One
Madisonian Budgeting, or Why the Process is so Complicated

The United States government's budget process is unique in its complexity. No other nation has a legislature so strong it actually dominates spending and taxing decisions, though, of course, without actually eliminating the important part played by the president and the executive branch. Congress itself is notoriously fragmented, partly by constitutional design and partly because its members and their constituents like it that way. Because majorities within Congress shift from election to election and program to program, winning coalitions often differ from month to month and issue to issue. Thus, the president's budget proposal—really his asking price—varies from being the center of negotiations to "dead on arrival." And Congress is quite capable of engaging in a series of exhausting votes that do not resolve matters.

There is more. In recent years, Congress has made many changes in budget procedures, partly to make up for its failure to follow past procedures. But nothing has been thrown out. Thus congressional supremacy coexists with the 1921 measures giving the president a bit more to say and the 1974 measures taking some of that back and trying to centralize consideration of the budget within Congress. Budget committees are loaded on top of appropriations and revenue committees; budget resolutions whose rules differ in each house are superimposed on the regular legislative processes. If one multiplies the number of independent actors times the number of different procedures times the variety of issues, bringing forth different alliances, no other national process can match the American for the sheer volume of considerations that must be taken into account.

We call this budget process Madisonian because it is designed not to secure efficiency but to prevent the abuse of power. Of course, we could

argue that, because contemporary budgeting has not been designed by anyone, our ability to govern, not merely to prevent others from governing, is at stake. However one looks at this matter in the light of disputes over the deficit, the path of change leads through the same budget labyrinth. If you don't know the rules, you can't play the most important political game in America.

"Budget" is both a noun and a verb, both a thing and an action. According to one dictionary, to budget is "to determine in advance the expenditure of (time, money, etc.) over a period of time." We would add to this the common understanding that budgeting also involves assessing resources and relating them to expenditure. By another definition, the budget as a noun is a "comprehensive work plan," projecting the activities of an individual or organization.

The federal government does considerable budgeting, which means making commitments about future activity. Most of these commitments are in the form of either laws or agreements behind the laws between the executive and Congress. These are promises to society that government will do something: pay for medical care for the poor, buy some number of jet fighters, or employ so many people to test new pharmaceutical products.

The U.S. government does not, however, consider all these matters at once, creating a comprehensive work plan. The president's budget, issued with great fanfare each January, is indeed comprehensive. Yet it combines proposals for new commitments and estimates of old ones; it makes no commitments of its own. The obligations of the U.S. government are those mandated or permitted by law, and Congress passes no comprehensive budget law.

Instead, Congress (and the president by signing) makes commitments in a great many ways, at a great many times. "The budget" is not one decision but many.

Types of Budget Commitments

Laws that commit the government to spend money create budget authority . Budget authority (BA) is just what it sounds like: authority granted to some agent of the government to spend money. The money spent is called an outlay . Outlays cannot be made without budget authority; but some small amounts of BA may never be spent if the government buys something more cheaply than anticipated.

In a given year some appropriated funds may remain unspent. The acts provide budget authority, but each year's outlays (i.e., actual spending) combine this year's and previous years' authority. For example, an

appropriation may allow the Urban Mass Transportation Administration (UMTA) to commit $200 million for a rapid transit extension in Chicago. This BA allows UMTA to enter into an obligation (i.e., contract) to spend that money. The money actually will be outlaid (spent) over a period of years as the extension is built and material, labor, and design are paid for. In fiscal 1980, the year before our story begins, 17.9 percent of federal outlays were based on such prior year contracts and obligations;^[1] a similar percentage of the budget authority obligated for fiscal 1980 would not be spent until later years.

This difference between budget authority and outlays is the primary source of confusion for people who follow federal budgeting. Congress votes on BA, but each year's spending—and thus fiscal policy and the deficit—depends on outlays. Congress has some idea how much outlay will result from its votes on BA, but the estimates can be controversial.

Budget authority itself takes a variety of forms, related mostly to the kind of activity being authorized. The major distinction is between annual and permanent appropriations.

