would expand the current powers of the President to propose rescinding appropriated funds under the Congressional Budget and Impoundment Control Act of 1974. Expedited rescission proposals (such as H.R. 2164, which passed the House in the closing days of the 102nd Congress) are the most limited in their grant of authority to the President. They would require the Congress to vote on proposed rescissions, with a simple majority prevailing on the vote. At present, the Congress can kill the proposed rescissions simply by failing to act on the proposal.

In the case of either the line-item veto or some increase in the

President's rescission powers, the Congress may want to consider three issues that opponents raise. First, because these measures would apply only to discretionary spending, which represents less than 40 percent of the budget, their ability to reduce the deficit is limited. Moreover, the ability of the veto to reduce spending is tied to the desire of an individual President to use the veto to reduce spending rather than to bargain for Congressional approval of his own spending priorities. Second, providing the President with the lineitem veto is clearly a shift in power from the legislative to the executive branch. Third, the line-item veto and its statutory substitutes could increase the Congressional work load. Much more Committee and floor time than is currently spent might be devoted to considering vetoes or rescissions.

Coverage of the Budget

Disagreements also exist over budget coverage. For example, some critics argue that programs currently in the budget have characteristics (such as special financing sources) that should preclude their inclusion in the budget. Others claim that government policies that are currently not recorded in the budget but that allocate nonfederal resources (such as regulation) should be in the budget.

Moving Trust Funds Out of the Budget. In general, federal trust funds are on-budget. Some argue that trust funds are not regular parts of the government and should be put off-budget, primarily because the funds are financed by revenues that are specifically earmarked, and because the intent of the Congress was to make these funds autonomous from the annual budget process. This argument was used to declare Social Security to be legally offbudget, effective in 1986. Proponents of this change continue to argue that including these funds in the budget clouds the true causes and size of the budget deficit.

Supporters of retaining trust funds in the budget oppose any movement to exclude trust funds, because they believe not including the activities of these funds inaccurately depicts the government's role in the economy and its

drain on credit markets. The assets of these programs are, in fact, IOUS from the taxpayer; financing these programs relies on the same source of funds that supports all other government programs. Further, opponents argue that putting trust funds off-budget puts them on favored political footing.

Increasing the Activities Covered by the Budget. Others propose increasing the coverage of the budget, primarily by including either governmentsponsored enterprises (GSES) in the budget, or by capturing in the budget the costs shifted by the federal government to state and local governments or private businesses.

GSEs are privately owned and managed, but have an implicit federal guarantee. Financial markets believe that if a GSE encountered financial difficulty, the federal government would take over responsibility for its financial obligations. As a result, some argue that the GSES' financial activities should be reported in the budget. Conversely, others argue that GSEs should not be in the budget because the budget should only record activities of government entities. If so, it is argued that GSEs, which are clearly privately owned, should remain off-budget regardless of the perception

of financial markets.

Another proposal is to record the cost of federal activities on nonfederal entities (state and local governments or private businesses) in the budget. Currently, federal regulation and mandates can have zero budgetary costs to the federal government, since any resources allocated by the private sector and state and local governments to meet these laws and rules are not included in the budget. But regulations and mandates clearly have real economic costs, and many believe the current level of regulation and mandates is too high, partially because mandates and regulations (unlike spending) appear "free" to policymakers.

Critics of current practice argue that the budget should include resources allocated by regulation and mandates. Others do not believe that these costs should be included in the budget for two main reasons. First, these costs can be quite difficult to measure accurately and cost-effectively in a way that makes them comparable with other spending. Second, these are not federal budgetary costs; no federal spending necessarily occurs as a result of mandates or regulations on nonfederal entities. To the extent that it does (that is, to the extent that mandates or regulations are "funded"), the budget already captures these costs.

Changing Budgetary Accounting or Information

Policymakers respond to the relative costs of various items in the budget, to the information provided in the budget process, or to both. For this reason, in order to correct some perceived bias, numerous proposals have focused on changing the way that the budget accounts for certain activities (or providing supplemental information on the implications of these activities). These reforms include adopting capital budgeting, expanding the use of accrual accounting, and embracing performance budgeting.

Capital Budgeting.

Current budgetary accounting does not distinguish between funds spent on investment, which create future benefits, and those spent on consumption, which create benefits in the present. Some analysts argue that the budget should record the cost of investing over the life of the

investment.

Supporters of that accounting treatment for investment suggest that the current cash treatment leads to a bias against federal investment. In their view, investment appears very expensive relative to consumption, as its full cost is recognized in the budget when the project is begun, even though the true economic costs (depreciation) occur over many years. They also argue that using cash-basis budgeting to account for investment leads to a