Revised March 16, 2001
How to Avoid Over-Committing the Available Surplus:
Would a Tax-Cut "Trigger" Be Effective or Is There a Better Way?
by Richard Kogan
View PDF version of this summary
View PDF of full report
View HTML of fact sheet
View PDF of fact sheet
If you cannot access the file through the link, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.
The Congressional Budget Office projects large surpluses over the next ten years $2.7 trillion excluding Social Security and Medicare. CBO says, however, that "the longer term outlook is unusually hard to discern at present. ... CBO's projections, like those of other forecasters, are based on just a few years' increased growth of productivity... projections of those recent changes as far as five or 10 years into the future are bound to be highly uncertain." Data CBO has provided suggest that even if Congress enacts no tax cuts and no spending increases over the next decade, there is a 35 percent chance the ten-year surpluses outside of Social Security and Medicare will be only half as big as projected and a 20 percent chance there will be no surpluses.(1)
In addition to the risk that current budget projections may be too optimistic, there is the risk that other, currently unforeseen circumstances will arise during the coming decade that increase budgetary pressure, such as faster-than-expected increases in health care costs or the possible emergence of a foreign military threat. Furthermore, even if the current projections prove precisely accurate, the projected surpluses are known to be temporary because the baby boom generation will start retiring before the end of this decade, ultimately leading to substantial increases in the cost of major federal programs such as Social Security, Medicare, and Medicaid. While projected surpluses may or may not materialize in the coming decade, long-term budgetary pressures that will arise when the baby boomers retire are more certain.
"Such enormous uncertainty about budgets just a few years in the future should influence policymakers' decisions about expensive, long-term commitments on the basis of mere projections." OMB, FY 2002 Economic Outlook, Highlights from FY 1994 to FY 2001, FY 2002 Baseline Projections, January 16, 2001, p. 26
Because of the great uncertainty in the projections, policy-makers could find that tax cuts or program increases enacted this year produce deficits outside Social Security and Medicare, rather than surpluses, an undesirable result given the need to increase national saving in preparation for the baby boomers' retirement. This paper examines and compares two strategies for minimizing the risk that enacting tax cuts or program increases will produce future deficits. (For ease of discussion, this paper uses the terms "surplus" and "deficit" to refer to the surplus or deficit in the federal budget excluding Social Security and Medicare Hospital Insurance (HI), except when the paper explicitly discusses the unified budget.)
- A "trigger," under which legislation cutting taxes or increasing mandatory programs also would include a statutory mechanism that automatically delays some or all of those tax cuts or program increases if the current budget projections turn out to have been too optimistic and deficits result.(2)
- A "budget reserve," under which this session of Congress would enact tax cuts or program increases that use up some but not all of the projected surpluses. To the extent that current projections hold true and surpluses materialize in later years, Congress could enact additional tax cuts or program increases in those years.
This paper concludes that a budget reserve would be much simpler than a trigger and far more likely to be effective in preventing a return to deficits. It is quite likely that a trigger will be evaded or will save far less than may be needed to avert deficits. Because of this likelihood, a trigger could lull policymakers into concluding incorrectly that a large tax cut that includes a trigger entails little or no risk of a return to deficits. Such a conclusion could lead to a bigger tax cut than would be enacted in the absence of a trigger, and therefore a greater risk of a return to deficits if, as is probable, the trigger proves largely ineffective for the reasons described below.
A Trigger vs. A Reserve: What Rudman and Rubin Say
On March 12, the Concord Coalition held a press briefing in which two former Treasury Secretaries (Pete Petersen and Robert Rubin), two former Senators (Warren Rudman and Sam Nunn), and Paul Volcker, former Chairman of the Federal Reserve, warned of the dangers of over-committing the projected surplus. Questioned about tax-cut triggers, Rudman and Rubin both argued that trigger would likely be ineffective and a reserve would be superior.
Senator Rudman remarked, "I probably have more experience with triggers than anyone in the room. We had a trigger in Gramm-Rudman, as some of you may recall. ... I can tell you from my experience that it's easy to mess around with triggers because people can fuss around with the numbers [or] they can ignore them. I mean, there was no criminal penalty for not pulling the trigger. If you didn't pull the trigger, the members of the Budget Committee weren't carted off to the federal penitentiary, so there was no forcing mechanism.
"A better thought, in my mind, rather than a trigger would be to ... put a more moderate tax cut in place over the next several years, and then ... if these surpluses pile up, [there is] no reason not to renew it and expand it. That would make a great deal of sense. It's certainly less risky.