Annual appropriations are enacted in the yearly appropriations acts, which allow agencies (e.g., the FBI or National Institutes of Health) to spend or contract to spend specific amounts of money. Annual appropriations acts are drafted and managed by special appropriations committees in the House and the Senate. Each committee in each house has the same thirteen subcommittees and identical jurisdiction over a group of federal agencies. Each subcommittee is supposed to produce a bill for its jurisdiction every year; thus there are thirteen annual appropriations acts.

Permanent appropriations generally are made by Congress in other legislation; the most important is the Social Security Act of 1935 with its numerous amendments. Almost all these are entitlements; the law says that a person or group is "entitled" to some payment if certain conditions are met.

Entitlements do not specify spending totals. Total spending under these programs is simply the sum of legislatively mandated payments applied for by recipients. Totals are not only not directly chosen, but they can also be known only in retrospect. One cannot know in advance either the number of unemployed or their previous base earnings, and therefore one cannot know the cost of unemployment insurance. Finally, while some entitlements are formally appropriated (for example, for food stamps, each year appropriations must be made so programs can draw funds from the Treasury), the government's obligations are created in the authorizing law. The appropriations committees cannot erase those obligations.

The differences between appropriation and entitlement spending mean that legal authority for spending is the product of decisions made by different committees at different times. Entitlements have been adopted and amended separately over the years. When one considers that over half the outlays comes from entitlements and that each year's appropriations create outlays over a period of years, it should be no surprise that only 27.3 percent of the outlays in the government's 1980 fiscal year resulted from that year's appropriations process.

What is true of spending is more true of taxing: tax law is an accretion of years of decisions. Like entitlement legislation, tax law is open-ended: individuals are obligated to contribute according to some criteria; the government is not guaranteed some specific sum of revenue. Revenue may be influenced and estimated but not decreed.

Tax legislation is considered by the House Committee on Ways and Means and the Senate Committee on Finance. These committees also control the many entitlement programs, such as unemployment compensation and the massive Old Age, Survivors, Disability and Health Insurance (OASDHI), that have special taxes to finance their benefits. OASDHI includes the old-age pensions we normally call social security, disability pensions, and medicare. Sometimes that system's taxes (FICA on your paycheck) are described as contributions earmarked for trust funds. They are still taxes: if you meet the criteria for paying them, do not pay them, and are caught, you may go to jail.

The tax committees have another type of jurisdiction that resembles spending in that it may allocate benefits among people and groups for social purposes: tax preferences . These provisions of the law can reduce a taxpayer's liability to the government so long as that person performs some act the government wishes to encourage. Tax preferences that benefit individuals greatly can yet be justified in wider social terms: the tax deduction for mortgage interest, for example, encourages both the construction industry and the social goal of widespread individual home ownership. Those who do not like a tax preference call it a loophole; the metaphor, suggesting money escaping or being diverted from its intended use, is misleading. Such loopholes, passed by Congress, are as intended as any other legislation.

Tax preferences and entitlements are similar in that people make commitments because of government policy—to invest under certain depreciation rules, to retire at a certain age given social security. Because people take action based on these promises, politicians are particularly reluctant to break them. The best-known entitlements and tax preferences are promises to huge numbers of people: the elderly (social security) or homeowners (the mortgage interest deduction). Both policies also increased substantially in the post-World War II era. On the spending

side, which received the most attention, entitlements grew in part from keeping old promises; as more and more people reached retirement age, the promised social security or civil service retirement pensions cost the government more. Entitlements also grew from making new promises: medical care for the aged (medicare), nutrition for the impoverished (food stamps), or increases in the benefits of big, old programs (social security).

Entitlement spending rose from 35.2 percent of federal outlays in fiscal 1967 to 53.6 percent in 1974 to 55.7 percent in 1980.^[2] Many commentators view this increase as an end-run by authorizing committees (spenders) around appropriations committees (guardians of the public purse). Yet the major entitlements would fit poorly into a system of annual appropriations. Medical care for the aged, for example, could be provided by annually funded government-run hospitals, but that is called socialized medicine, not considered politically feasible. Once the decision is made to reimburse patients for costs incurred in the private sector, spending cannot be planned in advance. In short, there are legitimate policy reasons for entitlement funding—efficiency or the political difficulty of alternatives or the desire to keep promises.

