will necessarily raise total health care spending in the short term. The real issues here are: (1) whether health reform includes provisions to cover the costs of these insurance expansions so that deficits and debt do not increase; and (2) whether health reform includes steps that begin to slow the rate of health cost growth so that total health spending is lower in the longer run than it otherwise would be. The House and Senate bills meet the first test, according to the Congressional Budget Office (CBO), with the House bill reducing deficits by $138 billion over ten years, the Senate bill reducing deficits by $130 billion over that period, and both bills continuing to reduce deficits for at least a decade after that. The bills also hold promise for the second test, CBO says, although policymakers will need to do more to slow health cost growth as we learn more about how to do it, such as by applying the knowledge gained in the coming decade from pilot projects in the health bills, comparative effectiveness research, and the like. (For a more in-depth discussion of these issues, see “Bills Would Begin to Slow Growth in Health Expenditures,” below.)

    On a related issue, critics complain that the CBO cost estimates showing that the bills would reduce the deficit are misleading and rest upon a gimmick — specifically, that neither the House nor the Senate bill includes a measure to permanently eliminate the SGR mechanism (as discussed in the second bullet, above). Since Congress likely will continue to prevent the SGR from taking effect, these critics say, Congress and CBO should consider the cost of such action as part of the cost of the health reform bills. Once that cost is added, they argue, the contention that the bills do not increase the deficit is false. Indeed, Congress likely will never let the full SGR cuts take effect, and it probably won’t offset the cost of scrapping them. But that cost is neither part of, nor in any way a result of, health care reform — the federal government will incur this cost regardless of health care reform, not because of it. This fact is undeniable: if health reform legislation were to die tomorrow, the full SGR cost would remain. To be sure, it would be better if Congress offset the cost of cancelling the SGR cuts. But that issue is separate from the question of whether the health care reform bills themselves add to the deficit or not.

Both Bills Contain Wide Range of Cost-Control Measures

The health reform bills passed by the House of Representatives (H.R. 3962) and introduced by Senate Majority Leader Reid (as an amendment to H.R. 3590) contain a wide range of measures to restructure the U.S. health system and slow the growth of health care costs, particularly Medicare costs. The bills begin to move in most of the areas that health policy experts consider promising avenues for reducing the growth of health care spending and where specific steps can be identified. “Pretty much every proposed innovation found in the health policy literature these days is encapsulated in these measures,” John Iglehart, founding editor of Health Affairs, recently wrote in the New England Journal of Medicine. (See Box 1 below.)

Similarly, Jonathan Gruber of MIT, one of the nation’s most respected health economists, recently said of the Senate bill, “It’s really hard to figure out how to bend the cost curve, but I can’t think of a thing to try that they didn’t try. They really make the best effort anyone has ever made. Everything is in here.”[2]

To be sure, some of these ideas could be pursued more aggressively than the bills would require. And continued intensive efforts will be needed for the foreseeable future, since we have much to learn about containing health care costs without compromising health care quality. Nevertheless, the House and Senate bills would take major steps to begin controlling health care costs.

Efficiencies in Medicare and Medicaid

The bills include a number of provisions that would make Medicare more efficient, providing significant savings that would help pay for health reform. [3] Many of these provisions are in line with the recommendations of the Medicare Payment Advisory Commission (MedPAC), Congress’ expert, nonpartisan advisory body on Medicare payment policy, on how to modify provider payment rates and encourage efficiency while assuring that payments are adequate so that beneficiaries continue to have access to providers. In recent years Congress has ignored or failed to act on a number of important MedPAC recommendations, but the current health reform bills would enact many of them into law, despite the opposition of powerful interest groups.

In dollar terms, the bulk of the bills’ spending reductions come in three areas:

  • Reducing or eliminating Medicare Advantage overpayments . MedPAC estimates that in 2009, Medicare will pay private insurers that participate in Medicare Advantage 14 percent more per beneficiary, on average, than it would cost to cover these beneficiaries in traditional Medicare. The Senate bill would scale back these overpayments, saving nearly $120 billion over ten years. The House bill would phase them out altogether over three years, as MedPAC has recommended, saving $170 billion. (All dollar estimates cited here are CBO estimates.)
  • Reducing the annual updates in Medicare fee-for-service payment rates . Medicare payment rates for covered services are updated annually according to formulas specified in law. Both the House and Senate health reform bills would reduce annual payment updates to hospitals, skilled nursing facilities, hospices, ambula­tory surgical centers, and certain other health care providers to account for improvements in economy-wide productivity. They would also reduce payments to home health agencies, skilled nursing facilities, and inpatient rehabilitation facilities, as MedPAC has recommended. The House bill would save $228 billion over ten years from such changes; the Senate bill would save $192 billion.
  • Reducing prescription drug costs . The House bill would require drug manufacturers to provide, at a minimum, the same rebates for drugs provided to dual eligibles (low-income Medicare beneficiaries who are also enrolled in Medicaid) under Medicare’s prescription drug benefit that Medicaid formerly required for those drugs. Even after devoting some of the resulting savings to filling the so-called “doughnut hole” in the Medicare drug benefit, this provision would produce net savings of $42 billion over ten years. Both the House and Senate bills would also increase the rebates that drug companies pay for the drugs that Medicaid covers, saving $25 billion and $38 billion, respectively.

Systemic Reforms in Health Care Payment and Delivery

The bills also take numerous important steps toward restructuring the health care payment and delivery systems to move away from paying providers for more visits or procedures and toward rewarding effective, high-value health care. Most of these provisions are not estimated to save much money in the next ten years because their effects — while promising — are unproven, but they constitute important initial efforts to slow the growth of health care costs and might well lead to larger savings than the official cost estimates suggest. Moreover, the proposed reforms are likely to reinforce each other and have a combined effect that exceeds the sum of the individual parts.

