chapter five Organizational Aspects
- A. Premchand
- Published Date:
- March 1989
… and they don't seem to have any rules in particular: at least, if there are, nobody attends to them. …
LEWIS CARROLL. Alice's Adventures in Wonderland
The previous chapters have been devoted to a consideration of theories and practices of budgeting. The success of budgeting as a system is crucially dependent on several organizational factors and on the attention paid to vulnerable areas. The first part of this chapter considers organizational aspects and the second part is devoted to a discussion of selected problem areas.
Features of Budgetary Systems
Financial management structures and related styles of administration have considerable influence on the actual working of budgetary mechanisms.1 In addition to providing an insight into the working of the institutions, the systemic features are of importance in assessing the cognitive abilities for identification, anticipation, and measurement of the fiscal issues. Budgeting, which initially reflected the legislative and executive relationships (and the efforts of the former to gain control over the operations of the latter) went through several stages of development, each leading to the succeeding one and resulting in a cumulative experience for those responsible for government finances. Budgetary systems do not necessarily imply rigid or established patterns, but over the years some broad groupings have developed that reflect common constitutional, legal, and administrative heritages. There are severe constraints in describing these types of budgetary systems because they have undergone many changes since the 1950s and distinctions between them have become blurred. Distinctive features of any one type may not be found in their entirety in one country, as several variations of the same practice are to be found. Also to be noted is the modernization of systems on different patterns by countries that originally started with the same type. Modernization might imply that the traditional systems have been discarded, but in practice few countries have totally discarded theirs; even where new techniques are introduced, the basic framework continues. The frequent changes may also imply that some of the features described here may not be found in that format.2 Also to be noted is that different terminology is used to describe the same process in various countries. Despite these difficulties, six broad types of budgetary systems can be distinguished in terms of their approaches and influences.
1. Among the original influences were the British type of budgetary system and the French financial management system. British budgetary procedures, which evolved over a long period, reflect empiricism and pragmatism. Although the United Kingdom itself had few budgetary laws,3 its colonies developed budgetary laws that continue to govern the current practices in several Asian, African, and Caribbean countries. The British type of budgeting largely consisted of a consolidated fund through which all revenues and expenditures flowed; and budgeting was undertaken primarily on the expenditure side, totally separated from the revenue side. The spending agencies were subjected to different degrees of control during the budget execution.
2. The French system, which is to be found in France's former colonies, is based primarily on two principles—a strong financial control system and a central treasury. The control system has three operational levels—comptrollers attached to spending ministries and overseeing their financial transactions, a cadre of public accountants responsible for the collection and disbursement of monies, and inspectors of finance having greater power and responsibilities and representing, as it were, the financial conscience of the Government as a whole. The French Treasury is not merely the cashier but also a banker. Its purview extends to acceptance of deposits from public agencies, in addition to normal treasury functions such as scrutiny of bills and payments. These twin features of the system are to be found in French colonies in Africa and Asia. Other African countries, such as Zaire and Rwanda, show strong similarities to the French system but are not fully comparable. Some of these influences, particularly those relating to expenditure control, are also to be found in the Middle Eastern countries.
3. A third group may be identified as the European systems. The major systems of this type are the Netherlands, Italian, and Portuguese systems, which are to be found in varying degrees in Indonesia, Somalia, and the former Portuguese colonies, respectively. These systems differ in several ways. The Netherlands system reflects greater reliance on the commercial principles of budgeting, including provision of depreciation allowances, and, until recently, accrual-based accounting. The Italian budgetary system shows the availability of funds beyond the fiscal year, with parallel operation of the preceding and current years' budgets and extensive controls by the Comptroller General. The Portuguese system relies more extensively than others on the legal approaches to budgeting and annual specification of budgetary allotments by law. Practices in the Federal Republic of Germany are not very different from those of other European countries; in any event, the German system did not have a significant influence on any other country.
4. The fourth type consists of U.S. practices, which are largely centered around a general budget and numerous trust funds that are not subject to legislative approval. More recently, the U.S. Government has, for analytical purposes, followed a unified budget, combining the features of general administrative budget, the cash budget, and the budget based on national income accounts.
5. A closely related practice, but forming a distinct type, is the Far Eastern budgetary system, which is influenced by Japanese tradition. This budgetary system comprises a general account, several special accounts dealing with specialized and quasi-trading activities, and extra budgetary accounts, of which the most important is the Fiscal Investment Loan Program, largely financed by borrowing. Budgets contain unconsolidated transfers from the general to special accounts. Similar practices are to be found in the Republic of Korea and, to a lesser extent, in the Philippines4 and Thailand.
6. The sixth category, although somewhat amorphous, is what may broadly be called the Latin American type. Latin American countries initially adopted procedures with a strong Spanish influence, developing in due course their own hybrid systems as a by-product of their administrative experience. The main features of the Latin American systems are extensive earmarking of funds, decentralization of government activities into autonomous agencies, reliance on a noncash basis, and combined accounting and audit machinery. Within these broad types of systems, various financial management practices have evolved. At the present time, several common practices and dissimilar features are found among countries.
Budget coverage differs from system to system. The primary differences relate to autonomous agencies, social security systems, and loans. In the U.K., European, U.S., and Far Eastern systems, budget coverage is fairly comprehensive. In the United Kingdom, it includes trading accounts and the coverage of the Public Expenditure Survey is even wider, including social security. Similarly, the American system also covers social security transactions, although, despite repeated attempts to the contrary, off-budget agencies have grown. In Italy and the Netherlands, the social security system is excluded. The Italian budget is limited to government departments only. The system in France has, however, undergone a process of “debudgetization,” under which certain lending and credit transactions carried out by the Economic and Social Development Fund have been separated from the budget. Also excluded are social security, road development expenditures, transactions with the local government, and loans for housing. This debudgetization has had only a slight effect on the former French colonies. In Latin American countries, budget coverage is generally narrower than that in other systems. It also indicates that the terms “surplus” or “deficit” should be taken as relative and, in any event, their meaning is far different from the surplus or deficit of the consolidated public sector, whose operations are important in assessing the impact on the economy. In several developing countries, social security is treated separately, as are government loans and, in certain cases, grants and loans received by the government.
The budget structures in these systems vary from unified (or unitary) budgets to dual and multiple budgets. Budget structures are, however, considered in detail in Chapter 10.
A related feature that should be mentioned is the procedure adopted for budget implementation, pending approval of the budget by the legislature. In numerous instances, the budget may not be approved before the fiscal year begins. The need for funds in the American system is met through an interim resolution by the legislature, and in the British systems by a vote on account, under which a fourth of the funds sought are approved. In the Netherlands, Portuguese, and Latin American systems, while awaiting approval the executive is empowered to spend for each month amounts not exceeding those spent in previous years.
The basis of budgets also differs from system to system. In the U.S. budget system, the appropriation acts specify the new budget authority, while the basis of budget making is the cash outlay during the year, representing the actual and likely expenditures from the previous and proposed budget authority. The U.K. budgetary system is based on cash, as are the Far Eastern budgetary systems. The Netherlands system, which formerly emphasized the “acquired rights” basis of expenditures, has recently moved to a cash basis. The French, Italian, and Latin American systems base their budgeting on expenditure authorization, program authorization and obligations, or appropriation authority. Revenues are similarly distinguished—some based on actual receipts and some on claims. In most systems, funds provided for the budget lapse at the end of the fiscal year. In the Italian budget, unspent and capital outlays and uncollected revenues are carried forward to the next and subsequent years. These constitute a parallel budget, as noted earlier. The amounts passed on (residui passivi) are fairly large, both in absolute amount and in relation to budgetary appropriations, but apparently have not led to major uncertainties in budgeting because both the amounts carried forward and the annual appropriations have remained stable.
Some distinguishing budgetary practices relate to borrowing. In the United Kingdom, until World War II, a legal limit was placed on borrowing, but this was later abandoned. More recently, however, borrowing became more important in the United Kingdom but has a self-enforced limit rather than a legal one. In several commonwealth countries similar limits are in force, but these are fairly flexible and are changed annually with the budget. The United States places a limit on government borrowing, while the Netherlands Government cannot borrow abroad to finance its budget operations. Differences are found in the inclusion of borrowing proceeds in the budget. In countries that have a legal requirement for showing balanced budgets, proceeds of borrowing are included on the revenue side of the budget. Indonesia, Japan, the Federal Republic of Germany, and the Philippines follow such procedures, while other countries show borrowing separately. In some developing countries, however, budgets may not reflect all government borrowings from abroad.
Another area in which differences have emerged relates to the use of policy budgets and concepts for measurement of the impact of the budget. These differences, however, do not reflect the budgetary system as much as they reflect the relative economic environment in which budgets are formulated. In industrial countries, as noted earlier, the legislative budgets, expenditure forecasts, and budgets based on national income are used for economic policymaking. Also, greater importance is attached, both in government and public policy discussions, to the amount of the budget deficit and the means of financing it. In the Federal Republic of Germany, the United Kingdom, and the United States, the budget deficit is a controversial political issue. To minimize public criticism, governments have on occasion attempted to conceal budget deficits through various arrangements. In Italy, a budget deficit is less controversial because the exact dimensions of the deficit do not emerge clearly. In the Netherlands and Sweden, economic policymaking aims at containing the budget deficits as a ratio of GDP. In the developing world, including a number of Latin American countries, budget deficits appear to attract less public debate, partly because of the difficulties in computing their size. Also, inadequate development of statistical systems have thwarted the formulation and analysis of the government budget in national income terms.