These good reasons should not obscure the consequences for government. First, more and more spending is not controlled by the institutions created for annual review of spending, the appropriations committees. Second, entitlement spending is subject to the vagaries of the economy; in providing certainty to the recipients, the government takes uncertainty upon itself. Third, the major entitlements are very difficult to cut.

A Madisonian Budget System

All modern governments are constrained in their budgeting by similar problems of budget composition. Most, however, have far simpler political systems. The main story of this book, the battle for control between Congress and the president, would make no sense in other nations. In England or Germany or Japan or Sweden the executive grows out of and dominates the legislature; an executive cannot exist without a legislative majority. If no majority exists, the government loses a vote of confidence, and a new election is held. In the United States the executive and the legislature may remain locked in bitter combat for years, each unable to remove the other.

This separation of the executive and the legislature is one of two distinguishing and, for budgeting, crucial aspects of the political system created by the Constitution. It is not just a structure but a value: we are taught in school James Madison's arguments that "separation of powers"

and "checks and balances" prevent tyrannical government and protect minorities. What is less obvious may be even more important: a system of checks and balances means that the legislature is as strong as the executive.

In other countries the executive assembles a budget proposal, which is difficult enough, and presents it to the legislature, expecting it to pass, as designed, in one bill. There may be some small changes at the margins, but the legislature is in no position to extensively revise the executive's plan. After all, the prime minister leads the majority party or coalition. Even in American state governments, however, the legislature is much more active than elsewhere; and Congress is, beyond doubt, the most powerful legislature of all.^[3]

Congress's unique role in budgeting is established by the Constitution, which states: "No Money may be drawn from the Treasury but in Consequence of Appropriations made by Law" (Article 1, Section 9). The president can only spend money if Congress lets him. This power of the purse was described by Madison, extrapolating from Parliament's battles with the king, as "the most complete and effectual weapon with which any constitution can arm the immediate representatives of the people, for obtaining a redress of every grievance, and for carrying into effect every just and salutary measure."^[4] Congress may not feel its power of the purse is quite so effective, but if members of Congress as a group can ever be said to agree on anything it is that their power of the purse must be retained if they are to maintain their independence.

That means power over little as well as big things. Editorialists—and presidents—continually criticize Congress for attention to details that are beneath the dignity of a national legislature. In his 1988 State of the Union message, President Reagan cited "such items as cranberry research, blueberry research, the study of crawfish and the commercialization of wild flowers." From Congress's perspective, however, the issue is not whether there will be programs with local "pork-barrel" benefits. The Department of Agriculture is going to do research on something, somewhere .^[5] The issue is who will decide which localities benefit. If the president decides, he has a substantial weapon to reward and punish legislators; if he can control the members, he can control Congress.^[6]

We have, then, a Madisonian budget system, based on ambition opposing ambition, as part of our Madisonian government. After two hundred years we still have a powerful legislature, divided internally and checked externally. If anything, the system of checks and balances has grown more elaborate, both internally through the committee system and externally as Congress has given the president powers to do things Congress feels necessary but difficult for itself.

The Executive Budget

The president's budget is issued at the beginning of each year, shortly after the president's State of the Union address.^[7] To this writing, it remains the only document that resembles our commonsense notion of a budget: a detailed summary of anticipated expenses and revenues, summed to totals that represent the fiscal policy of the U.S. government, a financial plan. The budget is produced by the entire executive branch in a process of planning and projection that begins more than a year before the president submits it to Congress. Its preparation is overseen by an elite executive agency of more than five hundred employees (once the Bureau of the Budget; now the Office of Management and Budget). The budget is a six-hundred-page book, accompanied by an appendix roughly as large as the Manhattan telephone directory. Agencies submit tens of thousands of pages of further documentation that are supposed to support presidential proposals and provide detailed plans for using the resources Congress agrees to provide. The president's program sets a standard; proponents of increases to one or another category have to explain not merely why that program needs more money but why it should be allowed to throw the whole budget out of whack. Agency heads must argue for the president's proposals even if they had originally requested greater spending.