A large number of the proposals involve Medicare, which has been a leader in developing and testing effective payment reforms that private insurers later adopt widely. As the largest U.S. purchaser and regulator of health care, Medicare exerts a major influence on the rest of the health care system; its reimbursement and coverage policies have served as models for private insurers and other public programs. For example, many private insurers follow Medicare’s lead in approving coverage of new medical technologies. Over the years, the private sector has also typically followed Medicare’s lead in adopting new payment mechanisms — including the prospective payment system for hospitals and fee schedules for physicians. [4]

Box 1: Experts Affirm That Health Reform Bills Include Important Cost-Control Measures

“The bills contain no shortage of ideas for reforming the delivery system, enhancing the quality of care, and slowing spending. Pretty much every proposed innovation found in the health policy literature these days is encapsulated in these measures.”

— John Iglehart, founding editor, Health Affairs
New England Journal of Medicine,
November 11, 2009

“The [House] legislation . . . takes important steps to ‘bend the cost curve.’”

— Henry Aaron, Brookings Institution; Jonathan Gruber, MIT and ten other prominent health policy experts
An Open Letter on House Consideration of H.R. 3962
, November 6, 2009

“[The House bill] contain[s] a laundry list of virtually every idea for improving the delivery, enhancing the quality, or controlling the cost of medical care now current. It is like they read through the table of contents of every Health Affairs for the past five years.”

— Timothy S. Jost, professor of health law, Washington and Lee University
Health Affairs
blog, October 31, 2009

While the House bill goes further in some aspects of cost control (particularly involving Medicare) and the Senate bill goes further in others (particularly its excise tax on high-end insurance plans), in many respects the two bills are comparable. Provisions of the bills that could start to “bend the cost curve” include:

  • Creating a health insurance exchange . Both bills would create a health insurance exchange that would offer eligible individuals and employers a choice of insurance options that are affordable and comprehensive. A well-structured exchange should reduce administra­tive costs and promote competition among insurers based on the cost and the quality of their products, not their ability to maximize profits by attracting healthy, less costly enrollees and avoiding sicker, more costly ones. (The House bill has more of the elements needed to produce this outcome than the Senate bill.) Economist Henry Aaron has termed the exchange a potentially “revolutionary innovation” that could become “capable of effecting real systemic change.” [5]
  • Establishing an excise tax on high-cost insurance plans. This provision, which only the Senate bill includes, would discourage insurers from offering, and firms from purchasing, very generous health insurance coverage that can encourage excess health care utilization. According to CBO, limiting the open-ended tax exclusion for employer-sponsored health insurance is one of “two powerful policy levers” the federal government has available to encourage changes in medical practice and slow health care costs.[6]
  • Reducing administrative costs. High administrative costs are one reason that health care is more costly in the United States than in other western industrialized nations, and both bills would reduce the amount that insurers spend on administrative overhead rather than health care. The bills take steps toward standardizing transactions between insurers and providers, including transactions related to enrollment, eligibility determina­tion, prior authorization, and claims. The House bill would also require insurers to spend at least 85 cents of each premium dollar on health care.
  • Researching comparative effectiveness . Both bills would establish an independent entity with a dedicated source of funding to conduct and synthesize research on the comparative effectiveness of different medical services, treatments, and items. According to CBO, “Such research holds the potential to reduce health care costs over the long term — possibly by substantial amounts if it is done rigorously and if its results are ultimately tied to changes in financial incentives for providers and consumers.”[7]
  • Promoting prevention and wellness . Both bills include provisions aimed at preventing disease and encouraging wellness and healthy behaviors, including coverage of additional preventive services in Medicare, Medicaid, and private health insurance.
  • Licensing follow-on biologics . Both bills would establish an abbreviated regulatory procedure for approving generic versions of biological drugs (drugs derived from living organisms). Public and private purchasers of drugs would save money from the availability of these new, lower-priced versions.
  • Strengthening primary care. Improving primary care and care coordination is widely viewed as a promising way of increasing value and achieving savings in U.S. health spending. [8] Both bills would increase payments to primary care providers in Medicare and provide incentives to increase the number of nurses and doctors in primary care. They would also expand efforts to assess the feasibility of making Medicare payments to qualified patient-centered medical homes — a model in which each patient has an ongoing relationship with a primary care physician who leads a team that coordinates and takes responsibility for his or her care.
  • Establishing quality measures and priorities . Both bills would direct the HHS Secretary to establish national priorities for improving the quality of health care services and health outcomes and to develop new patient-centered and population-based measures of quality. Such a foundation of information could provide essential guidance in moving toward a health care system that slows spending growth while improving value.[9]
  • Promoting high-value care. Experts agree that slowing the growth of health costs requires moving toward a payment system that rewards providers based on the value of their care, not just the volume of their procedures. The Senate bill would create systems of value-based payments in Medicare for physicians, hospitals, skilled nursing facilities, and home health agencies. The House bill would direct the Institute of Medicine (the health arm of the National Academy of Sciences) to study the extent and causes of geographic variation in health spending and how to promote high-value care. It would also instruct the HHS Secretary to develop a plan for modifying Medicare payments to implement these recommendations, which would go into effect unless Congress disapproves.
  • Establishing a center for innovation . Both bills would establish a center for innovation within the Centers for Medicare & Medicaid Services to test and evaluate different payment structures and methods to foster patient-centered care, improve quality, and reduce the cost of care in Medicare and Medicaid. The HHS Secretary would also be authorized to expand the scope and duration of approaches being tested, including implementation on a nationwide basis, without any congressional action, if they are found to reduce spending without reducing the quality of care.
  • Enhancing program integrity. The bills contain numerous provisions to improve program integrity and reduce fraudulent payments in both Medicare and Medicaid. These include strengthened legal requirements and authorities to prevent fraud and abuse and to facilitate its detection, enhanced penalties for violations, and increased funding for enforcement.
  • Reducing avoidable hospital readmissions. MedPAC has found that nearly 18 percent of hospital admissions among Medicare beneficiaries in 2005 occurred within 30 days after the individual was discharged from the hospital. MedPAC also found that hospitals could have prevented some of these readmissions and saved money by providing better care during the initial stay or better follow-up care after discharge.[10] The bills would reduce Medicare payments to hospitals with high readmission rates to encourage them to do a better job of preventing avoidable readmissions.
  • Promoting accountable care organizations. Both bills would create new Medicare payment models to reward accountable care organizations (ACOs) — physician-led organizations that take responsibility for the cost and quality of the care they deliver. Many analysts believe ACOs can significantly reduce costs. [11] The Senate bill would allow groups of providers who meet specified criteria to qualify as ACOs and thereby receive a share of any savings they achieve. The House bill would create an ACO pilot program and require the Secretary to expand its geographic scope if it proves successful in reducing costs and improving quality.
  • Examining payment bundling. Both bills would establish pilot programs for bundling payments for Medicare services that hospitals and post-acute care providers (such as nursing homes and rehabilitation facilities) provide during an episode of care. They would require the Secretary to develop a plan to implement bundled payments if the pilot programs prove successful. If providers received a single payment for hospital and post-acute services rather than a separate reimbursement for each service, they would have a greater incentive to coordinate and deliver more cost-effective care. Bundling payments, encouraging accountable care organizations, and reducing avoidable hospital readmissions would also help move towards a payment system that promotes high-value, high-quality care.