Traditionally, the fiscal years of governments have evolved over the years and have been influenced by crop cycles, political factors, convenience, the tradition of metropolitan governments, and relationship to the financial years of domestic industries. During the past three decades, several industrial and developing countries have changed their fiscal years. The United States changed its fiscal year in 1974, while the Federal Republic of Germany and Italy changed theirs in 1960 and 1964. Egypt changed its fiscal year in 1980, while Pakistan, Papua New Guinea, and Sri Lanka changed theirs during the 1960s and 1970s. Most countries have fiscal years that are based on calendar years; some fiscal years begin in July, while a few begin in April. Some countries—notably Nepal, Oman, and Saudi Arabia—have a floating system, whereby each fiscal year is determined on the basis of the lunar or solar calendar. Saudi Arabia prepares the budget in terms of the Hajra calendar, while India indicates the equivalent of the fiscal year, as supplementary information, in terms of the Saka year.
Discussion on the appropriate period of the fiscal year frequently takes place, often with recommendations being made for change of the fiscal year.5 The arguments for change stem from a variety of concerns. From a policy point of view, it is suggested in some developing countries that the original fiscal years largely followed those of the metropolitan governments and did not reflect the economic cycle of their operations. In India, for example, the revenue outturn and the estimates of expenditures are partly dependent on agricultural expectations, which are better known after the monsoon during May, June, and July. However, as the fiscal year starts in April, following the convention of the United Kingdom, the budget was always highly tentative and needed correction in the autumn. Thus, for greater realism in budget making, a change in the fiscal year (to July or October) was suggested in India. Similar considerations have led to a change of the fiscal year to begin in July in Pakistan. Supporting these policy considerations, arguments are also advanced based on statistical convenience. Policy-oriented data, such as national income accounts, balance of payments data, and company balance sheets are largely compiled on a calendar-year basis. Change of the fiscal year to a calendar year would result in greater uniformity in statistics and would minimize distortions in policymaking. It is argued that a common fiscal year for government, for tax purposes, for companies, and for compilation of national statistics would lead to greater harmonization, better comparisons, and, therefore, to improved policymaking. Other arguments in favor of a change relate to (1) conformity with other member countries in common market zones (for example, the European Community and the former East African Community)6 and (2) dependence on external aid, where change to the fiscal year of the donor country leads to a better understanding of the aid commitments. For example, Papua New Guinea changed its fiscal year to January in the mid-1970s for the latter reason.
It can be argued, on the other hand, that in the continuous process of budget making, the significance of the fiscal year is at best marginal.7 In the context of development plans or similar expenditure forecasts, cut-off dates are primarily determined for statistical convenience and the timetable of the legislature. In the sense of planning over medium term, as in most countries, it is believed that the fiscal year per se is not important. To that extent, the relevant considerations are statistical congruence and harmonization with other systems.
Another argument reflects disillusionment with the present budgetary systems and seeks to convey the difficulties in budgeting. Because of double-digit and triple-digit inflation rates, continuous fiscal adjustments are necessary in budgetary outlays and receipts. In such a context, budgeting on an annual basis, irrespective of the cut-off period, tends to lose significance. On the other hand, budget making for short periods may have unfortunate effects. It implies that the government has no plan for fighting inflation, leading to inflationary expectations, and such a budget might give the impression that it is far from complementing any monetary measures intended to counteract inflationary upsurge.
Further, any change in the fiscal year has a disrupting effect that may confuse both the public and government officials; its benefits may be negligible and the use of a transition period prior to any change is likely to affect the continuity of statistical series.
The debate is far from conclusive. The range of arguments illustrates the complexities of a change. It is clear, however, that a common fiscal year for budgets, tax purposes, and other national statistical series would lead to a better understanding of the behavior of economic variables and to improved policy coordination. Experience of countries that have changed their fiscal year indicates that the transition has been smooth and inexpensive.
Budgeting and related decision making are activities that take place throughout the year but which gain momentum during specific periods. Budget making begins with the examination of every policy proposal that has financial implications and then proceeds through the more formal stages of submission in a structured form by the administrative ministries, review by central agencies, and finally, to the inclusion of draft proposals in the budget estimates, their acceptance by the Cabinet, and submission of the budget to the legislature. These stages constitute the formal elements of a budget calendar, as illustrated in Chart 3.
Chart 3.Budget Formulation: Organizational Steps
The budget calendar is formulated with reference to the fiscal year, to the program of the legislature and related constraints, and to continuous expenditure planning which needs to be integrated into the main budgetary process. Reflecting these factors, different procedures are to be found in the formulation of budget calendars. In most countries, particularly the commonwealth countries, the budget calendar offers a basis not only for the new budget for the next year but also for the formulation of the revised estimates for the current year. The practices in a number of countries indicate that the formulation of the budget for the next year begins soon after the approval of the budget for the current year. In the United Kingdom and Sweden, for example, circulars for the preparation of budget estimates are sent to spending agencies soon after the beginning of the fiscal year or approximately one year ahead of the actual presentation of the budget. In the United States, the budget calendar starts about 18 months ahead of the presentation of the budget, and consideration by the Congress occurs about 11–12 months before the budget comes into effect. Similar procedures, although with less lead time, are to be found in other countries. In some commonwealth countries, such as India, the formal period of budget making is reduced to six months.
At the time when the main purpose of the budget was to ensure legislative accountability and when the economic environment was relatively stable, the budget calendar was considered a routine matter. During recent years, however, three major problems are associated with the budget calendar. First, its length is such that in some cases it could mean out-of-date estimates by the time of their transmission to the legislature. Second, the techniques adopted for submission of estimates could lead to fragmented decision making. Third, the time available for adjustment in the later stages might be too short to allow for viable policy changes.
The budget calendar is often viewed as a technical framework in which policy decisions already taken are converted into budgetary formats. On the other hand, if it is seen as an extended process in which major economic decisions are taken, then it would need a longer period. Such extended periods have, however, several practical limitations. The budget for the next year should generally take into account the financial problems during the current year and seek to introduce appropriate remedial measures by means of the following year's budget. The identification of the span of the budget calendar with the fiscal year implies that, in practice, time is too short to take into account the problems of the current year's budget. Agencies formulate their budgets in the light of the current year's actual expenditures. If these data are not available, they budget on the basis of rough estimates, which may introduce avoidable errors. More significantly, rough estimates would require more corrective action during the year. The problem is further compounded in countries with a high inflation rate that formulate budgets in current price terms or even in expected outturn prices.
The span of the budget calendar becomes important because the longer the period ahead of the actual starting point the greater is the possibility of discrepancies arising from out-of-date figures. The discrepancies and the need for further corrective action are likely to be reduced if the budget calendar is formulated to begin some time after the start of the fiscal year in order to take into account experience in the first part of the current fiscal year. It should be recognized that reductions in spans may not always be feasible where the legislature has, as in the United States, an elaborate role in the final formulation of the budget. Moderation of the budget calendar should also be viewed in terms of the effect on the participants in the process.
Both industrial and developing countries find that revenue estimates are usually submitted only in the later stages of the process. Similarly, different dates are set for submission of current or ongoing expenditures and capital expenditures, and new proposals at various periods are treated separately by finance and planning agencies. Although such separate submissions imply some administrative convenience, they also contribute to fragmented decision making, in which the revenue constraints and the interaction between current and capital or new and continuing expenditures are not fully recognized. In such cases, the onus of finding a balance among these aspects shifts to the technique of expenditure controls and the approaches of those who exercise them. It also implies that the administrative process itself does not contribute to smooth functioning.
A major problem is also faced toward the end of the budget calendar, in that the draft budget finalized at the technical level may undergo major changes at the cabinet level. Totally revised expenditure ceilings might be indicated, requiring changes throughout the budget. As time is short between the draft and final budgets, the agencies find it difficult to make adjustments within the constraints of the budget calendar. The result is an ad hoc budget, so that, in the first quarter of the year another budget is likely to emerge that is far different from the approved budget and that reflects the full adjustment. This, however, leads to confusion and, more important, time that should be devoted to budget implementation must now be spent in changing the budget approved by the legislature.
Budgeting involves the active participation of several agencies and individuals. If their efforts are to be coordinated within the time allotted, each participant must know his role as well as the roles of others. In the budgetary process, the budget circular is the vehicle for conveying the policy goals of government to the participants. Although often viewed as a routine affair, it facilitates communication before and during the entire budget process.
