FROM ISSUE NUMBER 21 ~ FALL 2014 GO TO TABLE OF CONTENTS

TEXT SIZE A A A

Reforming the Budget Process

JAMES C. CAPRETTA

The modern federal budget process is turning 40 years old this year. Created by the Congressional Budget and Impoundment Control Act of 1974, this process was established to address a specific set of concerns facing lawmakers at the time. It has helped address some (though not all) of those concerns, but it is certainly not well suited to helping us deal with the most daunting and significant fiscal problems the federal government now faces. Indeed, the budget process exacerbates those problems and contributes to Washington's larger paralysis and to voters' justified frustration with their elected officials.

The nation is facing serious fiscal challenges and has been for many years, but our elected leaders have done all they can to ignore the key problems, leaving them to fester. The result has been an unprecedented deterioration of the nation's fiscal position. From 1957 to 2008, federal debt never exceeded 50% of GDP. It now exceeds 70%, and the Congressional Budget Office's most recent long-term budget projections indicate it is heading up toward 100% over the next two decades. In the next decade alone — from 2015 to 2024 — the CBO estimates that the federal government will borrow another $7.2 trillion, and even that figure is based on the implausible assumption that spending on national security will fall to levels not seen since before World War II.

The fundamental problem with the nation's finances, and thus the problem our budgetary procedures should be oriented toward solving, is the skyrocketing expense of entitlement programs (known to budget watchers as "mandatory spending"). In 1962, the federal government spent 4.7% of GDP on benefit transfer and other automatic spending programs, such as Social Security, unemployment insurance, and welfare support. By 2013, spending on these mandatory programs had risen to 12.2% of GDP.

The current budget process does not force policymakers to confront fiscal and economic reality. The demographic trends and health-care cost pressures that are driving up entitlement spending have been with us for many years, and the projections showing that these trends will push federal finances past the breaking point have been widely understood for decades. But instead of forcing policymakers to confront the problem, the current budget process shifts financial pressures off of the welfare state and onto other portions of the federal budget. As spending on entitlement programs has soared since the early 1960s, funding for annually appropriated (or "discretionary") accounts, especially defense, fell precipitously — from 12.3% of GDP in 1962 to 7.2% in 2013.

Sooner or later — and probably sooner — the unsustainable nature of the current state of fiscal affairs will force significant change. It would be far better for the country if elected leaders did not wait for a crisis to act but instead confronted the nation's fiscal problems with foresight and planning. A reformed federal budget process might be able to establish conditions conducive to this kind of political leadership.

Changing the process is not a substitute for actual policy changes that will address the nation's budgetary problems, of course. Ideally, elected leaders would skip right to consideration of the long-term reforms necessary to provide a sustainable safety net consistent with strong economic growth. But in the absence of evidence indicating that serious reforms of the welfare state are around the corner, it is worth considering what kinds of changes in the budget process could encourage the adoption of the right policies.

Before we can think about such changes, however, it is important to understand what events and forces shaped the process we have today and why that process has proven inadequate.

THE BUDGET ACT

The Congressional Budget and Impoundment Control Act (generally known as just "the Budget Act") is arguably the most significant piece of budget-process legislation ever enacted in the United States. Passed in the final days of the Nixon administration, it was focused primarily on re-establishing constitutional balance in budgetary decisions. Congress had chafed at President Nixon's assertiveness in a number of areas, but especially with respect to the power of the purse. Nixon pushed against long-established constitutional boundaries by refusing to obligate funds for programs he considered low priorities, even though Congress had explicitly appropriated funds for these purposes. More than any other reason, Congress passed the Budget Act to rein in this practice of "impoundment."

The Constitution stipulates that no funds can be expended by the executive branch absent a clear appropriation by Congress. In the Budget Act, Congress wanted to make explicit in statute what was previously thought to be implied by the Constitution: that if the legislative branch appropriated funds for a designated purpose, then the executive branch had to spend the funds provided, regardless of how the president felt about the usefulness of the expenditure, unless Congress passed a new law explicitly cancelling the appropriation.