These reforms are likely to complement and reinforce each other and to be much more effective in combination than separately. If given the appropriate financial incentives, providers will find ways to strengthen the delivery of primary care, use the results of comparative effectiveness research to select high-value treatments, employ electronic health records (for which the February 2009 stimulus bill provides substantial funding) to manage and coordinate care, and make other needed improvements in the health care delivery system.

Some analysts have criticized the bills’ reliance on pilot projects, pointing out that many previous pilot projects have not led to changes in Medicare. Yet in most of those cases, the Administration and Congress did not expand the experiment because it failed to save money or improve care. [12] In some cases, political pressure from doctors, hospitals, or other providers caused Congress to block implementation, but the health reform bills attempt to avoid this roadblock by giving the Secretary authority to implement successful pilots without congressional approval. Congress could further strengthen these provisions to require prompt implementation of approaches that prove promising.

Some critics also fail to acknowledge that many highly touted proposals for systemic health reform are not ready for immediate large-scale implementation. As Drew Altman, president of the Kaiser Family Foundation and a leading health care expert, has written:

There are still many very serious question marks here. Can we figure out how to bundle payments? Will Accountable Care Organizations take off or remain limited to the relatively small number of integrated health systems that already exist today? Is this just the new buzz word in health care like so many before it? How long will it take for comparative effectiveness research to begin to pay off, and will we be able to figure out a system for incorporating it into coverage and payment decisions without raising the spectre of “rationing”? Can all of the promising delivery reform efforts now underway across the country to make the health care system more cost effective “scale up” in a way that is measurable, not just at [the] Mayo [Clinic] or for one company here or there but to really bend the curve?[13]

Like it or not, answering these questions will require time and testing.

Medicare Advisory Board

In addition to making specific changes in law affecting health care payments and delivery, the Senate bill includes a provision that would go into effect (or be “triggered”) if Medicare spending growth is projected to exceed specified levels.

The Senate bill would establish an Independent Medicare Advisory Board to develop and submit proposals to slow the growth of Medicare spending and improve the quality of care. The President would nominate the board’s 15 members, who would require Senate confirmation, for staggered six-year terms.

If the projected growth in Medicare costs per beneficiary in 2015 and thereafter exceeded a specified target level, the board would be required to submit a proposal to Congress to eliminate the difference. The required reductions could not exceed 0.5 percent of Medicare spending in 2015, 1.0 percent in 2016, 1.25 percent in 2017, and 1.5 percent in 2018 and thereafter. These limitations are prudent — larger reductions would run the risk of being too severe, adversely affecting beneficiaries’ access to providers, and might therefore be overturned by Congress. The board could not propose increases in Medicare premiums or cost-sharing or cuts in Medicare benefits or eligibility criteria; it would focus on proposals for savings in the payment and delivery of health care services. The board’s recommendations would go into effect automatically unless both houses of Congress passed, and the President signed, legislation to modify or overturn them.

SGR and Other Medicare Cuts — What Has Worked and Why

Despite CBO’s estimates that the House and Senate bills would modestly reduce deficits over each of the next two decades, a number of critics continue to assert that the legislation would increase deficits and debt. These critics claim that the Medicare savings that CBO estimates the legislation would produce (which would help offset the costs of expanding health care coverage) are unlikely to be realized because Congress and the President will weaken or repeal those provisions in coming years.

To support this claim, these critics point to the steps lawmakers have taken in recent years to prevent cuts in Medicare enacted in 1997 — particularly reductions in physician reimbursement rates required under the sustainable growth rate (SGR) procedures — from taking effect. The SGR experience, they argue, shows that lawmakers will eventually bow to pressure from affected providers and beneficiaries and make sure the promised Medicare savings in the health reform bills never actually occur.