Ideally, the budget circular should convey the national goals, state the economic prospects of the country, and reflect the policies proposed. It has an educational role in promoting better understanding of the budget itself. The circular should indicate the resource constraints and their impact on the budget. In that light, it should cogently state the proposed increases or reductions in allocations. For example, in Japan and in several other countries, limits of expenditure growth are communicated to each department and, at a later stage, a draft budget is presented as a basis for discussion with spending departments. In Denmark, allocations for continuing expenditures are indicated on a firm basis for the next four years. In countries with medium-term expenditure forecasts or development plans, similar indications are provided. The budget circular thus summarizes these indications in more specific detail and conveys changes in the economic forecast or policies.
The circular should also indicate the methods for computing the maintenance costs of completed projects and offer detailed guidelines on ceilings for new projects. Specific attention is to be paid to the various stages of project preparation in view of the substantial time lag between initial conception of a project and its consideration for inclusion in the budget. In developing countries, special attention is needed for projects and programs that are to be submitted to donors of aid. Indications should also be given on the price levels that are to form the basis for estimates, on adjustments for inflation, and on permissible increases in staff.
The circular has two other important functions. First, it is used as a medium for ascertaining the legal status of some policies. If legislation must be renewed or new legislation is needed, those facets come to light through the budget circular. Second, it is used for specifying the information that agencies need to submit for facilitating budgetary review.
Although the general importance of the above aspects is recognized by all governments, experience indicates that budget circulars have become routine and that they acquire notoriety for what is not covered in them. These gaps fortunately do not have any pattern and are best viewed as aberrations.
Revenue and Expenditure Bvudgets: Convergence
A related aspect concerns the submission of the expenditure budgets and revenue proposals to the legislature. In several commonwealth countries following the British tradition, expenditure budgets are presented to the legislature before revenue proposals are submitted. This practice has evolved over the years primarily in the belief that determination of expenditure is foremost in public finance because it is the basis for the revenues needed and also because revenue measures were generally well-kept secrets to prevent unseemly speculation. In retrospect, neither of these presumptions is entirely tenable. In democratic systems following the parliamentary procedures of the United Kingdom, the legislature cannot propose new expenditures or increase the proposed amounts. It can reduce expenditures, but such reductions are viewed as loss of confidence for the party in power. In such systems, expenditures are based on executive decisions and are formally approved by the legislature. The merit of the secrecy argument is also debatable, as most tax changes are anticipated by the public. However, lagged presentation of the revenue proposals has a major effect on determination of expenditures before it is known what revenues are needed for financing them. The revenue constraint may therefore not be fully realized, and the combined effect of the budgetary operations may not be readily ascertainable.
To minimize such distortions, legislative safeguards may be introduced—for example, those in the United States.8 In the U.S. system, an increase in expenditures beyond the amounts set forth in the preliminary resolution by congressional committees must be accompanied by a revenue proposal for that increase. In the United Kingdom, the Armstrong Committee9 suggested several changes in the legislative procedure to facilitate introduction of tax measures at the same time as expenditure plans. An important concern is whether this practice would lead to greater forestalling by the public in anticipation of tax changes, but the committee felt that the formulation of tax measures would still be governed by secrecy considerations and that any adverse economic effects would be minimal. Such a practice would have the major advantage of eliciting constructive comment from the legislature on tax measures rather than forcing their passage as a fiscal fait accompli. The timing and procedures, however, depend on the legislative approach and the role of the particular legislature.
Type of Expenditure Controls
Government budgets indicate the goals to be achieved with the given outlays. It is therefore essential to control the expenditure to serve the avowed purposes. Expenditure controls may be positive or negative. In a narrow and technical sense, they relate to the steps in formulating the budget and to achieving both economy and efficiency. Expenditure controls essentially reflect a managerial process that is both political and administrative, dealing with horizontal and vertical relationships within government organizations. Controls, as noted earlier, relate to both the present, in that the immediate budget has to be implemented, and the future, in that the current patterns determine the future shape of the budget.
The type of expenditure controls and their effectiveness are dependent on the external and internal environment of the central agencies and the spending agencies. The types of control are determined by the attitude of the legislature and by public opinion. Although their views are not always unanimous, their expressions of concern and the broad range of interests have substantial impact on the tenor of control. The internal relationships between central and spending agencies often have greater impact than public opinion. The budget represents a central vehicle through which policies are converted into actions. In setting goals, it must be decided what is the most appropriate form through which the joint responsibility of government can be exercised. This, in turn, is dependent on the prevailing value systems, the legal process for obtaining the goals, and the participants' adherence to universal rules of authority and exercise of power. Power is both formal and informal, and often informal power is more keenly felt.
Government, to recall Pigou's remark, is not a unitary being. Its financial activity involves thousands of decision-making centers and individuals. Given the same purposes and the same resources, however, organizations have different performance ratios and achieve different results. It could be argued, for instance, that the budgetary aims of efficiency and economy are universal and therefore a common institutional framework would be desirable for the various levels of government. In practice, however, this is an impossible task because each government has its own environment, in which it functions efficiently. This is not an issue that can be treated in terms of the differences between industrial and developing countries, but it is an issue governed by administrative heritage and the interaction of that heritage with economic forces, such as the overall magnitude of expenditures, the share of outlays directly spent by government, and the share of transfers of resources to other levels of government and to households.
The wide variations seen in the effectiveness of expenditure control can be explained in terms of the forces at work that are illustrated in Table 10. To start with, the objectives of government, which influence the approach to expenditure control, are determined at the political level. These political decisions may be influenced by the legislature or by public opinion.10 Within this framework, specific programs are evolved by ministries and central agencies. In the process of budget formulation, the partnership among central and spending agencies becomes vertical, the latter working within the former's guidance. The otherwise horizontal relationship among ministries changes in view of the inherently different responsibilities of the spending agencies from those of the central ministries. The spending agencies are responsible for the policies within the limits laid down by government and are responsible for their efficiency. Central agencies, on the other hand, have main responsibility for economic and financial policies and the management of the economy. For the spending agency, the budget is an administrative tool; for the central agencies, it is a medium for allocation of resources and a tool of macroeconomic management. The central responsibility can be discharged by central agencies only through the budgets and activities of the spending agencies.
|Control Component||Levels of|
|Determination of objectives|
|Functions or ministries||Political (Ministerial)||Political (executive and legislative)||Political (executive and legislative)||Administrative|
|Program or agency||Ministerial/administrative||Political/administrative||Mostly administrative||Administrative|
|Budget making||Political/administrative||Finance, planning, and Spending ministries||Finance, planning, and spending ministries||Spending ministry|
|Budget implementation and reallocation||Primarily administrative||Administrative (finance, planning, and spending ministries)||Administrative (finance, planning, and spending ministries)||Spending ministry and its agencies|
|Accounting||Administrative||—||—||Finance and spending ministries.|
|Political level||—||Partial identity of interests||Differential interests||Identity of interests|
|Administrative level||Central/fragmented||Partial identity of interests||Differential interests||Identity of interests|
|Inert areas||Political/administrative||Differential approaches||Differential approaches||Differential approaches|
|Cost reductions||Political/administrative||Identical approaches||Identical approaches||Identical approaches|
|Quality of services||Administrative||Mostly differential approaches||Mostly differential approaches||—|
|Normal||Decentralized approaches||Decentralized approaches||Decentralized approaches||Decentralized approaches|
The acceptance of this crucial relationship has, over the years, initially given greater power to finance, and later to planning. It has also enabled finance to generate power and to be the recipient of power generated at other levels. The budget agency or finance ministry cannot, however, work like a military mission. The cooperation it receives and the success of its own operations are dependent on the perceptions of other agencies. To a very large extent, partial identity of interests occurs in the allocation stage at the political and administrative levels, while in the context of stabilization it is likely that differential interests will emerge among central and spending agencies because of different perceptions on what expenditures can and should be reduced. Similar issues arise in the vertical relationships within spending ministries and their agencies. Differences in perception and advocacy arise, both horizontally and vertically, in the identification of inert areas. More often than not, spending agencies attach greater importance to their programs, which may not have the same priority on a national scale. But once areas of obvious waste and ways in which they can be remedied are established, ministries and spending agencies will have greater identity of interests. Objective criteria for determination lead to better control. When output measures and cost indices are available, control is objective and becomes less biased. Traditional factors may influence the central agencies to pay less attention to the quality of services in their zeal to achieve cost reductions, while spending agencies, being closer to their clientele, are likely to emphasize quality. The possibility of conflict therefore is as great as the entire budgetary process.
