Congressional budget reform: the unanticipated implications for federal policy making.
The Congressional Budget Process - Original and Subsequent Aims
The process of developing a budget was relatively uncoordinated in both the legislative and executive branches before 1921, when the Budget and Accounting Act made the President a central player in the process by requiring that he submit a unified budget to the Congress. This gave the executive branch the responsibility for defining the structure and the details of the federal budget. Prior to the passage of the Congressional Budget Act of 1974 "(c)urious and uncoordinated legislative procedures hid the implicit tradeoffs involved in fiscal legislation from both the Congress and the public, and there was a chronic failure to consider revenues and expenditures in relation to each other" (Stith, 1988; 515). Early each year, the President proposed a budget, which was considered by the Congress, primarily through the fragmented or uncoordinated committee process. The congressional "budget" was the cumulative (and somewhat accidental) result of legislation affecting annual discretionary appropriations, mandatory programs, and revenues. The Congress never examined or voted on overall spending or revenues or the appropriate stance of fiscal policy. Many members of Congress and observers of congressional budgeting were concerned that this failure to consider the whole was leading to irresponsible results, particularly given the increase in the proportion of the budget financed outside of the appropriations process - so-called backdoor spending. Irresponsible or not, however, members of Congress generally agreed that this piecemeal approach to the budget constrained Congress's ability to make comprehensive policy. At the same time, Congress was faced with a fundamental challenge to its spending priorities when President Nixon refused to spend funds appropriated by Congress for programs with which he did not agree.
These concerns prompted Congress in 1973 to create the Joint Study Committee on Budget Control. This process ultimately resulted in the enactment of the Congressional Budget and Impoundment Control Act of 1974. The act created a process by which budget decisions could be made comprehensively and could be protected from the normal parliamentary hurdles faced by other legislation. The process was neutral as to budget outcomes but could "be deployed in favor of higher or lower spending, bigger or smaller deficits" (Schick, 1980; 73). Although some members of Congress voted for the subsequent Congressional Budget Act because they believed that it would control spending or the deficit (some subsequent evaluations [Fisher, 1985] have deemed it a failure because it has not), the act itself contained no provisions which biased it one way or the other. It would have been biased, for example, if it had required supermajorities (such as two-thirds of the members in each house) to enact a tax increase. The act was outcome neutral, truly a process reform.
The act also embodied certain other principles, including preserving the ability of the Congress to act according to its budgetary preferences (hence, no points of order requiring supermajorities); substituting information for control (the Budget Act opted to make the Congress aware of what they were doing, rather than preventing them from doing it through spending limitations); preserving the participation of members in the budget process (rather than stacking the deck in favor of particular committees); allowing Congress to consider budget priorities comprehensively; and establishing a process to control presidential impoundments (Schick, 1980).
As the budget deficit grew substantially in the wake of the passage of the Reagan economic program in 1981, the Congress became increasingly aware that the budget process could not (nor, as noted above, was it intended to) serve as a constraint against these large deficits. Frustration with large deficits and the inability to contain them ultimately led to the passage in 1985 of the Balanced Budget and Emergency Deficit Control Act of 1985, popularly known as Gramm-Rudman-Hollings (GRH). GRH attempted to control the deficit through setting gradually declining deficit targets and was supposed to result in a balanced budget by fiscal year 1991. If the deficit targets were not met, automatic across-the-board spending reductions, or sequestrations, were to take effect.
The passage of GRH represented a fundamental change in the focus of the budget process. For the first time, the budget process was used to specify the end result to be achieved, rather than simply the rules to be followed that might lead to any number of different budget outcomes. As such, it was a switch from what Hanushek (1986; 6) has described as "process rules" (that is, rules governing decisions, timing, and priority setting) to "allocation rules" (which specify particular budget results, such as levels of spending and the deficit). According to Kate Stith (1988; 597), "[b]y establishing binding deficit limitations enforced outside of the legislative budget process, GRH sought to...amend our 'fiscal constitution,' which for over two centuries had permitted prevailing legislative minorities to spend without limitation."
The deficit, of course, did not come down as promised by the Gramm-Rudman-Hollings legislation. In fact, the fiscal year 1993 deficit (which would have been zero if the law, as revised in 1987, had met its goal) was actually $255 billion (Congressional Budget Office, 1994b). The act put a premium on short-term budgeting; under GRH, all that mattered was the single year for which the projections were being made. These annual targets were met through short-term fixes and budget gimmickry, including basing the budget on optimistic economic and technical assumptions, selling assets, and shifting costs between fiscal years.
