Journal Issue

Toward functional government accounting systems: A discussion of steps to be taken in modernizing governmental bookkeeping

International Monetary Fund. External Relations Dept.
Published Date:
December 1975
  • ShareShare
Show Summary Details

A. Premchand

Government accounting is often considered to be more routine than dramatic because it is, basically, a process of recording data on stocks, flows, claims, and other operations of governments. In relation to the formulation of fiscal policies and their periodic appraisal and evaluation, however, the infinite details of accounting, the rules and routines, the plan, form, and structure of accounts, and the necessary documents and reports, all acquire great meaning and interest. Notwithstanding the recent growth in the size of the public sector in developing countries and the consequent complexities of financial management, accounting systems have not received the study and reform that they needed or deserved. More than two decades ago, the United Nations pointed out in a study that “in spite of the importance of government accounting … relatively very little attention has been devoted to this subject in most countries in recent years.” While undoubtedly there was a growing awareness of, and even a concern about, the need for reform and development of accounting systems relevant to the new tasks, progress has been very slow. The UN statement is valid even now in the developing countries.

This article completes a trilogy by the author on government financial systems. His previous articles were “Financial control in developed and developing countries,” Finance & Development, June 1974 and “Budgetary reforms in developing countries,” Finance & Development, March 1975.

Much has been written since the Second World War about the normative aspects of government accounting, particularly about enterprise accounting and the system of national income accounts. Both of these are, however, in an infant stage of development in the developing world. As for government accounting proper, although some progress has been made in the mechanization of data recording and processing and in classification of transactions, a large number of problems remain and it is the intention to discuss here some of the issues, their current status, the options available, and the perspectives on progress in this area.

Existing accounting systems

While there are few established or rigid patterns of accounting, there have developed over the years certain systems reflecting a common constitutional, legal, and administrative heritage. Although there are several constraints in defining these systems precisely, in the developing countries they may be described as broadly following either a British, U. S., or French pattern. The procedures initially adopted by Latin American countries had a strong Spanish influence but later hybrid systems were developed as a by-product of their own administrative experience. There are variations, however, in individual practices world-wide which largely reflect the diverse political and administrative legacies. However, certain common features relating to the structure of funds, payment procedures, and recording of transactions may be noted.

The accounting systems have admittedly evolved with reference to a corpus of funds through which government transactions are carried out. In the British type, these funds usually are (1) a consolidated fund through which all transactions are carried out; (2) a contingency fund, which is more of an accounting entity than an operational fund, for meeting unforeseen expenditures pending legislative approval; and (3) a public account in which public monies are held in trust. In the U. S. type and other related practices there is usually a general account forming the hard core of the accounting system, and a number of other special accounts set up for specific purposes. In the French system, the basic factor is the treasury itself which maintains a sort of universal fund. The Latin American experience is somewhat more diversified in that some countries have only a single fund whereas others have a large number of funds. A general feature of most of these systems, however, is the maintenance of a single cash balance, although there are several exceptions to this in Latin American countries.

Budgeted funds are made available to the spending agencies in several ways. The British system involves exchequer issues and is usually controlled by the office of the Paymaster General. In the French system, an annual decree is issued indicating the amounts available for each agency. Funds are, however, retained by the treasury since payments are made by it. In the U. S. and Latin American systems, quarterly or similar periodic apportionments of funds are made to spending agencies. The amounts made available through this process, and the revenues collected by the agencies are recorded in different ways in different systems. Revenues are generally recorded on a cash basis, while in a few Latin American countries they are recorded both as payments due and actual payments received in & year.

Expenditures are variously recorded. In the French system, the recording of expenditures is done in two stages—by the spending agency, and by the treasury—the former based on the delivery of goods and services and the latter on the basis of the actual payments. In Latin American countries various procedures are in vogue; thus expenditures may be recorded with reference to firm obligations, payment orders, transfer of funds to spending agencies, delivery of goods, or checks issued. In the British and U. S. based systems expenditures are usually recorded on the basis of checks issued or paid and actual cash disbursements.