Such conflicts are sought to be resolved in different ways in different systems. In the United States, there is considerable decentralization of power to spending agencies, which have complete operational freedom within approved budgets. In continental Europe, too, spending agencies have a great deal of independence. In the French system, spending agencies have greater responsibilities for policy formulation than for financial and accounting controls, which are traditionally handled by the officials of the Finance Ministry. In Italy and Latin America, the general accounting offices similarly exercise financial and accounting controls. In the United Kingdom and in commonwealth countries, accounting controls are largely exercised by the personnel of the Finance Ministry, while policy and financial controls are assigned to spending agencies. Following the reforms in the United Kingdom in 1927, the principle that administrative policy formulation, on the one hand, and financial management and program implementation, on the other, should be integrated has gained greater acceptance. It was recognized that separation of these two functions could adversely affect financial management and might imply policy formulation without the necessary resource constraints. Administrative and financial responsibilities belong together. In some commonwealth countries, notably in India, the gradual devolution of financial responsibility was achieved by financial advisers. In the early 1950s, advisers were provided by the Finance Ministry to major spending agencies and the system was later extended to all ministries. Financial advisers, who initially served both finance ministries and spending agencies, were subsequently expected to serve only the latter.11
An analysis of the relationship between finance ministries and spending agencies reveals a cyclical pattern and a three-stage development. First, in the early years, when a government was small and its tasks were more manageable, finances were centrally controlled by royalty or their trusted representatives (as is still done in some Emirates in the Middle East). Second, with the growth of government, there was gradual decentralization of authority to spending agencies, subject to the observance of specified general principles. Such decentralization suited the managerial requirement and facilitated implementation. In some countries, the pace of decentralization was quickened by the emergence of coalition politics. Third, centripetal forces emerged again later, largely reflecting the requirements of a coherent economic management program. Administration of finely tuned economic management policies demanded a greater degree of centralization. Periodic economic crises strengthened this trend and, in some industrial countries, compensating measures were taken to strengthen financial management systems. For example, in Finland, financial controllers were appointed in spending agencies to investigate possible economies. In Canada, an Office of Comptroller General was set up for the entire Government. In the United States, drawing on European practices, Inspector Generals of Finance were appointed selectively in spending departments. In developing countries, particularly those at an initial stage of administrative development, the shortage of trained manpower contributed indirectly to a strengthening of the inherited centralized framework.
The significant issues pertaining to existing expenditure controls may be examined with reference to the relevance of control by direction, control by inducement or through a system of incentives, control by concertation, and the appropriate mix of policy and machinery. Control by direction implies that there is a strong central agency that is responsible for the economic management of the country and that it will provide direction to the numerous decision-making centers regarding their responsibilities. Such control also implies considerable centralization, and it is often suggested that central direction permits a coherent view of government policies, as well as facilitating decision making and implementation. The smooth functioning of control by direction requires a capable staff that can provide direction to the different areas of government. From the point of view of budgetary policy, data on the relation between input and output and costs should be available to enable the directing agency to give precise directions.12 In practice, however, central direction proved to be difficult and, as noted earlier, decentralization came to be adopted. Centralization stretched the limits of organization and became disfunctional. Furthermore, such control often became illusory because the lines of communication were not clear and because in reality action differed considerably from intent. As budgetary systems had not developed sufficient data on input and output relations and costs, central agencies came to be content with the specification of acceptable behavior in the spending agencies—in practice, setting accounting controls. Excessive dependence on central agencies contributed to a lack of effort in spending agencies and to making the Finance Ministry a whipping boy for ail omissions and problems in government policies.
To mitigate the problems of centralization, three major approaches have been suggested in recent years. Broadly, these advocate a reduced role for the Ministry of Finance—each for its own reasons. One suggestion is that the functions of the Finance Ministry should be limited to the formulation of policies and that implementation should be left to the spending agencies. Another suggestion is that the objectives of government should be stated clearly and periodically, while yet another suggestion is that the role of central agencies should be confined to prebudget scrutiny. The first and the third suggestions reflect variations of a theme. The second suggestion is carried out through medium-term development plans or financial forecasts and has been implemented in several countries in varying degrees. Decentralization of this type can be successful only when the spending agencies are fully capable of meeting their budgetary tasks. In practice, this was found to be a key handicap because spending agencies were more eager to pursue their policies than their finances. Partly this was due to the halfhearted implementation of decentralization itself. Moreover, spending agencies continued to rely on the Finance Ministry as a convenient feature of “redundance” or a backup system. In the United States, where decentralization was systemic, it developed a reverse dependence, in that the central agencies had to depend on the spending departments. If nondelegation of financial powers swamp the central agencies with an enormous workload and a range of issues that may stretch their analytic capabilities and effectively contribute to lack of financial management within spending units, it appears that liberal decentralization is not without problems either, for it reduces the capabilities of central agencies in carrying out their tasks. While the tasks of the central agencies have been helped by the introduction of information systems, problems still persist in both lines of approach.
Recognition of the drawbacks of control by direction has led to the advancement of control by incentives. It should be noted that this is largely a theoretical framework, as yet not fully refined. Schultze, who did much earlier work in this area,13 recognizes that the success of a majority of the programs is dependent on the response of the different decision makers in government, as well as on the extent to which the program benefits accrue to the intended beneficiaries—namely, the community. Therefore, Schultze suggests that incentives be included in the program design and in the formulation of budget itself. For example, irrigation programs of government can have appropriate tax and subsidy strategies to induce farmers to utilize the benefits. Similarly, certain average costs of services can be specified and agencies that keep within these costs can be rewarded. Schultze does not suggest that the whole range of government programs should be redrawn along these lines. Control by incentives, he believed, was to be applied selectively to supplement the traditional expenditure controls.
The introduction of incentives in government would require the specification of output, costs, and measures of performance—all areas in which little real progress has been made. There are two other limitations to this approach. First, it changes financial control from a bureaucratic technique to a broad philosophy that tends to change the character of public programs. In due course, it is likely to have an impact on the very nature and functions of the state. Second, provision of incentives to government employees outside their normal career opportunities is difficult. Such incentives do exist, ironically in centrally planned economies of the East European type. Notwithstanding these limitations, the selective application of this approach is worthy of further exploration.
Like the above controls, control by concertation has only been applied in a few situations. Basically, it recognizes that central controls have grave implications, while a decentralized system is not fully compatible with the central responsibilities of finance and planning agencies. Control by concertation views the relationship between finance and spending agencies as one of joint efforts in a common enterprise or a mutually compatible framework. The Plowden report in the United Kingdom14 pleaded for a “right balance and differentiation of function” in the unique relationship between these two types of agencies. Control by concertation emphasizes mutual consultation between the Finance Ministry and other agencies as an essential base for the formulation of policies and the functioning of controls, contributing to greater realism in central policies and to more effectiveness in the agencies. Practical experience with this technique, however, appears to have been limited to periods of financial crisis and its application during normal times is yet to be undertaken.
In the exercise of expenditure controls, the balance between policy measures and control techniques is important. In most programs, the role of the latter is peripheral and much depends on the policy premises. For example, absence of restrictions on subsidies, or less specific policies for entitlement payments, can hardly be expected to be compensated for by stringent controls. In fact, excessive reliance on control mechanics and their use in compensating for policy shortcomings could overload the machinery and build up undue expectations on their efficacy—and, when the realities are known, lead to frustration and to a general condemnation of controls. In such an atmosphere, public attitudes become even more critical, and a pervasive feeling might develop that outlays are uncontrollable. Financial discipline would be the first victim of such approaches. Pragmatic approaches to control should therefore seek realistic delineation of the role of policy measures, the role of controls and techniques, and their mutual complementarity.
The working of the budgetary organizations discussed above reveal several problem areas that merit more detailed consideration. Some of these are considered below.
In an ideal world, where all forces are at equilibrium and all individuals (and, by extension, organizations) work according to rationality models, there would hardly be a need for a Finance Ministry or a central controlling institution. The conflicts that normally arise would be settled by the working of the invisible hand. If all techniques, approaches, and procedures were applied as intended, the result would be a smoothly functioning system. The day-to-day world, however, is different from such expectations. Current discussions suggest a growing feeling of uneasiness about the growth of public expenditures and about the apparent inability of governments to cope with fiscal problems. It is alleged that policy controls of government have failed.
The forces contributing to this situation can be analyzed in terms of political, administrative, or financial controls. From a political point of view, decisions concerning finance are taken outside of the budget and the Cabinet. The Finance Ministry is not consulted before the Cabinet considers proposals for expenditures. Furthermore, the Cabinet itself is usually a divided body, with divergent views on spending and saving. It is even suggested that the ideal pattern of cabinet government is a rarity. While some issues relating to budget controls should be dealt with at a political level, the Cabinet must have a coherent view of the role of the budget. This implies that, even where the budget is organized to reflect only segments of the government, supplementary procedures would be needed to bring together the totality of public expenditure.
In several countries the Cabinet is subdivided into committees to facilitate the consideration of financial matters, but a major limitation in their functioning has been the lack of explicit recognition of financial and economic constraints. As a possible corrective, the Canadian Government has recently introduced an innovation, under which the Cabinet is divided into nine policy sectors (so-called envelopes), each being responsible for management within the resource constraints indicated for those sectors in the multiyear fiscal framework. As experience of this type is limited, no assessment can yet be made. However, it signifies the involvement of the Cabinet in matters of detail and its determination to keep the budget within recognized ceilings. While more experience, specifically during periods of financial stress, may reveal shortcomings, the potential of this approach cannot be overemphasized. Similar efforts, although less impressive, are being made elsewhere and only the future will tell whether they have been successful.