This criticism is misplaced, as a careful examination of the historical record and the specific Medicare policy changes in the health reform bills makes clear:

  • Every significant deficit-reduction package in the last 20 years has included Medicare savings, most of which have been implemented as planned. Virtually all of the Medicare cuts enacted in 1990 and 1993, which accounted for a significant portion of the savings in those large deficit-reduction packages, were implemented. And most of the savings enacted in 1997 other than the SGR cuts — nearly four-fifths — were implemented as well.

    Given that Medicare spending growth slowed significantly more than was anticipated after 1997 — in 1999, for the first time ever, it was actually lower than the previous year’s level — and the budget was balanced in 1998 for the first time in 28 years, it is surprising that Congress did not scale back even more of the savings enacted in 1997. There is little likelihood that the positive budgetary outlook that encouraged some easing of the 1997 cuts will return in coming years.

  • Congress did not anticipate or claim big savings from the SGR in 1997. CBO estimated that the SGR would save only $5 billion over five years and $12 billion over ten years, and would account for less than 3 percent of the total ten-year mandatory savings in the 1997 legislation that it was part of. Subsequently, the SGR turned out to have much larger and harsher effects than Congress intended and CBO anticipated, and Congress acted to prevent the much larger (and harsher) than anticipated cuts from taking effect.
  • It now is clear that the SGR provision was badly designed. Contrary to what policymakers intended in 1997, it turned out to be a blunt instrument that could have significant adverse side-effects. Congress’ decision to forestall large SGR cuts that it never intended to impose was justified on policy grounds.
  • The cost-saving Medicare changes in the health reform bills are similar to past Medicare changes that have taken effect. They generally are not similar to the flawed SGR provision.

The following pages discuss these points in more detail.

Most Previous Medicare Cost-Saving Provisions Have Taken Effect

Significant savings in Medicare have been a major component of every major deficit-reduction package enacted since 1990. [14] (See Box 2 below for details.) In the vast majority of cases, these savings were implemented with little or no modification, despite political pressures to keep them from going into effect.

Box 2: Medicare Cuts Have Been Major Parts of Deficit-Reduction Packages

Medicare savings have been a major part of every one of the 11 major deficit-reduction packages enacted since 1981.a The four most recent of these packages — those enacted in 1990, 1993, 1997, and 2005-2006 — are described below.b

  • 1990. CBO estimated that the Omnibus Reconciliation Act of 1990 reduced the deficit by $482 billion over five years, from fiscal years 1991 through 1995. c (Savings equal to the same share of the Gross Domestic Product would total $1.2 trillion over fiscal years 2011-2015.) Some $43 billion of the law’s savings came from Medicare, including $33 billion in lower reimbursement payments to physicians and hospitals and $10 billon in higher premiums and deductibles for beneficiaries. The law also raised the ceiling on wages and self-employment income subject to the Medicare payroll tax to $125,000, generating $27 billion. (In 1990 and 1993, CBO’s estimates covered only five years.)
  • 1993. CBO estimated that the Omnibus Reconciliation Act of 1993 reduced the deficit by at least $433 billion over five years (fiscal years 1994-1998). d Reductions in projected Medicare spending accounted for $56 billon of this amount. The Medicare savings primarily reflected lower payments to hospitals, physicians, and other providers, but the law also saved $8 billion by requiring Medicare Part B premiums to cover 25 percent of the cost of physician-related services through 1998.
  • 1997. CBO estimated that the Balanced Budget Act of 1997 reduced the deficit by $127 billion over five years (fiscal years 1998-2002). Medicare savings accounted for $116 billion of this amount. Over ten years, Medicare accounted for $394 billion of the bill’s $402 billion in mandatory savings. The legislation contained numerous provisions to slow Medicare costs, such as reducing the prices that fee-for-service providers received per unit of service, maintaining Part B premiums at 25 percent of the cost of services rather than allowing premiums to decline as a share of the costs, and slowing the growth in payments to capitated plans. The legislation also included the failed SGR mechanism, but this provision represented less than 3 percent of the projected Medicare savings.
  • 2005-2006. CBO estimated that the Deficit Reduction Act of 2005 (which was enacted in early 2006) would reduce mandatory spending by $39 billion in fiscal years 2006-2010, with Medicare savings accounting for $6 billion of the savings. Over ten years, Medicare accounted for $22 billion of the estimated $99 billion in savings. The largest savings came from reducing payments for a variety of services, particularly imaging and home health services, and limiting special payments to disproportionate share hospitals (DSH), which serve large numbers of uninsured patients.

a These include the Omnibus Budget Reconciliation Act of 1981, the Tax Equity and Fiscal Responsibility Act of 1982, the Deficit Reduction Act of 1984, the Consolidated Omnibus Budget Reconciliation Act of 1985, the Omnibus Budget Reconciliation Acts of 1986, 1987, 1989, 1990, and 1993, the Balanced Budget Act of 1997, and the Deficit Reduction Act of 2005 (enacted in 2006). See, Committee on Ways and Means, U.S. House of Representatives, 2004 Green Book: Background Material and Data on the Programs Within the Jurisdiction of the Committee on Ways and Means, March 2004, Table 2-45, pp. 2-144 to 2-145.

b The 1990 and 1993 budget reconciliation bills each included net revenue increases and spending cuts that reduced the deficit by roughly $500 billion over five years. In both 1997 and 2005, two separate reconciliation bills were enacted: one that reduced spending and one that cut taxes. In 1997, program reductions totaled $195 billion over five years, but the deficit was reduced by only $117 billion over that period when both reconciliation bills are considered, because of the tax cuts. In 2005, program cuts totaled $39 billion over five years, but the two reconciliation bills together increased the deficit by $37 billion over that period for the same reason.

c See Congressional Budget Office, Reducing the Deficit: Spending and Revenue Options, March 1994, p. 5.

d CBO, Reducing the Deficit, p. 7. This estimate depends on the baseline from which the reductions are calculated. CBO estimated that the legislation would reduce the deficit by $564 billion over five years relative to a baseline that assumed discretionary spending would grow at the rate of inflation from the 1993 level.