The successor to Gramm-Rudman-Hollings, the Budget Enforcement Act (BEA), was passed in 1990 and was designed to enforce the five-year deficit reduction agreement reached between the President and Congress in that year. The BEA effectively eliminated annual deficit targets, placed limits on the level of discretionary spending through fiscal year 1995, and established the pay-as-you-go (PAYGO) process to ensure that any tax or mandatory spending changes were deficit neutral. The original Budget Enforcement Act would have expired in 1995; the Omnibus Budget Reconciliation Act of 1993 extended both the discretionary spending limits and PAYGO until 1998. The BEA shifted the process away from deficit targets to controls on legislative changes in spending or revenues. By so doing, it focused attention on those actions that Congress and the President could directly control (spending and revenue actions), rather than holding them accountable for the size of the annual deficit, which can be influenced by the performance of the economy and other factors not controlled through the annual budget process (Joyce and Reischauer, 1992). As such, it has been described as a no-fault budget process. As long as budget rules are followed, the deficit can grow substantially without anyone being held responsible for the increase (Doyle and McCaffrey, 1991).
How Have the Changes Since 1974 Affected Policy Making?
Unquestionably, the process that exists today is different in many important ways from the process that was designed in 1974. The budget process has reasserted the congressional role in budgeting, has increased the information available to Congress on the budget and economy, and has curbed the President's impoundment powers. The process has not been viewed as an overwhelming success in other respects. The most frequent criticism is that it has not brought the order and timeliness to congressional budget action for which advocates had hoped. Deadlines for enacting budget resolutions and the passing of appropriation bills have routinely been missed.
Rather than comment further on issues such as these, this article focuses on three important changes to the budget process that have received less scrutiny, yet have fundamentally influenced the way budget policy is made. First, reconciliation, almost an afterthought in 1974, has become in many ways the most important part of the process. It has been used to make major policy shifts, most recently embodied in omnibus, multi-year, deficit-reduction packages. Second, while the Budget Act was policy neutral by design, this neutrality has been undone by subsequent reforms that sought to use the budget process to limit government spending or the size of the deficit. The failures (some real and some imagined) of GRH and the BEA to control spending and the deficit, coupled with the increasing complexity of the process, have contributed to widespread disillusionment, not only with the budget process but with government in general. Third, the increasing importance of budget enforcement has had important implications for policy making by forcing policy makers to take budgetary effects into account. The manner in which information provided for enforcement purposes is used has sometimes caused distortions, as policies are designed with enforcement in mind. Budget enforcement has also further empowered a minority of legislators who can often use budget rules to block policies.
Reconciliation - A Powerful New Tool for Centralized Control
In establishing its own budget process in 1974 and in the changes that have been made in that process since, Congress has been bucking its general trend toward decentralized decision making. While the Congress is generally viewed as an institution that has been moving away from centralized control - the declining influence of parties and the reduced influence of the congressional leadership are often presented as evidence of this - the congressional budget process attempts to move toward greater centralization of decision making (Ellwood, 1985; Schick, 1980). Nowhere is that centralizing trend more apparent than in reconciliation, where the Congress as a whole (through the budget resolution) is empowered to give instructions concerning changes in law to congressional committees. The centralizing tendency of the budget process in the face of the increasing fragmentation of the Congress has certainly contributed to the dissatisfaction of members with the state of congressional budgeting. In short, members of Congress and congressional committees do not like to be told what to do, but the budget process - and particularly reconciliation - is designed to allow the whole Congress to control its committees and members.
The framers of the Budget Act viewed reconciliation as unimportant; it was an optional process for tying the ceilings enacted in the second concurrent budget resolution (since eliminated) to the changes in laws governing taxes and spending (mainly appropriations) necessary to achieve them. Because the changes in the second resolution were only designed to accommodate revisions in political or economic circumstances since the passage of the first, and in any event came very late in the budgeting cycle, Schick wrote that "the reconciliation process is not likely to offer much opportunity for reconsideration of past actions" (Schick, 1980; 321).
Enter the Reagan revolution. President Reagan had proposed a radical budget restructuring that included large income tax cuts, increases in defense spending, and reductions in domestic spending. Amid a concern that the tax cut might not be enacted without first passing the budget cuts that made them appear affordable, the Reagan administration, led by Budget Director David Stockman, attempted to use reconciliation as a means to use the budget resolution to carry out the Reagan spending program by subjecting the program to a single up-or-down vote, thus paving the way for the tax cut. Although the tax cuts themselves were not enacted through reconciliation, the argument, as articulated by Stockman, was that the tax cuts might not have survived politically without first moving to cut spending (Stockman, 1986).