The smooth functioning of policy controls is impeded by the administrative approaches adopted by spending agencies in the budgetary process and in all financial matters. It is frequently alleged that spending agencies do not furnish the requisite information about their proposals. For several new policies, the financial implications may not yet have been fully worked out, but the policies, owing to time pressures, may still be introduced. It is also observed that spending agencies often do not furnish information even when it is available, test it might adversely affect the outcome.15 As a result, costs of projects that have initially gained budget support on the basis of underestimation have subsequently ballooned. This phenomenon, which is termed as the “camel's nose” or “tip-of-the-iceberg” approach to budgeting obviously has an undesirable impact on policy formulation. It is debatable, however, whether this approach is adopted deliberately or unwittingly because more detailed information is not available. The experiences of countries vary in this respect, and a number of cases may be cited to prove both contentions. If information is available, why cannot departments be requested to furnish it? Hartle believes that the government as a whole has “bigger fish to fry” and would not like to engage in arguments over nebulous issues.16 While this attitude may not necessarily be universal, it once again underlines the basic relationship between central and spending agencies and the need for concertation. At a practical level, spending and central agencies are often concerned only with the budgeted costs and not with the unbudgeted costs. Given the organizational structure of governments, provision for land and buildings for one department may be made in another department's budget as is true for pensions and many other benefits. The lack of appropriate market prices for some services of government adds further complexity, so that costs included in the departmental budget might often be understated, possibly leading to overuse and overexpansion of activities.
The purposes of policy control would not also appear to have been adequately served by the approaches to control. The review by the central agencies, it is argued with some vehemence, is often tantamount to “nibbling at the fringe,” the approach to control is tactical rather than strategic, intuitive rather than objective, short term in its orientation, and narrow in its focus. Control, which should be predictable, becomes arbitrary and ad hoc. The central agencies for their part would argue that the spending agencies have not been cooperative and have paid little attention to the financial aspects of policies. While arguments on both sides have merit, it is clear that overall control should be concerned not only with a program in itself but with its linkages with other programs, not only with annual changes but with the placement of those changes in government priorities, and not only with immediate visible costs but with indirect costs as well. These points are self-evident but their beneficial influence would be felt when they are ingrained into the budgetary process. It could be argued that in strengthening the budgetary process policy control is only a minor aspect and that, if controls are to become viable, political determination is more important. The debate, however, is not over whether political or administrative control is better; the two are complementary. Some budget analysts argue that budgetary systems are sufficiently well developed and that the tasks of control can be better met only when there is an ability to manage them in a political sense.17 Others hold the view that budgetary systems are still far from ensuring that the resources deployed in the budget obtain the maximum effect. Budgetary systems, both at political and administrative levels, have no real finality in their development, and their effectiveness depends on their responsiveness to changing requirements.
Earmarking of funds
The growth in the earmarking of budgetary revenues for specified purposes has been noted earlier. Arguments have been advanced over the period for greater earmarking, as well as for minimal earmarking or none at all. As with other issues of public policy, these arguments mix reason and emotion, the latter tending to obscure the realities. The choice of the extent of earmarking must be made, if at all, after a pragmatic consideration of the nature of earmarking, its spread, its strengths, and its weaknesses.
Earmarking is of two types—a specific allocation of a percentage of revenues or a specific amount allotted to special funds. Earmarking is done both by constitutional provisions and by specific legislation. It is substantial in Latin American countries and is found in Japan, Korea, the Philippines, and Thailand in the form of special accounts. However, it is rather moderate in the French type of budgetary system and is even less in the British type of system.
The basis for earmarking can be a broad one. For example, the Islamic Republic of Iran traditionally earmarked its oil revenues for the development budget. In some cases, the basis can be very narrow and specific. Typical examples are cesses and agricultural betterment levies for the development of crops or gasoline taxation for the financing of highway construction. In Peru, part of the proceeds of import taxation were similarly earmarked for financing defense imports. Variations on this theme are found in every country.
The enthusiastic supporters of earmarking point to the advantages, especially its philosophical basis and its inherent managerial facility. It is suggested that earmarking provides a direct link between costs and benefits of taxation and a strong case for additional taxation. From a legislative point of view, it provides greater accountability and is convenient to the decision maker because it reduces the political inefficiency inherent in resource allocation. In many Latin American countries and elsewhere, however, earmarking has emerged as an escape-mechanism from legislative purview. Confronted by legislative antagonism and prolonged delays in approval of the budget, the authorities those earmarking as a method of reducing legislative intervention and control over annual budget making. Ac a managerial level, earmarking is expected to gain greater efficiency because autonomous functioning of the budget is likely to contribute, with appropriate incentives, to minimization of costs and maximization of services. Further, earmarking reduces uncertainty in annual budget allocations and permits those responsible to plan ahead and to implement the budget.
In reality, each of those aspects appear to have led to greater problems rather than to solutions. The link between the payment of taxes and the benefits to the taxpayer has often been difficult, and game considerations may have contributed to taxpayers pretending that they have less concern for the proposed public goods than they, in fact, have. The remoteness of benefits and the lack of complete knowledge about the budget allocations has aggravated the feeling. Administratively, the growth of earmarking lent credibility to the view that the government had no real budget and that legislative controls were declining. The prevalence of earmarked funds contributed to the promotion of enclave mentality in those responsible for administering of such funds; when flush with funds, administrators engaged in overinvestment in their spheres of activities. In New Zealand, for example, the use of gasoline taxation for highways led to such a situation, and efforts were made in the early 1970s to reduce earmarked funds. Earmarking implied a rigidity: while some functions had more funds than needed, a few were starved of funds. National priorities were adversely affected. In tight fiscal situations retraction of allocations made to earmarked funds proved difficult. As an extension of this, the pursuit of anti-inflationary policies became a formidable task. As the prevalence of earmarking contributed to a slackening of central controls, agencies financed by earmarked funds tended to become a law unto themselves, contributing to a needless complexity of the budget.
These experiences should not, however, be considered a general rule but as instances where the effectiveness of earmarked funds was thwarted. Like any other technique, earmarking has its uses and abuses. Its usefulness is directly dependent on the spread of the earmarked funds, the flexibility in their management, and their relationship to the budget and national priorities. Whether earmarking is a lesser evil or not should be considered from these two aspects.
One of the main functions of the traditional Treasury or Finance Ministry is control over the personnel aspects of government so as to ensure economy in employment and parity and uniformity in government pay. Both these aspects have received increasing attention for several reasons. The public sector is the major labor-intensive industry in any country and the number of government workers has grown during recent years. In the industrial world, the extension of welfare services and, in some cases, the modernization of defense forces has brought about such a sharp increase in public services that now more than 20 percent of the labor force is employed by government. In developing countries, the implementation of development projects, the expansion of the armed forces, and the adoption of policies under which the state is obliged to provide employment to school leavers or graduates have contributed to a rapid expansion of government personnel. As numbers of such workers grew and as greater portions of government budgets came to be spent on wages and salaries, attention focused on problems encountered in formulating personnel budgets.18 The growth in numbers has also tended to confirm the traditional view that services provided by government tend to be costlier because of their labor-intensive nature. But the growth has also stimulated the feeling that employment as a goal in itself, unrelated to organizational requirements, contributed to considerable “slack” and to greater x-inefficiency. Simultaneously, the growth in numbers and the growth in civil service unions brought about new realities in budget making. In some countries, where civil service is represented by strong and militant trade unions, the role and decision making of central agencies have undergone considerable changes. These changes and related problems may be analyzed in terms of (a) the creation of additional posts, (b) the reduction in the number of posts, and (c) the government pay structure.
Creation of additional posts is a complex process that is generally expected to be completed before the beginning of the budget calendar. In theory, the staff requirements of each agency are expected to be reviewed and established with reference to work-load factors, which also determine the job classification pattern of each agency. The staff requirements are then reviewed during the formulation of the annual budget, and changes are determined with reference to new programs and projects. As an integral part of this process, appropriate changes are also made in job classifications to reflect growing work loads or Staff mobility requirements. The budgetary process also requires that ceilings on the growth of personnel be indicated as part of the resource ceilings communicated to spending agencies. Any targets specified for staff reduction must be indicated to the agencies, so that they can be reflected in their budgets. The smooth functioning of the process is, therefore, contingent on (1) the availability of criteria for the determination of staff strengths and (2) suitable indications to spending agencies on permissible rates of growth well ahead of the annual budget exercise. In practice, however, there are many problems.
First, with the gradual movement away from the object categories of expenditure to programs and objectives, it was expected that emphasis on personnel growth would yield to emphasis on controls over programs, and it was expected that personnel management would be an integral part of program management. The substantial growth in numbers has, however, contributed to a change in philosophy and in the controls over personnel, which were reintroduced even in countries that implemented program budgeting (for example, the United States). In effect, this meant that, regardless of the budgetary system and its degree of modernization, controls would be exercised over both programs and personnel, implying that work-load data would be fully related to personnel requirement.
This, however, has not been true in practice. For one reason, the objective criteria and measures for creation of posts have not been fully developed in a number of countries, although several countries have made noticeable progress. Improvements include the establishment of manpower units in agencies and formulation of standards in the productive, service, and administrative sectors of government. The formulation of standards and their constant revision to keep them up to date are, however, both formidable tasks in any organization. Where such standards exist, controls appear to have been exercised more for restraining future growth than for achieving reductions in staff. In fact, indication of man-year controls and related criteria might have had a perverse impact on spending agencies that persistently managed to show staff shortages relative to the work done by them. Agencies were reluctant to undergo central evaluation of their work, lest this might reveal where staff reductions could be made and could thus lead to reduced allocations.