Current issues in accounting

These accounting systems have undoubtedly endured over a long period—a feature that may be both an asset and a liability. They are an asset since they have provided some stability and have, by and large, been adequate in serving the accountability purposes. They are a liability in that they have served only narrow purposes and have not shown much adaptability in meeting the requirements of the development tasks. The current issues concerning accounting systems need to be seen in terms of what the systems are expected to do in the contemporary setting (see chart) and what they have so far been able to do. The main shortcomings of the existing accounting systems that have come to be felt and recognized arise from doubts as to their effective contribution to purposes and end uses, indicated in the chart. An exhaustive enumeration of all of them is not attempted here but the more important ones can be noted.

First, accounting developed over the years largely independent of the requirements of policy formulation and related administrative purposes. It concentrated mostly on providing a set of detailed appropriation accounts for meeting the accountability aspects. Owing to inadequate classifications, these accounts lacked the support needed for the functional and economic assessment of government operations. The systems were also characterized by emphasis on uniformity, legality, and propriety of transactions, and followed cumbersome and antiquated control techniques revolving around the old warrant procedures and regulations governing disbursements. This resulted in a deadening effect on the development of accounting as an integral part of the management responsibility of each agency.

Second, the systems laid far too much emphasis on recording cash disbursements and receipts rather than on measurement of costs. In the contemporary scene, a major imperative of financial management in government is measurement of cost, which forms the very core of the body of accounting. Yet this has been and continues to be one of the most neglected aspects of government accounting in developing countries.

Third, there were areas where accounting had not made effective inroads in the developing world and one such area was accounting of the assets acquired by government.

Fourth, even in the limited area of accountability, the inordinate delays in the compilation of accounts substantially reduced the effectiveness of these systems. A number of factors undoubtedly contributed to this situation. The working of the human factors (motivational influences, availability, and utilization of skills) and the process factors (systems, procedures, methods, and technology) led to a situation where the product factors (quantity and quality of data) suffered severely.

Fifth, experience indicates that the traditional accounting systems were weak in serving the public needs, particularly in matters relating to payment of taxes to government, and pension payments by government. The excessive rigidity, the absence of efforts to simplify the arcane procedures, and the considerable amount of paperwork resulted in bottlenecks in the transactions with the public. It is these and many other avoidable shortcomings that need to be remedied if accounting systems are to achieve the various purposes outlined for them.

Approaches toward reform

A discussion of some major areas such as the basis of accounting, aspects of fund control, classification, requirements of program management, and asset accounting may lead us to defining some moves toward reform.

• Accounting basis

One of the important considerations in the structuring of a government accounting system relates to the basis on which the accounts are to be maintained and reported. The choices in this regard are cash, obligation, and accrual. As the usage of these terms has led to semantic confusion, it is essential that they should be defined. The cash basis indicates a system of accounting in which receipts and expenditures are recorded at the time cash is received or paid out. Frequently, such a recording could be with reference to checks issued or paid. The obligation basis refers to a system in which transactions represent commitments to acquire materials or services or to make payments under specific conditions, and includes orders placed, contracts awarded, and related transactions requiring money disbursements, usually at a later date. The accrual basis has been variously defined, and includes, depending on the context, accounts maintained on an obligational basis to transactions in the liquidation periods. In a more precise way, the accrual basis is to be seen as a system in which revenues and expenditures are recognized as they are earned and incurred. Usually it means a system of recording receipts and expenditures at the time the liabilities are incurred as a result of services rendered; or, in the case of mass-produced items procured by government, the system refers to their delivery rather than to payments.