1981 was the second time that reconciliation had been used in concert with the first budget resolution - President Carter and the Congress had agreed to a small deficit reduction package enacted through the reconciliation process only a year earlier - but the use of reconciliation in 1981 was a watershed "because of the size and scope of the changes made and the threat they pose to hallowed congressional procedures" (Hartman, 1982; 389). Many analysts have argued that the large changes that took place in 1981 - changes that resulted in a legacy of large deficits - would not have been possible under the previously decentralized budget process (Fisher, 1990; Rivlin, 1986; Penner and Abramson, 1988). The use of reconciliation to drive the spending cuts on the front end in anticipation of the tax cuts was certainly an important part of that story.
Reconciliation has also proved to be an important part of the solution to the problem of large deficits. This was anticipated by Hartman, writing in 1982 about the results of the 1981 budgetary process (Hartman, 1982; 397).
Procedural changes directed toward strengthening budget control...make the most sense if one envisions the environment of the future as one of heightened attention to solving expenditure growth and cutting deficits. The crucial procedural innovation of the last few years - reconciliation early in the budget process - seems essential to any move to strengthen control. Because the predominant message for the congressional committees in this austerity scenario is that spending be cut and taxes raised, it is unrealistic to expect such measures to be put into effect unless committees are forced to act. This is the primary function of reconciliation: it allows Congress as a whole to order its committees to take actions.
In fact, while 1981 is remembered as a year when the congressional budget process was used to promote what are in retrospect viewed as irresponsible, deficit-exploding changes to the budget, the use of reconciliation in subsequent years has been for the opposite reason. Omnibus budget reconciliation acts (OBRAs) became commonplace in the 1980s, with the enactment of OBRAs in 1982, 1983, 1984, 1986, 1987, and 1989. The three most recent multi-year deficit reduction packages to pass Congress - the five-year agreements of 1990 and 1993 and the seven-year package passed in 1995 - were each implemented through use of the reconciliation process.
Reconciliation is so important for policy making because it permits actions to be taken in tandem that would arguably never survive separately, for two main reasons. First, combining spending and tax changes as part of one large package can communicate to members of Congress and the public that unpopular policies are part of an overall package intended to achieve some larger (and more popular) objective. In 1990 and 1993, the strategy focused on the argument that everyone's ox was being gored simultaneously (or, to use a less colorful phrase made popular during the legislative battles over OBRA-1993, they are engaged in shared sacrifice). In 1995, the spending reductions induced through the reconciliation process were marketed as necessary to fulfill the twin Republican promises of balancing the budget and cutting taxes.
Second, when done through the reconciliation process, the required changes in law are mandated. Under reconciliation, the details of legislation are fair game, but committees are required to report legislation that results in changes of the magnitude required by the budget resolution. If each change were considered without such a mandate, many of them would be picked off in committee and never see the light of day. In addition, bills considered outside of reconciliation could be filibustered in the Senate, whereas the Budget Act establishes a specific time limit for debating reconciliation legislation.
An End to Policy Neutrality
As noted above, the drafters of the Congressional Budget Act viewed it as very important that the process not be biased in the direction of particular budget outcomes. For this reason, they designed a process that could be used to increase or decrease spending (or revenues), or for increased or decreased deficits. (Recall that they wanted the Congress to play a role in the setting of short-run macroeconomic policy, a function of the budget that has all but disappeared today.) Beginning with Gramm-Rudman in 1985 and continuing through the Budget Enforcement Act, the focus of the process has changed markedly from a policy-neutral process to a process that is designed to be used to achieve particular policy objectives.
In the case of Gramm-Rudman, the purpose of the process was to reduce the deficit. This change can be explained politically by the need of legislators to engage in what Ellwood refers to as blame-avoidance behavior. When subsequent tax increases or spending cuts occurred, members of Congress could argue to aggrieved constituents that "the law made me do it." But, as he notes "the difficulty with expenditure limitations and balanced budget amendments is that they seek to cap the effects of politics rather than change the incentives that cause those effects" (Ellwood, 1988; 574). Regardless of the likelihood of success or failure, however, Gramm-Rudman certainly began the shift from a policy-neutral budget process to one where achieving a lower deficit was of paramount importance.
Gramm-Rudman also began the shift (expanded under the subsequent Budget Enforcement Act) to use the budget process to limit spending. GRH, on its face, was neutral as to whether the budget deficit was to be reduced by reducing spending or by increasing taxes. But the sequestration process was far from noncommittal on this point. In fact, by including only a sequestration of spending (as opposed to equally logical automatic tax increases), the GRH process imposed an implicit ceiling on federal expenditures (Stith, 1988). This set up lower spending as the default if other actions are not taken.