In some countries, notably in the United States, restrictions on staff growth appear to have been circumvented by the appointment of consultants. This tendency has, over the years, led to a situation where government functions are carried out by consultants who are not a part of government. The growth of this borderline activity has been substantial.
In a number of countries, an organizational dualism has developed where the creation of posts is done by the Civil Service Commission, the Public Service Board, and the Establishment Board, while funding is provided by the Finance Ministry. This division of duties, which is fairly common in both industrial and developing countries, has led to several practical problems. The separation implies that creation of posts will take place without resource constraint and that, once posts are created, either more resources will be provided or the posts reduced. As budget reviews take place at different times, confusion between the creation of posts and their funding is to be expected. When reduced financial allocations are made, spending agencies find it difficult to make adjustments and, to alleviate the problem, resort to overestimating their demands. Both the budgetary process and financial discipline are natural victims of this confusion.
Associated with the above and in connection with economies in expenditure, reductions in public employment and restraints in the growth of government pay have become more important. One reason for such importance is that, if personnel can be reduced or their increase restrained and if restraint or freeze is imposed on the pay structure, the need for economies in specific sectors can be obviated. This approach, however, may have the effect of leading to higher government outlays in the immediate short run. In industrial countries, reduction in employment is accompanied by compensatory payments through social security systems, and the transfer of personnel from the government payroll to the welfare rolls may have a longer-term impact on government. As the abrupt termination of services is difficult, redundancy in government can be achieved only through attrition and by short-term and medium-term measures. In the United States and several developing countries, greater reliance is placed on an across-the-board freeze on creation of posts and selective reduction either of posts whose occupants have retired or of posts that have been vacant for some time. In New Zealand a variant of this scheme (known as the “sinking lid”) was introduced, under which departments were directed to eliminate annually l½ percent of their staff. However, as tasks and services remained the same, departments resorted to greater use of capital and expenditures continued to grow at previous rates. Short-term measures without changes in the underlying policy cannot be expected to yield dramatic or durable results. In Japan, however, a medium-term plan comprising administrative reforms and personnel retrenchment has been initiated. It appears that planned attrition, undertaken jointly with medium-term expenditure planning, is likely to prove more beneficial than short-term measures.
The main issue in personnel budgeting is the role to be assigned to the Finance Ministry. Since institutional dualism has contributed to avoidable problems, the solution may be to bring the institutions together and arrange for their integrated functioning. Integration of personnel with finance would permit a coherent policy for the creation of posts and their funding. However, this approach has several limitations. Integration implies that the budget agency has a managerial function that most budget agencies are not equipped to perform. It can also lead to extreme centralization, with which the agency may find it difficult to cope, and the cure may prove worse than the disease.19 A preferred alternative is to retain the institutions but arrange for coordination of functions. The central themes of financial control and personnel planning should be coordinated, and greater opportunities need to be accorded to the finance agency in the budgetary process.
A similar central responsibility is needed in regard to the formulation of pay structure for government employees. This responsibility is traditionally a part of the Treasury's work, because decentralization would lead to chaos in the absence of parity between rates paid and kinds of work. But, by and large, this area is a neglected one. During recent years frequent adjustment for inflation, introduction of incentive awards, prevalence of too many grades, and the normal rate of grade or salary creep that adds to expenditure have led to increasingly greater discussion on the issue of determination of pay and its revision in governments.
Two preliminary considerations need to be noted. Government employees include military, civil, and industrial personnel. The civil service, in turn, comprises an enormous variety of workers, including administrators, teachers, engineers, doctors, and nurses. The complex varieties of work necessitate complex pay structures that aim at achieving parity. Government outlays consist of wages and salaries, and the concern here is with the latter. Wages are defined as remuneration for temporary and specific services, while salaries are defined as remuneration for longer-term employment that entails other benefits, such as pensions and perquisites. The determination of salaries may often be undertaken as part of the package of total benefits, while wage revision may not involve any benefits other than the wage itself.
The general principles of price determination—supply and demand—are not entirely applicable to the pay structure of the civil service. Although supply factors do have some implications, the major economic factors are the government's capacity to pay, the parity principles within government services, and, to a lesser extent, the opportunity cost of attracting personnel. In the early part of the twentieth century, the Royal Commission on the Civil Service in the United Kingdom enunciated as the primary principle of civil service that pay “should be fair comparison with current remuneration of outside staff employed on broadly comparable work, taking into account other conditions of service.” While the principle itself is unexceptional, two factors appear to have contributed to a moderate pay structure. One was the decisive approach to avoid the creation of social elites by high remuneration. Although, during periods of colonial administration, such an elite class was the avowed aim of the metropolitan government, subsequent emphasis on egalitarianism and the salary structure of the political class itself acted as dampers. Another factor was the capacity to pay. In most countries, there are differentials in pay for employees of central, state, and local governments although they might be working in the same city. While cost of living raises and similar compensations are offered on a uniform scale, the basic pay itself is different, partly because of differences in responsibilities and partly because of different fiscal capacities of the employer. In cities where the private sector is dominant, governments have to compete to attract talent, but in reality their pay structure cannot compete with that of the private sector because their capacity to pay is limited.20
However, the government structure cannot remain static. It needs frequent revision to assure continuation of the standard of life that formed the basis for the original pay structure. Such revisions have to be made within a given framework. First, any revision must be compatible with other national objectives, particularly growth of the economy, distributional concerns, and likely impact on the rate of inflation. Second, following this precept, it should be a part of national wage policy. Third, in keeping with the objective of containing inflation, it must ensure that the increase in the total wage bill is not more than the expected increase in the productivity of the public sector. Within this framework, consideration must be given to the ability to pay and the need for having wage structures that are comprehensive and easy to administer The revision must also have a degree of solidity about it, as too frequent revision or a revision limited to a specific sector will only bring demands for further revision.
Two other issues in personnel budgeting may be noted. Industrial and developing countries have found that decisions on pay matters are mostly the result of inquiries by ad hoc bodies and are not institutionalized within the government. Pay research units are rarities and their absence may have contributed, in part, to the growth of personnel problems. Also, pay revision is either too frequent and outside government control or too infrequent, resulting in a need to raise the pay level considerably at one time. Any pay revision incorporating such a precipitate rise will also lead to a substantial jump in the wage bill and to cutbacks in other areas if expenditures are to be maintained at projected levels.
An important fiscal policy issue relates to government regulations—difficulties in conforming to them, the costs involved, and their likely disincentive effects on industries. The government budget must take into account the revenue-raising and expenditure-incurring aspects of regulation. Revenues may be raised by assessing penalties for violations or they may be lowered when tax incentives and other concessions are provided. Expenditures may involve direct transfers of capital, subsidies, and implicit subsidies inherent in lower interest rates. In view of the growth of these tax concessions and expenditures, it is suggested that tax expenditures should be quantified and presented to the legislature. Such quantification of tax expenditures involves a systematic identification of tax relief and allowances with related items of public expenditures. The implicit belief in this approach is that elements of tax expenditures and direct expenditures are largely interchangeable and that consideration of both in the budgetary process would help to formulate economic policies designed to revitalize industry. More recently, the possibility has been explored of formulating a “regulatory budget”21 showing the impact of proposed regulations, benefits likely to accrue, and costs to government and to industry. Measurement of regulatory costs and benefits is, however, difficult and is further compounded by the lack of an accepted accounting basis. Moreover, some regulations may be issued by autonomous agencies whose budgets are not included in the government budget.
There has been limited success in preparing tax expenditure and regulatory budgets. In the United States an estimate of tax expenditures is included in the budget, while in the Federal Republic of Germany a report on subsidies accompanies the budget every other year. Such additional information, while not having a decisive impact on policymaking, permits a regular review and prevents the assumption that the benefits are permanent or normal. As expenditures of this type increase, it is likely that tax expenditure and regulatory budgets will be needed and that more attention will be paid to their formulation.
Public works budgeting
Government outlays on public works—comprising construction activity mainly on roads and bridges, airfields, drainage and sewerage systems, docks, harbors and canals, inland and sea defense works, multipurpose river valley projects, public housing, office buildings for government activities, and structural work connected with electricity systems and telecommunications—have greatly increased during the past three decades. Construction is a major visible index of government effort and development and its involvement in this major area consists of (1) direct construction undertaken by its own staff; (2) contracting of specified projects to the private sector while retaining responsibilities for design and supervision, as well as the early phase of commissioning; and (3) more recently, overseeing the ecological impact of construction in both the public and private sectors. Construction is also an area where governments have built up enormous inventories of heavy assets involving substantial outlays and where a large number of people are employed.22 A great deal of work is contracted out to the private sector, which has firms offering specialized services, such as designs for nuclear power systems, and vertically integrated firms owning quarries and workshops, mechanized plant, and standard building equipment. Governments have specialized agencies such as the Property Services Agency in the United Kingdom, the Army Corps of Engineers and Bureau of Reclamation in the United States, and the Public Works Departments in commonwealth governments, with total responsibility for construction of one kind or another.
