Paying for Health Care Reform : The Role of Cost Savings

94-666 E Paying for Health Care Reform : The Role of Cost Savings Nonna A. Noto Specialist in Public Finance Economics Division August 19, 1994 CRS gressi©nal Resear vice; The b ar • ;Gz~ gtess "§In IIIIIi IIIII IIII Iii IIIII IIIII IIII IIII PAYING FOR HEALTH CARE REFORM : THE ROLE OF COST SAVINGS SUMMARY To be financially viable in the long-run, any health care reform plan that guarantees universal coverage and generous health care benefits must either bring down average health care costs or be willing to finance ever-growing Government subsidies on behalf of low-income people and the Medicare population . Improved insurance coverage can be expected to lead to increased demand for health care services . The Clinton Administration reform plan proposes to pay for a large portion of this extended coverage through cost' savings in health care delivery and insurance administration, instead of increased dollar spending . As a backup spending control measure, the Administration plan provides for a cap on the rate at which insurance premiums could increase each year . A controversial issue in the debate over health care reform is whether or not to include enforceable spending caps at the outset, or wait and see whether voluntary efficiency gains can accomplish the desired cost savings before legislatingmandatory cost controls . According to estimates from several studies of the Clinton plan, savings of roughly $108 billion to $138 billion per year, or 16 to 20 percent of health care spending under the corporate and regional alliances, would be needed in the early years of the reform (1998-2000) in order to provide universal coverage and meet the lower global spending target envisioned by the Administration. Although the studies generally agree about the range of aggregate savings needed, they differ widely on the degree to which the savings can be achieved "voluntarily through efficiency gains instead of "enforced" through caps on insurance premiums and reimbursements to providers. The Administration expects that market reforms and competitive pressures in insurance administration and health care delivery should reduce private health expenditures enough that the proposed caps on the rate of growth of premiums would not become binding . But other analysts do not expect that the savings from managed care and voluntary reduction in cost-shifting would be sufficient . As a consequence, in order to meet the global spending caps spelled out in the Administration's proposed Health Security Act, health plans would have to cut payments to health care providers and/or cut back services to patients . If, instead, the caps were eased, that would mean higher premium payments from employers and families and larger Government subsidies . The Administration projected that the President's plan would decrease the Federal deficit by $38 billion in fiscal year 2000 . The Congressional Budget Office estimated that the plan would increase the deficit by $10 billion. The difference between the Administration and CBO estimates of $48 billion is relatively small, however, compared with the difference between those two sources, who both assumed that the premium caps would hold, and others concerned that they might not . Estimates of the added increase in the Federal deficit if there were no spending caps or efficiency gains reach $110 billion or more a year by FY 2000 . This raises concern about adopting a reform that does not include effective mechanisms for containing costs . TABLE OF CONTENTS WHY THE EMPHASIS ON CONTROLLING COSTS? 3 EXPENDITURES "ABOVE THE BASELINE" 4 CONSEQUENCES OF NOT ACHIEVING THE SAVINGS 5 COST-SAVING MECHANISMS UNDER THE CLINTON PLAN 6 SPENDING CAPS ON ALLIANCE PREMIUMS 6 PRE-EMPTION OF THE REIMBURSEMENT WINDFALL 7 LIMITS ON MEDICARE AND MEDICAID PAYMENTS 7 ESTIMATES OF THE NEEDED COST SAVINGS 8 CLINTON ADMINISTRATION 12 LEWIN-VHI 12 DRI/McGRAW-HILL 13 IN SUM 14 CONCERNS ABOUT THE EFFECT OF SPENDING CONTROLS ON HEALTH CARE DELIVERY 15 FEDERAL BUDGET EFFECTS WITH AND WITHOUT COST CONTROLS 16 HIGHER PREMIUM LEVELS 17 NO CAP ON FEDERAL ENTITLEMENT FOR PREMIUM SUBSIDIES 21 GLOBAL SPENDING CAPS NOT FULLY EFFECTIVE 22 IN SUM 23 APPENDIX A . SYNOPSES OF REPORTS ESTIMATING COST SAVINGS 25 CLINTON ADMINISTRATION 25 Federal Program Savings 26 Savings from Market Reform and Competition 28 LEWIN-VHI 30 GOLDMAN SACHS 33 Increased Utilization Due to Expanded Insurance Coverage . . . 33 Basic Alternatives in Paying for Reform 35 DRI/McGRAW HILL 36 APPENDIX B . NATIONAL HEALTH CARE SPENDING UNDER THE BASELINE AND THE CLINTON PLAN 40 REFERENCES 43 TABLES 1. Aggregate Cost Savings Needed to Meet the Spending 9 Objectives of the Clinton Universal Coverage Plan . Estimates of Components of Cost Savings 2 11 under the Alliance Plans 3. Effect of the Clinton Health Care Reform on the Federal Deficit : Alternative Estimates, With and Without Enforced Spending Caps, 1996-2000 . . 19 A.l . Changed Sources and Uses of Federal Funds in the Clinton Health Care Reform, Fiscal Years 1998 and 2000 (Clinton Administration estimates)27 A.2 . Savings from Market Reforms and Competition (Clinton Administration 29 estimates) A.3 . Projected Savings Relative to Expenditures (Clinton Administration 29 estimates) . Changes in Aggregate National Health Care Spending under the Health A.4 Security Act in 1998 (Lewin-VHI estimates) 32 A.5 . Increased Demand for Health Care Spending under the Clinton Plan 34 (Goldman Sachs estimates) A.6 . Increase in Efficiency Needed to Avoid Spending Caps (Goldman Sachs 34 estimates for the year 2000) A.7 . Consumer Spending for Medical Care Services (DRI/McGraw-Hill Estimates) 38 A.8 . Estimated Reduction in Nominal Health Care Spending (DRI/McGraw-Hill Spending Cap Simulation) 39 . A.9 Estimated Reduction in Real Health Care Spending (DRI/MeGraw-Hill Spending Cap Simulation) 39 B.I . Estimates by the Administration, the Congressional Budget Office, 41 Lewin-VHIL and DRI/McGraw-Hill PAYING FOR HEALTH CARE REFORM : THE ROLE OF COST SAVINGS Universal health insurance coverage and enhanced benefits can be expected to increase national demand for health care services . The question is how we want to pay for that . One way is just to pay more . This would mean greater aggregate premium payments by employers and individuals and a higher percentage of the economy devoted to health care . It would also mean higher taxes - or Government deficits - to subsidize the purchase of insurance by employers and individuals considered too poor to pay . Another approach is to reduce the average cost per person of providing health care . This could occur by lowering the per unit cost of providing particular services (for example, through efficiency gains in insurance administration and health care delivery or reduced payments to providers) and by placing some limits on the amount of care available to individuals . A financially workable health care reform plan is likely to include all of these elements, to some degree . The ClintonAdministration's proposal for universal health insurancei would extend the protection offered by standard health insurance, in terms of both the number of people and the range of health services covered . At the same time, the plan would restrain the overall rate of growth of spending on health care . To accomplish these two seemingly contradictory objectives, the Administration plan depends heavily upon cost containment mechanisms - if not voluntary, then enforced . A controversial issue in the debate over health care reform is whether or not to include enforceable spending caps at the outset, or whether to wait and see whether voluntary efficiency gains can accomplish the desired cost savings before legislating mandatory cost controls . The costs of both the Medicare and Medicaid programs were massively underestimated when the programs were introduced and seriously in subsequent years. The Government has tried to control the costs of these two programs by limiting the fees it will pay to providers for specific services. Nonetheless, projected increases in the cost of the Medicare and Medicaid programs continue to threaten deficit reduction after 1999 .' This experience leads to apprehension about introducing another health care entitlement program without an effective i The Clinton Administration's proposal was introduced in the first session of the 103d Congress as the "Health Security Act," under bill numbers H .R. 3600 in the House (by Rep . Gephardt) and S. 1757 (by Sen . Mitchell and others) and S. 1775 (by Sen . Moynihan) in the Senate . 2 U.S. Congressional Budget Office . The Economic and Budget Washington, January 1994 . p. 26, 28-29. Fiscal Years 1995-1999. Outlook : CRS-2 mechanism in place for either raising revenues or controlling costs if demand exceeds the initial projections . It is noteworthy that there is concern both about what would happen if cost controls such as those proposed by the Administration didn't work and what would happen if they did work. Opponents, skeptics, and neutral budget analysts of the Clinton plan have expressed concern about what will happen i f the controls did not work - or if there are no such controls in alternative reform measures : • Would the premium caps be lifted, raising the mandated costs to employers and individuals? • Would the Government cover the resulting higher pre subsidy costs? If so, would it raise taxes, cut other Government programs, or let the deficit increase? • Would the benefit package be scaled back or some limits placed on the amount of services an individual could expect to receive? Others are concerned about how health care providers would respond if the proposed controls were effective : • If payments to providers were restricted, would services to patients be cut back or rationed? • If there were controls on drug prices and other reimbursements, would the pace of private technological development and scientific research be slowed? The Nation is likely to face these issues under any health care reform plan that promises Federal subsidies to foster universal coverage and tries to contain Notably, these issues will arise even in the absense health care costs . comprehensive reform efforts . The numbers cited in this report refer principally to the Clinton Administration's health reform plan because the Administration's plan was issued earliest and in the most detail, and has consequently been subject to the most empirical analysis by Government and private sector analysts . Nonetheless, the general themes addressed are relevant to the evaluation of other health care reform proposals . This report examines the role of cost savings in achieving the objectives of health care reform as envisioned in the Clinton plan . The first section of the report briefly explains why controlling health care costs is central to the financial viability of the Administration's reform proposal . The second section explains how the expected increase in demand for health care "above the baseline" under a universal coverage system such as the Clinton plan means that the needed cost savings are larger than commonly indicated by simply comparing estimates of final spending under the reform plan to the baseline projections for the current CRS- 3 system . The third section outlines the major cost-saving mechanisms under the Clinton plan, with particular attention to the cap on the rate of increase of alliance premiums . The fourth section presents numerical estimates of the cost savings needed for the Clinton plan to meet its global spending targets . It draws upon estimates made by the Clinton Administration and two private organizations- Lewin-VHI and DRI/McGraw-Hill. The fifth section briefly discusses some of the concerns about what cost controls could mean for the quality and availability of health care services . The sixth section demonstrates the potential Federal budget exposure if a plan like the Clinton plan were enacted, guaranteeing health benefits and premium subsidies, but the projected cost savings were not achieved ; its draws upon estimates from the Congressional Budget Office in addition to the three sources already mentioned . The first appendix contains synopses of studies referred to in the main text, with emphasis on their method of estimating the cost savings needed to accomplish the Clinton plan . The second appendix presents the estimates of national health care spending under the baseline and under the Clinton plan, according to the different studies . WHY