June 11, 2004


by Robert Greenstein and Richard Kogan

PDF of analysis

View Related Reports

If you cannot access the files through the links, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.


When budget process legislation comes to the House floor later this month, an amendment may be offered by a group of conservative Members that would have adverse effects on Social Security, Medicare, veterans disability compensation, the Supplemental Security income program for the elderly and disabled poor, health and retirement programs for federal civilian and military personnel, and ultimately Medicaid and some other entitlements.  This proposal would be in addition to a proposal to place a “cap” on overall entitlement spending.[1]

Under the proposal, OMB would issue a 75-year cost estimate for any legislation that would increase the cost of any program subject to the proposal (except for Social Security, which would be treated somewhat differently and even more strictly, as described below).  Estimated cost increases over 75 years would have to be offset by cuts in the same program or another of the programs that would be subject to this provision, except Social Security.  Offsetting cuts in programs not subject to this provision — as well as offsetting measures that would raise more general revenues, such as measures to close abusive tax shelters or wasteful tax loopholes — would not be considered valid.  Legislation that did not meet this test would be subject to a point of order in the House and Senate and require 60 votes to pass on the Senate floor.

For Social Security, the rule would be even more stringent.  Cost increases would be projected not over 75 years but into infinity, and they would have to be offset into infinity through cuts that could come only from Social Security.  Legislation not meeting this test would similarly require 60 votes to pass in the Senate.

This proposal is part of a legislative package that the Office of Management and Budget transmitted to Congress on April 2.  It is included in a list of proposed amendments to the House budget process legislation that a group of House conservatives is currently circulating.

Departure from Traditional Pay-As-You-Go Rules

As noted, legislation that would increase the costs of one or more of these programs over the next 75 years (other than costs that would be offset by increases in revenues specifically dedicated to one of these programs, such as Medicare payroll taxes) would be barred, unless the legislation were accompanied by a measure making offsetting cuts in the same program or another one of the programs subject to the rule.  As a result, the proposal would represent a marked departure from the pay-as-you-go rules of the 1990s, not only in the proposal’s use of a 75-year cost estimating period but in other basic ways as well.  The pay-as-you-go rules of the 1990s required both entitlement expansions and tax cuts to be offset, and allowed the costs of such measures to be offset either by a reduction in any entitlement program or by raising any type of revenues.  Under this new proposal, by contrast, not only would tax cuts be exempt from budget discipline, and not only would measures to offset the costs of an entitlement improvement through an increase in general revenues be barred, but legislation that increased the long-term costs of any of the covered programs could be offset only by cuts in the same program or another of the covered programs.  Even offsetting cuts in other entitlement programs would not be considered valid offsets.

This proposal would have large effects.  Consider, for example, the need to address a significant problem in the SSI program for the aged and disabled poor.  The program’s asset limit is set at only $2,000 for an elderly or disabled individual, has not been adjusted to reflect inflation in 30 years, and is slated under current law to remain at $2,000 indefinitely.  Under the proposal, a measure simply to adjust the SSI asset limit for inflation would need to be accompanied by a measure that offset the cost of such an improvement over the next 75 years by making cuts elsewhere in SSI, unless cuts could be secured from another entitlement subject to this rule.  Since the other entitlements that would be subject to this rule all have more potent constituencies than SSI, the offsetting cuts likely would have to come from SSI itself.  Yet SSI already has income limits that are well below the poverty line, and the benefit that it provides to impoverished elderly and disabled individuals with little or no other income brings them up only to three-quarters of the poverty line.  Cuts in SSI would make the elderly and disabled poor still poorer.

Under the proposal, the Administration also would have the unilateral authority to add additional entitlements to the list of programs that would be subject to this new rule (i.e., to the rule that any legislation raising a program’s costs over the next 75 years must be offset by cuts in the same program or in other programs subject to the rule).  Once the Office of Management and Budget determined it could make long-term cost estimates for an entitlement, OMB would be free to subject the program to this rule.  Although OMB would consult with the Congressional Budget Office and the Budget Committees of the House and Senate, OMB alone would decide whether to make additional programs subject to this requirement.