The Budget Enforcement Act virtually abandoned the use of the budget process to stimulate deficit reduction, while expanding the emphasis on expenditure limitation even further. In the BEA, explicit statutory caps on both budget authority and outlays were set up for the discretionary portion of the budget, by fiscal year, for each of fiscal years 1991 through 1998. Because the caps do not permit discretionary spending to grow as fast as inflation in any one year, this cap system establishes a presumption that appropriated spending will shrink in real terms. Further, advocates of increased discretionary spending are not permitted to pay for this spending by increasing revenues, unless they can also change the statutory caps.
For mandatory spending and revenues, the PAYGO process set up a condition of deficit neutrality; that is, legislative actions in this area are to make the deficit no worse. Unlike discretionary spending, however, increased mandatory spending could be paid for by raising taxes without amending the BEA. The political difficulty of either raising taxes or cutting existing mandatory programs results in the PAYGO process effectively discouraging new (or liberalized) mandatory programs (Schick, 1991). Therefore, the effect of PAYGO is in the same direction as that of the caps - to discourage increases in government spending. (In principle, PAYGO also acts to deter tax cuts and so far it has done so. In a time when the public is seen as rejecting big government, however, the pressure to reduce taxes may outweigh the pressures for new spending. In such an environment, a prudent individual would not bet that PAYGO will act as an effective curb to tax cutting.)
Relative to the original goals of the Budget Act, the important distinction to make is one of control over the budget - that is, over total taxes and spending - versus control over only spending or control leading to a prescribed deficit number. The Budget Act was designed to foster control over the former - to permit a comprehensive evaluation of taxing and spending. GRH and the BEA created norms for budget outcomes concerning either the deficit (GRH), spending (the discretionary caps), or deficit neutrality (PAYGO).
Beyond the important implications of this shift for the budget process itself is the effect that this change in emphasis has had over the credibility of the process. After 1985, observers and participants were justified in judging the process according to budgetary outcomes, regardless of whether the process should have ever been expected to force those outcomes or not (Joyce, 1993). Seen in this light, the inability of GRH to reduce the deficit to the levels specified by the legislation was an indictment not only of Congress and the President (or the public's conflicting demands that they balance the budget without raising taxes or cutting spending) but of the budget process itself. This occurred in spite of the fact that some analysts claim that Gramm-Rudman did reduce both federal spending and the deficit, albeit not to the levels envisioned in the legislation (Reischauer, 1990;. Haem et al., 1992). it also occurred despite the success of the BEA on its own terms - the caps have held and the PAYGO process has discouraged new mandatory spending or tax cuts that are not paid for. These limited terms, however, create a public perception problem of their own. The public wonders how elected officials can judge a budget process to be successful if, in enforcing a deficit reduction agreement, it permits $200 billion deficits to be replaced with $300 billion deficits. In short, the continued large deficits are viewed by many as evidence that the budget process itself is broken and needs to be overhauled.
Aside from whether the outcomes promised by these budget reforms were achieved or not, there is the issue of the complexity of the process. The Congressional Budget Act layered a new process on top of the existing authorization and appropriations processes. The budget changes that have occurred since 1974 have either expanded the use of procedures (like reconciliation) that were not used much prior to the 1980s, or have created new procedures, reports, and rules on top of the old ones. GRH added deficit targets, sequestration, and sequestration reports. The BEA gave us spending caps and PAYGO, so that the possibility existed for not one sequestration, but three. Imagine the plight of the large number of freshman members of Congress elected in 1992 and 1994 when they first confronted a budget process which seemed to be written in a foreign language. And this is nothing compared to the confusion that the general public must feel. As Fisher has argued (1985; 24)
The power of the purse is not merely a means by which Congress controls the executive branch. It is also the way the public controls government. Any process that confuses legislators and the public, no matter how much it may delight the conceptual dreams of technicians, is too costly for a democracy.
It is important to note that the process has not become more complex because the drafters purposefully set out to confuse people (although, if they had, it is not clear how they could have done a better job). While some of the increases in complexity in the original system may have had to do with an unwillingness to do battle with committee chairs over existing turf, much of the increased complexity since that time has resulted from attempting to use the process to enforce particular outcomes. In addition, the process is focused much more on micro-level decisions - tracking the effects of individual bills or amendments on the budget - than was true under earlier iterations of the budget process. This, in turn, has occurred because we are in severe budget disequilibrium. The process would be simpler if the budget were in balance (Reischauer, 1993). So perhaps the complexity of the process is a necessary, but regrettable, result of current political and economic realities. It is nonetheless damaging to the credibility of the budget system that it can be understood by so few people.
The Increasing Importance of Enforcement in the Budget Process - The Ascendancy of Scorekeeping
Unquestionably, the framers of the Budget Act wanted congressional budgeters to take into account the consequences of their actions. For this reason, they not only created the budget committees and the concurrent budget resolution devices for coordinating congressional attention to the whole budget, but they also created enforcement mechanisms designed to discourage the Congress from breaching the discipline offered by the budget resolution. Congressional Budget Office cost estimates were designed early on to play this role - to answer the question "what does it cost?" when the Congress was debating a piece of legislation.