The provisions of the Budget Act regulating impoundments quickly resolved the constitutional crisis in Congress's favor, and after President Nixon's resignation on August 9, 1974, impoundments ceased to be considered a subject of national concern. The impoundment provisions are now largely irrelevant to current budget policy, though they were crucial in bringing about the new budget process.

The rest of the Budget Act, which was frankly given less thought in 1974, has proven to be of far more lasting significance. The non-impoundment provisions were also aimed at bolstering the legislative branch's role in spending decisions, especially in relation to the executive branch. But unlike the impoundment provisions, the rest of the Budget Act was focused on budget development, not on the execution of already-appropriated funds.

The most important institutional changes introduced by the Budget Act were the creation of the House and Senate Budget Committees and the Congressional Budget Office. The Budget Committees were charged with writing a budget plan for Congress — the annual congressional budget resolution — that would serve as a counter, or a response, to the president's annual budget submission. Crucially, the budget resolutions written by the Budget Committees would be based on projections and cost estimates produced by the independent and nonpartisan CBO, which, unlike the executive branch's Office of Management and Budget, could not be influenced by the policy preferences of the party in power.

The congressional budget resolution provides allocations of spending and taxing authority to the various congressional committees with jurisdiction over spending and tax laws. For instance, the resolution sets the maximum amount of discretionary budget authority — called the "302(a) allocation" after the section in the Budget Act designating its inclusion in budget resolutions — that the House and Senate Appropriations Committees are permitted to spend on the various bills that those committees produce each year. If one of the committees writes an appropriations bill that causes total spending within its jurisdiction to exceed its allocation, it commits a violation of the budget rules and puts the bill in jeopardy. Members of the House and Senate can raise objections to the further consideration of such bills. This mechanism is especially important in the Senate, where it typically takes 60 Senators to continue debating a bill that violates an allocation provided under the congressional budget resolution.

THE ELEPHANT IN THE ROOM

This system of budgetary allocations and enforcement operates differently for entitlement programs and taxes than for discretionary spending. The committees with jurisdiction over major entitlement programs, like Medicare and Medicaid, receive an allocation that can only be understood relative to a "baseline" of expected costs. These programs run based on complex formula provisions defining what will be paid, under what circumstances, to deliver the benefits promised in the statute. In practical terms, CBO looks at what the law requires, historical trends, and other factors such as shifting demographics to provide estimates of what the spending on a given entitlement program will be over the coming decade. A similar approach is used to provide a current-law tax baseline.

These baseline estimates for mandatory spending programs and taxes are determined by CBO, along with the Joint Committee on Taxation, and are absolutely critical to the construction of the congressional budget resolution. If the Budget Committees choose not to change mandatory spending or taxes, they allocate the baseline amounts to the respective committees with jurisdiction over these programs (the "authorizing committees"). If the allocations are equal to the baseline, then the authorizing committees have no budgetary room to increase spending on these programs (or decrease revenues), but they are also not forced to make any changes in these programs to cut expenses or raise taxes. In short, they could comply with the budget by doing nothing. (They could also approve budget-neutral bills, meaning that spending increases for one program would be offset with spending reductions in another; likewise, tax cuts would be offset with tax increases.)

The Budget Committees also have the authority to provide spending and tax allocations that differ from CBO's baseline projections. Such allocations are often accompanied by "reconciliation instructions," which direct the committees to produce new authorizing legislation that will "reconcile" the programs within their jurisdiction with the allocations provided to them under the congressional budget resolution. For instance, a resolution could attempt to achieve $100 billion in Medicare savings over a decade by cutting the Medicare allocation to the three committees with jurisdiction over the program (the Ways and Means and Energy and Commerce Committees in the House, and Finance in the Senate). With reconciliation, the committees are expected to produce legislation by a certain date that brings spending into line with what is provided in the congressional budget resolution. If multiple committees are part of a reconciliation instruction, their bills are pulled together into an "omnibus bill" for consideration in the House and Senate.