"My sense is if you were to do that, [and] if there [subsequently] were problems in the economy, you can probably live with a more moderate tax cut. If you put the whole tax cut into place and the problems develop, you would have a serious problem. So... a better way [than a trigger] would be to have a rolling tax cut that was at a lower level..."
Secretary Rubin added, "A trigger sounds good conceptually. I think once you get into the practicalities of it, it seems to me there's a pretty good probability it's not going to work. Therefore, I agree with Warren, you start with a moderate tax cut, and two, three, four years down the road, if the projections materialize, you can decide what to do."
A trigger approach was tried under the old Gramm-Rudman-Hollings (GRH) law, under which automatic spending cuts were supposed to be triggered if the budget headed off a predetermined track and missed specific deficit targets set forth in the statute.(3) GRH failed, however, to produce anything close to the budget results the GRH statute ostensibly required; the intentions of the GRH law were thwarted by widespread evasions and gimmicks that characterized its brief career and prevented its trigger mechanism from working as intended. The failure of GRH illuminates the practical realities that a trigger tied to a tax cut would likely encounter. Any trigger mechanism, no matter how well designed, is likely to suffer from most or all of the following five problems:
- If the trigger depends on whether the budget for the coming fiscal year is projected to be in surplus or deficit, the Office of Management and Budget can use a rosy forecast to produce the desired projection of a budget surplus, thereby insuring that the tax cuts continue in full force. The Supreme Court has ruled that automatic, triggered changes in law must be based on estimates made by Executive Branch officials, so any trigger that is based on budget projections will put OMB and the White House in charge, rather than more politically neutral officials at CBO or the General Accounting Office.
- If the trigger is based on actual budget results such as whether the budget for the prior fiscal year ended up in surplus or in deficit Congress can readily make last-minute shifts in the timing of billions of dollars in federal payments. Such timing shifts may suffice to restore a surplus for the about-to-be-completed fiscal year even though they would do nothing to restore fiscal health. For decades, Congress has been adept at inventing timing shifts; recently, more than $11 billion in expenditures were first shifted by statute from 2000 into 2001 to make the 2000 budget look better, and then shifted back from 2001 to 2000 to make the 2001 budget look better.
- The trigger may be politically unsustainable. No matter how technically sound a trigger mechanism is, Congress may be unable to resist the public pressure to deliver "promised" tax cuts or program benefits. The public may never understand (or may forget) that the promised tax cuts (or benefit expansions) were supposed to be contingent upon the accuracy of the surplus projections.
- Even without rosy forecasts, legislated timing shifts, or political interference, a trigger mechanism is unlikely to work well.
One risk is that a trigger will save considerably less than is needed. For instance, although the tax cuts proposed by President Bush are phased in over a number of years, the vast majority are fully in effect by 2006. Yet most of the projected surplus occurs after 2006, and it is the projections for these later years that are the most uncertain. If these projections prove to be incorrect and deficits were to return after 2006, the trigger would provide almost no remedy because it would be too late to delay most of the enacted tax cuts; the tax cuts would already have been phased in fully by this time.
There are additional problems with a trigger that is based on observed deficits in the just-completed fiscal year. Such a trigger could fail to delay tax cuts because no deficit was observed in the just-completed year, even when the budget was evidently sliding toward a major problem. Or the trigger could delay tax cuts even when the observed deficit was just a one-year aberration and did not signal a continuing budget problem. If the one-year aberration was the result of a temporary economic slowdown, then automatically delaying the tax cut might be the wrong fiscal medicine, possibly making a recession more likely, deeper, or longer.
- A trigger is likely to have unintended consequences, since the numerous new budget rules that would have to be implemented to enforce the trigger could end up affecting the balance of power between the institutions setting budget and economic policy. For example, the Executive Branch was able to manipulate parts of the GRH law to further its policy goals and reduce Congressional influence. Because OMB estimates were used to measure the size of the "problem" that had to be addressed to avoid triggering across-the-board spending cuts, the Administration was able to set the parameters of the budget debate. The Administration could, for example, propose a series of cuts in discretionary programs and then define the scope of the "problem" as being similar to the amount of savings the proposed cuts would produce. Under a new trigger tied to tax-cut legislation, Congress could be placed in the position of having to chose between delaying tax cuts or acquiescing to Administration pressure to cut various domestic programs (and being blamed for blocking the scheduled tax cut if it did not approve the budget cuts).