THE EMPHASIS ON CONTROLLING COSTS? One principal objective of the Clinton health care reform effort is to provide universal coverage . A major reason why people are uninsured is that health insurance is "unaffordable" for them . That is, the cost of insurance to them or to their employer on their behalf - is considered too high relative to their income . There is a concern, and a well-founded one, that more people are likely to become uninsured in the future if we continue the current system of private health insurance without some modification . If health care costs keep rising more rapidly than peoples' incomes, fewer people will find health insurance affordable . The Clinton proposal pursues at least three approaches to financing universal coverage : 1. bring down the average price of health insurance ; 2. have the Government subsidize the premium price for people considered too poor to pay; 3. pay for these subsidies primarily by cost savings elsewhere in the private and public health care system, and only minimally by levying new taxes. Constraining the growth of health care costs is a concern in its own right . However, controlling costs is especially important to achieving affordable CRS- 4 ersal coverage . The lower the average cost of health insurance, the more u people who can afford insurance without receiving subsidies, and the smaller the Government subsidies would have to be for those people who do need them . EXPENDITURES "ABOVE THE BASELINE" Estimates of national health care expenditures under the Clinton (or other) reform proposals are commonly compared with a "baseline" that represents what It is important to expenditures would be if current policies continued . understand, however, that these estimates of expenditures under the Clinton reform plan represent a net calculation of the projected increase in demand for health care minus the savings expected to result from the various cost control mechanisms . By expanding insurance coverage, the Clinton plan is expected to increase the demand for health care to levels above the baseline . (The Clinton plan would extend coverage to the uninsured and would improve the benefit package for many people previously insured, such as by covering preventive care without copayments, prescription drugs under Medicare, and long-term home and community-based care for the severely disabled .) Figure 1 illustrates how the Clinton plan depends on cost savings to pay for expanded coverage -and more . The middle line represents projected health care expenditures under the "baseline" of no reform - what is anticipated if the current health insurance system continues in place . The upper line represents the Clinton plan of universal coverage without savings or cost controls . These are the expenditures that would be expected if the demand for health care increased in response to enhanced insurance benefits, but the expected reductions in the cost of health care delivery and insurance did not occur . The lower line represents expenditures under the Administration's plan, assuming that the projected savings could be achieved or that the proposed cost controls would be effective . To keep total spending below the baseline, health care reform must accomplish savings sufficient to cover not only the difference below the baseline, but also the difference between the baseline and what spending demand would be under universal coverage in the absence of cost savings and controls .3 s This is succinctly explained by: Dudley, William . The Clinton Healthcare Plan: No Free Lunch . Goldman Sachs, U .S . Economic Research . New York, Goldman, Sachs & Co ., January 1994 . p . 6 . Summarized in Appendix A of this report . CRS- 5 FIGURE 1 . National Health Care Spending under the Clinton Plan With and Without Cost Reductions Clacton plan without cost reductions Billions of Dolt= f 1300 1200 1100 1000 , 900 800 1995 1996 1997 1998 1999 2000 Source : Based on numbers from DRI/McGraw-Hill . See appendix B of this report for the numbers and appendix A for a description of the estimation method . CONSEQUENCES OF NOT ACHIEVING THE SAVINGS There are as yet no official published Government estimates of what aggregate national health care expenditures would be under universal coverage (as defined under the Clinton plan, or any of the other reform plans) if there were no efficiency savings or caps on expenditures . This report gathered together what estimates were available from a variety of independent studies . Understanding the magnitude of the cost savings needed for the Clinton plan, or any of the other health care reform plans, to work financially is important to assessing the degree of change being expected from the health insurance and health care delivery system . A serious concern is that if the increased demand for health care services cannot be more than fully offset through "voluntary" efficiency gains in health care delivery, it would need to be paid for in other ways . If the proposed spending caps are not enforced under the Clinton plan or are omitted from other reform plans, this could mean higher premium payments for employers and families, and larger Government subsidies and deficits . If the proposed spending caps were enforced, this could mean reduced payments to providers, the rationing of health care services, and/or scaling back the benefit promises of the standard insurance package . CRS-6 COST-SAVING MECHANISMS UNDER THE CLINTON PLAN There are two conceptually different sources of expenditure savings under the Clinton plan . This report refers to one as "redesign" savings and the other as "enforced" savings . "Redesign" savings include the type of savings the Clinton Administration expects will occur as a byproduct of reforms and increased competition in the health care and insurance markets . They include, for example, the spending reductions expected from an increased reliance on managed care and more efficient provider behavior, the streamlining of insurance administration, and the voluntary reduction by providers of the cost-shifting factor previously added to their prices . These can be thought of as "efficiency gains" which would help produce the same - or improved - health outcomes for less money . "Enforced" savings include the impact of the global spending cap on the rate of growth of the alliance premiums; limits on contributions on behalf of former Medicaid recipients ; restraints on the growth of Medicare spending; and the pre-emption of any reimbursement windfall in the setting of the initial fees and premiums . These can be referred to as "spending caps" that would in effect limit the payments to providers for the agreed-upon package of insured health care services . The "enforced" spending cap mechanisms will now be described briefly . SPENDING CAPS ON ALLIANCE PREMIUMS In what is referred to as "global budgeting," under rules set forth in the Health Security Act, the National Health Board would set the target premium for a regional alliance for its first year of operation . The Act further spells out the limit on the rate at which regional alliance (and corporate alliance) premiums could rise each year thereafter . In recent years, the rate of inflation in health care costs has been substantially higher than the overall consumer price index (CPI) . Under the Administration plan, the permitted rate of increase in the per capita premium for the standard benefit package would be brought down to the CPI over a three-year period, between 1996 and 1999 . Several other reform plans have espoused the goal of reducing the rate of growth of health care spending to the CPI within a few years . Only a few have included an enforceable cap . Section 6001 of the Health Security Act defines the permitted "general health care inflation factor" for the period from 1996 through 2000 as the increase in the CPI plus specified amounts - 1 .5 percentage points in 1996, 1 .0 percentage point in 1997, 0 .5 percentage point in 1998, and zero in 1999 and 2000 . After 2000, if the Congress did not specify new inflation factors, the default factor would be the percentage increase in the CPI combined with the percentage growth CRS-7 in real gross domestic product (GDP) per capita during the preceding three years 4 (This default factor is roughly equal to total growth in nominal GDP . Its effect would be to hold national health care spending to a constant, rather than rising, percentage of GDP .) The Clinton Administration repeatedly stressed the belief that its spending goals could be met without having to use what it considers "backup" protection - the cap on the rate of growth of alliance premiums . In contrast, most other studies o f the plan reviewed in this report conclude that spending demands would exceed the global budget limits and thereby make the spending caps binding . According to some estimators, the spending cap on alliance premiums is the largest source of projected savings under the Clinton plan . PRE-EMPTION OF THE REIMBURSEMENT WINDFALL Under the Administration's health alliance system (and other universal coverage proposals), providers would be paid the full approved fees for all patients, including people previously uninsured (who had in the past been provided with "uncompensated care") and those who were previously covered by Medicaid with its low reimbursement rates . Unless fee levels were reduced by the amount of the "cost-shifting" component previously charged to full-paying customers, providers might collect a "reimbursement windfall" of extra income . Only part of this fee reduction is expected to happen voluntarily on the part of providers . The Clinton plan provided for setting first-year premium levels so that any potential windfall would be offset or "pre-empted ." The ability of health care reform to lower the average private insurance premium depends importantly on removing the cost-shifting component from current fee levels . Some of the differences in the estimates of premium levels reflect differences in analysts' assumptions about the ability to reduce costshifting . Analysts who did not believe that the cost-shifting component would be fully eliminated tended to have higher estimates of the initial premiums . LIMITS ON MEDICARE AND MEDICAID PAYMENTS Savings under Medicare would come primarily from reduced reimbursement of providers - including hospitals, physicians, and other service providers . Savings under Medicaid would come from restricting the payments to providers that would continue to be made under the Medicaid program on behalf of cash-assistance recipients and by capping the payments to the alliances on behalf of other Medicaid beneficiaries below previous levels of expenditures . In 2001 adjustments would also be made to reflect the increase in the actuarial value of the benefit package that would occur when added benefits were introduced . 4 CRS- 8 This report does not pursue the examination of proposed savings under the Federal Medicare and Medicaid programs . It concentrates instead on savings in health care spending under the domain of the private health insurance alliances . ESTIMATES OF THE NEEDED COST S VINGS Numerous studies of the Clinton plan have now been made by Government and private sector analysts . This section of the report summarizes and compares the main conclusions of three studies with regard to the size and composition of cost savings needed for the Clinton plan to meet its objectives . Included are estimates made by the Clinton Administration and two private organizations Lewin-VHI and DRI/McGraw-Hill . 