The president's budget is only a proposal, without legal force, but it gives him a loud voice in the budget debate. It is the president's program, endorsed by the only official "elected by all the people." The effort expended in its preparation argues for its support; the executive claims to have assembled and considered far more information than Congress could ever manage. If the executive agencies behave and support the budget (a huge "if"), the president can claim that the people who run the programs have asked for these amounts. Because the president ultimately hires and fires, the agencies have reason to go along.

Until 1975, the president could also claim that only his set of choices was calibrated to the needs of fiscal policy, managing the economy through manipulating the deficit. He had economic specialists, in the Treasury Department and the Council of Economic Advisers, to justify his choices. Congress had no comparable set of experts to respond, no means to develop its own fiscal policy.

The president's budget-making powers stem from the Budget and Accounting Act of 1921. Before that act, agencies submitted their estimates directly to Congress. Although, like all legislation, it had multiple causes, the 1921 Act passed largely because, as Frederick C. Mosher explains, the First World War, "with its tremendous expenditures and

debt, magnified the enthusiasm among the public and particularly in the Congress for any measures that promised reduction of alleged governmental extravagance and taxes, and this was exactly what supporters of a budget system offered."^[8] Reformers interested in making the government more businesslike had argued for several decades that the process of review and coordination of a national budget would make the government more efficient. The idea had strong popular support, and apparently many members of Congress who objected to giving the president new powers felt forced to go along.^[9]

Congress, however, took steps to limit the grant. Most important, Congress still had to pass the laws; it had the last word. The 1921 Act also took the accounting oversight function away from the Treasury and vested it in the General Accounting Office (GAO), under a comptroller general appointed for a fifteen-year term with the advice and consent of the Senate and explicitly not subject to removal by the president. GAO was to be Congress's check on executive operations. Finally, in separate action, Congress strengthened its appropriations committees, in form at least, centralizing itself in response to the new central power within the executive.

This new set of procedures had one major advantage for Congress: it gave the president primary responsibility for proposing cuts, viewing him as better equipped for such a task because he can impose priorities within the executive branch far more easily than any group in Congress can impose its will on colleagues. (The executive is in principle, and in part in fact, a hierarchy; Congress is anything but a hierarchy.) Once the president imposed priorities, members of Congress could then respond to the budget, changing it where it differed too much from their own priorities or where constituency pressures were too great and leaving the president with the blame for other decisions.

In order to justify altering the president's budget, however, Congess had to claim comparable knowledge and ability to look at the whole picture. It opposed to the president's massive budget, documents the equally copious record of appropriations hearings. And the appropriations committees, while broadly decentralized in operation, somehow always managed to stay within the president's totals.^[10]

Appropriations: The Old Congressional Budget Process

Thirty years ago, the "power of the purse" and "budgetary process" meant appropriations. Entitlements were smaller and, at any rate, they were not in the mainstream of government. The real work—building roads, testing drugs, forecasting the weather, defending our allies, paying salaries, and buying uniforms, tanks, and laboratory equipment—all

went through the appropriations committees. Because, as one current member puts it, "nothing happens without the money," those two committees were and remain among the most powerful and prestigious in Congress.

Their power and huge jurisdiction explains their structure. With fifty-seven members in the House and twenty-nine in the Senate, the two committees are easily the largest in Congress. They are big because they have so much work to do and because their members must have personal contacts throughout their houses. Appropriations committees have more subcommittees, which are particularly independent, because only the strongest of committee heads, with great knowledge and advantages in staffing, could hope to know enough to argue much with subcommittee leaders.