To avoid triggering off a tax cut, Congress might even feel compelled to grant emergency power to the President to impound funds or delay expenditures. Alternatively, Congress could enact across-the-board cuts to annual appropriations bills, even if the fiscal problem that threatened to delay the next phase of a tax cut was caused by the surplus projections being overly optimistic (and the tax cut consequently having been made too large in the first place) or by the tax cut having unanticipated side-effects that resulted in its costing more than was anticipated. When it comes to last-minute budget cuts, appropriations bills often are the only available target.
If Congress were to cut the appropriations bills for a given year to avoid triggering off the next phase of a scheduled tax cut, however, that also would pose another problem. It would be making one-year budget cuts to enable a permanent tax cut to take effect.
Finally, a trigger could interact with the Federal Reserve's operation of monetary policy. When the Fed raises interest rates to stave off inflation, the normal result is an increase in federal interest costs, i.e., an increase in federal expenditures. Would it be appropriate for such an increase to trigger a delay in a tax cut? Would it be appropriate for such an automatic result to influence the Fed's decisions on monetary policy?
Does Tying the Trigger to Debt Levels
Rather Than Surplus Levels Solve These Problems?
Under any trigger, Congress must meet a specified budget target to avoid automatically delaying tax cuts (or mandatory program increases). Three possible types of budget target are:
- a specified surplus or a balanced budget excluding Social Security and Medicare Hospital Insurance;
- a specified surplus including Social Security and Medicare Hospital Insurance;
- a specified level of the debt held by the public.
The type of budget target chosen in a trigger makes almost no difference to this analysis. If Congress is required to meet a budget target based on OMB projections, OMB can create rosy scenarios to project the target will be met. This is true whether the target is for a specified level of surplus or for a specified level of publicly held debt. Likewise, if Congress is required to meet a target based on actual outcomes (such as the actual surplus or publicly held debt at the end of the just completed fiscal year), shifting the timing of billions of dollars in outlays or receipts a few days may suffice to avoid pulling the trigger. If a target cannot be evaded by optimistic forecasting or timing shifts, political pressure might induce Congress to abandon the trigger mechanism and allow the full tax cut or mandatory program increase to take place as scheduled. Even if a target is not evaded through rosy scenarios or timing shifts and is not abandoned because of political pressure, the trigger mechanism delaying those aspects of a tax cut or mandatory increase not yet phased in will likely save far less than needed to meet the target.
All these aspects of this analysis are equally applicable whether the budget target is the surplus excluding Social Security and Medicare, the surplus including those trust funds, or the debt held by the public.
Other aspects of the analysis may be affected to a modest degree by the type of target Congress chooses. The nature of gimmicks such as altered scorekeeping might differ slightly depending on the type of target. The unintended consequences also might differ.
Altered Scorekeeping. One simple way to evade a target tied to a specified level of the surplus is to declare by statute that a certain type of federal spending does not count against the target. This expedient will not help evade a target based on the debt held by the public, since it will not change the amount the Treasury has to borrow. On the other hand, there are other ways in which scorekeeping can be altered to evade a target for the debt. For instance, Congress could mandate that the term "debt held by the public" be limited to Treasury debt, i.e. that it exclude debt instruments issued by other agencies. About $28 billion in federal debt that currently is outstanding consist of debt was issued by agencies other than the Treasury, such as through FHA debentures and TVA bonds. If Congress acted to exclude agency debt from the debt target, it would achieve an immediate $28 billion windfall and would likely induce more federal borrowing to be undertaken by agencies and less by the Treasury. (This would increase the government's interest costs since markets demand higher yields on agency securities.)
More obliquely, disguised agency borrowing can be expanded and can be redefined as not constituting agency debt. Disguised borrowing occurs when agencies issue lease-purchase contracts, monetary credits, or other binding promises as a way of financing government activities, rather than simply paying cash up front. Currently, CBO and OMB attempt to see through such disguises and classify such transactions as "outlays financed by borrowing," but Congress or the administration could alter the scorekeeping so that such gimmicks would no longer "count" against the debt. Proposals for disguised borrowing were common during the Gramm-Rudman-Hollings era.
Unintended Consequences. For simplicity, this analysis has been written as though the goal is to avoid a deficit outside of Social Security and Medicare. If, instead, the goal is to hit a specified target for the debt, budget transactions involving Social Security and Medicare Hospital Insurance will become part of the calculation. One result would be that legislation to cut Social Security or Medicare benefits, or legislation to increase payroll taxes, would help meet the targets.