Cost estimating and scorekeeping played a somewhat limited role in influencing the policy process from 1974 to 1985 compared to what has happened since GRH and BEA. This is primarily because the consequences of failing to adhere to the requirements of the budget process have become more real. After the passage of GRH, enacting legislation that would cause the deficit targets to be breached became an event that could trigger across-the-board cuts in spending. Note that this gets the attention of the whole Congress, because actions under the jurisdiction of one committee can lead to cuts in programs under the jurisdiction of others. Since the BEA was enacted, the existence of explicit spending limits (the discretionary caps) and explicit assumptions of deficit neutrality (PAYGO) has made the question, "How will you pay for it?" the first one asked of proponents of costly new spending. How much they will have to pay is tied up in the enforcement mechanisms established as part of the budget process.
This development was undoubtedly necessary given the change in the focus of the process. Once the decision was made to use the budget process to return the budget to (or closer to) balance through enacting deficit limits and spending limits, it was necessary to have some kind of mechanism in place to enforce those strictures. This has given rise to a complex set of administrative rules and procedures that govern the "scorability" of various policy changes. In response to these rules, advocates of particular policies sometimes adjust the parameters of their policy proposals to reflect how OMB and CBO will score them. This allegedly happened in the design of President Clinton's health reform plan, where caps on insurance premiums were included not because the administration thought they were a good idea, but because they believed that the Congressional Budget Office would score them as reducing spending.
A discussion of budget enforcement carries with it some natural bias, depending on the actor, since the world of federal budgeting is generally divided into enforcers and victims. Those who participate in enforcing the rules - primarily the budget committees, CBO, and OMB - are constantly defending their decisions against attack from those who feel aggrieved by them - primarily other congressional committees and federal agencies. For this reason, rather than discuss whether enforcement is inherently good or bad (which seems a moot point since, in the current environment, the option of abandoning enforcement seems neither desirable nor likely), I would like to discuss several limitations of the current emphasis on enforcement that might be considered either when applying the current budget law or considering its revision.
I start with a caveat. The budget rules that have been created work well in the majority of cases. They have been developed over time in order to be able to treat all policies consistently; that is, to subject them to a common set of criteria. This is necessary in order for the rules to be viewed as fair and nonarbitrary. But the rules cannot anticipate every exception or every unanticipated impact (they also cannot anticipate the ingenious ways that people will devise to try to get around them, which is probably one reason that they are not all written down). But despite the good intentions of the rules and their enforcers, there are nonetheless situations where the budget rules as they have been developed may have unanticipated or distortionary consequences. When considering the problems that the current enforcement system creates for the policy process, three issues stand out - the narrow scope of activity to which the rules apply, the artificial division of the budget that prohibits tradeoffs between mandatory and discretionary spending, and the "budget" issues that often give rise to additional opportunities to block legislation for partisan political purposes.
Narrow Scope of Budget Enforcement. By design, the enforcement procedures created as a part of the budget process have a narrow focus. The question that is asked under the BEA, for example, is, "What is the effect of this policy (bill) on federal taxes and spending for each of the next five (or fewer) fiscal years?" This is the right question in many cases, although it does create incentives to push costs beyond the five-year enforcement window in others. Relying solely on the information created as a result of these rules may distort decision making in cases where the important effects on the federal budget occur many years in the future, or where the more important effects to be considered are the overall economic effects, independent of the federal budgetary effects. In other words, because enforcement focuses on short-term federal budget effects and ignores costs and benefits to nonfederal actors, there is some danger of over-emphasizing only one part of the story.
The best recent example of this is health care reform. As most people are aware, President Clinton proposed, and Congress considered, a comprehensive overhaul of the nation's health care system. The Clinton plan was only one of many that was the subject of debate during the 103rd Congress. Many of these plans differed radically from each other in the method that they proposed to address health care, but all faced a common hurdle - the necessity of having the bills creating these plans scored by the Congressional Budget Office.
The effects of a major change like health reform can only be reasonably evaluated according to their long-term effects on national health care and on the economy as a whole. So the budget process, left alone, would ask the wrong question here, What effect will this health reform plan have on the federal revenues and spending over the next five years? Other questions might be raised in the context of calculating the federal costs as well, including, What counts as federal revenues and spending and what does not? (as was true with the analysis of the Clinton plan's health alliances) (Congressional Budget Office, 1994a). But these are not close to the major questions that policy makers considering health care reform should focus on. Certainly questions of overall economic costs and benefits (not just those that happen to show up in the federal budget), quality of care, and access to insurance are more important than these.