The reconciliation process has played a very large role in the nation's recent economic history, in large part because reconciliation bills enjoy privileged status in the Senate. Debate on them is limited, meaning they cannot be filibustered and therefore can pass with a simple majority rather than the 60 votes often necessary to get other major legislation through the chamber. These reconciliation rules have paved the way for some of the most significant pieces of legislation signed into law during the past 35 years — including President Reagan's first round of spending cuts in 1981; the budget deals of 1990, 1993, and 1997; the Welfare Reform Act of 1996; and President Bush's tax cuts in 2001 and 2003.

The reconciliation procedures provide a current-law roadmap for pursuing entitlement and tax changes in Congress under expedited rules, but it is important to understand how different the budgetary restraints on entitlement programs are from those on discretionary spending. For appropriated accounts like defense, the National Institutes of Health, or the National Park Service, the budget resolution can establish a hard upper limit on the total amount of appropriated spending, and this upper limit is rather easily enforced (if the will to do so exists in Congress). An enforceable budget for entitlement spending is far more elusive because it is based entirely on estimates, both for baseline projections and assessments of what new legislation might do to the baseline forecast. In effect, entitlement spending is never held to a firm budget. If spending rises in the programs because of higher enrollment than expected, or higher average benefits than expected per enrollee, nothing in the current budget process can force Congress to enact corrective steps to limit spending. The money simply comes out of the treasury to cover the costs, and the treasury often has to borrow the necessary funds. These programs enjoy wide popular support, so Congress is generally reluctant to open them up for amendment anyway. But the budget process makes it very easy for elected leaders to do nothing and allow entitlement spending to rise "naturally" as more people seek benefits, and as the rules governing benefits grow more lenient based on political and interest-group pressures.

Although inflation can also be built into projections for discretionary accounts, it has more often been the case that the default option for appropriated accounts is a freeze or a near-freeze in spending in the budget resolution, or adherence to upper limits previously agreed upon by the Congress and the president. In successive rounds of bipartisan budget deals going back to at least the 1990 budget agreement, both major political parties have found it much easier to apply spending restraint (even if it is modest) to the discretionary accounts of government rather than entitlement programs. In effect, as budgetary pressures have risen with the growth of entitlement spending since the 1970s, successive Congresses and presidents have found the path of least resistance for budgetary restraint to be placing ever tighter caps on annually appropriated spending programs. This partially explains why spending on these accounts is now at 6.8% of GDP, down from 10.5% in 1972, while entitlement spending continues to grow.

This inclination to restrain appropriated accounts rather than entitlement programs is also related to the mismatch between the time frames typically contemplated in budgeting and those that are necessary to really address the nation's fiscal imbalances. The fundamental problem is not that the United States may, in any given year, have to borrow a large amount of money due to a temporarily high deficit. The problem is that rapid growth in entitlement spending has created the prospect of a chronic imbalance that can be corrected only with structural reforms of the programs in question — or ever higher rates of taxation. The current budget process is ill-equipped to help policymakers see the real problem or do anything about it.

The Budget Act requires Congress to write budget plans addressing only the upcoming fiscal year and the four that follow it. In recent years, both the president and Congress have tended to write budgets that cover ten years rather than the required five. But not even a ten-year budget captures the information necessary to see the fundamental fiscal problem or the positive results that could be produced from genuine structural reforms, which may take years to implement.

It is telling that one of the most consequential fiscal corrections of recent decades was enacted entirely apart from the regular budget process. In 1983, President Reagan and Congress agreed on a long-term framework to close the financing shortfall in Social Security. Among other things, the reform increased the age at which beneficiaries could start drawing full benefits without an early-retirement reduction. This reform was phased in very slowly, starting in 2000, an amazing 17 years after the law was enacted. The two-year bump in the normal retirement age — from age 65 to 67 — will not be fully phased in until 2027, more than four decades after the law was enacted. The normal budget process therefore showed no benefits to these reforms, even though its long-term effect is significant.