Reflecting the crucial role of public works and being constantly in the public eye, contract performance and management, in particular the financial aspects, have come to receive a steady stream of criticism in the media, although construction contracts have been neglected in the literature on budgeting.23 These criticisms cover both the financing and the management of contracts. From a financial point of view, it is stated (on the basis of empirical evidence) that costs of public works tend to escalate, that the final cost of the project is often a multiple of the original estimate, that schedule slippages are too many, and that projects are completed too long after the envisaged date. The gaps between expenditure plans and realities for government construction are too wide and outlays do not fully reflect the quality of works completed, leading to the feeling that there is an expenditure illusion. Some outlays are considered economically unproductive, and some projects have more visibility and prestige than priority and financial viability. From a management point of view, it is noted that maintenance of government buildings and other works is poor and expensive, and the whole area of contracts is perceived as one that seethes with corruption at all levels in government.
The above criticisms are not exhaustive but are cited here to illustrate the nature of the problems involved. Some of the criticisms are applicable to financial management systems as a whole, but they have even greater relevance to public works budgeting. The experiences of a number of developing countries reveal that very few government agencies have in-house facilities for designing works or, when construction is performed by outside consultants, for reviewing the cost estimates indicated in the feasibility study. In order to obtain approval in the initial stages, estimates appear to be pegged low and then revised upward during the construction stage. Apart from inflation and the triggering of escalation clauses in contracts, final expenditures turn out to be more because of frequent changes in designs and additions after work is commenced. Administrators tend to retain control of a project by keeping its requirements incompletely defined and subject to change. Many contractors prefer to cooperate with them, as loose arrangements present less business risk and probably more work. There is also the problem of implementation being held up for shortage of funds or slowed down as a result of natural calamities, monsoon ravages, and physical shortages. Many developing countries do not have the financial capability to maintain “redundant” supplies of equipment as a reserve, and any breakdown will lead to prolonged delays in the completion of the work. Unproductive investments are made and wrong technologies are chosen, partly because of lack of investment planning and partly because of politicization of the budget-making process. Cost benefit analysis, which has a great applicability for the public works programs (its origins were primarily in the area of multipurpose river valley projects), is not extensively applied. Because public works are highly visible and tangible, political representatives frequently add their favorite projects to those proposed by government. In the United States and the Philippines (prior to 1972), experience conclusively indicates that “pork barrel” considerations lead to bloated public works budgets that are spread over a number of years and that some queuing is arranged to reflect political interests. Even in administrative systems oriented to the cabinet form of governments, public works budgets receive considerable attention in the Cabinet and additions are often made. Australia in the early 1930s and Papua New Guinea since its independence have found that projects are added by the Cabinet and that works programs are revised. When such considerations dominate, hasty judgments on projects are inevitable.
Management considerations are also important. Reflecting the dualism in policy approaches, more emphasis appears to be placed on the construction of new public works than on their maintenance. Maintenance is viewed as less capable of generating income or is the casualty of austerity measures.24 Another management consideration that dominates discussion of public works relates to the award of contracts. Contracting is viewed in different ways, depending on the viewer's vantage point. In the developing world, particularly in countries that are dependent on imported technology and manpower, officials see the process of contracting as a complex one in which private contractors have the upper hand. Budgetary estimates of costs are considered by bidders as the floor rather than the ceiling, and contracts are often rigged, revealing informal cartel arrangements among bidders. The bidders, however, argue that payoffs are common and that contracts are awarded without adequate regard for technical quality, schedule and price, and the experience of the contractor. All these arguments have some validity. While the practices described are not universal, they are not infrequent. Some contracting problems can only be remedied by attempting reforms in the overall field of public administration, while others can be mitigated by strengthening the budgeting of public works.
In regard to strengthening the budgetary process, a preliminary consideration relates to the option between use of in-house facilities and outside contractors. Construction is an area where the private sector has a great deal of expertise and experience. Establishment of in-house facilities implies direct employment of workers and frequently direct construction. The budgetary system envisaged is dependent on the choice made. Reliance on outside contracts implies a budgetary process with a pronounced accent on the review of contracts. Building up of in-house facilities implies a budgetary process geared for direct construction and private contracts. Whatever the choice, the budgetary process may need to be strengthened if the criticisms mentioned earlier are to be minimized. The first choice involves a further choice—between renting accommodation or building anew. These conscious choices have a direct influence on the extent of contracting to the private sector and whether such contracting is more economical than direct involvement.
Budgeting for public works has several components and the approaches to these vary. With regard to maintenance, budgeting concerns relate to its periodicity and its impact on the longevity of the buildings or structures. These concerns translate, in practice, to the outlays that are included in the current budget and those that are included in the capital budget. Approaches to the current budget tend to emphasize the element of continuing outlays, while approaches to the capital budget stress the total cost of the project and the annual profile of expenditure. Both types of outlay involve long periods from design to completion of the public works and therefore need medium-term advance planning. Formulation of these budgets requires advance consultation with spending agencies and an indication of resource ceilings. Consultations with other levels of government are also needed, as construction activities of regional, state, and local governments are frequently financed by conditional transfers from central governments.
The budgetary process should also resolve issues as to how and when to fund projects. In funding the projects, the major issue is whether the approved funds should be made available on an obligation basis to project authorities. Supporters of managerial autonomy believe that a firm indication of funds and their availability to project authorities reduces uncertainty, prevents resort to alternative strategies, and generates financial responsibility and better management. In support of this, it is further argued that public works are frequently financed by borrowed funds or by foreign aid in developing countries and that earmarking of such funds with built-in flexibility for management will be appropriate. On the other hand, the outlays involved are generally massive and are significant for demand management and market borrowing by government. Relinquishing control over them and earmarking them may prevent effective functioning of economic policies and, from a technical point of view, may lead to immobilization of funds. Selection of projects implies that a number of other projects are available and will be taken up if funds immobilized by approved projects are released to them in the event of progress that is less than expected. In the context of annual budgeting, emphasis is on annualizing the outlays on projects in progress and new projects so as to achieve maximum utilization. Budgetary processes will therefore have to aim at achieving a reasonable compromise between the requirements of central agencies and those of the project authorities.
Another issue of funding relates to the stage when provision is to be made in the budget. Several procedures are in vogue. In some countries, a token budget provision is first made to obtain legislative and executive approval. After detailed costing and preparation, further amounts are provided in the budget by supplementary appropriations. In some countries, budget provision is made on the basis of full internal costing, while in a few others such internal costs and budget estimates provide a floor for bidding. Where tendered costs differ substantially from estimated costs, budget provisions tend to become dated even before the start of the project and lead to a distorted image of cost increases. In some countries if the tendered costs vary by more than a specified margin from the estimated costs, projects are withdrawn and design modifications are made to comply with overall resource levels. When such modifications are not feasible, project viability is reviewed again and decisions are then taken. The possibility of the variations between estimated and tendered costs has led in some countries to the adoption of a “limit of the cost” approach so that financial limitations could be kept in mind. In such cases, the budget process may advantageously be divided into a two-step process, the first step providing a “token” amount for fulfilling administrative formalities and the second step providing for tendered costs to reflect the likely actual costs.
The requirements of public works budgeting at a normative level and on the basis of country experience may be considered in terms of a series of steps in the formulation, financing, and implementation of projects. The policy function of the budget implies a certain degree of priority planning within the resource constraint and in terms of the desired level of public investment to serve economic policy. In procedural terms, this is equivalent to an annual request for the compilation of public works projects. Such a compilation should not be open ended but should induce the spending authorities to undertake priority planning. The planning undertaken in the spending agencies will differ from that undertaken in central agencies. Both are needed, with a view to achieving the necessary trade-offs. Formulation of designs and estimation of costs enable a clearer examination of the environmental aspects of the project and its role in the overall plan. Other considerations are the relative tempo of construction in various sectors and regions of the country, a recognition of intersectoral dependencies, and acceleration or deceleration where needed. From an economic point of view, the application of cost-benefit analysis and the ranking of projects within and among sectors provides a different dimension to priority planning.
The program function of the budgetary process consists of the preparation of the annual budget and of making provision for works in progress, for new works for which designs have been prepared and tendered, for minor works, and for maintenance expenditures. The annual budget is based on a work plan drawn up for each major construction project. Priority planning within the annual budget frame is dependent on the availability of sites,25 estimated total outlays for a year and their implications for demand management, and some balance between outlays for maintenance and those for new projects. Recent experience suggests that budget provision for maintenance is done in the aggregate without a firm basis for the actual requirements. Such notional estimates receive instant acceptability from the budgeteers but prove to be major problems later. In all these phases, organizational capabilities and their interaction with project costs need therefore to be kept under review. The annual budget should also have contingency funds to adjust for inflation and related cost escalations. During the year relevant budgetary controls need to be exercised either to ensure smooth functioning of projects or to arrange diversion of funds from projects lacking progress to other needy projects.