In fact, in response to a concern about the narrow scope of the budget rules, CBO attempted to answer broader questions in its analyses of various health reform plans (such as the effects on the federal budget 10 years in the future, on overall health spending, on administrative feasibility, or on employment). These effects, while they may be considered, are not scorable under the budget process. For this reason, the PAYGO rules - which require budget neutrality in each fiscal year through 1998 - led to a redrafting of plans to ensure that they did not increase the federal deficit in that time period. The question, of course, is whether this focus on the more narrow question of short-term federal costs and benefits might contribute to the making of bad health policy.
This is not only a question faced under health care reform. In 1994, the bill to ratify the General Agreement on Tariffs and Trade (GATT) created a PAYGO problem, since the reduction in tariffs included in the agreement would decrease revenue in the short run, with the purported benefits of increasing economic growth in the long run. This is an essential part of the policy - low tariffs are part and parcel of free trade. Once again, it is short-term budgetary effects that count. The long-term economic effects do not. The Senate was forced to waive its budget rules in order to pass the GATT legislation, setting up an additional hurdle to the ratification of the agreement. By the same token, if the 104th or some subsequent Congress reforms the welfare system, it may well consider the nonfederal effects of welfare reform and the budgetary effects beyond the five-year window. Welfare reform in 1995, however, was part of a budget-cutting exercise, suggesting that the need to save money in the federal budget dwarfed any other concerns. In these and other cases, the broader implications are considered in spite of the current budget rules, not because of them.
Artificial Division of Spending Discourages Tradeoff. Because the Budget Enforcement Act, for the first time, divided the budget into three categories - Social Security (which has its own budget rules), PAYGO (including other mandatory spending and revenues), and discretionary spending - it also created walls between them. The divisions were set up with good reason. The drafters of the act wanted to hold congressional committees accountable for their own actions. Therefore, an overage on the PAYGO scorecard results in a sequestration of mandatory spending (which affects authorizing committees), while a breach of the discretionary spending caps results in an across-the-board cut in appropriated spending (which affects appropriations committees). The exclusion of Social Security from PAYGO was for a separate reason - the myth that Social Security, as a "self-financed" program, should be walled off from the rest of the budget.
However good the reason, this kind of a process has one major drawback from the perspective of setting overall budgetary priorities. It does not permit, without changes in law, tradeoffs that would cross the divide between the categories. Therefore, revenue increases can be used to finance increases in mandatory spending, but cannot be used (without increasing the caps) for increases in discretionary spending, even if there is substantial agreement that these increases would have positive effects that more than offset the cost of raising taxes. Take well-chosen public investment programs as an example. A President or a Congress seeking to increase spending on infrastructure or other "investment" spending, even if there was general agreement concerning the worthiness of such spending (and a corresponding willingness to ante up new taxes to pay for it), could not do so under the existing budget process without recommending corresponding decreases in other discretionary programs. Indeed, this is the dilemma that has faced the Clinton administration in getting the President's investment programs funded, given a shrinking discretionary pie.
This can also be illustrated by considering a very different problem that faced Congress and the President in 1995. Both pledged to enact tax cuts, yet the budget law declares reductions in discretionary spending as off-limits to pay for them. This created a major dilemma for proponents of these policies, since both the President and the Congress apparently desired to use discretionary spending cuts to finance their tax reduction proposals. They could overcome this by using the same implicit process that was used in 1990 and 1993 - that is, to pass a budget reconciliation bill that includes changes in tax law and additional constraints on discretionary spending. The walls between taxes and discretionary spending cannot be crossed without such an omnibus bill, however, creating an additional hurdle to the enactment of these policies.
Further, if a policy decreases spending in one category of the budget, while increasing it in another, the decreases cannot be counted as offsets against the increases. One recent example of this surfaced with regard to the 1993 Federal Workforce Restructuring Act, which authorized federal agencies to offer buyouts as inducements to early retirement for federal workers. The purpose of the buyout legislation was to save money associated with the salaries of federal workers - a savings that would affect the discretionary portion of the budget. But the bill created costs as well, mainly in the form of earlier retirement payments made to bought-out workers. This cost showed up in the mandatory portion of the budget. The savings to be gained from salary savings could not be counted as an offset against the increase in retirement payments, therefore, the bill had to include offsets to make up for the overage on the PAYGO scorecard. This was almost enough to scuttle a bill that was widely supported and that everyone agreed would lead to a net reduction in federal spending.