The heads of the thirteen appropriations subcommittees are known as the "College of Cardinals." The name testifies to the sense among other members of Congress that the Appropriations Committee is a priesthood of sorts, with its own rites and norms and with leaders wielding great power. That sense of appropriators as being different has always been much stronger in the House. Strength remains in the House because Appropriations is exclusive; a member of Appropriations cannot sit on any other House committee (except, under the new budget act, for five members who represent Appropriations on Budget). Before the vast expansion of congressional staff, members of House Appropriations did much of the budget review themselves. Time spent together and their involvement in a different kind of work built committee members' self-identification as appropriators, and nonmembers' view of them as forming an arcane priesthood. Even now the House committee, particularly its staff, remains an unusually unified and distinctive organization.^[11]

The appropriations and authorizing committees are inherently in conflict. House rules distinguish appropriations from legislation and forbid legislation on an appropriations bill, but the separation is murky. Not funding an activity or funding it only under certain conditions (e.g., under what circumstances medicaid will pay for abortions) are policy decisions. If the military construction appropriation does not provide funds for housing American soldiers in the Sinai Peninsula of Egypt, then Congress is obstructing the Camp David agreement; an appropriation is legislating policy. Yet a military construction bill that says nothing about what will be built would be a bit skimpy.

In general, Congress expects the appropriators to make policy if, and only if, that is necessary to do their job; that is, no authoritative statement of current policy exists. An appropriations subcommittee chairman expressed the distinction: "Appropriations says the most efficient way to

spend money … [authorizations] should say what the need is." Authorizations can provide limits on the need (e.g., a billion dollars for mass transit), and appropriations above those limits are not in order. Likewise, appropriations are not in order for a program that has not been authorized. These rules are weakened, however, by the fact that appropriations are laws like any other laws: they supersede older legislation. If Congress chooses to override its own rules about appropriations, it is within its rights.

Members of Congress frequently legislate on appropriations bills for various reasons. The authorizations committees may refuse to report legislation that would be supported on the floor. In cases of disagreement between the committee and chamber majorities, the latter can express its will through the appropriations. Thus U.S. aid to South Vietnam was ended in an appropriations bill. Opponents have extensively restricted federal funding for abortion through versions of the "Hyde amendment." Sometimes, authorizing legislation for a program has lapsed because of disagreement over some terms of its authorization. Everybody knows that the dispute will be settled but not on what terms or when. By funding the program (Housing, the Department of State, the Department of Justice) anyway, the appropriations committees technically legislate. Sometimes the appropriators just want some program change (rarely large) and can ram it through.

Appropriations, through which Congress exercises its power of the purse over the executive, can easily constitute a parallel legislative process for the president as well. A president who wants to kill programs would be crazy to push for new legislation; instead, a Nixon or Reagan proposes to zero out the appropriation. He can use his veto against appropriations, but he cannot force legislation.

Yet the veto is weakened for the same reason that members are tempted to use appropriations to legislate: the bills must pass. Almost every bill contains enough "must" items for enough members to make failure unthinkable.^[12] Opponents of postal subsidies generally want to keep the IRS and Customs Service, all from the Treasury-Postal-General Government bill. Opponents of housing tend to like the National Aeronautics and Space Administration or the National Science Foundation and the Veterans Administration, all in the HUD (Housing and Urban Development)-Independent Agencies appropriations act. If those bills fail, parts of government important to many people may fail with them.

Appropriations bills are targets of extraneous "riders" precisely because the appropriations train is going to get through—maybe not on schedule, but eventually. Ultimately the norms that say appropriations should be matters of economy, good management of programs, and routine financing of existing obligations conflict with the opportunity

that appropriations provide to get something done in a Madisonian system of checks and counterchecks.

The old system of House, Senate, president, authorizations, and appropriations was difficult enough to operate. Good times in the 1950s and 1960s helped, and so did a system of mutual expectations and roles that reflected substantial agreement (in retrospect) between Congress and the president on matters of budgetary priorities.^[13] From 1966 to 1973, the system broke down in what Allen Schick called "The Seven-Year Budget War." Congress responded in the Budget Control and Impoundment Act of 1974.

Creating a New Budget Process

Like a family that expected good times and then was disappointed, the federal government in the early 1970s found itself without enough money to meet all its commitments. Entitlements grew faster, and the economy (and hence revenue) grew slower than expected. The deficit rose from 5.5 percent of federal outlays in fiscal year (FY) 1967 to 11 percent in FY71. The problem was more difficult than it sounds because the growth of entitlements meant that the old budget process of appropriations covered less of the budget. To eliminate the deficit through that system would have required cutting 8.4 percent of appropriations in 1967 but 19.6 percent in FY71.^[14] The appropriators, and many other people, wanted some way to bring entitlements into the purview of annual budget choice.