5 Although the studies generally agree on the amount of savings needed in the aggregate, they differ considerably about how much of the savings would come from "redesign" savings versus "enforced" savings . Table 1 s mmarizes the findings from three sources with respect to the aggregate savings in health care expenditures needed to meet the global budget objective of the Clinton reform plan . The DRI numbers correspond to the graphical presentation in figure 1 on page 5 . The first row contains the estimate of what health care spending would be under the universal coverage proposed by the Clinton plan if there were no cost savings from either efficiency gains or enforced spending caps . The second row presents the "baseline" estimate of what health care spending would be if there were no comprehensive health care reform and current policy continued . The third row presents the estimate of what spending would be under the Clinton plan with the cost reductions planned under the terms of the Health Security Act . The fourth row is "needed savings," the difference between spending with and without cost controls . The fifth row presents this measure of needed savings as a percentage of what expenditures would otherwise be under universal coverage, according to the reference measure used in that particular study . The resulting percentage is a rough measure of the degree of savings that would need to be achieved through "efficiency gains" in order to "voluntarily" meet the global budget targets set by the Clinton plan . s The Congressional Budget Office analysis is not included in this section . For the purposes of its estimates, CBO accepted the Clinton Administration's assumption that the proposed caps on the rate of increase of the premiums would be effective . CBO did not separately estimate the magnitude of cost reductions that would be required in order to meet the caps . CBO did, however, discuss possible consequences of the premium caps . U .S . Congressional Budget Office . An Analysis of the Administration's Health Proposal . Washington, February 8, 1994 . p . 70, 74 . CRS- 9 TABLE 1 . Aggregate Cost Savings Needed to Meet the Spending Objectives of the Clinton Universal Coverage Plan (in $ billions or percent) Estimating group (year of reference) Clinton (alliance portion only)' (2000) Clinton (national health expenditures)b (2000) LewinVHF (acute care only) (1998) DRI/ McGraw -Hill' (NIPA) 1,541 .5 1,299.8 (2000) 1. Universal coverage under Clinton plan without cost reductions 674 2. Baseline reform NA 1,653 1,395 .0 1,200.4 3. Clinton plan with cost reductions 566 1,597 1,394.4 1,150 .9 4. Cost savings needed (line 1 - line 3) 108 NA 147.1 148 .9 5. Percent cost savings needed ( line 4/line 1) 16.0% NA i out 9.6% 11.5% Notes and sources : ' The Clinton Administration projections include expenditures for services in the guaranteed benefit package delivered to all those in regional and corporate alliances . This includes privatized Medicaid acute care . It does not include Medicare or public health expenditures. Source : See Appendix table A .3. b See Appendix table B .1. ° The Lewin-VHI numbers include spending for acute care only . They exclude spending for long-term care, public health, research and construction . See Appendix table A-4 . d DRI/McGraw-Hill uses National Income and Product Accounts (NIPA) definitions of health care spending . The NIPA accounts do not include spending emanating from Federal, State, or local hospitals, or prescription drugs . The NIPA health care expenditures are about 80 percent of expenditures under the HCFA (Health Care Financing Administration) definition of the National Health Accounts (NHA) . See Appendix table A .7. CRS- 1 0 Caution should be used in interpreting and comparing the estimates of needed cost savings . The Lewin-VHI estimates are for 1998, while the others are for 2000 ; savings are projected to get cumulatively larger, the longer the reform is in place . Each study uses different economic assumptions, estimating methods, and data . Furthermore, each of the studies uses a different definition of health care expenditures (see notes to table 1) . Their magnitudes vary substantially, as can be seen by reading across the first row . This affects the percentage estimates . In particular, the percentage measure of needed savings is substantially larger when the denominator is smaller . The base for the Clinton Administration's cost savings estimates (column 1) is spending under the private regional and corporate alliances, including Medicaid patients who would be enrolled in the alliance system . This alliance segment represents about 35 percent of total projected national health care expenditures under the Clinton reform plan .' In contrast, the private sector estimators of the Clinton plan have looked at measurements of national health care spending that also encompassed Medicare and other public spending, with some omissions explained in the notes to table 1 . Table 2 attempts to standardize the comparison among the studies by focusing on savings under the regional and corporate health alliances only (the measure shown for the Clinton Administration estimates in table 1) . Even so, it is difficult to generalize among the studies . Each study measures different components and there are large differences in the size of savings attributed to a particular category . The diversity of the estimates serves as a warning not to accept as precise the numerical estimates from any single study . In an effort to calculate a savings percentage with respect to a standardized base, the concept of "benchmark spending" is created as a point of reference (the denominator) . Benchmark spending has been defined in table 2 as $566 billion plus the amount of estimated spending reductions under the alliance plans . As shown in table 1, $566 billion was the Administration's estimate of spending under the alliances in 2000, with cost reductions . Thus benchmark spending is a rough approximation of spending under the alliances with universal coverage but without cost reductions . Total spending reductions are the same in table 2 as table 1 for the Clinton Administration estimates . For Lewin-VHI and DRI/McGraw-Hill, the savings totals are somewhat lower in table 2 . The Lewin-VHI total cost saving estimates in table 1 also include $13 .1 billion in net savings for Medicare and an increase in spending of $3 .5 billion for Medicaid . (See appendix table A .4 .) The DRI total 6 A comparison of two sets of Clinton Administration numbers (columns 1 and 2, line 3) suggest the regional and corporate alliances would account for $566 billion out of $1,597 billion, or 35 percent of total health care expenditures in 2000 . According to CBO's estimates, expenditures through the health alliances would be $585 billion or 37 percent of $1,583 billion in total national health expenditures in calendar year 2000 . Calculated from CBO, p . 26 . CRS- 1 1 TABLE 2 . Estimates of Components of Cost Savings under the Alliance Plans (in billions of dollars) Estimator (Year of reference) Spending under alliances after cost savings (Administration estimate) Clinton Administration (2000) LewinVHla (1998) $566 $566 DRI/ McGrawHillb (2000) $566 Redesign savings $108 $58 .0 $17.4 Managed care 14 .9 17 .4 Changes in provider behavior 57 Consumer switching to low cost plans 24 Administrative savings 17 6 .7 Elimination of small group market 11 Voluntarily reduced cost shifting 36 .4 Enforced spending caps 0 79.8 107 .6 Alliance premium caps 0 47 .3 107 .6 . . ....Pre-emption . . . . . ... .... .. . . . . .. .. .. . .of . . . . . ..cost .. .. . . . . .. ....shifting . . . . . . . .. ..... . . . . . . . ... ... . . . . .. .. . . . . . . . . .. .... .. .... .... . .. . ... . . . . . .. . . . . . . . . .. . . . . . . . . . . 32 . . . . . . .5 . . . . . . . . . . . . . ...... . . .. .... . . . .. . . . . . . .... Total spending reductions under 108 137 .8 125 .0 alliance plans Benchmark spending for alliances 674 703 .8 691 before cost reduction -----------------------------------------------------------------------------Spending reductions needed as a 16% 20% 18% percent of benchmark spending Percent of savings from redesign 42% 100% -14% ---------------------------------------------------------------------------Percent of savings from enforced caps 0% 58% 86% Notes : The absence of an entry in the table does not necessarily mean that the estimators did not include that factor in their model or believed its value to be zero . It only means that they did not publish an estimate for that factor . For example, both the Clinton and DRI estimators took the reduction of cost shifting into account but did not publish an estimate of the magnitude . The individual items may not sum exactly to totals due to rounding . Source : For further explanation, see appendix A's discussion of each of these sources . CRS-12 cost saving estimates in table 1 also include $24 billion in Medicare savings . (See appendix A.8.) The findings on cost savings from the three studies will now be described briefly . A more detailed explanation of the studies and their estimating methods is presented in appendix A. CLINTON ADMINISTRATION The Clinton Administration expected that market reforms and competitive pressures would reduce the rate of growth in private health expenditures sufficiently that the legislated caps on premium growth would not become binding. The Administration estimated that "redesign" savings of $108 billion in 2000 would offset the $107 billion in expenditure reductions needed to meet the premium caps . The $108 billion represents a 16 percent savings relative to what expenditures under the regional and corporate alliances would otherwise be. (See appendix tables A .2 and A .3 .) LEWIN-VHI Using a microsimulation model, the consulting firm Lewin-VHI made detailed estimates of specific components of spending increases and decreases under the Clinton plan . Their estimates suggest that if the proposed enhancement in private insurance coverage and public health programs were made, then in 1998 national health care spending would be $147 billion higher without the anticipated cost savings and spending caps . Overall savings in national acute care health expenditures of 9 .6 percent would be needed to bring spending down to the level of the global budget target (table 1).' Spending reductions associated with the alliances only were $137 .8 billion or roughly 20 percent of the benchmark (table 2) .8 'In another measurement, Lewin-VHI estimated that per capita costs in the health alliances would be about $201 per person per month in 1998 without the "enforced" cost controls, and $182 per month with cost controls . Comparing these two numbers implies that the budget cap would be enforcing a 10 .4 percent cost-squeeze in 1998 . Lewin-VHI, p . 25-26 . Separate estimates for the four enrollment classes imply a cost-squeeze of 9 .8 percent as a result of the premium cap . The savings expected from the "redesign" of the health care system are incorporated in the premium estimates . Lewin-VHI, Appendix D, table D-17, p. D-50 . 'The measurement for Lewin-VHI is particularly rough because the starting point for the benchmark, $566 billion, is a measure of spending in 2000, while the Lewin spending reduction estimates are for 1998 . CRS- 1 3 Lewin-VHI did not expect that savings from managed care and streamlined administration would be sufficient to avoid the spending caps ; out of total spending reductions under the alliances of $137 .8 billion, $21 .6 billion or 16 percent was estimated to come from such savings . Lewin estimated that the caps on alliance premiums would save over twice as much - $47 .3 billion or 34 percent of the total spending reductions under the alliance plans . Note that the Lewin estimates are for calendar 1998, the first year that the reform plan was expected to be in effect in all States and before managed care or the premium caps could be expected to have much cumulative savings effect . Although the other estimators are known to have taken reduced cost-shifting into account, Lewin-VHI was the only one to publish an estimate of the reduction in cost-shifting that would need to be accomplished if providers are not to reap a "reimbursement windfall" from universal coverage . The combined total reduction in cost-shifting of $68 .9 billion accounted for half of Lewin's total estimated spending reductions . Of that amount, slightly over half, or $36 .4 billion, was projected to come from a voluntary reduction in cost shifting by providers . The remaining windfall to be forcibly "preempted" when the initial levels of the alliance premiums were set in the first year of the plan was estimated at $32 .5 billion . (See table A .4 and the accompanying discussion in the appendix.) Adding the estimated savings from voluntary reduction in cost shifting to the savings from managed care and administrative streamlining, the total savings from "redesign" are $58 .0 billion or 42 percent of the total spending reductions under the alliance plans . Adding the estimated savings from pre-empted costshifting to the savings from the alliance premium caps gives a total of $79 .8 billion in "enforced savings," or 58 percent of the total estimated spending reductions under the alliance plans . DRIIMcGRAW-HILL The economic forecasting firm DRIIMcGraw-Hill (DRI) analyzed the Administration's reform plan in the context of its quarterly macroeconomic model of the U .S . economy . For the year 