Hearings of the Committee on Rules
"Biennial Budgeting: A Tool for Improving Government Fiscal Management and Oversight"
Professor Charles Whalen
Senior Institute Economist, Institute for Industry Studies, Cornell University
1. In your prepared testimony you state that "an international perspective on this issue generates two findings worth noting. One is that legislators devote considerably less time to budgeting in parliamentary democracies - interminable budgeting is largely an American phenomenon. The other finding is that annual budgeting is a product of custom and tradition, not fiscal theory." Can you elaborate more on these two findings?
i) Parliamentary democracies feature a more unified government structure. Critical budget decisions are negotiated within the cabinet; parliaments usually debate and enact a budget in only one or two months.
Moreover, reliance on a parliamentary structure is not the only way other nations achieve budgetary stability. Nearly all employ multiyear budgets for certain expenditure categories such as military spending. In addition, some nations, such as Sweden, place annual budgets in the context of multiyear budget frameworks [see, for example, Allen Schick, The Capacity to Budget (Washington, DC: Urban Institute, 1990, p. 220]. ii) The roots of annual budgeting are not found in economic or political theory. Rather, this fiscal period was initially employed because it corresponds with a full cycle of nature's seasons. The practice of annual budgeting may have considerable merit in an agrarian society. It loses much significance, however, in a highly industrial nation.
At the state level, trends in fiscal practices moved from an initial heavy reliance on annual budgeting to almost exclusive use of biennial budgeting (and biennial legislative sessions) by the early Twentieth Century. After World War II, many states reverted back to annual budgeting. Since 1986, four states have turned once again toward biennial budgeting. Today, 23 state budgets have at least some biennial components.
2. You discussed the Defense Department experience. Since the Defense Department continues to be required by law to submit a biennial budge request, would the Pentagon be a good first candidate for a transition from an annual to a biennial process? Also, since this submission requirement still exists, has any of the benefit of this preparation remained for the Department even though they have received annual appropriations?
In the mid-1980s, Representative Les Aspin and Senator Sam Nunn – respective chairs of the House and Senate Armed Services Committees – concluded that a two-year defense budget could help their committees focus less on the details of budgeting and more on defense oversight, direction-setting and policy analysis. Defense Department managers also saw biennial budgeting as an opportunity to devote greater attention to program evaluation and agency planning. (See Robert Art, "The Pentagon: The Case for Biennial Budgeting, Political Science Quarterly, 1989.)
According to research by Professor Robert Art (cited above), the Defense Department's submission of biennial budgets yielded benefits in the form of increased agency evaluation, better planning and reduced procurement costs – despite the continued reliance on annual appropriations (see my testimony and the Art study for further details). A study that takes an updated look at the Defense Department experience would indeed be worthwhile.
Two comments are warranted in response to the question of whether the Pentagon's budget would be a good first candidate for a transition to biennial budgeting. On the one hand, oversight and planning needs seem to make the Defense budget a good candidate – and there are certainly many highly predictable elements within the defense budget. On the other hand, overall defense spending is widely considered one of the most difficult items to place on a two-year cycle, since some military expenditures are often neither predictable nor deferrable.
3. You suggest that government contractors, grantees and program recipients will benefit by program managers having more resources and ability to devote to focusing on improved program results. Could you give some specific examples of how this would be the case?
Contractors, grantees and program recipients benefit from greater certainty with respect to funding over time. As indicated in Professor Art's study, contractors "can strike more advantageous deals" with subcontractors and parts suppliers – and thus reduce procurement costs – when operating with a multi-year time horizon. Moreover, instability of weapons-system production also increases costs for both the government and its contractors.
In my testimony, I make reference to remarks offered by Dr. Alice Rivlin. On this topic, she writes: "Transfer payments, especially for the elderly, need to be predictable, so that people's lives are not disrupted. Military capability suffers if signals change too frequently. . . . Moreover, service delivery – whether roads, schools, health, or whatever – requires planning time. Frequent change and uncertainty of funding are costly and disruptive. Annual review of research grants increases the ratio of proposal writing to actual work. Indeed, almost all programs would work better if authorizations and appropriations were enacted for several years. Exceptions would appear to be few: natural disasters; military actions; and counter-cyclical programs where triggering mechanisms are not appropriate" (Challenge, March-April 1981, p. 36).
4. In your prepared testimony you state that "as for the possibility of a correlation between biennial budgets and increased reliance on supplemental appropriations and rescissions, evidence from older studies is mixed while the two most recent studies (published in the mid-1990s) find no correlation." Can you elaborate more on the findings of these studies?
In 1972, a report by the Council of State Governments indicated legislators in states that switched to annual budgeting identified a need for fewer emergency appropriations as one reason for keeping the new system (Annual or Biennial Budgets?, p. 15). A 1984 study, conducted by researchers at Texas A&M University, interviewed officials and observers in states that moved to annual budgeting. That study reported no discernable link between the budget period and the need for supplemental appropriations (Annual versus Biennial Budgeting?, section 3, pp. 13-14).