Deficits alone, however, could not have united a broad bipartisan majority behind the Budget Act of 1974; for that, Congress needed Richard Nixon, who gave liberals reason to go along. Congress and the Nixon administration got into a vicious and enervating struggle over deficits and, more important, budget priorities. Nixon tried to slow the expansion of the welfare state and other functions by keeping budgets down, particularly for new programs. At the same time, he wanted to wind down ("Vietnamize") the war more slowly than did the war's critics and to employ any savings to remedy claimed military weaknesses that had been ignored while resources were poured into Indochina. Liberals particularly could not see why the military deserved a funding bill created for a war that was a mistake in the first place.

The story of these battles has been told extensively elsewhere.^[15] For our purposes, three points are crucial:

Impoundment was an old device based on an understanding that if events changed, making an appropriation either no longer necessary or unserviceable, the executive did not have to spend it, so long as most concerned members of Congress agreed. It was supposed to be a tool for better management, and it presumed that the players agreed on policy. Instead, as Allen Schick wrote,

far from administrative routine, Nixon's impoundments in late 1972 and 1973 were designed to rewrite national policy at the expense of congressional power and intent. Rather than the deferment of expenses, Nixon's aim was the cancellation of unwanted programs…. When Nixon impounded for policy reasons, he in effect told Congress, "I don't care what you appropriate; I will decide what will be spent."^[16]

"The aim of impoundment was," Schick adds, "to change the mix, not merely the level, of expenditures."^[17]

The policy stakes in Nixon's impoundments were striking enough, but the political stakes were decisive. Save during a major war, no president had ever so bluntly asserted his primacy over Congress. If Nixon could get away with massive impoundments, what could he not do? If the power of the purse could be defied, what was left for Congress?

The world had been stood on its head. Since the time of royal governors and their civil lists, the legislature's problem had been to restrain the executive by limiting its funds. Now it faced a chief executive who wanted to spend too little, who defied the legislature (which, as far as Congress was concerned, meant the people) by refusing funds for the bureaucracy.

The Budget Act of 1974

The 1974 Budget Act represented a wholly unprecedented approach to budgeting: the legislature itself, debating freely and openly and considering

relevant information, would make programs fit fiscal policy; the parts would fit the whole. Congress would make a budget.

Republicans hoped that "spenders," forced into the glare of public scrutiny, would retreat from their nefarious schemes. Democrats hoped that the conservative priests of budget balance—forced to relate generalities about waste or states' rights to the body of programs that built the roads, fed the poor, and supported the farmers—would no longer be able to deceive the public with platitudes. Each believed that God, which in a democracy means the public, was on its side. The public, as our story will show, was on both sides and neither side.

A new procedure formalized and limited impoundment.^[18] If Congress were to make fiscal policy, it needed its own economists. If it were to consider program costs, it needed neutral analysts, separate from its committee staffs, just as the executive had an OMB independent of the agencies. Therefore the Act created a new staff institution, the Congressional Budget Office (CBO). If CBO said a program would cost twice what its proponents claimed, the CBO could be believed. Its director, appointed jointly by the Speaker of the House and Senate president pro tem for a four-year term, has extensive authority over the office.

Under Dr. Alice Rivlin, CBO gained a reputation for both competence and neutrality. Because estimates of future spending cannot be known with certainty, CBO, like everybody else, errs. In regard to estimates of both the economy and individual program costs, however, CBO usually has less reason to compound lack of knowledge with a policy-political bias than do agencies or the OMB. In its first few years, especially, CBO's estimates proved the more accurate.

CBO and the impoundment-control process easily found their places in the congressional process. The budget committees and new budget resolutions fit far less well.

The Senate Budget Committee (SBC) was established with sixteen members (now twenty-two), chosen by party caucuses and serving indefinitely. The House Budget Committee (HBC) was structured in an unusual manner. Five of its members were to come from the Appropriations Committee and five from Ways and Means. One would be a member of the Democratic leadership, and one would come from the Republican leadership. The other thirteen members were appointed through the usual House procedures (and the numbers were increased as the committee grew more popular). The committee is a mixture, therefore, of regular members and those who represent power centers within the House. In addition, membership is rotating, rather than permanent; no member could serve on the Budget Committee for more than four (now six) years out of every ten. Rotation decreases the chance that committee members will become isolated or parochial in their viewpoints.