2000, DRI estimated that nominal medical care expenditures under the full Clinton plan would be $49 .5 billion below the baseline case of no reform, a reduction of 4 .1 percent. This decrease, however, reflects the net outcome of an increase in spending of $99.5 billion (above the baseline) resulting from improved insurance coverage, offset by savings of $149 billion from spending controls . The $149 billion in savings represents 11 .5 percent of spending under the Clinton plan without spending caps, measured according to the NIPA definition of health expenditures . (See table 1 in the text and table A .7 in the appendix .) Spending reductions associated with the alliances only were $125 .0 billion or 18 percent of benchmark spending without spending reductions (table 2) . According to the DRI estimates, $17 .4 billion or 14 percent of the $125 billion in savings under the alliance plans in 2000 would come from managed care . The CRS- 1 4 remaining $107 .6 billion or 86 percent would come from from the alliance premium caps . (See table A.8 in the appendix .) In its modeling, DRI assumed that roughly one-third of the windfall gains would not be eliminated under the standard State Government reimbursement guidelines .' However, DRI did not publish an estimate of reduced cost-shifting . IN SUM There is general agreement among these three studies of the Clinton plan that a large amount of aggregate savings would need to be achieved in annual national health care spending in order to avoid the caps on the growth of alliance premiums specified in the Health Security Act . For the early years of the plan, 1998-2000, the needed savings have been estimated at from $108 billion to $138 billion per year under the alliance portion alone . These numbers suggest that if the health care delivery system under the corporate and regional alliances could not achieve "efficiency gains" of approximately 16 to 20 percent within the portion by the very earliest years of the reform, then in order to meet the global spending targets set forth in the Clinton plan the cap on the rate of growth of premiums would need to be enforced . There is considerable difference among the estimators about the degree to which the savings under the alliance plans can be achieved "voluntarily" through efficiency gains, instead of "enforced" through caps on insurance premiums and reimbursements to providers . The Administration expected that market reforms and competitive pressures in insurance administration and health care delivery would reduce private health expenditures sufficiently that the global spending caps would not become binding . In contrast, the private sector estimators suggest that of the total needed savings under the alliances, only a minority would come from "redesign savings" or efficiency gains from managed care and streamlined insurance administration, while the majority would come from enforced spending caps (caps on the growth of alliance premiums and the pre-emption of cost shifting) . 10 DRI's estimate of "redesign" savings from managed care was roughly comparable to Lewin-VHI's ($17 .4 billion in 2000, compared with $14 .9 billion in 1998, respectively) . Both were far lower than the Administration's estimates of $57 billion from changes in provider behavior plus $24 billion from consumers switching to low cost plans in 2000 . Roughly speaking, the Administration expected 100 percent of the spending reductions to come from redesign savings ; Lewin-VHI 42 percent ; and DRIIMcGraw-Hill just 12 percent . (See table 2 .) s DRI, p . 17 . to The Administration, CBO, and the private estimators generally agree that most of the savings under the Medicare program would come from the enforced spending caps . CRS- 15 These large estimates of needed savings raise concern about about the financial implications of adopting a reform plan that guarantees universal coverage but does not have an identified mechanism for cost containment . CONCERNS ABOUT THE EFFECT OF SPENDING CONTROLS ON HEALTH CARE DELIVERY The Clinton Administration believes that in the present health care system there is a large amount of expenditure on "unnecessary care" (medical care that does not cost-effectively contribute to improved health) and expenditures that are higher than they need be because the medical problem was not addressed at an earlier or more preventive stage . Consequently, the Administration reasons, increasing the efficiency and effectiveness of service delivery can reduce total health care spending without lowering the quality of health care outcomes . No doubt, some efficiency savings can be made in the health care delivery system without seriously affecting the nature and quality of care . A key question is, will these savings be large enough - and soon enough - to meet the Clinton Administration's proposed global budget target? As explained in the previous section, the other estimating groups surveyed anticipated that the spending caps would be triggered . If so, what are the implications? Economists as a group are skeptical that binding price controls or spending caps can be imposed on any product or service without having supply fall short of demand at the controlled price . The question economists would ask is whether health care providers can be expected to simply accept being paid less for supplying the same services ." More likely, providers will accept some decrease in fees but will also adjust their behavior to curtail the quality or quantity of services they provide in exchange for lower payments . Currently the Medicare and Medicaid programs reimburse providers at rates well below the posted prices . But, because Medicare and Medicaid represent only a part of their business, providers have been able to offset some portion of this underpayment by refusing to serve some of these patients, increasing the quantity of services, or by raising the prices charged to their full-paying customers (known as cost-shifting) . Under the Clinton reform plan, the percentage rate of reimbursement to Medicare providers would be reduced below current levels . The Clinton plan would fold many Medicaid recipients into the private alliance plans but would have the Government pay the alliances less than what previous Medicaid expenses had been . Thus, the cost-shifting from Medicare and Medicaid to the rest of the population would continue, at an amplified level . Under the Clinton plan, controls on payments to insurers or providers would apply to most of the health care system . There would be few remaining 11 That is, is the supply of health care services really "price-inelastic"? Or is it more likely to be somewhat "price-elastic"? CRS-16 customers to whom providers could shift costs . A big question is how providers would respond . DRI/MeGraw-Hill pursued the analysis of what the cap on the amount of money flowing into the alliances might mean . DRI expected that both provider incomes and utilization would be reduced as a result of the spending caps . By the year 2000, unit prices of health care services (also referred to as provider incomes) would be about 4 .7 percent lower than under baseline conditions, and 6.8 percent lower than under the Clinton plan in the absence of caps . 12 In the aggregate, real health care services would be cut by approximately 5 percent starting in the years 1998-2000, once the spending controls started to take effect . Per household, the reduced utilization of services in 1999 and 2000 would be equivalent to $300 per year measured in 1987 dollars, or $461 in 1994 dollars ." These estimates cover the population as a whole, including those previously uninsured . Consequently, for the previously insured portion of the population, the decrease in services would be even larger . Sacrifices and tradeoffs are an inevitable part of any effort to reform the health care system . In this case, paying for expanded insurance coverage by controlling costs can be expected to lead to changes in the way health services are delivered, in order to reduce the average amount spent on health care per person. FEDERAL BUDGET EFFECTS WITH AND WITHOUT COST CONTROLS The preceding discussion focused on measurements of national health care expenditures . This section focuses on measuring the effect of the health care reform plan on the Federal deficit . Note that while estimates of health care spending are typically made on a calendar year basis, estimates of the effect on the Federal deficit are made on a fiscal year (FY) basis . The Clinton Administration estimated that its health reform plan would decrease the Federal deficit - by $37.7 billion in FY 2000, with an extra $13 .0 billion "cushion" to pay for additional premium discounts, if necessary .14 None of the other studies had as optimistic an estimate of the likely effect on the deficit and none had any cushion remaining . The major difference between the Administration and other analysts in their estimates of the effect of the Clinton reform plan on the Federal deficit arises 12 Calculated by CRS from the unpublished DRI medical price deflator series . a DRI, p . 4 . 14 See the shaded portions of appendix table A .1 for the Administration's estimates of the cushion and the effect on the deficit in 1998 and 2000 . The cushion was approximately 15 percent of the projected premium discounts . CRS- 17 from the amount of "discounts" or subsidies they estimate the Government would be obligated to pay on behalf of the employer share of the insurance premium . 16 The Administration's estimates of the net budget cost 16 of these premium discounts reflected certain assumptions : • about the level of the premium needed to cover the guaranteed benefit package ; • that the total amount the Federal Government would spend on premium discounts each year was a "capped entitlement," limited to the dollar amounts specified in the Health Security Act ; and, • that the global budget limits set forth in the Health Security Act regarding the annual rate of growth in the private premiums would be effective if needed but would not have to take effect at all because savings from market reforms and competition would be sufficient to bring expenditures below the global limits . All of these assumptions have been challenged . The other analysts of the Clinton plan made different assumptions regarding the dollar level of the premiums, whether the cap on the Federal entitlement for premium discounts would hold, and whether the global spending caps would hold . These led to substantial differences in their projections of the effect of the plan on the deficit . When the Congressional Budget Office released its analysis of the Clinton health care reform plan in February 1994, considerable attention was given to the fact that CBO showed the President's plan increasing the Federal deficit slightly, while the Administration had shown it decreasing the deficit slightly, over the period 1996 to 2000 . What the numbers summarized in table 3 reveal is that the difference between the Administration and CBO estimates is relatively small compared with the difference between those who assumed that the spending caps would hold and those concerned that they might not . HIGHER PREMIUM LEVELS The Clinton plan promises to subsidize the premium payments of employers once they exceeded a certain percentage of payroll and the family share of lu For an explanation of the sources of the differences, see CBO, p . xiii and Table 2-4, p . 36 . Also, see Lewin-VHI, Table ES-3, p . ES-15 or Table 12, p . 56 . 16 Gross discounts are offset by transfers from the Medicare and Medicaid programs . CRS- 18 premiums once they exceeded a certain percentage of family income ." As a consequence, if the estimated level of the premiums rises, so does the estimated Federal obligation to pay subsidies. Higher average premiums also mean higher premium payments for employers . This means less potential increase in the taxable base for income and payroll taxes and consequently lower additional tax revenues than the Administration had projected . Thus, if premiums rise above the levels projected by the Clinton Administration, the reform plan is more likely to increase the Federal deficit - both by raising the Federal Government's expenditure obligations and reducing its revenues . All of the other studies concluded that, in order to cover the promised benefits package, the average health insurance premiums would have to be higher than the levels suggested by the Clinton Administration in the Fall of 1993, by 15 to 20 percent ." (CBO estimated the premiums would be 15 percent higher 19 and Lewin-VHI 17 percent higher ." DRI/McGraw-Hill, based on outside data, assumed premiums would be 20 percent higher .") 