For example, nearly all of the substantial Medicare savings in the 1990 and 1993 budget reconciliation bills went into effect. Along with a vibrant economy, they played a significant role in producing budget surpluses in 1998 through 2001.  Also, virtually all of the savings enacted in the 2005 Deficit Reduction Act took effect.

Congress and the President did moderate some of the Medicare cuts enacted as part of the 1997 Balanced Budget Act (BBA), but even there, the vast majority of the intended cuts were allowed to take effect.  For example, CBO estimated that the law’s Medicare provisions (other than the SGR cuts, which accounted for a small fraction of the total savings) would reduce Medicare spending by $56.7 billion in 2007, the last year the CBO estimate covered.  According to CBO estimates, Medicare changes enacted in the subsequent nine years (other than changes in the SGR and the addition of the new Medicare prescription drug benefit in 2003) increased program spending in 2007 by a total of $11.3 billion. [15] By this reckoning, even if all of this increase represented backsliding on the savings enacted in 1997, [16] $45.4 billion — four fifths — of the Medicare savings promised for 2007 in the 1997 legislation survived. This finding belies the contention that Congress and the President are not willing to let promised cuts in Medicare take effect.

Furthermore, the decision to relax a modest portion of the savings enacted in 1997 must be considered in light of what was happening with Medicare and the budget as a whole. In the ten years before 1997, Medicare spending grew at an average rate of about 10 percent per year. [17] But, after 1997, spending slowed dramatically — much more than was anticipated as a result of enactment of the BBA.[18] In 1998, Medicare spending grew by only 1.5 percent, and in 1999 — for the first time in the program’s history — Medicare spending was lower than in the previous year. In 1999, Medicare spending was $24 billion (or 10 percent) lower than CBO’s September 1997 projection for that year, which took into account the savings enacted a month earlier in the BBA.

In addition, after 28 years of deficits, the federal budget was in surplus in 1998, 1999, 2000, and 2001. In 2001, both CBO and the Office of Management and Budget projected that if policies remained unchanged, the federal government would amass $5.6 trillion in surpluses over the next ten years. Then-Federal Reserve Board Chairman Alan Greenspan warned Congress that the projected surpluses were too large to be healthy for the economy because the United States would pay off the entire national debt and start using remaining surpluses to buy parts of private companies. Lawmakers’ focus turned from deficit reduction to proposals that would reduce revenues and increase spending. Congress and the President enacted large tax cuts in 2001 and 2003 and a new Medicare prescription drug benefit in 2003, none of which they paid for.

Given these developments, it is not surprising that lawmakers were receptive to requests from certain Medicare providers to modify some of the changes they had enacted to slow Medicare growth in order to reduce deficits. In 1999 and 2000, Congress provided relief from planned cuts in payments to various providers and increased payments to private capitated plans (plans that are paid a set annual amount per beneficiary) serving Medicare beneficiaries under the part of Medicare now called Medicare Advantage.[19] Further changes in payments to some providers and to Medicare Advantage plans were part of the 2003 law that established the prescription drug benefit. Yet despite these changes, nearly four-fifths of the Medicare savings enacted in 1997 outside of the SGR mechanism (which, as noted, was intended to produce only very small savings) survived.

Moreover, it is highly unlikely that the budgetary situation that made Congress more receptive to these changes will be repeated. There seems to be virtually no chance that sustained budget surpluses will reappear anytime in the foreseeable future or that the growth of Medicare spending will slow to near zero, let alone decline.

Finally, some observers have suggested that the Medicare cuts in the pending health reform bills are much larger than the cuts enacted in previous years. Relative to the size of the Medicare program, however, the proposed cuts are not unusually large.

CBO estimates that the net Medicare savings in the bills total around $450 billion over ten years, or about 6 percent of the total Medicare spending (excluding premium income) CBO projects over that period. That is only slightly more than the Medicare savings in the 1990 and 1993 deficit- reduction bills (which equaled about 5 percent of projected spending over five years) and is less than the savings in the BBA of 1997 (which equaled 9 to 12 percent of projected spending over ten years, depending on whether savings from higher Medicare premiums are taken into account). There is no reason to think that the size of the cuts in the health reform bills makes them less likely to be sustained than the previous cuts were. (Even after one adjusts for savings that Congress did not allow to take effect, the savings in the BBA were larger than those in the current health reform bills.)

SGR Was Not Intended to Produce Substantial Savings

Critics who now point to the SGR as an example of Congress claiming large savings from a provision that it then does not allow to take effect have not carefully examined the historical record. CBO estimated that the 1997 BBA would reduce mandatory spending by $119 billion over five years and $402 billion over ten years, with savings in Medicare representing the bulk of these promised savings — $116 billion over five years and $394 billion over ten years.

