be made one year in advance. At present, federal grants for a given year are often appropriated after state and local governments have enacted their budgets, making it almost impossible for them to plan their federally aided programs. (b) Programs should be shifted to a three-year authorization basis. Some are now authorized annually, some for two years, and some for three years or longer. The programs in the civilian budget should then be divided into roughly three equal groups, with one group authorized each year. A complete review of each program every year is impractical, and facing up to this fact by shifting to three-year authorizations might produce more realistic and effective program reviews by both the Congress and the executive branch. (c) Annual appropriations should be converted into three-year appropriations where this is consistent with the nature of the program involved. In some programs, multiyear appropriations would serve little purpose. For example, in programs that pay benefits based on statutory entitlements-such as public assistance, veterans' pensions, and Medicaid—the size of the appropriation does not determine how much is spent; the appropriation simply provides the funds that must be paid to beneficiaries under the law. Similarly, where discrete projects are being funded-a dam or a federal building -appropriations should cover the cost of the entire project regardless of how long it takes to build. But in many other programs of a continuing nature, in which the level of the program can be controlled by the appropriation, conversion from annual to three-year appropriations might be a desirable accompaniment to the three-year authorization process. (d) Every appropriation bill, as reported to the House and Senate by the Appropriations Committees, should include a fiveyear projection of expenditures on the programs financed by the appropriation, assuming the continuation of present policies and practices.

6. Periodically throughout the congressional session, estimates should be prepared and published showing how prior congressional actions on authorizations and appropriations during the session up to that time had affected the five-year budgetary outlook submitted by the administration. At present, a little-known group-the Joint Committee on Reduction of Federal Expenditures, composed of the ranking members of the Senate Finance Committee, the House Ways and Means Committee, and the Senate and House Appropriations Committees, the secretary of the treasury, and the director of the Office of Management and Budget-issues a periodic "scorecard" showing the effects of congressional action on the President's budget requests,

mainly in terms of the current budget year. Perhaps this committee, with appropriate staff resources, could modify its procedures to perform the task described above.

Converting annual into three-year authorizations and appropriations would raise several questions. Would the major substantive and appropriations committees (or subcommittees) of the Congress have nothing to do for two years and then face a major legislative agenda in the third year? Since each committee deals with several programs, the three-year authorizations and appropriations could be so scheduled that each committee would have a work agenda every year. Would the conversion reduce flexibility to adjust expenditures in the light of economic circumstances in accordance with fiscal policy? Even now, with annual appropriations, most expenditures cannot quickly be varied. Through the use of budget amendments to raise or lower previously enacted three-year appropriations, the administration or the Congress could provide the same degree of counter-cyclical flexibility as now exists.

Would a three-year authorization and appropriation cycle hamper an incoming administration in carrying out the objectives for which it was elected? Again, the new procedures would not be an obstacle. A new administration would not have to wait two or three years to make modifications; it could submit budget amendments. More important, however, the new procedures would encourage incoming administrations to realize that effective changes in priorities can be made only in a long-term context, and would help concentrate their attention on making such changes. It is possible, of course, that annual amendments would become so commonplace that annual authorizations and appropriations would, in effect, be restored. Self-restraint in the use of annual amendments by the Congress and the executive branch would be necessary to prevent subversion of the new system. But any procedural reform encounters this difficulty. There is no magic formula to ensure that improved budget procedures will withstand a joint effort by both the Congress and the executive branch to thwart them.

These proposed changes have two basic objectives:

• To provide the Congress and the public with a framework of information about total revenues and expenditures over a five-year period, within which to judge the consequences of particular decisions on taxes or expenditures.

• To shift attention from the very immediate, and often less im

portant, consequences of specific budgetary decisions to their longerterm impact on national priorities.

The information provided by these procedural changes would not bear directly on the merits of any particular tax or expenditure proposal. But it would help the Congress to make explicit decisions about priorities, so that long-run changes in the patterns of expenditures and taxes might become more nearly the result of deliberate choices rather

than the unintended consequences of short-run decisions.

92-496 - 73 - 37