Hearing of the Committee on Rules
"Biennial Budgeting: A Tool for Improving Government Fiscal Management and Oversight"
Associate Professor,
Department of Political Science, University of Maryland, Baltimore County
Thank you, Mr. Chairman Dreier and members of the committee, for the invitation to testify on biennial budgeting. Twelve years ago, when I was working for the Congressional Budget Office (CBO), I wrote a long paper on this topic that was published by the Senate Governmental Affairs Committee. Discontent with the budget process was then at its peak, as was the deficit. Now, deficits are only a vanishing memory, yet complaints are still raised about the budget process. Some are as perennial as the daffodils that grace Washington this month: "the budget process is too time-consuming," "deadlines are usually missed," and "budget decisions are often repetitive."
In my opinion, such discontent is sometimes misdirected. A budget process that allocates one-fifth of the nation's gross domestic product cannot run as smoothly as the Swiss train system. That would also be undesirable, for democracies work only when elected officials argue about how government spends, taxes, and borrows. To paraphrase the wisdom of Finley Peter Dunne's fictional Mr. Dooley, "budgeting ain't beanbag." I suggest that we should celebrate that.
Yet the budget process does have serious flaws, and the Congress should consider how to improve it. Twelve years ago, my paper used the typical "on the one hand, on the other hand" approach of the CBO, but on balance it suggested that biennial budgeting had serious disadvantages. In contrast, my testimony today is somewhat more favorable, because budgeting has changed in two significant ways: there is much less interest in using the budget for discretionary macroeconomic stabilization, and multiyear budget agreements have become common practice.
Responding to Unanticipated Events
Opponents of biennial budgeting often argue that it would prevent the Congress from responding quickly to unanticipated events. Since this is implausible regarding national security crises, the typical specter is a recession hitting hard in the second-year of the biennium, with the Congress having to wait a year to respond. But think back to 1993, when the Congress rightly killed the President's so-called "economic stimulus" bill. That proposal was minuscule in amount and offset by deficit reduction proposals, poorly targeted, and timed too late in the recovery from the recession. In fact, the long record of discretionary counter-cyclical fiscal policy is dismal; economists showed long ago that relying instead on the budget's automatic stabilizers is a preferable response, and most elected officials now agree. If biennial budgeting would further deter the Congress from displacing the Federal Reserve's critical role in stabilizing the economy, this is a strong argument for its adoption.
Note in this regard that some opponents of biennial budgeting also observe that more states have switched from biennial budgeting to annual budgeting than in the reverse direction, in part to provide maximum flexibility to meet the states' balanced budget rules during a recession. From the macroeconomic perspective, however, state balanced budget rules are strongly procyclical and therefore dysfunctional for stabilization. State balanced budget rules should not serve as a model for the federal government. Nor is it likely that most Members of Congress would want to copy a typical mechanism for state fiscal flexibility–many governors have impressive powers to cut authorized spending without legislative involvement; should the President have such powers as well?
Promoting A Medium-Term Perspective
The second reason that biennial's time may have come is that the country now has enough experience with multiyear budget agreements to understand how long they should last. You may remember the harsh argument in 1995 and 1996 about whether 7 years or 9 years was the right time frame for a balanced budget agreement. In retrospect, the extraordinary economic performance of the past few years makes this dispute seem a bit overdone. Nevertheless, that many-year agreement was preferable to a one-year agreement, because it acknowledged the political need for a fiscal glide path. But now that deficits have been eliminated, a biennial preiod may provide even greater political realism than a many-year period. The two-year electoral cycle for the House is the timing belt for our political system; comprehensive political agreements do not usually last much longer than the next election.
Biennial agreements may also be more realistic from the economic perspective. Budget experts have long cautioned that the annual budget is myopic, so it is good that budget proposals now project effects for five, ten or more years. However, it is clear that a desire for good budgeting is not the only cause of this practice; assuming a strong economy over a long period provides more-than-enough resources to finance many proposals. The problem is that the great sensitivity of these long-run projections to some critical assumptions is being ignored. When projections are too optimistic, the government can approve too much spending and cut taxes too much, quickly frittering away the fiscal discipline that has been purchased at high cost. Using long-run forecasts poorly may be worse than the myopia of the annual budget; biennial budgeting's implicit focus on the next two years may be more prudent than current practice.
The Inevitability of Late Decisions
If biennial budgeting would promote budgetary planning over the medium-term, it is still unlikely to make budgeting timely. It is human nature to postpone difficult decisions until a deadline approaches. Since a two-year budget would require more decisions than a one-year budget, biennial budgeting threatens to increase the problem of missed deadlines.
Now it is true that a single-issue bill is sometimes more difficult to pass than an omnibus bill. And it's possible to imagine that some legislators would oppose a one-year budget because their projects weren't included, but would support a two-year budget which added their projects in the second year of the biennium. The second budget year would permit the budget's managers to make commitments that are more credible than "wait until next year." However, it is doubtful that this benefit would exceed the costs of negotiating another year of budget allocations.