Proliferation of Points of Order Expands Use of Budget Rules to Block Legislation. Another area where the evolution of the budget process has led to substantial changes in policy making concerns the use of points of order to block legislation, particularly in the Senate. Points of order, which are procedural devices that can be used to block legislation if all or part of that legislation would violate the law or congressional rules, have become more important with the use of the budget process for deficit reduction and enforcement. Not only have new points of order been created but, beginning with GRH in 1985, a three-fifths requirement was established in the Senate for the first time to waive many Budget Act points of order. This requirement was expanded and extended by the BEA, both in the 1990 act and when it was renewed in 1993. In fact, while only 10 Budget Act points of order were considered in the Senate in the 12 years between 1975 and 1986, more than 120 were considered between 1987 and 1994. (Bach, 1989; Congressional Research Service, 1995).
The proliferation of points of order has provided more tools to a minority that can be used to change or block legislation. This has the effect of both altering the drafting of legislation to attempt to eliminate possible sources of points of order and providing opportunities to kill bills once they have come to the floor.
The former result occurs both in the House and in the Senate. In the House, where Budget Act points of order are typically waived in the rules enacted governing the floor consideration of legislation, attempts are nonetheless made to avoid points of order since the minority party has historically been inclined to tally the waivers of rules permitted by the Rules Committee for use against majority party candidates in congressional campaigns. In the Senate, the threat of a point of order can also cause legislation to be changed in the drafting stage, since proponents know that a challenge on point-of-order grounds requires them to get 60 votes for passage instead of the usual 51.
An example of the use of the Budget Act to affect the content of legislation before it reaches the floor is the so-called Byrd rule. The rule, which was codified in 1990 as Section 313 of the Budget Act, authorizes a point of order against a reconciliation bill, amendment, or conference report that includes "extraneous" material. Extraneous is hard to define precisely, but the definition includes provisions that would increase the deficit or make changes that are fundamentally nonbudgetary. During consideration of the 1993 reconciliation bill, hundreds of separate House provisions were reportedly removed from the bill because their inclusion in the conference report would potentially have violated the Byrd rule. The practical effect of this was to tilt the results of reconciliation to the Senate's advantage.
Of course, points of order are also used to block legislation once it reaches the floor, again mainly in the Senate. Points of order have a subtle effect on legislating in the Senate. A minority of senators trying to defeat a piece of legislation, after all, does have the filibuster available. In general, though, it is politically easier to oppose legislation on a Budget Act point of order than to filibuster, if only because a budget point of order connotes fiscal responsibility, while a filibuster connotes obstructionism.
A recent example of this came during consideration of the crime bill in 1994. Senators in opposition to the conference report on this bill could have chosen to employ the filibuster as a dilatory tactic. Instead they turned to the budget process by opposing a new trust fund that was created by the bill. Technically, the creation of such a trust fund was under the jurisdiction of the Budget Committee, which had not passed on it; thus the trust fund provision became subject to a point of order under Section 306 of the Budget Act. This occurred in spite of the fact that the bill, including the same trust fund, had passed the Senate by a vote of 95 to 4 only months earlier. It is, of course, an open question whether opponents would have decided to filibuster the bill if the point of order was unavailable to them. But it is a question they did not need to answer.
Conclusion - What Does the Future Hold?
The point of the preceding discussion is not to suggest that the deficit-driven changes that have occurred in the budget process were unnecessary or even, on balance, for the worse. The point is that they have fundamentally changed the process from what was intended in 1974. Perhaps this was inevitable following the explosion of the deficit after 1981. But it is worth considering that the damage that has been done - in terms of the credibility of the process, its complexity, and the opportunities that it presents to influence or block policy - is a byproduct of the use of the process to control federal deficits and spending.
Someday, the problem of large federal deficits may be behind us. At that point, it will be worth reconsidering what we have done to the budget process and evaluating whether the procedures established under the 1974 act were fundamentally correct and should be reestablished. Agreement is certainly not universal on that point. For example, Fisher has argued forcefully that the 1974 budget process has fundamentally damaged budget policy making by allowing the President to escape responsibility for budget outcomes (Fisher, 1990). Regardless of whether the 1974 budget process was an improvement over what preceded it or not, it is clear that the changes that have occurred since 1985, even if necessary, have had effects that would need to be reconsidered if, as the 104th Congress desires, there is to be a return to equality between federal spending and taxes.
Acknowledging the limitations of the current budget process is not the same as prescribing a new one. Many budget observers would agree that the current process is necessarily too complex, too rule-bound, and too biased. The question, however, of what constitutes a good budget process has continued to elude budget scholars. This in part stems from precisely the problem, as noted by Roy Meyers, that the budget process is expected to achieve a large number of diverse (and sometimes contradictory) goals simultaneously, including "preventing insolvency, making efficient allocations of the government's limited financial resources, approaching intergenerational equity, contributing to fiscal stabilization, and being responsive to public demands" (Meyers, 1995). Establishing a good budget process requires not only defining these goals but prioritizing them. Further, the priorities set, as reflected by the experience of the last 20 years in the federal process, do affect budget outcomes and budget decision making in important ways.