Most other significant entitlement reforms won't require a 40-year phase-in schedule. Indeed, our largest entitlement programs will be in the red long before then. Nonetheless, because of the need to give plenty of notice to beneficiaries about upcoming changes in program rules, it is often not possible to achieve significant savings from serious reforms within the typical ten-year window of today's budget process. Moreover, some reforms require complex implementation adjustments, which also can mean years will pass before the most significant budgetary savings become visible.

The resulting budgetary myopia provides a strong bias against reforms like "premium support" in the Medicare program. Premium support would substantially modify Medicare's financing system by harnessing the power of competition and consumer choice to drive quality and efficiency gains in the program. Private insurance plans and the traditional government-administered Medicare benefit would compete against one another by submitting bids for how much they would charge to provide Medicare-covered services. Beneficiaries would get their entitlement in the form of financial support — the amount of which would be based on the bids — that they would then apply to the insurance plan of their choosing.

Restructuring Medicare in this way has the potential to transform the program and greatly improve its long-term financial outlook. The Congressional Budget Office has found that such a reform could yield major savings in Medicare spending without increasing costs for beneficiaries. But it is a complex reform that would require some time to implement and show its effects. Moreover, to avoid disrupting the insurance arrangements of current beneficiaries, most proposals to move in this direction provide for a lengthy transition period. The result is that premium support — arguably among the most important steps that could be taken to improve the nation's fiscal position — would produce very little savings in the normal ten-year budget window.

TOWARD A FUNCTIONAL BUDGET PROCESS

Today's budget process, then, tends to contribute to rather than mitigate the federal government's budget problems. But, in the absence of the will to transform our entitlement system, an improved budget process offers the best opportunity to build some political momentum for real reform. Several key reforms to the budget process could both make a significant difference and actually stand a chance of enactment even amidst today's partisan divisions.

The first reform would seek to change the way the elected branches interact in the budget process. A key characteristic of a congressional budget resolution is that it is not a law. It is a concurrent resolution, which means it is only relevant for Congress. Presidents do not have to sign budget resolutions, they are in no way bound by them, and in fact they have often denounced them as containing misplaced priorities — just as opposition parties in Congress regularly criticize the president's budget submissions.

The parallel budgetary processes of the executive and legislative branches are a reflection of our constitutional structure. As co-equal branches of government, each has a substantial role over the federal budget, and there is no legal requirement that they ever fully come to an agreement with each other. Indeed, with some exceptions, it can be said that the federal government never truly operates within a budget because the legislative and executive branches rarely agree on one.

This structure has practical consequences. It is possible — in fact it frequently happens — that Congress will proceed based on one budgetary framework, defined by a congressional budget resolution, that differs substantially from the framework the president supports. The differences between the two branches are addressed, if at all, only when actual spending or tax legislation that is a product of the budget-resolution framework makes its way all the way through the legislative process and reaches the president's desk. At that point, the president can sign or veto the legislation. Usually, if there is an ongoing disagreement, the anticipation of a veto is enough to bring the entire process to a standstill. This dynamic is an important reason why there are regular, drawn out budget fights between Congress and the president as the fiscal year draws to a close each September 30. The two branches spend most of the year working from different budgetary plans, and then they have to scramble at the last minute to arrive at an ad hoc resolution to the disagreement so that the government can continue operating.

Even if the two branches can cobble together a temporary solution, however, the situation is very different from having in place a budget framework that lasts multiple years and provides financial structure and stability to government finances. The current process does not apply any countervailing pressure to offset the institutional and political tendencies toward budget stalemate that are built directly into our constitutional order.

The idea of joint budget resolutions offers a possible, partial antidote for such budgetary drift, rising entitlement spending, and the endless inertia in our federal budgeting practices. Unlike a congressional budget resolution, a joint resolution must be agreed to by the president — and therefore would be a law. It thus has the potential to facilitate, and perhaps even pressure, the legislative and executive branches into coming to an agreement early in the process on key budgetary aggregates that would govern later decisions by both branches.