But the SGR provision would save only $5 billion over five years and $12 billion over ten years, according to CBO, less than 5 percent of the bill’s total five-year Medicare savings and 3 percent of its ten-year savings. In other words, if the SGR provision had not been included in the 1997 law to begin with, 97 percent of the planned Medicare savings would have remained. Clearly, Congress’ failure to allow the SGR cuts to take effect after 2002 did not undo a large portion of the Balanced Budget Act’s Medicare savings.

SGR Approach Is Fundamentally Flawed

The President and Congress have refused to allow the reductions in physician reimbursement rates required under the SGR formula to take effect since 2002. The primary reason is that the SGR approach is fundamentally flawed, and the cuts it would have imposed far exceeded what Congress had anticipated in 1997 — or what would represent defensible policy. Although the opposition of physicians played a role in these decisions, other, better designed and more defensible Medicare cuts that politically powerful providers also opposed were allowed to take effect.

The fundamental flaw is that the SGR formula attempts to control the rate of growth in total Medicare spending for physician-related services just by limiting the rates at which physicians are reimbursed for the services they provide. But overall spending for care provided to Medicare beneficiaries through physicians depends only in part on physician reimbursement rates — it also depends on the number and intensity of the health services provided (that is, by the average number and type of services provided per beneficiary). [20] Under the SGR formula, if the volume and intensity of services increase faster than the rate that the SGR targets allow, then the amounts paid to physicians for services they provide must be cut sufficiently to move overall physician-related Medicare spending back toward the targets. As a result, the SGR formula can cause the rates that physicians are paid for their services to be cut in dollar terms below the previous year’s rates, even though the costs physicians incur in providing these services have increased.

In general, changes in the volume and intensity of services that physicians provide to Medicare beneficiaries track trends throughout the U.S. health care system, in privately funded as well as publicly funded care. This is not surprising, since Medicare beneficiaries are treated by the same physicians that provide care to other patients and are supposed to receive the same level and quality of services that other patients do. Moreover, there are no specific mechanisms in Medicare to reduce the volume and intensity of services provided to beneficiaries below what patients in private insurance plans generally receive.

One of the characteristics of U.S. health care is that the volume and intensity of services provided have increased significantly over time. But in the late 1990s, when the SGR formula was devised, the rapid increase in the volume and intensity of services in the 1980s and early 1990s had abated considerably, and many forecasters assumed the slowdown would be permanent. CBO’s estimate that the SGR provision would save only $5 billion over the 1998-2002 period reflected this assumption. And it meant that CBO and Congress expected the SGR formula would not significantly lower physician payments below the levels needed to keep pace with increases in costs.

For the first few years after 1997, this assumption proved accurate; the increases in reimbursement rates for physicians under the SGR formula equaled or even exceeded the increase in the actual cost to physicians of providing those services, as measured by the Medicare Economic Index (MEI). [21] By 2002, however, the increase in the volume and intensity of physician services began to return to its long-term trend, and the inherent flaw in the SGR formula began to become apparent. In 2002, the SGR formula called for a 4.8 percent reduction in the basic reimbursement rate for physician services — which amounted to a 7.2 percent reduction after taking inflation into account. Congress and the President allowed this hefty reduction to go into effect.

In 2003, however, policymakers balked at allowing an additional 4.4 percent reduction (an additional 7.2 percent reduction in real terms) to take effect on top of the 4.8 percent reduction instituted the year before, and they instead enacted a modest 1.7 percent increase. Since then, Congress and the President have enacted legislation five times that overrode cuts of 4 percent or more prescribed under the SGR, in some cases substituting small increases in the reimbursement rate and in other cases freezing the rate at its current level. Most recently, in 2008, Congress set aside an 11.5 percent cut that the SGR formula would have required in 2009 and provided for a 1.1 percent increase. Overall, the average nominal increase since 2002 has been 1.0 percent a year, compared with an average annual increase in the MEI of 2.5 percent. The SGR formula calls for a 21.5 percent reduction in physician reimbursement rates in January 2010, but Congress is expected to prevent that cut from taking effect. [22]

Despite these actions to keep the full cuts required under the SGR formula from going into effect, physician payment rates have been substantially reduced in real (inflation-adjusted) terms since 2001. Even if Congress acts in coming weeks to disallow the deep cut the formula would trigger in 2010 and keeps the reimbursement rate at its current level, the reimbursement rate for physicians next year will still be 17 percent below the rate paid in 2001, adjusted for subsequent increases in the costs that physicians incur in providing services as measured by the MEI.[23]

If Congress were to allow the cut to take effect in January 2010, which it clearly will not do, the reimbursement rate would be 35 percent below its real 2001 level, an outcome that cannot be justified on policy grounds and one that would risk inducing large numbers of doctors to stop accepting Medicare patients.

MedPAC summed up the problems with allowing the required SGR cuts to take effect as follows:

[T]he Commission is not satisfied with the current [SGR] physician payment update mechanism. It does not provide incentives for individual physicians to control volume growth, and is inequitable to those physicians who do not increase volume unnecessarily. And it continues to call for substantial negative updates through at least 2016. Such reductions in physician payment rates, if they take place, would threaten beneficiaries’ access to physician services.[24]

In short, the SGR formula attempts to control physician reimbursements with an extremely blunt tool that cannot reasonably achieve that goal. [25]

SGR Differs from Medicare Provisions in Health Reform Bills

In requiring adherence to a target without providing the tools necessary to achieve it, the SGR mechanism has more in common with failed procedural efforts to force deficits to hit pre-ordained targets — such as the discredited Gramm-Rudman-Hollings law enacted in 1985 and repealed in 1990 — than with other Medicare policy changes contained in prior legislation or in the current health reform bills.