Rotation aids HBC in gaining information from other committees and also ensures that the committee's power is not hoarded by a small group of representatives. But rotation also weakens the committee, especially its head, because committee members, aware of the rotation, can afford to cross him.

The budget committees were to write a pair of budget resolutions. The First Resolution (to be passed by May 15) would set targets for other committees and Congress as a whole to meet; it is a formal counterpart to the president's budget. The resolution recommends totals for budget authority (BA), outlays, revenue, and thus the deficit and total public debt. It provides spending targets for each budget function (into which the president's budget is also divided), such as Function 150, International Affairs, or Function 350, Agriculture. Under Section 302(a) of the Act, after a resolution is passed, the budget committees report to every other committee the total outlays and new budget authority it could appropriately provide under the terms of that year's resolution. Under Section 302(b) each committee (most important, Appropriations) subdivides its 302(a) allocation among its subcommittees. The First Resolution was thus half a version of the president's budget. It divided spending into functions; 302(a) reports related the functional division to actual bills, and the Budget Committee staffs, at least, had a good idea of how much should be spent at the program and subcommittee levels. But the real allocation among programs was not in the budget resolution; it would be made by the other committees.

The budget committees used the president's budget, CBO analyses, views and estimates of need that other committees were required to submit by March 15, and their own extensive staff and hearings to develop the policy and political information needed to draft first resolutions. Procedures in passing that resolution were much like those for any other bill, except that, because it was a rule for Congress, it did not need the president's signature. Nor did it have the force of law or appropriate funds. The First Resolution was only a recommendation, less detailed than the president's budget but, as the product of a lengthy process of discussion and accommodation within Congress, more likely to reflect what Congress would actually do. And, like the president's budget, its estimates of revenue and debt and entitlement spending were as much wishful thinking as plan, dependent upon estimates of the future course of the economy.

By early September all relevant authorizations and appropriations were to have been passed. The budget committees were to review economic developments and legislative actions and then report out versions of the Second Concurrent Resolution on the budget. The Second Resolution

had the same components as the First, except that its totals were supposedly binding. Thus, after passage of the Second Resolution, any legislation considered that would cause limits in that resolution to be breached could be objected to and ruled out of order. As the Act was written, if the limits in the Second Resolution would not be met as a result of legislation already passed, then reconciliation instructions could be included in that resolution. These would order the relevant committees to report legislation reconciling spending or taxing to the budget totals. The necessary legislation was supposed to be passed by September 25, in time for the new fiscal year to begin on October 1.

Special procedures expanded both the appropriations committees' ability to question entitlement growth and their control over various other types of backdoor spending. CBO's independent analysis was expected to restrain new entitlements by making their costs more prominent; no new entitlements could be created before the start of a new fiscal year; and the appropriations committees received a limited right to propose amendments to new entitlement legislation that exceeded the committee of jurisdiction's allocation under the First Resolution.

In form, therefore, the 1974 act created a real budget. Congress would choose totals, look at programs, and make the details conform. Entitlement spending would be confronted while totals were being considered. The process of adopting resolutions would draw attention to questions of the relative sizes of revenue and expenditures and their effects on the economy far more explicitly than had been possible in the past; and voting would force members of Congress to take stands in a way not previously required. Debate on the floor would inform legislators and the public about the choices made. The procedures that related taxing and spending to the size of the budget would force Congress to take the totals seriously.

How the New Process Worked

In 1980, as this book begins, Congress had been working within its new process for five years.^[19] One cannot say the system was working as intended because its sponsors disagreed on the intent. A few developments, however, were worth noting.