There are several ways to facilitate the consideration of joint budget resolutions. The most straightforward option would be to build upon the current process by amending the current Budget Act rules to allow an optional joint-resolution "spin-off" from any congressional budget resolution that was agreed to by both the House and Senate. Congress would not have to pursue a joint resolution, but if it did so then legislation would automatically be sent to the president upon adoption of the congressional budget resolution. The joint resolution would reflect the key budgetary aggregates agreed upon: total discretionary spending, total mandatory spending, revenues, deficits, and debt. The president could then either approve or veto the legislation.

If the president vetoed the joint budget resolution, the process would revert back to the same one in place today under the Budget Act. Congress could proceed under the terms of the budget resolution, and engagement with the executive branch would be postponed until later in the year, when the spending and tax bills flowing from that budget were transmitted to the president. If, however, the president agreed to the joint budget resolution and signed it into law, the budget framework contained within it would have the force of law, and both branches would be bound by it.

It is critical that a joint budget resolution have the capacity to place an upper limit on mandatory program spending, along with caps on discretionary spending and a floor for revenues. This would ensure that Congress and the president truly engage in budgetary decision-making. There would be clear tradeoffs between the key budget categories: Congress and the president could choose to put more pressure on mandatory spending programs and thus perhaps ease the pressure on discretionary accounts, or vice versa. In addition, proposals to cut deficit spending with tax hikes would be clearly identified in the budget plan.

A joint budget resolution would not make the political disagreements that often divide the two branches of government disappear, so this revised process is not a guarantee of budgetary stability. But neither Congress nor the president wants to be seen as fiscally irresponsible, so there would be some political pressure to participate honestly in coming to an agreement on a budget framework. A joint resolution would also force the two branches to begin coordinating early in the process by reconciling the differing estimates from CBO and OMB in order to come up with an agreeable joint plan.

PROMOTING ENFORCEMENT

The joint budget resolution concept would work only if there were a sensible enforcement system attached to it. The purpose of establishing a budget framework through a joint budget resolution would be to encourage additional legislation that brings programs and taxes into line with the agreed-upon budget totals. In other words, the joint resolution would be a catalyst for a genuine reform agenda, because inaction would no longer be the consequence-free path of least resistance. Presumably, large changes in entitlement spending and taxes contained in a joint budget resolution would be assigned to the authorizing committees in the form of reconciliation instructions. This would allow fast-track consideration of the reforms implied in the resolution's top-line numbers.

But Congress will feel the pressure to act on tough legislative reforms only if the joint budget resolution's limits are enforced even without additional legislation. That means making the caps on spending — including entitlement spending — binding once the joint resolution is agreed to by Congress and the president.

Over the past three decades, the two branches have settled on an acceptable and well-understood process for establishing caps on discretionary spending. These caps, first adopted in the 1990 budget agreement, place a hard upper limit on spending for programs with annual, discretionary appropriations. If the caps are breached, the executive branch is required to impose across-the-board cuts (called a "sequester") on essentially all programs covered by the caps to ensure that total spending stays within the agreed-upon amount. These caps have been effective at controlling discretionary spending (although "emergency" designations have served as an end-run around the caps in certain years). The enforcement of a joint budget resolution would employ such caps, but it would also need a new corollary enforcement system for mandatory spending and tax collection.

Unfortunately, the existing sequester mechanism for mandatory programs, enacted in the Budget Control Act of 2011, is woefully inadequate. For starters, too much mandatory spending is exempt from the sequester altogether, including the entirety of Social Security and Medicaid. Further, while Medicare is subject to across-the-board cuts (up to 2% of program spending each year), the cuts are irrationally designed and exacerbate some of the worst features of the current program. Medicare is a heavily price-regulated program, and the cuts simply lower the prices paid to all providers of services to Medicare's beneficiaries. Instead of promoting real reform of the program through competition and market incentives, these cuts further cement the failed approach of cutting Medicare costs with ever more stringent payment regulation. The result is not cost-efficient health care but more distortions in the marketplace.