In material accompanying the transmittal of this proposal in April, OMB emphasized its intention to add Medicaid as soon as possible to the list of programs that would be subject to this rule.  Once that were done, proposals to expand Medicaid to cover more of the uninsured would have to be offset by cuts elsewhere in Medicaid (or in one of the other entitlements subject to the rule).  In contrast, proposals to write new health-insurance-related tax deductions or similar health-insurance-related tax breaks (other than refundable tax credits) into the tax code would not have to be offset at all, even through such proposals would necessarily be less well targeted on the uninsured population and thus be less effective at reducing the ranks of the uninsured.  As a result, the 75-year pay-as-you-go proposal would favor tax cuts for health insurance costs borne by already-insured high-income taxpayers — who would receive the largest benefits from health-related tax deductions because they are in the top tax brackets — over measures to extend health care coverage through Medicaid to uninsured working-poor families.

Indeed, this proposal for a 75-year pay-as-you-go rule narrowly applied to selected entitlements appears to have a highly ideological nature.  In conjunction with the broader proposal to establish pay-as-you-go rules from which tax cuts would be exempt, this proposal would tilt the “playing field” so that Congress could not consider competing policy proposals in an evenhanded manner.  In the example just noted, proposals to reduce the ranks of the uninsured through an expansion in a health care program would have to overcome formidable obstacles, while proposals ostensibly to address the same problem through tax cuts would not.  This type of imbalance would diminish the ability of policymakers to engage in thoughtful efforts to find the most economical and effective solutions to national problems.

The rest of this analysis examines the proposal in greater detail.


How the Proposal Would Affect Programs Other than Social Security  

When legislation affecting any of a specified group of entitlement programs other than Social Security was considered, the Office of Management and Budget and the Congressional Budget Office would produce an estimate of the effect of the legislation on the “unfunded obligations” of those programs over the next 75 years.  A program’s “unfunded obligation” would be defined as the program’s cost minus any “dedicated” revenue that goes to the program, such as payroll taxes.  For programs that are funded entirely by general revenue and have no dedicated funding sources, the “unfunded obligation” would simply be the program’s cost.[2]

Legislation that would increase the long-term “unfunded obligation” of any of the specified entitlement programs by more than a very small amount — more than 0.01 percent of the Gross Domestic Product over the 75-year period or one percent of program expenditures over that period, whichever is lower[3] — would have to be paid for through offsetting cuts in the same program or in another of the programs that would be subject to this rule (or through offsetting increases in dedicated receipts for the program or another of the covered programs, such as an increase in Medicare payroll taxes).  If legislation did not comply with this rule, it would need 60 votes to pass the Senate.

This essentially is a highly restrictive version of the traditional pay-as-you-go rule that uses a 75-year measuring period (rather than a 10-year period) and mandates that the only permissible “pay-fors” are cuts within the same program or another one of the programs subject to this requirement.  Cuts in entitlements outside the group would not be considered acceptable offsets.  Nor would increases in general revenues.

The affected programs would include:

The Director of OMB would have unilateral authority to add other entitlement programs to the list once the Director determined that OMB could make long-term cost estimates for those programs.  In documents accompanying the proposed legislation, OMB stated that it would add Medicaid to this list of programs in the near future.

Problems with the Proposal for Programs Other than Social Security

The proposal would unduly restrict Congress’ ability to make decisions about budget priorities and to shift funds across a broader array of entitlements in response to changes in need.  Offsets would not be considered valid unless they came from the same program or another of the entitlements subject to this rule.

This proposal, in conjunction with the proposal to apply broad pay-as-you-go rules to entitlements but not to taxes, also would distort policy debates.  Policymakers wishing to examine options to reduce the ranks of the uninsured, for example, would be faced with an unlevel “playing field.”  Once Medicaid was added to the list of programs subject to this rule, proposals to reduce the ranks of the uninsured by broadening Medicaid coverage, such as by extending coverage to low-income working parents, would have to be paid for with cuts made elsewhere in Medicaid itself unless a cut could be instituted in another entitlement within the group of entitlements that would be subject to this requirement.  That would likely foreclose consideration of such an option.  But proposals to reduce the ranks of the uninsured by writing more tax breaks related to health insurance into the tax code would not have to be offset at all; they would be “free” for policymakers, even though they would add to deficits just as unpaid-for entitlement expansions would.  The new budget rules thus would confer an enormous advantage on tax-based approaches to health insurance regardless of their relative efficiency and effectiveness.