Even a budget system that has evolved to meet ideal specifications cannot ensure successful completion of construction activities unless it is accompanied by well-developed methods of project management. Project management is not meant to be independent of the other phases of administration, but is closely interwoven with them. Specifically, in regard to public works, it consists in the initial stages of developing feasibility studies on the proposed projects. This step is followed by a review to approve the designs and to ensure that costs are within the estimates. Project schedules, indicating network milestones and annual work plans, are also needed. During the implementation stage, actual progress has to be reviewed and revisions made.
An important function in project management relates to the overview of the contractual services and to the discipline enforced on them. Experience of both industrial and developing countries indicates that the implementation of this function is a common problem of varying magnitude. Part of the problem arises from the administrator's quest for flexibility, part from the contractor's search for fallback positions, and part from the launching of projects without complete designs and firm decisions. It seems that important decisions are made after the start of the project and are revised frequently. Inevitably, projects are delayed and become more expensive. Project management that puts adequate emphasis on comprehensive planning of designs, has firm commitments, and enforces discipline on contracts should eventually mitigate this problem.
Conflict vs. cooperation
Throughout the budgetary cycle, different functions devolve on the central and departmental agencies. The manner in which functions are allocated has often proved to be a controversial issue. The first question is whether there should be a central agency that is responsible for all governmental construction. Experience shows that a central agency would be ideal, in order to reap the benefits of scale. In several countries, distinctions are made between civil and defense construction, with construction of roads and bridges being handled by specialized agencies. Such a division implies that the scale of construction has grown too large to be efficiently handled by a single agency. Central agencies permit a high degree of coordination, flexible use of resources, priority planning, economic development of specialist skills, and promotion of government-wide standards. These special agencies have their miniature counterparts in the spending agencies and in the central agencies. A unit in the administrative agency is responsible for initiating action on projects and for processing them through the works agencies and the Finance Ministry. A unit in the Finance Ministry is expected to have sufficient financial expertise to monitor contracts.
In the exercise of the respective functions, conflicts have arisen between the Finance Ministry and the spending agencies. In some countries, the central agencies are concerned with the economic aspects of projects but detailed cost estimates form part of the responsibility of the administrative agencies. In others, finance ministries have specialized staff for the purpose and are associated with the spending agencies from the inception to the completion of a project. Both these types of systems have led to problems. The first type implies the delegation of such an extensive power that the requirements of the central agencies may only be partly met, while the second type reflects such extensive control that it may prove to be an irritant. The resolution between the central tasks and the administrative responsibilities of the spending agencies is not easy. The spending agencies cannot reconcile the many parts of the budget in a coherent scheme without some direction from central agencies. The absence of a framework and a decisive approach on the part of the central agencies leads to a great deal of administrative effort without result. Procedures emphasizing cooperation and joint efforts are clearly indicated.
Unlike systems, styles of management do not lend themselves to neat classifications. There is little empirical work in this regard and what is available is mostly anecdotal in nature. For an attempt at analyzing the approaches to economic research (which need not necessarily influence the management style), see Harry Johnson (1973).
There is not much literature on changes in a country's budgetary system, but evidence of changes may be found in studies done at different times. For European countries, see, for example, United States, Staff Papers and Other Materials Reviewed by the President's Commission (1967b). For studies of the United Kingdom, the Federal Republic of Germany, France, the Netherlands, and Sweden, and for a later view on the same countries, see United Kingdom, Memoranda on the Control of Public Expenditure (1978a). For a description of the European countries, see also David Coombes (1970).
Molonier points out that, like other British institutions, its budgetary system was not founded “on the Cartesian principle tabula rasa. In this sense, British budgetary laws faithfully reflect the British mentality: the idea of system is alien to it” (see his contribution in David Coombes (1970). It has to be noted, however, that what might have been alien in Great Britain came to be firmly organized in its colonies. Indeed, the financial system formed, in addition to law and order, the hard core of British rule.
The Philippine budgetary system was changed to a U.S. type in the early 1950s but retained in its fund structures some of the previous influences.
See, for example, the report of the Indian Administrative Reforms Commission's Study Team, India (1966). An exception, however, is provided in the report of Lord Armstrong's Committee, Budgetary Reform in the U.K. (1980). After considering the issue, the Committee those “not to take a position for or against it” (p, 16). C. Johnson (1980, pp. 29–30) argues vigorously for a change in the U.K. fiscal year.
The United Kingdom, a member of the European Community, has a different fiscal year. Denmark and Ireland have a common fiscal year with the United Kingdom. The three members of the former East African Community have the same fiscal year.
Peru and Spain formerly prepared biennial budgets but now have annual budgets. Bahrain has moved to a two-year budget system from 1978. In analyzing the U.S. budget for 1973, a group from the Brookings Institution advocated a proposal for three-year authorizations and appropriations, primarily to permit more intensive but less frequent review of major programs. See Charles L. Schultze (1972), pp. 464–68.
Arthur Smithies was among the many who advocated a simultaneous consideration of revenue and expenditure budgets. Instead of the fragmented approach of the congressional committees, he suggested the formation of a Joint Budget Policy Committee (see Arthur Smithies (1955)). Further suggestions on the same lines were made over the years. For a discussion of this issue prior to the introduction of revised congressional procedures in the United States, see William A. Niskanen (1973).
The Armstrong Committee (1980) noted that advance indication of tax measures could take three forms: (1) a statement of the change in revenue considered necessary for the next year; (2) a statement of the required changes, together with alternative policy packages; or (3) a statement of the revenue change, plus a set of policy proposals. Appropriately, it considered the first as ineffective and the third as a rigid extension of the present procedures. It therefore recommended the second for comment from the legislature. (It labeled such a new budget as a “unified system.” This is distinct from the unified budget system proposed by the U.S. President's Commission in 1968, which essentially referred to the integration of consolidated cash and administrative budgets. The Armstrong Committee's proposal is for a unified budget of tax and expenditure plans.)
In some countries, public interest groups compile their alternative budgets for influencing public opinion. Examples of this are to be found in countries where fiscal policy and the magnitude of the government budget are highly sensitive issues. In Japan, political parties play an important role in the process of budget compilation, with the government party formulating a guideline and each opposition party publicizing its own principle. In the United Kingdom centers of learning, such as the Institute of Economic Affairs, and in the United States the Conference Board and the Brookings Institution publish special studies and alternatives to the budget strategy.
For a detailed review of these aspects, see A. Premchand (1966b).
In a sense, such a central agency is similar to the unitary form of firm in which complete direction is provided. Firms are classified, for purposes of control, into unitary, holding company, multidivisional, transitional, and other forms. See O. E. Williamson (1975), pp. 151–54. Decentralization in government is comparable with controls in a multidivisional firm.
In the United States, congressional committees have repeatedly pointed out that if full information had been made available, decisions would have been different.
Douglas G. Hartle (1978), p. 46.
The Danish Government has, for example, felt that “the budget and planning cooperation between central and local governments is today so well organized and the systems of information are so well developed that both management and control can take place in a reasonable way. … What—in the perspective of history—can be doubted is the ability to manage, in a political sense, the structural adjustment that is necessary” now? (Budgetredegϕrelse [Multi-Year Budget White Paper], Copenhagen, 1980). The Swedish Budget 1980/81 (Stockholm, 1980, p. 57) views the issue as a technical one and believes that “the question to consider in the current perspective for fiscal policy was how the resources that are already being channelled via the Budget can be deployed to maximum effect.” The points of view have undoubtedly been influenced by the respective policy and institutional settings.
Control of personnel by the Treasury has been an old problem. The traditional concerns of the Treasury have been growth in numbers of workers, problems in the determination of posts, monitoring of actual trends, and attempts at reducing the number of posts during periods of financial strain. For a historical discussion of this aspect in the United Kingdom, see Maurice Wright (1969).
The Lambert Commission Report recommended, the inclusion of staffing and training functions in the Canadian Treasury, to be named the Management Board. The Public Service Commission's role would be restricted, in its view, to see that recruitment is done on a sound basis (Canada, Report of the Royal Commission on Financial Management and Accountability, 1979).
In some countries, civil service within government may not have a uniform pay structure. For example, in Indonesia, employees of a few departments receive more pay because of their greater involvement with the public and the fear that, unless they are suitably compensated, they might be easily corrupted.
The term budget is viewed by some as a misnomer, for what is essentially sought is an application of cost-benefit techniques to regulations. Similarly, tax expenditures are also considered as a device for classifying tax concessions in terms of expenditure categories.
For example, in the United Kingdom, about 20 percent of the workers engaged in construction are employed by public authorities at various levels (United Kingdom, Britain 1977: An Official Handbook (1977), pp. 49–52).
Contracts have been generally considered as a part of general project management and not as a separate sector. Some pilot studies on performance budgeting have, however, paid attention to the application of these techniques to public works. See S. S. Viswanarhan and A: V. Chaturvedi (1972).
The picture is by no means uniform. In the United Kingdom, 41 percent of the workers employed by government in 1977 were occupied in building maintenance (United Kingdom, Britain 1977: An Official Handbook. 1977).
In a number of developing countries, procurement of sites is difficult and long lead times are needed. Underexpenditure in the public works budget, which is common, arises from an incorrect estimation of this factor.