The Gramm-Rudman-Hollings law failed primarily because it set fixed, unrealistic deficit targets and ignored the fact that many factors beyond policymakers’ control exert a heavy influence on a given year’s deficit level, such as the health of the economy. [26] When the economy did not perform as well as had been hoped for, deficits far exceeded the law’s targets, and the automatic spending cuts required under Gramm-Rudman-Hollings when deficit targets were missed became unacceptably large. For fiscal year 1991, for example, the law called for a 34.5 percent cut in non-personnel defense spending and a 31.6 percent cut in non-exempt domestic spending, which included virtually all non-defense discretionary (or annually appropriated) programs. Inevitably, Congress and the President found ways to prevent these crippling cuts from going into effect.

In contrast to the SGR and Gramm-Rudman-Hollings, the Medicare provisions in the health reform bills seem well designed to accomplish their assigned tasks and are not based on crude formulas likely to result in unanticipated, unacceptably large cuts. Many of the provisions of the bills do impose limits on, and in some cases call for reductions in, the amounts that Medicare pays for various services. But those limits are based on specific assessments — by MedPAC in many cases — of the appropriate level for such payments in light of the costs that providers incur in providing them, rather than on changes in overall Medicare spending that are driven in substantial part by other factors.

To be sure, it is possible that some of the proposed reductions will prove too ambitious and have to be modified. But there is no reason to believe that most of them will fail in the way that the SGR did when growth in volume and intensity of physician services across the U.S. health care system reverted to its long-term trend, drove Medicare physician costs well above the SGR targets, and essentially turned the SGR formula into a blunt meat ax.[27]

Bills Would Begin to Slow Growth in Health Expenditures

Because people who lack insurance use fewer health care services, expanding insurance coverage will, by itself, increase health care spending in the short term. CBO estimates that covering all of the uninsured would increase total health spending by between 2 percent and 5 percent. [28] It is therefore no surprise that the actuary at HHS’s Centers for Medicare & Medicaid Services (CMS) has estimated that the House bill — which, like the Senate bill, would extend coverage to up to two-thirds of the uninsured — would increase national health expenditures by 1.5 percent in 2015, when its coverage expansions would be fully phased in. [29]

In this light, it is noteworthy that both the House and Senate bills would achieve these significant coverage expansions without adding to the federal deficit. In fact, CBO says both bills would reduce deficits over the next decade (the House bill by $138 billion, the Senate bill by $130 billion) as well as after that.

Moreover, under the Senate bill, the total federal cost for all health care spending and tax subsidies in the decade after 2019 would be no higher than if we continued current law, according to CBO, a major accomplishment for a bill that extends coverage to more than 30 million of the uninsured. That’s because the bill finances its expanded health coverage by redirecting existing spending and tax subsidies, mainly from less productive uses elsewhere in the health sector.

Although covering the uninsured will necessarily increase the level of national health expenditures at first, the key question is what will happen to the rate of growth of health expenditures thereafter. Even a modest slowdown in annual cost growth will more than offset the initial cost increase within a short period of time.

Fortunately, as explained above, both the House and Senate health reform bills include an extensive array of provisions that hold considerable promise for slowing the growth in health care costs over the long haul. In crafting the final legislation, Congress should be sure to include those provisions from each bill that hold the greatest potential for cost control and, where possible, make them even stronger. After health reform is enacted, continual efforts will be required to learn more about delivering high-quality health care at lower cost and to put what is learned promptly into practice.

End Notes

[1] Horney is Director of Federal Fiscal Policy at the Center on Budget and Policy Priorities and former Chief of the Budget Projections Unit at the Congressional Budget Office. Van de Water is a Senior Fellow at the Center and former Assistant Director for Budget Analysis at CBO.

[2] Quoted in Ronald Brownstein, “A Milestone in the Health Care Journey,” The Atlantic Politics Channel,

[3] Edwin Park and others, House Health Reform Bill Expands Coverage and Lowers Health Cost Growth, While Reducing Deficits, Center on Budget and Policy Priorities, November 6, 2009; Chuck Marr and others, Senate Health Reform Bill Is Fiscally Responsible, Center on Budget and Policy Priorities, November 19, 2009.

[4] Rick Mayes and Robert A. Berenson, Prospective Payment and the Shaping of U.S. Health Care, Johns Hopkins University Press, 2006.

[5] Henry J. Aaron, “Remarks for the 20th Anniversary of the Department of Health Care Policy, Harvard Medical School,” April 29, 2008.

[6] Paul N. Van de Water, Excise Tax on Very High-Cost Health Plans Is a Sound Element of Health Reform, Center on Budget and Policy Priorities, November 10, 2009.

[7] Peter R. Orszag, “Comparative effectiveness options in health care,” Congressional Budget Office Director’s Blog, December 18, 2007.

[8] Cathy Schoen and others, Bending the Curve: Options for Achieving Savings and Improving Value in U.S. Health Spending, The Commonwealth Fund, December 2007.

[9] Mark B. McClellan and others, Bending the Curve: Effective Steps to Address Long-Term Health Care Spending Growth, Engelberg Center for Health Care Reform at the Brookings Institution, September 2009.

[10] Marr and others.

[11] Kelly Devers and Robert Berenson, Can Accountable Care Organizations Improve the Value of Health Care by Solving the Cost and Quality Quandaries?, Urban Institute, October 2009.

[12] Christopher Weaver and Kate Steadman, “Medicare Experiments to Curb Costs Seldom Implemented on a Broad Scale,” Kaiser Health News, November 3, 2009.

