Although the idea of budget balance in the administrative budget has been the dominant consideration in the budgetary policy of most countries, it has gradually been realized that such a concept may be inappropriate when external shocks such as exchange rate movements or a world recession occur. Because varying levels of unemployment are a major reason why expenditures may change without comparable change in the public sector output, the concept of a full-employment budget has emerged. This type of budgeting is based on receipts and expenditures that would prevail under conditions of full employment. The approach views the actual expenditures and receipts for the coming year as of secondary importance; it assigns primary importance to the influence of the budget on the national economy. In time of recession a budget deficit may thus be presented as a necessary step toward achieving a balanced budget at full employment. Ideally, the budget should include estimates of expenditures and revenues at full employment, and also estimates of the same items at the anticipated level of employment. These ideas have been extensively used in the United States.

An analogous procedure could be used with respect to inflation, but this idea is still far from acceptance, because governments are no less reluctant to anticipate inflation than they are to budget for unemployment.

The U.S. full-employment budget was developed during World War II and has been regularly published in the president’s annual Economic Report since 1962. Other countries adopted similar measures as an aid to policy making—for example, the Netherlands’ “structural budget margin,” developed in the early 1960s, and West Germany’s “cyclically neutral budget,” calculated by its German Council of Experts beginning in the late 1960s.

An analogous procedure could be used to correct for the impact of inflation. When inflation is rapid, interest rates are correspondingly high and a government may appear to run a deficit as a result of high debt servicing costs even if the real value of the debt is declining. The United Kingdom, for example, has seen a government deficit in almost every year in the postwar period, even though its debt has been a diminishing fraction of national income, because growth and inflation have been increasing the level of national income. Although inflation adjustments have been widely advocated and often adopted in private sector accounts, governments have been reluctant to adopt them for public finances because of the element of uncertainty in prediction.

As the emphasis in budgetary policy has shifted away from mere authorization of government spending and toward more public scrutiny of what government accomplishes, the idea of appraising value received for money spent in government finance has grown in importance. This has led to an increasing variety of measurements of public sector efficiency. In general terms, taxpayers need to be satisfied that their money is being used wisely. Because of the wide variety of items within even a single program, however, it is often difficult to identify precisely what is spent on the provision of each service, and the services that are provided rarely have well-developed private sector counterparts to act as a basis for comparison.

In some programs, governments have developed efficiency measures that relate observable facts, such as the quality of national health or the number of operations performed, to the cost of providing the service. The use of such measures is by no means widespread, however, and their basis is often open to question. The principal difficulty is that there is either no meaningful measure of the output of a public service—defense, for example—or output is complicated and multidimensional—as with education or health. The result is that any method used to measure efficiency is open to debate and challenge.

Attempts to control public expenditure, particularly since the mid-1970s, have led to some reexamination of which programs should remain in the public sector. In the United Kingdom many services (for example, hospital cleaning) have been transferred from public sector agencies to private contractors, in the search for more cost-efficient purchasing.

There are three principal bases for public expenditure planning: cash, volume, and cost. The cash basis is concerned simply with the projected money expenditure on the services involved. Making such projections is difficult because what the cash expenditure will buy depends on what happens to prices over the planning period. Moreover, many public expenditures cannot be planned in cash terms, because legislation prescribes the output. Most social benefits, for example, must be paid to anyone who is entitled to receive them, and this means that the government cannot control directly the amount of the expenditure.

The volume basis is concerned with the planned output of public services. The difficulties of measuring output, however, have already been noted. More often the planning process, assuming that changes in inputs are associated with changes in outputs, operates with reference to the cost basis of programs.

All countries have an annual program of public expenditure allocation, in which those responsible for individual programs argue for greater allocations for their activities and those responsible for raising the money attempt to control the amount allocated. In practice, the results of this process depend as much on the political weight of individuals in charge of a spending program as on an objective assessment of its desirability. The normal practice is to take as a base what each program spent the previous year and then argue about incremental changes, rather than (as under zero-base budgeting) to consider each program in its totality. This creates perverse incentives, in that departmental heads who have saved money in one area in a particular year have an incentive to spend more in other areas in order to protect next year’s total budget.

The basis for most expenditure planning is therefore the number of public employees already in place and the volume of goods and services purchased in the base year. This, multiplied by base year prices, gives the input volume in the base year. In the late 20th century many countries (particularly the United Kingdom) have been abandoning this approach, largely because it gives inadequate control of total expenditure. One reason for a given volume’s costing too much to supply is the so-called relative price effect. This arises because goods and services bought by the public sector (labour, medical care, or defense equipment) may rise in price more quickly than commodities generally. Once this has been determined, volume can be expressed in cost terms. The relative price effect is somewhat subjective, however, because of the difficulty of measuring the quality of goods and services. In the case particularly of health care and defense, the relative price effect will often contain the increased price of services and improved equipment, which are actually a volume increase.

Cost measures, however, merely reflect the cost of a given input; controlling public expenditure in cost terms without taking full account of the relative price effect’s change may lead to inappropriate volume responses or, more commonly, spiraling costs as existing input volume is maintained. Hence many countries have moved one stage further, attempting to monitor and control public expenditure in purely cash terms. The United Kingdom’s public expenditure programs, for instance, are now “cash limited.”

Although planning in cash has a superficial simplicity, at times of significant inflation it is not a very appropriate tool, and differential price rises may lead to a balance of expenditure provision somewhat different from the intended plan. In practice, although cash planning is presented as the base on which decisions are taken, those countries that have adopted this approach in fact allow informal flexibility in cash budgets, with volume measures being implicitly, if not explicitly, adopted.