Skip Navigation

Chairman Nussle, Ranking Member Spratt, and Members of the Committee, thank you for inviting me to this hearing. I am happy to be here this morning to discuss mechanisms of budgetary discipline—something that I have been actively involved with for more than 20 years.

During those years, I have seen the budget process change dramatically: from the one-year budgets of the late 1970s, to the budget deals between Congressional leaders and President Reagan of the early and mid-1980s, to the aggregate deficit control mechanisms of Gramm-Rudman-Hollings (GRH) in the late 1980s, and finally to the three versions of the Budget Enforcement Act (in 1990, 1993, and 1997).

As an active participant in devising and enforcing many of those mechanisms for budgetary discipline—and, more recently, as an observer as other people have tried their hand at enforcement—I believe that I have a sense of what characteristics make a mechanism more likely to be effective or more likely to fail. On the basis of that experience, I feel that four principles are required for any budget enforcement mechanism to succeed.

• Shared Goals. For a system of discipline to be effective, its overall goals must be broadly shared: by the Congress and the President, by Republicans and Democrats, by the Senate and the House, even by the major committees (budget, appropriations, tax, and authorization). An example of the need for shared goals can be found by comparing the Budget Enforcement Act (BEA) at its inception in 1990 with virtually the same BEA seven years later. The goal of eliminating large and growing deficits through limits on spending was shared by the majority of political players in 1990 and during the first half of the decade. As the economy continued to expand, revenues came in at unexpectedly high rates, which—combined with the end of the Cold War, the end of the thrift bailouts, and the BEA’s limits on spending—produced unanticipated surpluses. Suddenly, the shared goal of deficit reduction had been achieved, and the willingness of the President and the Congress to adhere to the restraints of the BEA withered, even though the law was virtually unchanged.



• Realistic Assumptions. Budgetary assumptions—particularly for five or 10 years into the future—cannot be much more than educated guesses. Some demographic trends can be projected with good accuracy, but precision in forecasting has never been possible over more than a very short period of time. Nevertheless, overly optimistic assumptions about economic or technical factors (such as the timing of spending) can discourage policymakers as the unrealistic targets are missed by wider and wider margins.



• Appropriate Sanctions. The BEA's mechanism of spending caps and pay-as-you-go (PAYGO)—and especially the sequestrations available to enforce it—was a big improvement over GRH's mechanism of deficit caps and sequestrations, largely because of the appropriate application of the sanctions for violating the limits. Under GRH, a sequestration might have been required of discretionary programs, for example, because of economic factors unrelated to the caps, even if the appropriators had not exceeded their appropriation targets. Under the BEA, by contrast, no sequestration of discretionary programs would occur unless the appropriations committee exceeded its limits. Note that the enforcement mechanism for the spending caps is weakened when lawmakers use special provisions to protect one class of programs from sequestration. Limiting the programs subject to sequestration makes the burden on the remaining programs heavier, thus providing a greater incentive to waive the sequestration entirely.



• Availability of Safety Valves. Exemptions for emergency spending (for events such as natural disasters, wars, and recessions) can strengthen a mechanism for budgetary discipline if the exemptions are applied fairly and honestly. For example, during the first seven years of the BEA, emergencies were limited to natural and economic disasters, as had been defined in 1991. That system broke down in the late 1990s, however, as the definition was discarded and any semblance of discipline abandoned.

If the goal of the current Congress is to retain the discipline of the caps and PAYGO mechanism but gear the specific targets so that all surpluses stemming from Social Security receipts are used to pay down debt (in other words, so that there is no on-budget deficit), I would remind the Committee that an earlier model exists. The 1990 BEA was drafted with that same goal in mind. Besides the spending caps and the PAYGO mechanism, it also contained a provision that established declining targets for the on-budget deficit. That provision could be used as the foundation for procedures to enforce on-budget balance.

The precise caps for a new mechanism would need to be worked out, of course, and the political agreement required to implement the new regime of budgetary discipline would not be easy to obtain. However, the laws to implement a new agreement that protects the Social Security surplus already exist. And, except for the aggregate deficit mechanism, they were used successfully for the first several years of the BEA.

A final point: I spoke to this Committee last year about a number of conceptual problems that I thought needed to be addressed by a new budget concepts commission. None of those problems have been addressed, and some may have gotten worse. No matter what new regime of budgetary discipline results this year, I continue to advise that a new budget concepts commission is necessary.