17 Under the Clinton plan the Government would obligate itself to pay subsidies or premium "discounts" to the alliances to make up the difference between the cost of the premium and the capped contribution of certain employers and individuals . For the 80 percent employer share of the average premium, the financial liability of all employers participating in the regional alliances would be limited to 7 .9 percent of their payroll, in the aggregate. For small employers of low-wage workers, the employer share would be capped at a much lower percent of payroll, as low as 3 .5 percent . For most early retirees, the Government would pay all of the employer share . For the family share of the premium, low income people would be subsidized based on a percent-of-income cap . For a more detailed summary description, see U .S. Library of Congress . Congressional Research Service . Health Care Reform : President Clinton's Health Security Act . Report No . 93-1011 EPW, by Beth Fuchs and Mark Merlis . Washington, Nov . 22, 1993 . p. 28-36 . Also, CBO, An Analysis of the Administration's Health Proposal . p. 11-12 . 18 For those premiums see : The White House Domestic Policy Council . Health Security: The President's Report to the American People. Washington, October 1993 . p. 112-113 . 19 CBO, An Analysis of the Administration's Health Proposal, p . 30, 36. 20 Lewin-VHI, Inc . The Financial Impact of "The Health Security Act ." December 9, 1993. Fairfax, Virginia, Lewin VHI, Inc ., 1993 . p. 25. 21 DRI/McGraw-Hill . The Administration's Health Care Reform Plan : National Macroeconomic Effects . Prepared for : Citizens for a Sound Economy Foundation . Washington, February 1994 . p. 9. CRS- 1 9 TABLE 3 . Effect of the Clinton Health Care Reform on the Federal Deficit : Alternative Estimates, With and Without Enforced Spending Caps, 1996-2000 (change in deficit and CBO baseline deficit, in $ billions) Source Administration, with caps on premiums and Federal entitlement a Lewin-VHI, with caps on premiums and Federal entitlement' CBO, with premium caps' Fiscal year 1998 1999 1996 -3 .2 1997 2000 1996-2000 6 .9 4 .8 -18 .2 -37 .7 -47 .4 -3 .9 0 .8 11 .1 -5 .3 -14 .4 -11 .7 1 .0 20 .0 32 .0 21 .0 10 .0 84 .0 38 .1 DRI with premium caps d -2 .1 -3 .6 26 .1 15 .9 1 .2 2 .2 10 .0 58 .5 88 .8 114 .8 274 .2 DRI without premium caps -----------------------------------------------------------------------CBO baseline deficit 166 182 180 204 226 958 without health care reform' Note : A negative number means decreasing the means increasing the deficit . eficit ; a positive number Sources : a U .S . Executive Office of the President . Office of Management and Budget . Budget of the United States Government for Fiscal Year 1995 . Washington, U.S . Govt. Print . Off., February 7, 1994 . p . 190 . ' Lewin-VHI, Inc . The Financial Impact of "The Health Security Act ." December 9, 1993 . Fairfax, Virginia, Lewin-VHI, Inc ., 1993 . p . 45 . 'U .S . Congressional Budget Office . An Analysis of the Administration's Health Proposal . Washington, February 8, 1994 . p . 29 . The Administration's Health Care Reform Plan: National Macroeconomic Effects . Prepared for : Citizens for a Sound Economy d DRI/McGraw-Hill . Foundation . Washington, February 1994 . Appendix C, p . 3-4 . U .S . Congressional Budget Office . The Economic and Budget Outlook : Fiscal Years 1995-1999 . Washington, January 1994 . p . 29 . e CRS-20 Starting with 17 percent higher premiums than the Clinton Administration, Lewin-VHI estimated that net Federal subsidy payments would be $37 billion higher ($153 billion in subsidies rather than the $116 billion estimated by the Administration), measured cumulatively over the 1996 to 2000 period ." Also, because employer spending on health care would increase, taxable wages would decrease, and Federal tax revenues would be lower - all in the opposite direction from what the Administration projected. For the 1996-2000 period, Lewin-VHI estimated that tax revenues would decrease by $18 billion ; in contrast, the Administration had projected a $23 billion increase in tax revenue from an expanded taxable base of employee wages . The net difference is $41 billion less in revenue under the Lewin-VHI estimates . Added together, the $37 billion in higher subsidies and $41 billion in lower revenues mean a differential effect on the Federal budget of $78 billion over the five-year period, relative to the Administration estimates23 This uses up the $45 billion cushion allowed by the Administration . In addition, it uses up $23 billion of the ($58 billion in) deficit reduction hoped for by the Administration ." The Congressional Budget Office estimated that, instead of the deficit decreasing by $37 .7 billion as the Administration had estimated, the Clinton reform plan would increase the deficit by $10 .0 billion in 2000, a difference of $47.7 billion . Approximately $15 billion of this difference can be attributed to the 15 percent higher level of premiums assumed by CBO 25 The additional cost to the Federal Treasury (and in parallel fashion to private payers) would be even larger if premiums are higher than assumed by these studies . Premiums could be higher if the health reform plan adopted promises a more generous package of benefits or if it does not reduce cost-shifting to the degree envisioned in the Clinton plan . 22 The subsidy does not increase by as much as the premiums because some individuals and employers would be below the percent-of-income and percent-ofpayroll caps needed to qualify for a subsidy ; they would pay the increased premium on their own, without Federal subsidy . 2a Lewin-VHI, p. ES-15 or p . 56. 24 The Lewin-VHI study makes its comparisons to Administration estimates presented in testimony to the Senate Finance Committee by Alice Rivlin, Deputy Director of the Office of Management and Budget, Nov . 4, 1993. The Administration's estimate of deficit reduction was subsequently revised downward . 25 CBO, An Analysis o f the Administration's Health Proposal, p . 36-37, 38-39. CRS-2 1 NO CAP ON FEDERAL ENTITLEMENT FOR PREMIUM SUBSIDIES Section 9102 of the Health Security Act describes as a "capped entitlement ." the system of Federal payments to regional alliances (to make up for premium discounts on behalf of low-income families and small, low-wage employers and for administrative expenses) . The Act specifies the dollar amount that can be spent on those alliance payments for fiscal years 1996-2000, and provides an indexing formula for subsequent years 26 The Congressional Budget Office expressed the belief that the caps on payments to the alliances would not be legally binding . CBO felt that the section 9102 limitation did not diminish the Federal Government's responsibilities under the Administration's reform proposal : other provisions of the Health Security Act would still oblige the Government both to make subsidy payments on behalf of employers and families and to ensure health coverage for all eligible people . 7 Furthermore, the proposed legislation contains no provisions for limiting those entitlements in the face of a funding gap, other than providing for expedited congressional consideration of the matter ." The Act does not offer any substantive guidelines for how to deal with a shortfall ." (Similar questions should be asked about the likely effectiveness of other proposals to the limit the Government's obligation to subsidize health insurance, once the promise is made to guarantee coverage .) 26 According to section 9102(e) of the Act, the maximum total Federal payments to the alliances would be limited to $10 .3 billion in FY96, $28 .3 billion in FY97, $75 .6 billion in FY98, $78 .9 billion in FY99, and $81 .0 billion in FY2000 . For fiscal years after 2000, the limit would be the previous year's limit inflated by the increase in the CPI multiplied together with the average annual percentage change in the population for the previous three years and the average annual increase in real GDP per capita for the previous three years . This is equivalent to the growth rate in nominal GDP . 27 CBO points out, "Because the Congress has the constitutional right to make and change its own rules, however, procedural mechanisms cannot guarantee that an issue will be considered . If the Congress took no action, the courts might be asked to decide which portion of the legislation took precedence-payments to the alliances to ensure coverage of the specified benefits or the limits on federal U.S. Congressional Budget Office . An Analysis of the payments ." . 8, 1994 . p. 24. Administration's Health Proposal . Washington, Feb 28 29 Ibid . The Health Security Act provides that when the Secretary of Health and Human Services anticipates that the capped amount is likely to be insufficient for a fiscal year, the Secretary is to notify the President, the Congress, and the regional alliances . The President has 30 days to submit to Congress a report containing specific legislative recommendations for actions to eliminate the The Congress is to give these recommendations "expedited shortfall . consideration ." C -22 The Health Security Act provided for $274 .1 billion n subsidies over the 5 years FY1996-2000, comfortably exceeding the Fall 1993 Administration projection of $116 billion and the Lewin-VHI estimate of $153 billion in needed subsidies . Consequently, it did not matter for the Administration's or Lewin-VIII's estimates that they had accepted the entitlement cap . CBO estimated that, without the cap on entitlements, Federal subsidy payments owed to the regional alliances would be $82 billion in 1998,$108 billion in 2000, and $173 billion in 2004.3° Those estimates exceed the capped Federal alliance payments specified in the Health Security Act31 - by $27 billion in FY2000 alone, and $56.9 billion over the FY1996-2000 period . CBO concluded that the cap on the entitlement was unlikely to be enforceable, as did DRI/McGraw-Hill . GLOBAL SPENDING CAPS NOT FULLY EFFECTIVE Like the Administration and Lewin-VHI, CBO assumed that the global spending caps would be effective . The DRI/McGraw-Hill estimators challenged that assumption . In practice, the concept of the caps not being effective means that because the consequences of enforcing the spending caps would be so unpopular - with consumers and/or providers - the Congress would relax them, permitting an increase in private alliance premiums and Federal premium subsidies (as well as Medicare and Medicaid spending) . DRI estimated the effect on the deficit assuming that the global spending cap was effective and, alternately, assuming that the cap was not effective but that all other major elements of the Clinton plan were adopted . With the spending cap, DRI estimated that the deficit would increase by $1 .2 billion in 2000 ; without the spending cap, the deficit would increase by $114 .8 billion, a difference of $113 .6 billion. Lewin-VHI made an alternate estimate of the effects on the Federal budget assuming that premiums would be allowed to grow at a rate 1 .5 percentage point higher than permitted under the Health Security Act . They note that this represents a midpoint between the rate of growth allowed under the Act and the currently projected rate of growth in health spending . 32 s° U .S. Congressional Budget Office . An Analysis of the Administration's Health Proposal . p . 33. si See footnote 7 on p . 8. 