Expecting a biennial schedule to speed up budgeting forgets that legislatures are deliberative bodies–they usually take time to reach conclusions. You may know that a very significant percentage of health care spending occurs in the last week of a typical person's life, even if that person has had good preventative care. Similarly, legislatures make many important decisions the week before sine die–not earlier, regardless of good intentions.
Limits on "Emergency" Supplementals
Predicting other potential effects of biennial budgeting is more difficult. Some of the claimed effects–both positive or negative–appear overstated to me. This is because the biennial bills would make no changes to other aspects of the process, or more importantly, to critical incentives faced by Members, Senators, or the President.
The most popular approach to biennial budgeting would use odd-numbered calendar years to pass a budget resolution and appropriations bills, both of which would cover the next two fiscal years. In even-numbered calendar years, the Congress would turn to oversight, authorizations, and campaigning. However, unanticipated situations would surely require the Congress to spend part of the even-numbered calendar year adjusting some of the appropriations made in the previous year. The risk is that some would be turned into most. A nearing election day, and the desire to take popular positions just before it, would stimulate many Members and the President to propose significant budget amendments in hopes of catching the last train. Consequently, there is a very real potential for a huge supplemental that would dominate the spring and summer of the even-numbered calendar year. Therefore–and I hesitate to suggest this because of the budget process is already overburdened with rules–the committee should consider supplementing any biennial budgeting bill with the part of H.R. 853 that dealt with emergency spending.
A Joint Biennial Budget
Another proposal in H.R. 853--converting to a joint budget resolution–should also be considered in conjunction with biennial budgeting. The concurrent budget resolution was invented with the hope that the Congress could plan as well as enact budgets, but history has shown that planning is a more demanding task for the Congress than it is for the executive branch. Similarly, while the traditional view of appropriations bills is that they are "must-pass vehicles" which equalize the bargaining positions of the Congress and the President, I understand that many in the Congress now feel otherwise. One reason the Congress has great difficulty passing the budget resolution and the appropriations bills is that the Congress does not seek and reach an early accord with the President over the budget's general parameters. I will admit that a joint budget resolution process would not guarantee that the branches could reach such an accord, but there is little risk in trying. It may be overly ambitious for the Congress to go it alone for a two-year budget.
Committee Reorganization and Earmarks
A final hope for biennial budgeting is that it would enable the Congress to better design programs and oversee agency performance. These would be extremely valuable outcomes, but they will not be produced simply by reducing the time spent on budgeting. In fact, as Members of the Appropriations Committee have correctly pointed out to your committee, authorizing committees are not the only committees that oversee agencies, and appropriations bills are powerful message senders to agencies.
Members of the Rules Committee know better than anyone on Earth the institutional disadvantages of turfing between authorizing and appropriations committees and between different authorizing committees. The logical implication is that some committee overlaps should be reorganized out of existence, as Chairman Dreier has suggested in the past. A more radical approach would be to combine authorization and appropriation activities within committees specialized by functions of government. This would help the Congress link plans, performance measures, and budgets as envisioned by the Government Performance and Results Act (GPRA). It would also better connect committee allocations in the budget resolution to the resolution's consideration of priorities among budget functions. While I understand the huge political barriers in the way of this proposal, I submit that moving to a logical committee structure would more likely improve policy design and oversight than would changing the budget schedule.
Advocates of biennial budgeting often complain that excess time spent on budgeting reduces oversight activity in the Congress. While the Congress is actually the world leader in quality legislative oversight, there is no question that it can improve oversight of agency performance, and also its earlier authorization of programs. Again, GPRA provides a logical framework for increasing the clarity and consistency of agency goals, measuring how well agencies attain these goals, and connecting goals and performance to budget allocations.
Some of the biennial budgeting bills would adjust GPRA by linking two-year availability of funds to two-year performance reports. While this is a necessary condition for success, it is insufficient. At least as necessary is good faith between the branches. Over the past few years, the Congress has released some critical reports of agency plans and performance measures. In many cases, the complaints were justified. From the agency perspective, however, Congressional complaints would be more credible if the Congress spent less time stuffing appropriations and authorizing bills and reports with earmarks that are guaranteed to reduce agency effectiveness and efficiency. Members naturally want to deliver benefits to their individual districts, and to some degree, it is useful for the Congress and the country when they succeed. But if self-restraint is the only major barrier to earmarking, the practice can get out of hand. If the Congress wants to promote better program authorizations and oversight, it needs to make earmarking much more difficult. Without that easy option, more legislators will develop the expertise in programs and agencies that is essential for our democratic system to function well. A simple change to the timing of the budget process may help promote oversight, but it is more important to create the conditions under which oversight is highly valued by both Members and their constituents.
Conclusion
In conclusion, the committee should seriously consider biennial budgeting. Multiyear budgeting can properly focus attention on the controllable medium-term, but passing a simple biennial budgeting bill will not ensure this. Any shift to biennial budgeting should be coordinated with other complementary changes to the budget process, and a modified H.R. 853 would be a good starting point for such an omnibus bill. Biennial budgeting could also better connect decisions about budget allocations to related decisions about program goals and agency performance. However, it is unlikely that biennial budgeting will have this effect unless the Congress revisits the question of whether committee jurisdictions and responsibilities should be modified.
Thank you, and I'd be glad to answer any questions you might have.