Another result is that loan guarantee programs would gain a favored position relative to direct loan programs. Currently, outlays for direct loan and loan guarantee programs are both calculated based upon the estimated subsidy they provide to a borrower. If the subsidy and administrative costs are essentially the same (as with direct student loans and guaranteed student loans), there is little budgetary reason to prefer one method of providing benefits over another. But although federal outlays and the surplus are the same whether direct loans or loan guarantees are chosen, the debt is not. With direct loans, higher federal debt occurs up front as the loans are made and lower debt occurs later when they are repaid; with loan guarantees, the same net cost does not show up until years later, when default claims are paid. Converting from direct loans to loan guarantees would make it easier to meet near-term debt targets even though such a conversion may have no effect on the surplus or on the government's long-term financial position. In essence, this would be another timing gimmick.
A third result is that the Treasury might be pressured to change its cash management strategies to avoid triggering a delay in tax cuts. Although the debt generally declines by the amount of the unified budget surplus, the relationship is not exact. One reason for this is that the Treasury keeps tens of billions of cash in banks, earning interest, rather than redeeming debt. The cash is needed to cover the difference between daily outlays and receipts, which are not fully predictable. If the Treasury finds itself close to the debt target at the end of the year, it may be under great pressure to run down its cash balances to an unusually low level (posing some risk of delays in payment on government checks) or to sell gold or foreign currency to avoid issuing debt securities (or to allow faster redemption of debt securities). Such cash management distortions would be another form of timing gimmick. They also could be costly if gold or foreign currency were sold at less than optimum prices.
A Budget Reserve
The case for a budget reserve was made recently by former Treasury Secretary Robert Rubin. In a New York Times opinion editorial, Rubin wrote, "[I]t is wise to be prudent we should avoid committing ourselves to dramatic courses of action that are hard to reverse... even if you favor a very large tax cut as the preferred use for [the] available surplus which I emphatically do not even a moderate degree of prudence would suggest waiting a few years to see whether or not the projected surpluses are actually occurring..."(4)
History suggests it is far easier to enact additional tax cuts and budget increases in coming years if the projections hold up or improve than to enact tax increases or budget cuts in future years if the projections prove to have been too optimistic.
A budget reserve does not suffer from the five defects of a trigger, as outlined above. A reserve merely requires that tax cuts or program increases be enacted partly in this year and partly in other, future years, rather than all at once; Congress can continue enacting tax cuts or program increases in future years so long as current projections prove to be valid and the surpluses actually materialize.
It is difficult for Congress to repeal tax cuts or mandatory program increases once they have been enacted. And for the reasons discussed here, it is unlikely a trigger would succeed in delaying tax cuts or mandatory program increases if a need to do so arose. In contrast, it entails much less political and budgetary risk to enact more prudent tax cuts or program increases this year, and then to enlarge them to the extent that future surpluses materialize. Holding some of the currently projected surpluses in reserve is both simpler and likely to be more effective than a statutory trigger in minimizing the risk of a return to deficits.
1. For a discussion of the degree of uncertainty surrounding CBO's projections, see Surplus or Deficit? Projections of a Large Budget Surplus Are Surrounded by a High Degree of Uncertainty, Richard Kogan, Center on Budget and Policy Priorities, Feb. 6, 2001. Also see Chapter 5 of The Budget and Economic Outlook: Fiscal Years 2002-2011, CBO, January 31, 2001.
2. A trigger based on a specified level of debt rather than a specified surplus or balanced budget is discussed in the text box on pages 5 and 6. This analysis concludes that whether the target is a specified level of the surplus or a specified level of the debt makes little difference in the operation of the trigger. For simplicity, this paper assumes that the event that would trigger a delay in enacted tax cuts or mandatory benefit increases would be a deficit outside of the Social Security and Medicare Hospital Insurance trust funds. This approach is consistent with a budget plan that would use the entire $2.7 trillion surplus on tax cuts, program increases, and attendant increases in interest costs. Alternatively, a budget plan might choose to use only part of the projected $2.7 trillion surplus for those purposes, in which case the plan would call for surpluses during the period 2002-2011. Under such a budget plan, the event triggering a delay in enacted tax cuts or mandatory benefit increases could be a surplus smaller than that called for in the budget plan.
3. Gramm-Rudman-Hollings refers to the Balanced Budget and Emergency Deficit Control Act of 1985 (P.L. 99-177 and its successor, P.L. 100-119)
4. A Prosperity Easy to Destroy, New York Times op-ed, Sunday, Feb. 11, 2001.