32 KPMG Peat Marwick conducted estimates of the Clinton plan assuming that the spending caps would be 50 percent effective, which is comparable to this Lewin-VHI assumption . Peat Marwick also estimated the impact on national (continued. ..) CRS- 23 Lewin-VHI estimated that, cumulatively over the period 1996-2000, with a 1.5 percentage point higher rate of increase in the premium, the Federal deficit would increase by $17.4 billion, rather than decreasing by $24 .6 billion, as they estimated under the terms of the Health Security Act . This $42 billion swing in the projected effect on the deficit is made up of three components : an increase in new Federal program expenditures of $22 .9 billion, a decrease in savings to current programs of $6 .2 billion, and a decrease in tax revenues of $12 .9 billion, all measured over a 5 -year period ." Other proposals have been more lenient than the Clinton plan in their goals of reducing the rate of growth of health care spending . This suggests a substantially larger Federal deficit effect, unless the subsidy promises are significantly cut back . IN SUM These estimates of the change in the Federal deficit as a result of the Clinton health care reform plan range from the Administration's projection of the deficit decreasing by $38 billion in FY 2000, to the DRI estimate of the deficit increasing by $115 billion in the absence of efficiency gains or cost controls . This is a range of discrepancy of $153 billion for a single year . The CBO baseline estimate of the total Federal deficit without health care reform is $226 billion for FY 2000 . Part of the difference between the Administration's and other estimates of the effect on the deficit is attributable to the higher premium estimates by all the other analysts . The effect of higher premiums is compounded by rejecting the assumption (made by the Administration and Lewin-VHI) that the proposed cap on the Federal obligation to pay premium subsidies would be enforceable . By far the largest part of the difference, however, is attributable to the projection by some private researchers that health care demand will exceed the global spending targets and that the proposed spending caps will not prove effective. Estimates of the added effect on the deficit if the spending caps are 32( . . .continued) health expenditures and Federal subsidy costs assuming that the premium caps were effective at the rates of 25, 50, and 75 percent . They did not publish estimates for 0 and 100 percent effectiveness, however . Consequently, their estimates could not be compared directly with those from the other studies included in this report. Relative to the studies reported here, the Peat Marwick analysis suggests much larger increases in national health care spending and the Federal deficit if the spending controls are not effective . See: KPMG Peat Marwick . Analysis of H.R. 3600, The Health Security Act of 1993 . March 28, 1994 . Also, the Technical Appendix . Washington, April 7, 1994 . 33 Lewin-VHI, p. 54-55 . CRS-24 not fully effective run as high as $110 billion a year or more by FY 2000 . This raises concern about adopting a reform that does not include effective mechanisms for containing costs . CRS-25 APPENDIX A. SYNOPSES OF REPORTS ESTIMATING COST SAVINGS This appendix describes four of the studies of the Clinton health care reform plan referred to in the main text . Special attention is given to their method of estimating the cost savings that need to be accomplished to meet the spending goals of the Clinton plan . The emphasis is on private sector savings more than savings under Government programs . The Administration, in the Federal Budget for Fiscal Year 1995, projects the savings in Federal programs anticipated as sources of funding for the health care reform, in addition to new tax revenues. This is the most publicly available measurement of the Clinton plan . A Clinton Administration briefing handout explains how the needed savings in private sector programs can be achieved without resorting to premium caps. The Lewin-VHI study addresses changes expected in both Government and private health care spending. It identifies and includes numerical estimates of parts of the health care system where increased spending can be expected, as well as offsetting areas where reduced spending can be expected, including both voluntary and enforced savings . The Goldman Sachs report is included because it provides a broad overview of the basic issues . It explains the main components of expected increases in health care utilization, provides aggregate estimates of the efficiency gains needed to avoid the spending caps, and lays out the major policy alternatives if the desired cost savings are not achieved . DRI/McGraw-Hill measures the effects of the health care reform in both "nominal" and "real" dollar terms . They estimate the savings attributable to managed care, Medicare spending caps, and private sector spending caps . They then allocate the total nominal savings between cuts in prices and cuts in real services . They estimate the effect on the Federal deficit with and without spending caps . Not included is the CBO study which focused on the e ect of the reform plan on the Federal budget . CLINTON ADMINISTRATION The Clinton Administration provided separate estimates of savings for Government-financed health care programs and for the private alliance system . The two sources cannot simply be added together . The President's Budget for Fiscal Year 1995 contains estimates of savings under Government health CRS-26 programs as part of its "sources of funds" to pay for the health care reform 34 A handout distributed at one of the White House briefings on the reform plan presents estimates ofthe savings that the Administration anticipates from market reform and competition in the private health care system 35 Federal Program Savings Table A .1 presents the Clinton Administration's estimates of the effect of the reform plan on the Federal budget . For simplicity of discussion, detailed line items have been aggregated into a few broad categories, under sources and uses of funds. 36,37 Estimates are presented for fiscal years 1998 and 2000 . (Calendar year 1998 is the first year the program is expected to be implemented in all States.) Approximately half of the $74 .7 billion in sources of funds for 1998 is expected to come from cost savings in Federal health care programs, and the other half from increased tax revenues . On the uses of funds side, approximately half would go to new Federal health care programs . The other half would subsidize alliance insurance premiums and reduce the deficit with what is left over . The Administration proposes a substantial curtailment in the rate of growth In addition, the of Federal spending on Medicare 3 S and Medicaid . Administration expects to realize savings in Federal health programs for the Department of Veterans Affairs, the Defense Department, Federal Employees 34 U.S . Executive Office of the President . Office of Management and Budget . Budget of the United States Government for Fiscal Year 1995 . Washington, U.S. Govt . Print . Off., February 7, 1994 . Section 4, `Reforming the Nation's health care system to provide health security for all Americans,' p . 188-190 . 35 U .S . President, 1993- (Clinton) . Using Resources More Efficiently: Anticipated Savings from Health Reform . White House briefing handout . Washington, February 1994 . as Underlying the net summary figures reported in this table are offsetting transfers of payments from other Federal programs, payments required from the States, and payments required from beneficiaries . For more detailed budgetary estimates, see Clinton Administration . Distributional Analysis . Chart II-F, p . 16 . Also: Clinton Administration . Description of Proposed Financing Sources. Nov. 2, 1993 . Reproduced in Bureau of National Affairs . Daily Tax Report, no . 212, Nov . 4, 1993 . p. L-4 to L-5 . Also: Lewin-VHI, Inc . The Financial Impact of "The Health Security Act." December 9, 1993 . Fairfax, Virginia, Lewin-VHI, Inc., 1993 . Table 6, p . 39. s7 For a concise description of the sources and uses of Federal funds under the Health Security Act, see Lewin-VHI, p . ES-4 to ES-6 . 3a Thirteen percent of the Medicare savings would be financed by increased payments from beneficiaries . CRS-2 7 Health Benefits program, and the Public Health Service, as well as Medicaid in part by shifting enrollees to the alliance plans that are subject to various spending controls . The contribution ofprogram savings to sources offunds is expected to grow over time - from 51 percent in 1998 to 61 percent in 2000 . The Administration expects the cumulative effect of the cost controls to generate considerably more savings on Federal program expenditures ($79 .2 billion in 2000 compared with $38 .4 billion in 1998) and more revenues from a broadened taxable base (contributing $13 .7 billion in 2000, compared with $4 .4 billion in 1998) . On the uses of funds side, although total discounts paid out are expected to increase by $20 billion, the net cost of premium discounts is projected to fall slightly (from $31 .4 billion to $28 .8 billion), largely as a result of increased offsets from the Medicaid program . The biggest projected increase in the use of funds is for deficit reduction . For 1998, the Administration projects that their health reform plan would increase the deficit by $4 .8 billion ; for 2000, it would reduce the deficit by $37 .7 billion . The cushion for discounts is also projected to rise slightly, from $10 .4 billion in 1998 to $13 .0 billion in 2000 . TABLE A .l . Changed Sources and Uses of Federal Funds in the Clinton Health Care Reform, Fiscal Years 1998 and 2000 (Clinton Administration estimates) (in $ billions) Sources of Funds 1998 2000 Uses of Funds SAVINGS 38 .4 79.2 NEW PROGRAMS Medicare savings 22 .1 39 .2 Medicaid savings 9 .2 27 .1 Other Federal program savings 6 .9 10.9 Reduced debt service from deficit reduction 0 .2 2.0 1998 2000 36 .1 47 .6 Medicare drug benefits Long-term care 15.0 17.2 12.2 20 .1 Public health/ administration/miscellaneous 8 .9 NEW REVENUES 36 .2 50 .6 INSURANCE SUBSIDIES Tobacco tax Corporate assessment 11 .1 10 .9 Tax deduction for self-employed Other revenue effects 20.0 5 .1 5 .2 Net cost of premium discounts 34 .5 Cushion for 74 .7 .6 (subsid discounts DEFICIT REDUCTION TOTAL SOURCES 10 .3 129.8 TOTAL USES (4 .8) 37 .7 74 .7 129 .8 Source : U.S . Executive Office of the President . Office of Management and Budget . Budget of the United States Government for Fiscal Year 1995. Washington, U .S . Govt . Print . Off., February 7, 1994 . Format adapted from Chart 4-1, p . 188 . Numbers for 1998 from Tables 4-2 and 4-3, p . 189-190 . Columns may not add to totals due to rounding. C S-28 Savings from Market Reform and Competition The Clinton Administration expects that insurance and health care market reforms and competitive pressures will reduce the rate o f growth in private health expenditures sufficiently that the legislated caps on premium growth would not become binding . All together, the savings detailed in table A .2 total $108 billion in 2000 ; 39 the Administration estimates that $107 billion in savings would be needed in order to meet the expenditure budget under the premium caps set forth in the Health Security Act ." (See the last line of table A .2 .) Relative to expenditures for universal coverage under the regional and corporate alliances before these savings, estimated at $674 billion in 2000, $108 billion represents savings of 16 percent. In earlier years of the reform, savings would be lower . For 1998, the estimated savings is 11 .1 percent of baseline expenditures . (See the last line of table A .3 .) The Administration expects savings from the high administrative costs now associated with non-group and small group insurance policies to amount to approximately $11 billion per year in 2000 . As a result of the single claims form and standardized benefits, universal coverage, and better coordination between providers and health plans, they expect that the administrative costs facing hospitals and physicians' offices can be reduced by about $17 billion in 2000 . This encompasses a reduction of 0 .7 percent of total hospital expenditures and 3 .4 percent of total physician expenditures . The Administration also expects that when consumers have a choice among plans, information with which to compare standardized plans, and a price incentive, they will choose lower-cost plans, for a savings of $24 billion in 2000, or 3 .6 percent of baseline expenditures . The largest source of anticipated savings is from changes in provider behavior toward using more cost-effective practice patterns and reducing The Administration suggests that expensive, unnecessary procedures . productivity improvements of 1 .75 percent per year are reasonable, and would reduce expenditures by 9 percent at the end of 5 years . They estimate savings of $57 billion in 2000 from more efficient provider behavior . Table 2 in the text compares the Administration estimates of components of cost savings to the estimates made by Lewin-VHI and DRI/McGraw-Hill . as These savings estimates do not include potential savings from malpractice reform, workers compensation reform, reductions in fraud and abuse, and other proposed reforms . 