A better enforcement mechanism on mandatory programs would not be a substitute for reform of the programs, but neither would it reflect a bias toward protection of the status quo. In allowing for a joint budget resolution, the budget process should also be amended so that the sequester of mandatory spending reflects five key considerations.

First, some programs for very low income Americans, such as Supplemental Security Income, should be exempted from the cuts. But it is not unreasonable to subject some other income-support programs to the enforcement of spending control. For instance, if spending breached the upper limit, food-stamp eligibility might be lowered modestly for the highest-income participants. Similar adjustments could be made in other programs.

Second, Medicaid should be explicitly included in the enforcement mechanism. The federal matching payments to the states should be reduced commensurate with the reduction in spending necessary to stay within the mandatory-spending cap. States will rightfully complain that this will burden their budgets, so they would need to be given relief from existing federal Medicaid mandates to provide them with the budgetary flexibility necessary to accommodate the federal cut.

Third, the automatic cuts in Medicare should be designed to promote reform rather than hinder it, which means promoting more cost-conscious consumption of services. For starters, higher-income beneficiaries should be required to pay more for their services. Further, all beneficiaries should be required to pay something when they get care. Achieving this will require adjustments to the rules for supplemental insurance plans (known as Medigap plans) and providing pro-rata adjustments in deductible and co-insurance payments.

Fourth, the sequester process should be recalibrated periodically so that actual spending is brought in line with the joint-budget-resolution levels based on revised estimates. In addition, the sequester should leave room to allow the necessary savings to be achieved over a period of, say, five years, to avoid abrupt year-to-year adjustments.

Finally, if mandatory spending was expected to come in below the joint-budget-resolution cap, there would be no automatic process to increase spending. Congress would need to pass new legislation to make the desired changes and achieve the increased spending called for in the budget plan.

Opponents of this approach will argue that designing the enforcement mechanism this way would tilt the playing field toward reforms favored more on the right than on the left, thus giving Republicans more leverage in budget battles. But that is unlikely to be true. These automatic cuts would be indiscriminate and blunt, and they would not constitute the kind of fundamental entitlement reform that most conservative-leaning politicians favor. Cutting entitlements with a sequester mechanism is unlikely to satisfy anyone — their inadequacy as actual budget policy is part of their purpose.

A more difficult problem for conservatives will be how to handle enforcement of the tax numbers in a joint budget resolution. It will be difficult to reach broad agreement on the joint-budget-resolution concept if strict enforcement is applied only to the spending side but not to taxes. Still, applying a parallel enforcement mechanism to tax collection should be strongly resisted. For starters, the long-term fiscal problem is a spending problem, not a tax problem. Entitlement spending will push federal expenditures well above the historical rate in the coming years, according to CBO. Revenue, on the other hand, is already expected to exceed the post-World War II norm due to tax hikes enacted in recent years.

In addition, revenue estimation is so imprecise that small misjudgments in assumptions can produce big swings in actual tax collection compared to projections. Automatic enforcement of tax levels could thus produce very large pro-rata tax hikes. Entitlement spending can be volatile too, but, unlike revenues, the total is divided up among a large number of programs, so the likelihood of a systematic misjudgment is much reduced.

Addressing the enforcement of the tax side of a joint budget resolution could, for instance, involve triggering an expedited process for revenue-increasing tax reforms or an automatic delay of any forthcoming tax cuts in current law. But it would not be reasonable to allow for automatic tax increases as part of the budget process.

LONGER-TERM BUDGETING

While a joint budget resolution with sensible enforcement provisions would be a significant improvement over the current process, it would not, by itself, solve the problem of inattention to the longer-term budget picture. An important first step toward addressing that problem would be to agree on how to measure the long-term fiscal obligations of the government. That information could then be used to inform current legislative debates and, eventually, to spur legislative action.