The author wishes to thank the following individuals for their comments and assistance on previous drafts of this article: Jim Blum, Bill Danster, Sandy Davis, Gail Del Balzo, Bob Hartman, Jim Horney, Roy Meyers, Marvin Phaup, Bob Reischauer, and Irene Rubin.
Bach, Stanley, 1989. "Points of Order and Appeals in the Senate." Congressional Research Service.
Congressional Research Service, 1995, Waivers of the Budget Act Considered in the Senate. unpublished.
Doyle, Richard and Jerry McCaffery, 1991. "The Budget Enforcement Act of 1990: The Path to No-Fault Budgeting." Public Budgeting and Finance, vol. 11 (1), 25-40.
Ellwood, John, 1985. "The Great Exception: The Congressional Budget Process in an Age of Decentralization." In Lawrence Dodd and Bruce L. Oppenheimer, eds., Congress Reconsidered 3rd ed. Washington: Congressional Quarterly, pp. 315-342.
-----, 1988. "The Politics of the Enactment and Implementation of Gramm-Rudman-Hollings: Why Congress Cannot Address the Deficit Dilemma." Harvard Journal on Legislation, vol. 25, 553-575.
Fisher, Louis, 1985. "Ten Years of the Budget Act: Still Searching for Controls." Public Budgeting and Finance, vol. 5 (Autumn), 3-28.
-----, 1990. "Federal Budget Doldrums: A Vacuum in Presidential Leaderhip." Public Administration Review, vol. 50 (November/December), 693-700.
Haem, Sung Duek, Mark S. Kamlet, David C. Mowery, and Tsai-Tsu Su, 1992. "The Influence of the Gramm-Rudman-Hollings Act on Federal Budgetary Outcomes, 1986-1989." Journal of Policy Analysis and Management vol. 11 (2), 207-232.
Hanushek, Eric, 1986. "Formula Budgeting: The Economics and Analytics of Fiscal Policy Under Rules." Journal of Policy Analysis and Management vol. 6 (1), 3-19.
Hartman, Robert, 1982. "Congress and Budget-Making." Political Science Quarterly, vol. 97 (Fall), 381-402.
Joyce, Philip G., 1993. "The Reiterative Nature of Budget Reform: Is There Anything New in Federal Budgeting?" Public Budgeting and Finance, vol. 13 (Fall), 36-48.
Joyce, Philip G. and Robert D. Reischauer, 1992. "Deficit Budgeting: The Federal Budget Process and Budget Reform." Harvard Journal on Legislation, vol. 29, 429-453.
Meyers, Roy T., 1996. "Is There a Key to the Normative Budgeting Lock?" Policy Sciences (forthcoming).
Penner, Rudolph and Alan Abramson, 1988. Broken Purse Strings: Congressional Budgeting 1974-1988. Washington, DC: The Urban Institute.
Reischauer, Robert D., 1990. "Taxes and Spending Under Gramm-Rudman-Hollings." National Tax Journal vol. 63 (September), 223-232.
-----, 1993. Testimony before the Joint Committee on the Organization of Congress. March 4.
Rivlin, Alice, 1986. "The Need for a Better Budget Process." Brookings Review (Summer), 3-10.
Schick, Allen, 1980. Congress and Money. Washington, DC: The Urban Institute.
-----, 1991. Testimony before the House Committee on the Budget. October 10.
Stith, Kate, 1988, "Rewriting the Fiscal Constitution: The Case of Gramm-Rudman-Hollings." California Law Review, vol. 76 (May), 593-668.
Stockman, David, 1986. The Triumph of Politics: Why the Reagan Revolution Failed. New York: Harper and Row.
United States Congressional Budget Office, 1994a. An Analysis of the Administration's Health Proposal Washington, DC: U.S. Government Printing Office.
-----, 1994b. The Economic and Budget Outlook: An Update. Washington, DC: U.S. Government Printing Office.
Philip Joyce is an assistant professor of public administration at the Maxwell School of Citizenship and Public Affairs, Syracuse University. Between 1991 and 1995, he analyzed budget process reform issues for the Congressional Budget Office.
|Printer friendly Cite/link Email Feedback|
|Author:||Joyce, Philip G.|
|Publication:||Public Administration Review|
|Date:||Jul 1, 1996|
|Previous Article:||Response to Fine, Reeves and Harney's article on employee drug testing.|
|Next Article:||Violence in the American workplace: challenges to the public employer.|