40 This refers to the private sector and Medicaid savings assumed by premium caps and does not include savings in the Medicare program . CRS-2 9 TABLE A .2 . Savings from Market Reforms and Competition (Clinton Administration estimates) (in $ billions) Source of Savings 1996 1997 1998 1999 2000 Elimination of small group market 1 3 9 10 11 Streamlined administration in hospital and physician offices 3 6 10 15 17 Consumer switching to lower cost plans 1 4 14 22 24 Change in provider behavior 8 18 29 42 57 Total savings from health reform 14 31 63 89 108 Expenditure reductions needed to meet premium caps 13 28 50 77 107 TABLE A .3 . Projected Savings Relative to Expenditures (Clinton Administration estimates) (in $ billions and percent) 1996 1997 1998 1999 2000 Total savings from health reform 14 3 63 89 108 Clinton plan expenditures before savingsa 482 534 570 620 674 Savings as a percent of expenditures before savings 2 .9% 5 .9% 11 .1% 14 .4% 16 .0% a Projected expenditures for services in the guaranteed benefit package delivered to all those in regional and corporate alliances, including new expenditures to insure the uninsured and improved coverage for the underinsured . (Does not include Medicare or public health expenditures .) Source for tables A .2 and A .3 : U .S . President, 1993- (Clinton) . Using Resources More Efficiently: Anticipated Savings from Health Reform . White House briefing handout . Washington, February 1994 . Tables on last two pages . Percentages calculated by CRS . CRS-30 LEWIN-VII The consulting firm Lewin-VHI used its Health Benefits Simulation Model (HBSM), a micro-simulation model of health expenditures ." The model is based upon detailed data regarding actual insurance coverage of individuals, patterns of health care utilization, and health care expenditures by sources of payments for U.S. households . Their published estimates focus on 1998, the first year that the plan is proposed to be in full operation. Lewin-VHI estimates that in 1998 aggregate national health care expenditures would be essentially equal under the Clinton plan ($1,394 .4 billion) compared with continuation of current policy ($1,395 .0 billion under the baseline) 41,43 The near-equivalence in total spending before and after reform masks an increase in spending of approximately $147 billion that Lewin-VHI projects to be offset by an equal amount of spending reductions under the Clinton reform plan . Table A.4 rearranges the information presented in the Lewin-VHI study to emphasize three broad categories of tradeoff under the Clinton plan . The first is the tradeoff between the increased utilization of health care services due to expanded insurance coverage (+ $64 .0 billion), and the spending restraint mechanisms embedded in the plan (-$71 .5 billion) . The net effect is an estimated $7.5 billion savings . Managed care accounts for $14 .9 billion of the savings, offsetting 23 percent of the increased utilization of $64 billion . The alliance premium caps contribute approximately three times more savings-$47 .3 billion or 74 percent of the increased utilization . The second tradeoff is recouping the reimbursement windfall that providers might otherwise collect once they were being paid by the health care alliances for people previously uninsured (uncompensated care) or covered by low Medicaid reimbursement rates (+ $68 .9 billion). This would be accomplished by eliminating the cost-shifting component in the fees that had previously been charged to "paying customers" (- $68.9 billion) . Only part of this is expected to occur voluntarily on the part of providers ($36 .4 billion). The remainder ($32 .5 billion) would be recouped by setting the reimbursement rates and initial premium levels so that any potential windfall is completely offset . The net effect on health care expenditures is, therefore, assumed to be zero . " Lewin-VHI, Inc . The Financial Impact of "The Health Security Act ." December 9, 1993 . Fairfax, Virginia, Lewin-VHI, Inc ., 1993 . 42 The Lewin-VHI aggregate of $1,395 billion measures acute care spending, not national health spending ; it includes Medicare but excludes spending on nursing homes . 43 The Administration had estimated that in 1998 expenditures under the Clinton plan would be $15 billion above the baseline of no reform . CRS-31 The third is a tradeoff between increased administrative costs on the part of the Federal Government and the States to manage the newly created system of health care alliances (+ $13 .6 billion), versus a decrease in the administrative costs borne by health insurers and providers, mostly as a result of the standardization of the health insurance system (- $6 .7 billion). The net effect is an estimated increase in administrative costs of $6 .9 billion . The projected net increase in administrative costs (+$6 .9 billion) largely offsets the net savings from spending controls (-$7 .5 billion), leaving an overall net savings of $0 .6 billion under the Clinton plan in 1998 . For later years Lewin-VHI projects the cost controls would bring aggregate spending significantly below the baseline . (See appendix B of this report, page 40 .) CRS- 3 2 TABLE A .4 . Changes in Aggregate National Health Care Spending under the Health Security Act in 1998 (Lewin-VHI estimates) (in $ billions) Baseline National Health Expenditures 1,395 .0 INCREASED SPENDING NET CHANGE REDUCED SPENDING Increased Utilization 64 .0 Spending Restraints (71 .5) (7 .5) Redesign : Previously uninsured Expanded coverage for already insured 41 .6 Long-term care Public health activities (including WIC) 11 .6 5 .4 5.4 68.9 Uncompensated care savings Increased reimbursement for Medicaid beneficiaries 23 .2 TOTAL INCREASE Spending caps Medicare spending limits Alliance premium caps Medicaid (net of offsets) Recovery of Improved Reimbursement Redesign : Reduction in cost shifting (56 .6) (13 .1) (47.3) +3.8 (68.9) 0 (36 .4) Enforced: 45 .7 Increased Federal and State Administrative Costs State alliance Alliance administration Guarantee fund reserve accumulation (14 .9) Enforced: Enhanced Provider Reimbursement Federal operations Program administration Medical education Veterans hospitals Managed care 13.6 4 .7 1.7 1.3 1.7 Pre-emption of reimbursement windfall Reduced Private Administrative Costs Redesign : Insurer administration (includes newly insured) Provider administrative savings (32 .5) (6 .7) +6 .9 (4 .8) (1 .9) 8 .9 5.0 3.9 146.5 Redesign : Enforced : TOTAL REDUCTIONS National Health Expenditures under Clinton Reform Plan (58 .0) (89 .1) (147 .1) NET CHANGE (0.6) 1,394 .4 Source : Adapted from : Lewin-VHI . The Financial Impact of "The Health Security Act ." Executive Summary, p . ES-3 ; table ES-l, p . ES-4 ; and Appendix A, table A-l, p . A-3 . CRS-3 3 GOLDMAN SACHS Increased Utilization Due to Expanded Insurance Coverage William Dudley, an economist at Goldman Sachs, the international investment firm, 44 expects that the expansion of insurance coverage promised under the Clinton plan would significantly increase the demand for health care services, as the out-of-pocket costs of obtaining certain kinds of care is reduced for many people . His estimates are based on rough, aggregate calculations, as explained below . He estimates that by 2000, relative to total spending on those health care services within the purview of the Clinton plan, personal health care spending would increase by 11 .0 percent, made up of the following components, summarized in table A.5 : • Universal coveraae . Previously uninsured people could now seek health care knowing that insurance would pay most of their bill . Based on the assumption that demand from the uninsured would increase about 75 percent, Dudley estimates that this would increase aggregate personal health care spending by 4 .6 percent . • Improved benefits . Even people previously covered by insurance would find that their new insurance now fully covered certain preventive care and reduced their per visit or per prescription out-of-pocket copayments . Assuming that demand from this group increases by about 6 percent, this adds 2 .7 percent to health care spending . • Expansion of benefits . Medicare beneficiaries would now have insurance coverage for prescription drugs . New services for home and community-based long-term care would be available to the elderly and disabled, regardless of a person's income . Assuming that demand for prescription drugs in the Medicare population would increase about 15 percent and that demand for long-term care at home would double, this adds another 2 .7 percent to health care spending . • Improved access for Medicaid recipients . Medicaid recipients would now have the same level of insurance purchasing power as other patients seeking health care . Assuming that the demand for health care services among Medicaid recipients increases by 5 percent, this adds 1 percent to personal health care spending . Dudley uses as abase for his percentage calculation the Lewin-VHI estimates of health care spending under the Clinton plan . (Lewin-VHI assumed that initial premiums would be 17 percent higher than the Clinton Administration had estimated, but accepted the Administration's limits on the growth rate of premiums and Medicare and Medicaid spending .) Nonetheless, Dudley's rough Dudley, William . The Clinton Healthcare Plan : No Free Lunch . Goldman Sachs, U .S . Economic Research . NewYork, Goldman, Sachs & Co ., January 1994 . 44 CRS-34 estimate of increased utilization is substantially higher than Lewin-VHI's ($143 billion for Dudley in 2000, versus $64 billion for Lewin-VHI in 1998) . TABLE A.5. Increased Demand for Health Care Spending under the Clinton Plan (Goldman Sachs estimates) Percent increase in total personal health Source of increase demand care spending' 4.6 Universal coverage Improved benefits 2 .7 2.7 Expansion of benefits Improved access for Medicaid recipients 1.0 11 Total .0 'Excludes nursing home and dental expenditures, which Dudley concludes would not be significantly affected by the Clinton plan . Source : Dudley, William . The Clinton Healthcare Plan : No Free Lunch . Goldman Sachs, U .S . Economic Research . NewYork, Goldman, Sachs & Co ., January 1994 . p. 2. For the year 2000, Lewin-VHI estimated that national health care expenditures under the Clinton reform would be 3 .5 percent lower than the CBO baseline . Dudley estimates that roughly 20 percent of national health care expenditures would not be affected by the Clinton plan (e .g., most dental care and nursing home care) . Consequently, to achieve overall savings of 3 .5 percent would call for savings of 4 to 5 percent on the other 80 percent of expenditures that are under the purview of the Clinton plan . TABLE A.6 . Increase in Efficiency Needed to Avoid Spending Caps (Goldman Sachs estimates for the year 2000) Percent change in spending To cover to be offset 11.0 Increased demand from improved insurance coverage Lower aggregate spending target 4.0 Total 15.0 Dudley thus concludes that to meet the lower spending goal of the Clinton plan without having to enforce the global budget caps, there would need to be efficiency gains of 15 percent . That is, efficiency gains would need to compensate for both the 11 percent increase in utilization above the baseline and the 4 percent reduction in aggregate spendingbelow the baseline . (See table A .6 above .) CRS- 3 5 Basic Alternatives in Paying for Reform Dudley sees two main ways to reconcile increased demand for health care services with lower aggregate expenditures : increased efficiency and/or rigorously enforced spending caps4 5 Dudley points out that 2000 would be only the third year that the Clinton plan would be in full effect . He concludes that it is unreasonable to expect efficiency gains as large as 15 percent to occur so quickly, and perhaps not even in the longer run . Consequently, he expects that the global budget caps would become binding . Enforcing the spending caps could take the form of rationing health care services to patients (including reduced quality or timeliness of service) or cutting the incomes of health care providers . If instead, the caps are eased, that would mean higher private premiums and Government subsidies . Dudley is concerned about what would happen to the Federal deficit if the global caps were not effective in controlling expenditures or were overturned because they were not politically viable in the face of the sacrifices required . He estimates that the Government is at risk for about 80 percent of any cost overruns, because of the Clinton plan's caps on the contributions required from employers and families and the promised premium subsidies . Dudley reached the same conclusion later issued by the Congressional Budget Office that the capped entitlement limit on the subsidy payments would probably not be feasible in practice . Dudley lays out three alternative parameters for financing the promises of the Clinton plan : accomplish efficiency gains of 15 percent by the year 2000, enforce the budget caps rigorously by cutting health care services and/or health care providers' income, or see the Federal budget deficit worsen dramatically . He starts from the Lewin-VHI estimates that health care spending under the Clinton plan would be $1,573 .8 billion and that the Federal deficit would decrease by $14 .4 billion for the year 2000, assuming the spending caps were 100 percent effective . Dudley calculates a matrix of 20 estimates of the effect on the deficit, assuming efficiency gains ranging from 0 to 15 percent, and spending cap effectiveness ranging from 0 to 100 percent . In the "worst case," if there were no efficiency gains and the spending caps had no effect, Dudley estimates that total health care expenditures would be $188 .9 billion higher than the Lewin-VHI baseline . If the Government paid 80 percent of these higher costs, Federal outlays would increase by $151 .1 billion . If none of this were offset by increased tax revenue, the deficit in the year 2000 would increase by $136 .7 billion, rather than decrease by $14 .4 billion as projected by Lewin-VHI . 