The budget committees had developed very differently in the Senate and the House. Senate Budget Chairman Edmund Muskie (D-Maine) and ranking minority member Henry Bellmon (R-Okla.) worked to develop resolutions that could command substantial bipartisan support. In the House, by contrast, Republicans viewed the resolutions as the place to demonstrate the difference between the two parties. Because resolutions

became partisan battles, the committee heads had to win majorities entirely within the Democratic party. Party leaders therefore became key actors in the House process. House resolutions tended to be more liberal than the Senate's, given the different coalitions needed to pass them.

First resolutions had not significantly constrained spending. If anything, they allowed liberal majorities in 1975 and 1976 to justify their opposition to President Ford's proposed spending cuts. The heavily Democratic majorities of 1975–1978, however, should not have been expected to want to limit spending. By 1979, in the wake of the Proposition 13 tax revolt in California, inflation fears, and a Democratic president's attempts to restrain spending, the First Resolution was assuming some spending cuts.

But the system had no teeth. Reconciliation was too late in the schedule; no one could expect committees to draft, debate, report, pass, and then confer on reconciliation bills in the ten days allowed—from September 15 to September 25. The big stick of the Second Resolution was the point of order against legislation that breached its totals, but this weapon was less than it seemed. The point of order could be waived; although HBC and SBC in different ways influenced that decision, the committees could be overridden on the floor. The proposals for increased spending could be virtually irresistible (e.g., food stamps running out of money or a Mount St. Helens blowing its top). Most important, even when a set of spending acts passed after the Second Resolution, the overall total was not likely to be exceeded until the last one or two bills. And that straggler was likely to be totally guiltless (like foreign aid, always late and always slashed before it reached the floor). It did not make much sense to savage one bill because of failings on others.

Although the new process was weak, it nonetheless was used in the battles over policy. Program opponents (or supporters) could claim that the First Resolution did not (or did) leave room in the budget for funding. Seeking to satisfy constituents, members could propose extra funding for pet programs in budget debate, where the results were not binding. Thus to the regular legislative process of authorizations and to the alternate legislative process of appropriations now was added a shadow legislative process of budget resolutions. We say "shadow" because its forms were produced by real bodies and real conflict, but themselves had no substance.

Whatever their effect, budget resolutions stood forth as visible statements about the direction of the nation. Therefore, the seven-year budget war did not end; only the field and weapons changed. Most Democrats and most Republicans, egalitarian liberals and individualist conservatives, fought over the size of government and, within that, over

emphasis on military or social welfare spending. The budget resolution figures for spending and revenues, for defense and social functions, became battlegrounds even though they were not binding.

Coda: A Budget Is Many Things and One of Them Is a Performance

The budget is many different policies. It is fiscal policy, designed to stimulate or restrain the economy, to fight unemployment or inflation. The budget summarizes the balance of public and private sectors of the economy—the proportion of GNP taken by taxes or consisting of federal government spending—in short, "how much" federal government we have. The distribution of spending in very broad categories describes the kind of government we want: one that emphasizes military might or protects the middle class or helps the poor. The assumptions in budget resolutions, and action on appropriations or entitlements or tax expenditures, make the budget also a package of thousands of specific program policies: how much to invest in airport safety; which people, if any, should receive special nutrition benefits; how many F-16s the Air Force needs.

For partisans, particularly leaders, of the Democratic and Republican parties, the aggregates in the budget resolutions represent their party's influence on the course of American government; short of the actual organization of the two houses (election of the Speaker, committee assignments), no other action is potentially of as great import to the party leadership. The battle of the budget tests their generalship. The ability of the parties to stay together in the final encounter, apart from the vote to organize the houses along partisan lines, is now their ultimate test of cohesion.

To those members of Congress who identify with the institution—which, depending upon the issue and challenge, ranges from a few to all—budgeting tests Congress: Can Congress choose? Can it enforce its will? In short, can Congress govern? The president asks the same questions, slightly changed: Can I govern the agencies? Can I govern Congress? Can I govern responsibly and maintain public support? Finally, both president and Congress must ask: Do we control policy, or do the policies control us? Can any of us control events?

Republicans and Democrats alike had agreed that the great issues should be faced directly and without obfuscation. How the people they represent would fare under a budgetary process that compels great choices without agreement on what those choices should be is an integral part of the deficit problem, for it is one thing to agree that the deficit is too large and another to agree on how to reduce it.