One approach would be to establish a uniform definition of the federal government's long-term unfunded liabilities. In general, this would mean calculating the present value of expected future tax receipts and expected future spending commitments. The INFORM Act, promoted by economist Laurence Kotlikoff, is one version of this kind of approach. Long-term forecasts are already done annually for both the Social Security and Medicare programs, and so a large portion of the federal budget can be assessed on this basis with current information (the latest projections indicate the combined unfunded liabilities for these two programs amount to over $40 trillion). These forecasts could be supplemented with assessments of other parts of the budget, especially those that create long-term obligations on behalf of the government, such as pension programs and guarantees.

Some disputes will inevitably arise over what constitutes an "unfunded" federal commitment. For instance, one part of Medicare is financed mainly from the government's general revenues and not from a dedicated payroll tax. But drawing on "general revenues" can mean running a deficit and borrowing the funds from the public to pay the bills. In most cases, references to Medicare's unfunded liability omit "general revenue" contributions and look solely at revenue generated from taxes and premiums compared to the program's expected spending. Similar judgments would need to be made for other programs.

Once a common measure was established, it could be used to assess legislation in Congress and perhaps be incorporated into a joint budget resolution. For instance, the budget process could be amended to require CBO assessments of how significant new budget legislation would alter the unfunded-liabilities calculation. Bills that would increase those liabilities could be ruled out of order and thus pulled from consideration unless a supermajority of members chose to override the rule.

In addition, a process could be established to "reconcile" spending and revenue proposals by the committees with jurisdiction over the major programs with large unfunded liabilities. The reconciliation tag would give the proposals privileged status, expediting corrective legislation to narrow the fiscal gap. This kind of legislation would not need to show deficit reduction over a ten-year period but rather a reduction in the government's unfunded liabilities measured over 50 or 75 years.

FIXING THE SYSTEM

Together, the joint-budget-resolution concept and a new emphasis on long-term challenges would constitute serious reform of the existing budget process. But there are problems with the existing approach that would not be addressed directly by these reforms. A few additional changes to address these glaring problems should also be a part of the budget-process reform agenda.

First, Congress should repeal pay-as-you-go budgeting. During most of the period since 1990, Congresses and presidents have operated under a budget procedure known as "pay as you go" (or "PAYGO"), which is distinct from the Budget Act process. Prior to 2010, PAYGO was a temporary provision, with a sequester enforcement mechanism similar to the 2011 Budget Control Act model. It was also part of the congressional budget procedures and governed the consideration of tax and spending legislation. In 2010, Congress passed a permanent PAYGO law into statute.

The PAYGO concept is designed to force new entitlement or tax legislation to be "budget neutral." That is, any expansion in entitlement spending, and any tax cuts, must be accompanied by other tax and entitlement changes to ensure that the legislation does not result in an increase in the deficit over some set number of years.

The real aim of PAYGO, however, is to prevent tax cuts. PAYGO does not place any constraint on the "natural" (and inexorable) growth of entitlement spending that occurs under current law. It does, however, put a big hurdle in the way of tax cuts that might be promoted as part of a pro-growth economic agenda. For instance, President Reagan's 1981 tax cuts would have had to clear a much higher political hurdle if PAYGO had existed at the time.

PAYGO is the embodiment of the view that fiscal responsibility entails "paying for" newly enacted spending commitments. That's very different from the view that sound fiscal policy is focused on spending control to allow private actors to keep and use as much of their own money as possible. The ideological bias that PAYGO represents should be eliminated from the budget process.

Second, Congress should allow expedited consideration of Social Security reforms. Social Security is treated as an "off-budget" program because of its self-financed structure. In theory, the program is paid for by payroll-tax contributions, so Congress believes that the program should be addressed on its own terms and not based on its implications for overall government finances.

That sentiment (however questionable) can be preserved while still allowing Social Security reform to proceed more readily. Under current rules, Social Security changes cannot be included in reconciliation measures. This means that Social Security, which has an unfunded liability totaling $10.6 trillion over the next 75 years, gets less favorable consideration under the budget rules than do small changes to much less financially significant programs.

