Skip Navigation

Thank you and good morning.

The budget environment today is reminiscent in many ways of the state of the budget process in the late 1960s. Then, as now, there was (and I quote from the 1967 Report of the President's Commission on Budget Concepts) criticism of the budget based on:

• Confusion arising from the number of competing concepts of budget totals currently used or stressed in the President's budget message and the relationships between them;
  
  
• The appropriate accounting treatment of individual items or groups of items and the effect of such treatment on the budget totals; and
  
  
• The search for better congressional and public understanding of the budget program and more up-to-date availability of budget information.

At that time, confusion about the competing budget totals was caused by the fact that budgetary information was presented in three separate ways: in a national income accounts budget, an administrative budget, and a consolidated cash budget. Today, confusion is just as pervasive about differences between the on-budget budget, the off-budget budget, and the unified budget. In the late 1960s, technicians and policymakers were dissatisfied with the budgetary accounting of federal loans, business-type governmental receipts, and means of financing the deficit other than borrowing. Their counterparts today are equally concerned about the budgetary treatment of tax credits, such as the proposed universal savings accounts, and a large number of purely financial transactions, including transfers of Treasury securities to trust funds and federal purchases of private marketable securities. Finally, on the need for better understanding of the budget by the Congress and the public, it is difficult to believe that there could have been less understanding then than now of, say, trust funds generally and of the Social Security, Medicare, and transportation trust funds in particular.

The 1967 commission effectively addressed the most pressing budget issues of that day. And we continue to benefit from its legacy, perhaps most especially from its forceful case for a single federal budget that comprises all federal fiscal activity. But the commission completed its work over 30 years ago. As our children often remind us, that was then, this is now. The budget process and the environment in which it functions have changed so considerably in the intervening years that I believe we need a new, nonpartisan commission on budget concepts now.

In 1967, there was no explicit Congressional budget process and, of course, no Budget Enforcement Act. We had not yet encountered the fiscal stress of a runaway budget deficit. The notion of uncontrolled mandatory spending was primarily of academic interest. Baby boomers were entering the labor force, not preparing to retire. The tax code and most federal policies seemed much simpler then.

Today, the need for a comprehensive, single budget remains evident. Yet our commitment to that concept does nothing to settle new questions of when and how federal activities are to be scored in the unified budget. Consider, for example, three conceptual issues that I regard as critical but currently unanswered.

First, what is the "right" budget horizon? It is clear that the choice of budget horizon makes an enormous difference to policy decisions. For the 1967 commission, and for Presidential budgets thereafter, it may have seemed natural that the budget would present data for the past year, the current year, and the budget year. Experience has demonstrated, however, that a single budget year combined with emphasis on a single, bottom-line measure of the deficit or surplus creates incentives to craft legislative proposals that defer budget costs until the next year. The Congress responded to that development by lengthening the budget horizon to five years. For a variety of reasons, both the Office of Management and Budget and the Congressional Budget Office extended the horizon to 10 years. In his budget proposal for fiscal year 2000, the President adopted a 15-year horizon. For Social Security, the House of Representatives now uses a 75-year horizon.

Depending on the budget period, one can argue that today the United States has a substantial surplus, a small surplus, or an unsustainable deficit. One can similarly argue that the government either has resources available for new spending initiatives and tax cuts or faces a fiscal black hole. Informed public and policy discourse cannot easily handle differences of that magnitude for a single, unified budget. That strikes me as closely analogous to the situation that policymakers and the public faced in the late 1960s.

Second, should we continue to selectively add more accrual concepts to the predominantly cash basis of accounting in the budget? Much of the budget is now allocated to deferred payment programs, including Social Security, Medicare, federal retirement, and insurance programs. For such programs where monies are collected in one year in exchange for promises to make payments later, cash-basis accounting has the potential to mislead policymakers and the public alike. The issue of cash versus accruals is also closely related to the problem of which budget horizon to use. We can focus on single-year budgets, provided that those budgets reflect accurately the future effects of events and actions in the budget year.

Accrual accounting, however, may be more susceptible to manipulation and distortion than cash accounting. To warrant consideration today, an accrual system would need to be able to measure accurately the long-term cost of the current year's accumulation of Social Security, earned retirement, and Medicare benefits. Such a system should also be able to recognize the expected costs of promises extended today for deposit, pension, and disaster insurance programs. That is a tall order with substantial downside risks, but it is also worthy of the best efforts of a new commission on budget concepts.

Third, how do we address the task of distinguishing policies that increase federal expenditures from policies that reduce federal receipts? Thirty years ago, the division between tax expenditures and spending programs seemed much cleaner. Expenditures were payments made to nonfederal entities, either as payments for goods and services or for grants and other transfers. Tax cuts were reductions in revenues stemming from a drop in tax rates or a narrowing of the tax base. In the past 30 years, however, we have developed and refined the tax credit to the point where it is an almost perfect substitute for a direct spending program. The current requirements for turning a spending program into a tax credit seem minimal. All that is needed is legislation that amends the tax code and uses the tax return to establish eligibility for benefits and to calculate their amount. Legislation can be classified as a tax reduction even if the benefit is unrelated to the amount of taxes otherwise due. Accordingly, tax credits are now provided for such diverse activities as producing "alternative" fuels, reforestation, education, and income support.

My concern here is not simply to call things by their right names but, consistent with the recommendations of the 1967 commission, to disclose accurately the size of the government relative to that of the economy. Classifying spending programs as tax reductions lowers both outlays and receipts and, thus, understates the size of the government by potentially significant amounts. For example, if all tax credits were reclassified as expenditures, then spending and receipts would rise by about 0.4 percent of gross domestic product (GDP). For fiscal year 2000, projected spending and receipts would be about 19.8 percent and 21.1 percent of GDP, respectively. Under this concept, receipts could reach a historic high, exceeding their peak during World War II. I am not asserting that tax credits should be classified as spending; I am only suggesting that the issue warrants consideration by a new commission on budget concepts.

In addition to the tasks of specifying the appropriate time horizon for the budget, choosing between a cash or accrual basis of accounting, and providing a basis for distinguishing changes in receipts from changes in outlays, there are many other budget issues that need to be resolved. They include:

• The disparate treatment of seigniorage on currency (counted in the budget as a receipt) and on coinage (counted as a means of financing);
  
  
• The appropriate budgetary accounting for potential arbitrage gains and losses to the government from borrowing at the risk-free rate and investing in risky assets;
  
  
• The general applicability of probabilistic budget scoring (of so-called one-sided bets);
  
  
• The budgetary treatment of assets acquired by the government through long-term leases; and
  
  
• The possible inclusion in the budget of nonappropriated fund instruments of the federal government, such as military PXs.

I claim no monopoly on budget issues that need to be resolved by an expert and authoritative body. My guess is that all of you here today could fill the agenda for a dozen such commissions.

In the current budget debates, we are hearing a lot about various gimmicks to move spending out of fiscal year 2000. Public scrutiny of those proposals is important to make sure that the substance and consequences are fully understood. But as important as our collective review of those proposals may be, that importance pales in comparison with the need to assess the larger conceptual issues: What should be measured in the federal budget and when? What exactly is an outlay and what is a receipt? What is a federal trust fund? How should the budget recognize long-term commitments?

Questions of such importance cannot be answered by technicians alone. The policy implications are too broad and far-reaching. Establishing a new, nonpartisan commission on budget concepts would greatly enhance and bring order to the current debate on reforming the budget process.