[13] Drew Altman, “Pulling It Together: The ‘Third School’ for Controlling Health Care Costs?,” Kaiser Family Foundation, October 29, 2009. Available at

[14] Significant Medicare savings also played an important role in deficit-reduction legislation enacted in the 1980s, including the Tax Equity and Fiscal Responsibility Act of 1982, and budget reconciliation legislation enacted in 1981, 1984, 1985, 1986, 1987, and 1989.

[15] The bills enacted after the 1997 BBA that contained provisions (other than SGR-related provisions) affecting Medicare spending in 2007 were: the Medicare, Medicaid, and SCHIP Balanced Budget Refinement Act of 1999; the Medicare, Medicaid, and SCHIP Benefit Improvement and Protection Act of 2000; the Medicare Prescription Drug, Improvement, and Modernization Act of 2003; the Deficit Reduction Act of 2005; and the Tax Relief and Health Care Act of 2006.

[16] While a large portion of the cost does represent changes in provider reimbursement cuts enacted in 1997, there are some costs unrelated to the 1997 legislation. It is difficult, however, to separate out those costs precisely.

[17] Carolyn L. Merck, Jennifer O’Sullivan, Madeline Smith, and Sybil Tilson, “Medicare Provisions of the Balanced Budget Refinement Act of 1999 (P. L. 106-113),” Congressional Research Service, RL30860, February 22, 2001, p. 1.

[18] See, for example, Stuart Guterman, “Putting Medicare in Context: How Does the Balanced Budget Act Affect Hospitals?,” Urban Institute, July 2000.

[19] The Medicare Advantage changes are particularly noteworthy. Prior to 1997, Medicare paid private plans more on behalf of their enrollees than it would have paid if they had remained in Medicare’s fee-for-service sector, because the payment formula did not adequately adjust for the enrollees’ better-than-average health status. The 1997 legislation attempted to reduce this disparity by slowing the growth in payments to private plans. But the 2003 legislation turned this idea on its head by requiring Medicare to pay Medicare Advantage plans more than the average cost of serving comparable enrollees in traditional Medicare. This occurred not only because the President and congressional leaders had largely lost interest in deficit-reduction efforts (although deficits had reappeared starting in 2002), but because they wanted to change Medicare fundamentally by encouraging large numbers of beneficiaries to move into private insurance plans paid by Medicare. See Howard Gleckman, “The Coming Showdown over Medicare,” Business Week, June 16, 2003.

[20] CBO describes “intensity” as “the complexity of services utilized in caring for patients. For example, use of a computerized axial tomography (CAT) scan rather than an x-ray represents an increase in intensity.” Congressional Budget Office Director Peter R. Orszag, “Medicare’s Payments to Physicians: Options for Changing the Sustainable Growth Rate,” statement prepared for hearing before the Senate Committee on Finance, March 1, 2007, p. 2.

[21] The MEI measures the changes in the cost of physicians’ time and operating expenses, adjusted for changes in productivity.

[22] The SGR formula generally limits annual reductions to about 5 percent (depending on inflation) below the level the formula called for in the previous year. However, legislation enacted in recent years to forestall required SGR cuts has provided that the level of the reimbursement rate for the following year should be calculated as if the cuts called for in the previous years had taken effect. This leads to scheduled cuts that are more than 5 percent below the previous year’s level and become larger with each passing year.

[23] This is calculated by comparing the 2010 physician payment conversion factor with the conversion factor for 2001, adjusted for the growth in the MEI since 2001 and for changes in the relative values assigned to each service that also affect the total reimbursement rate. The conversion factors for 2009 and 2010 and the MEI estimates and projections are taken from Tables 5 and 6 of the CMS document “Estimated Sustainable Growth Rate and Conversion Factor, for Medicare Payments to Physicians in 2010,” . The 2001 conversion factor and “budget neutrality adjustments” used to account for changes in relative values are found in previous SGR conversion factor update reports published in the Federal Register.

[24] MedPAC, “Report to the Congress: Improving Incentives in the Medicare Program,” June, 2009, p. 253.

[25] Reducing the rates paid to physicians for Medicare services not only fails to constrain the other factors that affect total Medicare spending for physicians’ services — including the volume and intensity of services — but indirectly increases volume and intensity. As Peter Orszag pointed out in testimony he delivered in 2007 as CBO director, “evidence suggests that fee reductions such as those implied by the SGR mechanism would result in a partially offsetting increase in the volume and intensity of services provided by physicians.” Orszag testimony, p. 2. Similarly, researchers in the Medicare Office of the Actuary concluded that “reduced fees are likely to be met by a combination of an increase in volume and a shift in the mix or intensity of services furnished to Medicare beneficiaries.” “Physician Volume & Intensity Response,” Office of the Actuary, August 13, 1998,

[26] For a discussion of the failure of GRH, see Congressional Budget Office, Economic and Budget Outlook: Fiscal Years 1994-1998, January 1993, Chapter 6, “The Budget Process and Deficit Reduction.”

[27] CMS Chief Actuary Richard S. Foster has questioned whether the annual productivity adjustments to the updates for institutional providers included in the House bill could be sustained while maintaining their participation in the program (“Estimated Financial Effects of the ‘America’s Affordable Health Choices Act of 2009’ (H.R. 3962), as Passed by the House on November 7, 2009,” November 13, 2009). The productivity adjustments would reduce annual updates by only 1 to 1½ percentage points, however, much less than the sharp reductions that the SGR formula would require.

[28] Congressional Budget Office, Key Issues in Analyzing Major Health Insurance Proposals, December 2008, p. 76.

[29] Richard S. Foster, “Estimated Financial Effects,” p. 12.