46 45 Increased efficiency - providing the same service at lower cost - could take the form of reducing the unit costs of a particular service (including savings in administrative costs) or eliminating unnecessary medical procedures that do not improve patient outcomes . 46 Dudley, The Clinton Healthcare Plan, table 4, p . 8 . CRS- 36 DRI/McGRAW HILL The economic forecasting firm DRI/MicGraw-Hill analyzed the Administration's reform plan in the context of its quarterly macroeconomic model of the U .S. economy ." DRI developed a five-stage simulation to examine the differential effect of adding separate layers of the Administration's plan . DRI measured the effects of each stage on numerous macroeconomic variables, including national output (GDP), employment, inflation, interest rates, as well as on the Federal deficit and both nominal and real spending for medical care services ." DRI based its analysis on the September 22, 1993, version of the Clinton plan . The forecasting period was (calendar years) 1994 to 2000 . DRI uses National Income and Product Accounts (NIPA) definitions of health care spending, not National Health Accounts . Based on outside data, DRI assumed that the initial premiums are approximately 20 percent higher than the Administration's estimate .49'5° DRI did not accept the cap on the Federal obligation to pay premium subsidies . Starting from a baseline of no reform, DRI first modeled universal coverage, assuming Federal deficit financing of the increased utilization . Second, they added the employer mandate, including the requirement for employers and families to pay premiums, offset by Government subsidies for some . Third, they added the corporate assessment on companies that formed their own alliances . Fourth, they included the other tax changes in the Clinton plan, including the tobacco tax, the limits on flexible spending accounts (cafeteria plans), and the changes in medical expense deductions . Fifth, and finally, they modeled the effect of the spending caps - in which they include caps on the rate of growth of alliance premiums and Medicare, and the shift to managed care . This CRS report focuses on the differences found between DRI's fourth simulation, which measure the effects of all major elements of the Clinton plan without the spending caps (table A .7, line 2), and the fifth simulation, which incorporates the spending controls (table A .7, line 3) . 47 DRIIMcGraw-Hill . The Administration's Health Care Reform Plan : National Macroeconomic Effects . Prepared for : Citizens for a Sound Economy Foundation . Washington, February 1994 . 48 DRI also estimated the employment effects for individual industry sectors . 49 5° DRI, p. 9 . DRI assumed that employers would pass the mandated premiums forward in higher product prices (rather than shifting them backward to their workers in the form of lower cash wages or other benefits) . They note that subsequent estimates they have made suggest that employers will shift half the premiums forward in higher prices and half back in lower wages . DRI, Appendix F, p . 2 . CRS- 3 7 DRI measured the costs to the economy if the Administration's plan was adopted without spending caps and the added demand had to be financed through employer-paid premiums, the corporate assessment, and the other proposed taxes . Without cost controls and with higher payments required to cover health care, output (real GDP) and employment would be lower, and the inflation rate and Federal deficit would be higher . a 1 (See table 3 in the text .) DRI believes that the caps would become binding and that "enforced" health care spending reductions would be necessary if the cost containment objectives of the Clinton plan were to be met . DRI assumes that both provider incomes (prices paid for medical services) and utilization (quantity of medical services provided) would be reduced . DRI allocated between these two components based on historical relationships between prices and utilization . Of the $149 billion projected reduction in nominal spending in 2000 as a result of spending caps, DRI attributed $84 billion (56 .4 percent) to a lower medical price rate and $65 billion (43 .6 percent) to a lower delivery of medical services ." DRI found that, despite the expansion of the insured population under the Administration's health care plan, the level of health care services (as measured by real consumer spending for medical care services) is virtually unchanged relative to the baseline of no reform . (Table A .9, line 3 compared with line 1 .) This implies that consumption of health care services by those who were already insured must drop to accommodate expanded care for others ." 51 DRI . Table 1 .2, p . 2 . 52 DRI, Appendix F, p . 4 . 53 DRI, p . 2 . CRS- 3 8 TABLE A .7 . Consumer Spending for Medical Care Services (DRI/McGraw-Hill Estimates) (in billions of current or nominal dollars) 1996 1997 1998 1999 2000 Total spending 1 . DRI Baseline 883 .4 959 .2 1,034.2 1,115 .0 1,200 .4 2 . Clinton plan without spending caps (all taxes, sim . 4) 888 .9 983 .9 1,110 .8 1,203 .0 1,299 .8 3 . Clinton plan with spending caps (sim . 5) 887 .0 978 .0 1,029 .9 1,084 .0 1,150 .9 Differences in spending 4 . Increase in spending under the Clinton plan without controls relative to the baseline (line 2 - line 1) 5 .5 24 .7 76 .6 88 .0 99 .4 5 . Decrease in spending under the Clinton plan with versus without controls (line 3 - line 2) -1 .9 -5 .9 -80 .9 -119 .0 -148 .9 6 . Spending under the Clinton plan with controls relative to the baseline (line 3 - line 1) 3 .6 18 .8 -4 .3 -31 .0 -49 .5 7. Savings from spending controls as a percent of Clinton plan without caps (line 5/line 2) 0 .2% 0 .6% 7 .3% 9.9% 11 .5% Source : DRI, Appendix D, p . 1 . Simulation results for Additional Taxes (simulation 4) and for Spending Caps (simulation 5) . CRS-39 TABLE A .8 . Estimated Reduction in Nominal Health Care Spending (DRI/McGraw-Hill Spending Cap Simulation) (in $ billions) 1996 1997 1998 1999 2000 Managed care 0 .0 0 .0 14 .9 16 .1 17 .4 Medicare spending caps 2 .0 6 .0 13 .0 18 .0 24 .0 Private sector spending caps 0 .0 0 .0 53 .1 84 .9 107 .6 Total Spending Reductions 2 .0 6 .0 81 .0 119 .0 149 .0 Source : DRI, Table 2 .6, p . 20 ; p . 21 . TABLE A .9 . Estimated Reduction in Real Health Care Spending (DRI/McGraw-Hill Spending Cap Simulation) (in billions of 1987 dollars or percentage) 996 1 . DRI real baseline 997 998 1999 2000 544 .2 562 .2 579 .1 597 .2 613 .6 629 .1 2 . Real spending under Clinton plan without spending caps 525 .9 555 .9 3 . Real spending under Clinton plan with spending caps 527 .6 560 .9 583 .2 597 .7 +5 .0 -30 .4 -31 .4 4 . Change in real spending as a result of caps e 2 - line 3) 5 . Change in real spending as a % of health care spending before caps (line 4 / line 2) +0 .3% +0 .9% -4 .2% -5 .0% -5 .0% Source : DRI, Table 2 .6, p . 20 ; p . 21 ; unpublished baseline numbers . CRS- 40 APPENDIX B . NATIONAL HEALTH CARE SPENDING UNDER THE BASELINE AND THE CLINTON PLAN The Administration estimated that from 1995 until 1998 (the year when the plan is first intended to be in effect in all States) total national health care expenditures would be slightly higher than under the current system ; by the year 1999 and thereafter, expenditures would be lower than the baseline . Although other estimators cited in this report may agree on this general pattern, they differ over which year the crossover of the baseline would occur. For example, Lewin-VHI and DRI show the crossover occurring one year earlier, by 1998, and CBO one year later, by 2000 . It is also worth noting that aggregate national health care expenditures are projected to be substantially higher in 2000 than in 1994 whether or not the plan envisioned in the Health Security Act takes effect . The Clinton reform plan promises only to restrain the rate of growth of total health expenditures, not to lower them absolutely . CRS- 41 TABLE B .1 . Estimates by the Administration, the Congressional Budget Office, Lewin-VHI, and DRI/McGraw-Hill (in $ billions, calendar years) 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Administration Baseline 982 1069 1168 1275 1392 1517 1653 Clinton plan 1072 1179 1290 1407 1492 1597 Difference 3 11 15 15 -25 -56 CBO Baseline 982 1069 1163 1263 1372 1488 1613 1748 1894 2052 2220 Clinton plan -1176 1285 1411 1489 1583 1700 1820 1942 2070 Difference 13 22 39 1 -30 -48 -74 -110 -150 Lewin-VHI Baseline 998 1.098 1185 1288 1395 1510 1631 Clinton plan 998 1098 1206 1316 1394 1477 1574 Difference 0 0 21 28 -1 -33 -57 DRI/McGraw-Hill Nominal baseline 812 883 959 1034 1115 1.200 Clinton plan with caps 887 978 1030 1084 1151 Difference 4 19 -4 -31 -49 Clinton plan without caps 889 984 1111 1203 1300 Difference 5 25 77 88 99 (continued) CRS- 42 TABLE B1 . Estimates by the Administration, the Congressional Budget Office, Lewin-VHI, and DRI/McGraw-Hill-continued Sources : Clinton Administration . The Health Security Act : A Financial and Distributional Analysis . December 1993 . Chart I-A, p . 3 . Chart I-B, p . 4 . U.S . Congressional Budget Office . An Analysis of the Administration's Health Proposal . Washington, Feb . 8, 1994 . Table 2-1, p . 26 . Lewin-VHI, Inc . The Financial Impact of "The Health Security Act ." December 9, 1993 . Fairfax, Virginia, Lewin-VIII, Inc ., 1993. Table 6, p. 39 . DRI/McGraw-Hill . The Administration's Health Care Reform Plan : National Macroeconomic Effects . Prepared for: Citizens for a Sound Economy Foundation . Washington, February 1994 . Appendix D, p . 1, simulation results for Additional Taxes and Spending Caps . Unpublished baseline numbers . CRS- 4 3 REFERENCES DRI/McGraw-Hill . The administration's health care reform plan : national macroeconomic effects . Prepared for : Citizens for a Sound Economy Foundation . Washington, February 1994 . Dudley, William . The Clinton healthcare plan : No free lunch. Goldman Sachs, U .S . Economic Research . New York, Goldman, Sachs & Co ., January 1994 . KPMG Peat Marwick . Policy Economics Group . Analysis of H .R . 3600, The Health Security Act of 1993 . Prepared for The Congressional Institute, Inc . Washington, March 28, 1994 . Analysis of H.R. 3600, The Health Security Act of 1993, Technical Appendix . Prepared for The Congressional Institute, Inc . Washington, April 7, 1994 . Lewin-VHI, Inc. The financial impact of "The Health Security Act ." December 9, 1993 . Fairfax, Virginia, Lewin-VHI, Inc ., 1993 . U .S . Congressional Budget Office . An analysis of the Administration's health proposal . Washington, February 8, 1994 . U .S . President, 1993- (Clinton) . The Health Security Act : A financial and distributional analysis . Briefing handout . Washington, December 1993 . Using resources more efficiently : Anticipated savings from health reform . White House briefing handout . Washington, February 1994 . U .S . Executive Office of the President . Office of Management and Budget . Budget of the United States Government for Fiscal Year 1995 . Washington, U .S . Govt . Print . Off., February 7, 1994 . Section 4, "Reforming the Nation's health care system to provide health security for all Americans," p . 177-93 . crsphpgw