Another problem is that 75-year cost estimates in Medicare and Medicaid are too speculative to be used for a purpose such as this.  Cost estimates in Medicare and Medicaid over a 75-year period are much more uncertain than 75-year estimates in Social Security.  Long-term cost estimates for Medicare and Medicaid are dependent not only upon projections of future birth and death rates, as long-term Social Security projections are, but also upon projections of future trends in health care costs.  Health care costs in coming decades will be heavily influenced by medical breakthroughs and advances in medical technology that are impossible to predict today.  Long-term cost projections in Medicare and Medicaid are subject to a much wider range of uncertainty than long-term cost projections in Social Security are.

Federal Reserve Chairman Alan Greenspan made this point in recent Congressional testimony.  He emphasized that long-term Medicare cost projections are much more uncertain than long-range Social Security projections.  “We don’t have such confidence on Medicare [cost projections],” he stated.  “We do not understand the processes that will evolve over time, we cannot anticipate the processes that are going to occur in medical technology and in medical application which gives us any reasonable way to come at what the costs are going to be.”[4] 

Seventy-five year cost estimates in Medicare and Medicaid thus are not sound enough to use as a basis for such a far-reaching budget rule.

A final problem is that the proposal would likely have harsh effects on vulnerable people.  This can be seen by examining the effect of applying the proposal to the Supplemental Security Income program, the nation’s basic cash assistance program for the elderly and disabled poor.  SSI is a parsimonious program; it is largely limited to elderly and disabled people who live in poverty, and the maximum federal SSI benefit for an elderly or disabled individual equals only 76 percent of the poverty line.

As noted above, the SSI program has very low asset limits.  When the program was established in 1974, based on a proposal from President Nixon, the asset limits for the program were set at $2,000 for an individual and $3,000 for a couple.  Thirty years later, the asset limits are still $2,000 and $3,000.  Had the asset limits that Congress established in the early 1970s simply kept pace with inflation, those limits would now be $7,586 for an individual and $11,379 for a couple.

Under the Administration’s proposed budget legislation, these asset limits would likely remain stuck at $2,000 and $3,000 for another 75 years, with the result that fewer impoverished elderly people and people with disabilities would be able to qualify over time.  Any attempt to address this matter — and to try to keep these severe asset limits from falling even further behind inflation — would be subject to a requirement that the cost of such an adjustment over the next 75 years be offset by cuts made elsewhere in SSI, unless sufficient cuts to offset the cost could be made in another entitlement program subject to this rule, which would be unlikely.  The probable result would be either that nothing would ever be done to address this problem, and over time fewer and fewer impoverished people would qualify for SSI, or that the asset limits would be adjusted but SSI benefit levels and eligibility limits would be pushed further below the poverty line to produce the requisite savings.

Such an outcome would not have to occur under the pay-as-you-go rules of the 1990s (or under the pay-as-you-go provision included in the Senate budget resolution this year, which would resurrect the rule in effect in the 1990s).  Under those rules, the resources to pay for an adjustment in the SSI asset limits would not have to come from within the SSI program itself or from a narrow group of entitlements that each have strong constituencies protecting them.  Under the pay-as-you-go rules of the 1990s, offsetting savings could be secured from any entitlement other than Social Security or from any revenue-raising measure, such as a measure to scale back modestly the recent capital gains and dividend tax cuts, which are conferring lavish tax breaks on the wealthiest of the elderly while doing nothing for elderly and disabled people living in poverty.  The proposal in question appears designed to preclude such options from being considered.

More broadly, the proposal would have regressive effects.  General revenues, which are raised primarily through progressive taxes, are currently used to finance 45 percent of all entitlement expenditures, and 77 percent of entitlement expenditures outside Social Security.  Nevertheless, this proposal would demand that any future improvements in the affected entitlements be financed either through entitlement cuts or through increases in regressive taxes such as dedicated payroll taxes.