A 1987 General Accounting Office (GAO) report provides mixed results on the basis of interviews conducted with state legislative and executive officials. "No strong pattern" on the use of budget adjustments emerged from the interviews. Nevertheless, the report did find "a slight indication" that some states which moved to annual budgets experienced fewer adjustments while some which moved to biennial budgets experienced an increase in adjustments (Budget Issues: Current Status and Recent Trends of State Biennial and Annual Budgeting, pp. 14, 19, 25).
Another GAO report with information relevant to this question was released in 1993. After stating that 49 states must keep their budget balanced, the report indicated 85 percent of annual budgeting states took action to close a budget gap in the prior budget period, while only 47 percent of biennial budgeting states did so. (See Balanced Budget Requirements: State Experiences and Implications for the Federal Government; percentages based on Tables 3.1-3.3.)
According to a 1994 report by Ronald Snell of the National Conference of State Legislatures, supplemental appropriations since 1989 have been driven by unpredictable national economic developments that affected all states regardless of the budget period employed (Annual and Biennial Budgeting: The Experience of State Governments, p. 8).
5. In your testimony you point to the need for "provisions allowing for changes in response to major unanticipated developments and emergencies." Could you elaborate on what these "provisions" might look like?
Supplemental appropriations and rescissions should be permitted each year. Just as the current process permits budget adjustments during the fiscal period, so to should biennial budgeting.
6. One argument against biennial budgeting is that it would exacerbate the need for supplemental spending bills, which can tend to balloon as they move through the process. Do you have any suggestions for the Committee about how to control the supplemental process to ensure that we do not slide back into a defacto annual process should a biennial budget be implemented?
Avoiding over-use of the supplemental appropriation process requires minimizing the need for "must-pass" supplemental bills and controlling legislators' ability to attach extraneous amendments to such bills. One way to avoid the need for "must-pass" supplemental bills is to create "rainy day" funds for certain events such as natural disasters. (Representative Obey has even taken this idea a step further and suggested creating a system by which states could ensure themselves against such disasters.) Dan Crippen, Director of the Congressional Budget Office, suggests another alternative worth considering: including procedures in the budget resolution that would allow budget totals and allocations to be updated automatically for changing conditions.
7. You state in your testimony that you agree with Dr. Alice Rivlin in a conclusion that, under a biennial budget model, we should "minimize unexpected changes in tax and spending policies." Does this mean that you would argue that the process should seek to prevent the consideration and enactment of reconciliation bills in the non-budget year?
Reconciliation bills should generally be passed in the budget year. Freestanding revenue measures and authorization legislation would be considered in the non-budget year -- and their changes would become effective (except in extraordinary cases) at the start of the next biennium.
8. As you note in your remarks, a major argument in support of biennial budgeting is that it would free up the Congress and its committees to focus on more in-depth oversight during the non-budget year. We have plenty of evidence that under the current system, major portions of the budget go from year to year without authorization. Do you think that we need to make additional process changes, in conjunction with enacting a biennial cycle, to strengthen the authorization process?
There are presently a number of ways to circumvent House rules that were designed to ensure authorization precedes appropriation. With biennial budgeting, it may be desirable to make such circumvention more difficult.
9. Some have suggested that biennial budgeting will cause agencies to seek greater spending cushions for the second year of a biennial budget. You discussed this in your written testimony and stated that "published examinations of state budgets do not support this contention." Could you comment further on the findings of these examinations in this area?
In 1984, Texas A&M researchers conducted interviews with officials and budget observers in states that had moved from biennial to annual budgets. They report "no agreement among the respondents as to the effects of [a switch to] annual budgets on the need for agency cushions" (section 3, pp. 13-14).
In 1994, Ronald Snell of the National Council of State Legislatures (NCSL) considered whether the budget period affects state spending. In particular, he examined evidence summarized in a prior NCSL report and a more recent academic study. The prior NCSL report, which reviewed the Texas A&M research and four additional studies, produced the conclusion that the budget period does not have any clear effect on state spending. Snell added that the more recent study "failed to find strong evidence on either side of the issue" (Annual and Biennial Budgeting: The Experience of State Governments, 1994, p. 8). [Snell's references: Barbara Yondorf, Annual versus Biennial Budgeting: The Arguments, The Evidence, NCLS, 1987; Paula Kearns, "State Budget Periodicity: An Analysis of the Determinants of the Effect on State Spending," unpublished paper, Department of Political Science, Michigan State University.]
10. What role do you see the authorization process currently playing in the budget process and do you think that role should change? Do you think changing or strengthening its role would improve the role and condition of programmatic oversight?
I would agree with former Representative Lee Hamilton, who recently argued that oversight is currently underused and its value is often underestimated. The two-stage authorization-appropriation process makes sense only if authorization committees engage aggressively in oversight. Biennial budgeting provides an opportunity for congressional committees to conduct such oversight – and to introduce more long-term thinking into policymaking.
Biennial budgeting enables Congress to make full use of agency performance data required by the Government Performance and Results Act. A legislative session devoted primarily to oversight, evaluation and authorization is consistent with the bipartisan goal of devoting greater attention to measures of program outputs and effectiveness. In short, biennial budgeting is consistent with the goal of making government more efficient.