It is possible to preserve Social Security's off-budget status while still allowing serious reform to advance under procedures similar to reconciliation. One approach would be to provide expedited consideration of Social Security legislation that achieved the elimination of the program's unfunded liability, as certified by CBO. Under such an approach, Social Security reform legislation that brought long-term financial stability to the program would be considered in the Senate under terms similar to reconciliation legislation. The only difference would be that the Social Security reform legislation would be required to be considered as a stand-alone bill, and the savings generated by it could not be used to meet deficit-reduction goals established in a congressional budget resolution or a joint budget resolution.

Third, Congress should provide for the automatic continuation of government operations. Since 1977, there have been only four years in which all 12 appropriations bills have been passed by the start of the fiscal year. Instead, continuing resolutions (or CRs) are passed to temporarily fund programs requiring an annual appropriation. On occasion, when Congress and the president could not agree on individual spending bills, a CR has been used to fund programs for a full fiscal year (this happened in fiscal year 2007 for nine bills and in fiscal year 2011 for 11 bills).

The prospect of an impending government shutdown has been used in recent years to stoke partisan tensions and extract concessions that otherwise might not have been provided in the budget process. It is apparent from recent budget battles that the prospect of a shutdown generally works in favor of those opposed to applying additional spending discipline. A better approach to these situations would be to provide for automatic continuation of government services, but under rules that still provide both branches of government with an incentive to reach agreement on full-year funding levels.

Senator Rob Portman, a Republican from Ohio, has introduced S.29, the "End Government Shutdowns Act," in this Congress to provide for this kind of "automatic CR." If no appropriation measure or CR for a set of government agencies were in place by the start of the fiscal year, S.29 would provide automatic funding until a full-year appropriation bill was approved. For the first 120 days of the fiscal year, programs governed by S.29 would receive 100% of the previous fiscal year's funding. For each subsequent 90-day period, the funding would decrease by one percentage point.

Recent budget showdowns have made it apparent that government agencies will eventually get funded, and that shutting down operations for a time yields very little leverage to those pushing for more significant change. It would be better to ensure that the government remains open with an automatic CR but at a level that is not too comfortable for the executive branch and its agencies. That would ensure continued pressure for agreement on a larger, full-year budget plan.

A BETTER BUDGET PROCESS

The United States is entering a new era of economic and fiscal policy. Rapid demographic changes and rising entitlement costs are creating significant pressures that must be addressed soon to avoid the risk of substantial economic dislocation. Reform is coming for the welfare state, in one form or another.

Such reform is exceedingly difficult to enact thanks to the natural tendency of elected leaders in advanced democracies to protect incumbent benefit programs at the expense of longer-term economic and future generations of voters. It is too easy for our politicians to put off political pain today by increasing economic pain tomorrow. And the sharing of budgetary responsibility between two co-equal branches of government makes large-scale change particularly difficult to achieve in the United States.

These are realities that a budget process cannot, by itself, ever fully overcome. If a majority of elected leaders are determined to avoid significant changes to the welfare state at all costs, then it is likely that only a large-scale crisis of some sort will force the needed course correction. Ironically, a carefully considered reform plan, put in place before an emergency is upon us, is more likely to preserve and strengthen the nation's safety net than a plan enacted in the haste of an impending economic crisis.

The task is difficult, but not hopeless. In recent years, a number of political leaders have stepped forward to offer serious reform plans for the nation's entitlement programs that would begin to address the fundamental problems in the federal budget. A better budget process could help these leaders by lowering some of the political and legislative obstacles to reform, even if it cannot remove them altogether.

James C. Capretta is a senior fellow at the Ethics and Public Policy Center and a visiting fellow at the American Enterprise Institute. He wishes to thank Sam Sussman for research assistance. This essay is derived from a research project commissioned by the Mercatus Center at George Mason University.