Social Security Provision Also is Problematic

The proposal also would bar legislation that would increase long-term unfunded obligations in Social Security.  This aspect of the proposal differs in two respects from the part of the proposal that would apply to Medicare, SSI, and other specified entitlements.  First, the costs of Social Security legislation (other than costs defrayed by increases in dedicated revenues) would be estimated not over 75 years but into infinity (i.e., “over an infinite horizon”), and the projected costs into infinity would have to be offset.  Second, the offsets would have to come from Social Security itself; they could not come from any other entitlement.

Given Social Security’s long-term financing shortfall, the idea of erecting a barrier against legislation that would increase the program’s long-term unfunded obligations and requiring that offsets for such legislation come from Social Security itself has merit.  Unlike programs such as veterans disability compensation, SSI, or Medicaid — which are funded by general revenues rather than dedicated revenue streams and to which the concept of “unfunded obligations” is no more applicable than it is to the defense budget — Social Security does have unfunded obligations, and they are substantial.  Moreover, the use of 75-year cost estimates in Social Security is well established and is not subject to as much uncertainty as 75-year cost estimates are for health care programs.  Indeed, a point of order already exists in the House against legislation that would increase Social Security imbalances over 75 years. 

Yet this component of the proposal is designed in a way that makes it seriously flawed as well.  Rather than creating a 60-vote point of order in the Senate against legislation to increase Social Security’s unfunded obligation over the next 75 years, the Administration is proposing a barrier against legislation that would increase Social Security’s unfunded obligations over an infinite horizon.  Use of Social Security cost estimates that seek to extend into infinity would run directly counter to the advice of leading experts in the field.

Making projections of Social Security costs and receipts over an infinite horizon entails making guesses about birth rates, death rates, and productivity growth rates centuries into the future.  In December 2003, the American Academy of Actuaries, the nation’s leading professional organization of actuaries, warned against relying on Social Security projections made over an infinite horizon.

In a letter to the trustees of Social Security system, the American Academy of Actuaries emphasized that infinite-horizon projections in Social Security “provide little if any useful information about the program’s long-range finances and indeed are likely to mislead anyone lacking technical expertise in the demographic, economic, and actuarial aspects of the program’s finances into believing that the program is in far worse financial condition than is actually indicated.”  The Academy also stated that: “Given the uncertainty of projections 75 years into the future, extending the projections into the infinite future can only increase the uncertainty, rendering the results of limited value to policymakers…”  Finally, the Academy warned that “Anomalies and incongruities inevitably arise from extending any set of long-range actuarial assumptions to infinity.”[5]



Action to address projected increases in the costs of Social Security and healthcare entitlements in coming decades will ultimately be necessary.  However, the ideological and unbalanced nature of this proposal — with its exemption of all tax cuts from budget discipline, its indiscriminate favoring of tax-based approaches to national problems over programmatic approaches regardless of their relative efficiency and effectiveness, its severe constraints on the types of “offsets” that would be allowed, and its reliance on cost estimates that often may be unreliable — make the proposal both unsound and inequitable.

End Notes:

[1] For a discussion of the entitlement cap, see other Center on Budget and Policy Priorities analyses, including “Entitlement Cap Proposal Would Require Cuts of $1.8 Trillion Over the Next Ten Years,” March 16, 2004, and “Claims That Proposed Entitlement Cap Would Shield Medicare and Not Cause Massive Cuts Are Incorrect,” June 9, 2004.

[2] Each calculation would be made on a “present-value” basis, to account for the effect of interest earnings or interest costs over long periods of time.

[3] This rule also would apply to legislation whose 75-year cost is below both of these thresholds but whose cost in the final year of the 75-year period (i.e., in the 75th year) exceeds 0.01 percent of projected GDP or one percent of projected program costs for that year.  In addition, in the case of Medicare, this rule also would apply to legislation projected to increase costs by more than 0.02 percent of taxable payroll over 75 years or in the 75th year.

[4] Testimony of Alan Greenspan before the Joint Economic Committee, April 21, 2004.

[5] Letter from Eric J. Klieber, Chairperson, Social Insurance Committee, American Academy of Actuaries, to Trustees of the Social Security System, December 19, 2003.