Thank you for inviting me to testify before your subcommittee this morning to offer my views on S. 1434, the “Biennial Budgeting Act of 1995,” as well as Senator Thompson’s draft legislation to reform the budget and oversight processes of the Congress.
I admire your efforts to seek changes in the budget process and in the broader legislative process that would improve the efficiency, timeliness, and productivity of Congress. As co-director of the Renewing Congress Project and as a student of Congress for many years at the Brookings Institution, I have invested a good deal of time seeking improvements in the legislative process. But I would be less than candid if I failed to convey my disappointment with the major proposals under consideration by this committee. While there is understandable sentiment in both the legislative and executive branches for trying to reduce the time and energy expended on budgetary decisions, a biennial budgeting process is unlikely to produce the desired benefits at acceptable costs. Similarly, while the slogan “cut their pay and send them home” may have broad popular appeal, proposals that link congressional pay to the timely completion of the budget and that mandate shorter congressional sessions are, I believe, based on faulty diagnoses of the underlying problems.
The fiscalization of policy debate has been a prominent feature of our national politics since the late 1970s. Substantive debates are regularly transformed into budgetary debates; conflict and delay are the norm. The natural temptation is to blame this syndrome on the budget process: budget issues overwhelm the system because of cumbersome, duplicative, time-consuming procedures in both the executive branch and Congress.
I believe this diagnosis is wrong. Budget deficits, and the lack of consensus among politicians and the public on how to deal with them, are largely responsible for the institutional pathologies identified at both ends of Pennsylvania Avenue. There is no guarantee that simplifying the process will produce the desired effects if large budget deficits remain a prominent feature of our public finance. It is worth noting here that some of the states operating with biennial budgeting have seen fiscal stress force them into large-scale budgeting exercises in what are intended to be off years.
This is not to say that process changes are irrelevant. But it is crucial that reforms be crafted with an eye toward the underlying fiscal realities and the politics produced by those fiscal realities. It may well be worth trying to adopt process changes that reinforce and complement such constructive features of the present system as multi-year deficit reduction agreements between the president and Congress as well as the caps on discretionary spending and the pay-as-you-go (PAYGO) rules governing mandatory spending and taxes. Success in recent years in bringing the deficit down to more manageable dimensions, however, cautions against wholesale process changes whose consequences can’t be reasonably forecast. And since the major fiscal pressure in the years ahead, especially once the baby boomers begin retiring, is from mandatory programs not controlled by the annual appropriations process, a shift to biennial budgeting will do nothing to encourage an early and thoughtful consideration of adjustments needed in our social insurance programs.
The case for biennial budgeting is familiar and, at first blush, compelling. Compared with the present system driven by annual budget resolutions, appropriations, and (oftentimes) authorizations, a comprehensive biennial cycle is seen as encouraging long-term planning and evaluation over incremental budgeting, promoting greater predictability and efficiency in federal programs, eliminating excessive duplication and delay in congressional decision making, ensuring more vigorous congressional oversight of federal programs, reducing the enormous amount of busy work associated with the preparation of annual budgets in the executive branch, and discouraging congressional micromanagement through line items, earmarks, and other restrictions in appropriations bills.
Each of these objectives has merit, but as countless hearings and reports over the last fifteen years have made clear, there i s no reason to believe that they would be byproducts of biennial budgeting. Indeed, most careful students of budgeting have concluded that a switch to a biennial cycle is unlikely by itself to reduce either the workload or the deficit or to increase the quantity and quality of congressional oversight. For example, fiscal pressure and forecasting errors, to say nothing of the political and policy interests of members of Congress, could easily transform mid-cycle adjustments through supplemental appropriations into major budgeting exercises. By raising the stakes for members in the first year, a two-year cycle could boost the amount of time spent and controversy engendered, making the timely completion of budget legislation even less likely. And the loss of annual appropriation controls might well intensify pressure for congressional micromanagement rather than replace it with careful evaluation.
It is important to distinguish between the appropriations process, which every industrial democracy in the world continues to operate on an annual cycle, from other features of the budget process. As my colleague Allen Schick, one of the leading authorities on the budget process, has noted, other countries have experimented successfully with interyear flexibility and multi-year arrangements in which agencies are given discretion to carry funds over from one year to the next or funds sufficient to operate for several years. In the United States many funds appropriated in one year are available for obligation beyond the budget year. And with most of the budget exempt from the annual appropriations process and the rest constrained by multi-year spending caps, it is clear that an annual budget cycle is something of a myth.
Some changes in the process strike me as reasonable. For example, Congress might insist that all authorizing legislation extend beyond a single year. In addition, the budget resolution could be shifted to a two-year cycle, with basic fiscal policy guidelines set only once per Congress, it its first year. Biennial budget resolutions and reconciliation fit nicely with the discretionary spending caps and PAYGO provisions that were a part of recent budget packages. Taking a serious run at deficit reduction every two years seems sufficient. I am not pollyanish about the likely impact of these changes. Setting economic projections 33 months instead of 21 months out will necessitate frequent adjustments in the numbers, adding to the temptation to manipulate them for partisan or ideological purposes. But on balance, it would create the possibility at least of better time allocation.
The case for shifting the appropriations process to a biennial cycle is substantially weaker. I believe that it is desirable to retain an annual capacity to fine tune government policy in response to changing circumstances, mistakes in writing laws, and agency failure to carry out legislative intent. In reality most appropriations accounts are not thoroughly reviewed very year; those programs that are technically predictable and politically stable receive de facto multi-year funding. It is not obvious that formalizing this pattern would provide any value-added for long-term planning in the agencies. But retaining the capacity to deal with problems and to provide a guaranteed minimum amount of oversight through the annual appropriations process is worthwhile. As the same time, greater flexibility in management could be achieved through roll over authority and experiments in performance budgeting.
I am even less taken by proposals to dock members’ pay when the budget process is not completed on a timely basis and to mandate shorter congressional sessions in the second year of a biennial budget cycle, with members’ pay suspended if Congress fails to adjourn as required. These proposals are undoubtedly crowd pleasers, but would they improve the content and process of budget policy? The first proposal is based on the assumption that delays in completing the budget cycle result from the sloth and irresponsibility of individual members, the second on the assumption that policy making (read expansion of government) expands to fill the time in session available. I think both assumptions are flat out wrong and that the proposals, if enacted, would do more harm than good. They would reward wealthy members for whom congressional pay is merely a supplement to their annual income while reinforcing the myth that the government overspends because members of Congress out of touch with ordinary Americans are corrupted by the evil ways of Washington. Delays in the budget process reflect public dissensus over how to cope with deficits; they can’t be avoided without dealing with the underlying sources of stress. We have a full-time, year-round Congress because we have a large, complex federal government. If a smaller government is desired, it will take long, hard work by members of Congress, building public support, and writing new law to achieve it. Cutting their pay and sending them home is no substitute.
In sum, I support a biennial budget resolution and multi-year authorizations combined with annual appropriations bills. But I do so fully recognizing that these and other changes in the process are no panacea and that the difficult problems of budgeting will remain with us for the foreseeable future.