Several schools of thought have developed during recent years, each advocating a particular basis as an ideal one for governments. For example, those that advocate a cash system have urged it because it would: (1) permit an easy identification of the impact of fiscal transactions on the economy since actual payments made by government tend to increase money incomes and activate the economy, while payments made to the government decrease the liquidity and the community’s demand for goods and services; and (2) facilitate an assessment of the cash position which the treasury in its role as a banker should know. A cash system does not, however, measure the unpaid claims or liabilities to the government which are equally strong in exerting an influence on the economy. As for an obligational basis, it is pointed out that it provides a midway base between legislative appropriations and actual payments and indicates at any given stage of budget execution the liabilities incurred and awaiting liquidation. Thus, it is suggested, the obligational system of government accounting should lead to better financial management. In terms of budgetary flexibility for managing aggregate demand during a fiscal year, the obligational system also provides a better basis for control either for increasing or reducing expenditures. It is, however, inadequate for the measurement of departmental performance or costs.

Accrual accounting favored

The system of accrual accounting has recently come to gain increasing advocacy, in view of the fact that it is considered by some to be the best measure of the impact of the budget, indicating the time at which the government actually incurs a liability or registers a claim. In addition, the concepts of accrued expenditures, referring to the actual resources acquired, and program costs referring to the resources used, are also advanced as integral parts of this system. These offshoots, together with the basic system, are intended to facilitate budget formulation and program management. The critics of this system point out that it is generally difficult to decide the exact point of time at which claims or liabilities can be said to have influenced the community’s economic decisions. Accrual based systems also may not provide an adequate framework for certain procedural aspects of financial management, such as purchase authorizations and related control procedures. Moreover, the implementation of this accountancy basis involves the deployment of considerable technically skilled manpower, which is one of the major scarcities in the developing world. Even in developed countries, the implementation of this technique encountered snags. It has not yet been employed in its entirety by any government. Indeed, the experience of the United States indicates that considerable difficulty was experienced in regard to the switchover to the new system in areas such as corporate taxes (because of their volatility), grants to other levels of government where goods and services accrue, and in the assessment of work done for which progress payments are made over a period.

The issue facing governments is, however, one of recognition of basic complementarity between these systems. Each one serves a basic purpose and in some cases the distinction between some systems may not be significant enough to be considered mutually exclusive. For example, in a vast number of cases such as services, there is unlikely to be any difference between the incurring of a liability (accrual) and its payment (cash). Moreover, maintenance of information on the obligation method in addition to the cash or the accrual method would not be difficult if adequate decentralized procedures are followed. Thus the more important move for the developing countries would be to initiate a selective approach of supplementing their existing cash-based accounting systems with features of obligational and accrual accounting in those areas which they feel are important for immediate control.

• Control of funds

Another important area of accounting that has received and continues to receive a good deal of attention relates to the increasing proliferation of funds and problems of their control. Although governments previously tended to transact through a main consolidated fund, over the years an increasing resort to expedients and a growing practice of providing autonomy to certain operations has led to a widening of the corpus of funds. In addition, prevalence of practices such as backdoor spending, borrowing without legislative approval, and contract authority, tended to encourage a greater spread of funds.

The existing fund structures in the developing world may be broadly classified into limited systems (5 funds or less), intermediate systems (5 to 20 funds), and extensive fund systems (see table). While it cannot be conclusively stated that intermediate and extensive fund systems have always contributed to the compounding of problems, it can be said that they have given rise to numerous avoidable issues. Experience indicates that there is an inverse relationship between the number of funds and the exercise of financial controls and that in situations where funds are widespread, considerable difficulties are experienced in the consolidation of accounts. This is so because of (1) problems in the identification of relevant accounts and elimination of double counting; (2) absence of uniformity in classification practices among the funds; and (3) different fiscal years and complementary periods for various funds.

Structure of government funds in selected developing countries in Africa, Asia, and the Middle East1

General fund accounts

Special fund accounts

Revolving fund accounts

Debt retirement sinking funds

Deposit and trust funds

Total number of funds

Limited fund systems
Yemen Arab Republic11
Intermediate fund systems
Sri Lanka2518
Extensive fund systems

Limited to central government; autonomous agencies of the government ere not Included.

Includes, In several cases, contingency funds maintained as accounting entities.

The data are not complete In that several trading accounts maintained by the departments are not always centrally consolidated.

Separate sinking funds are maintained for each debt issue. For this purpose, however, they are considered as a single entity.

Includes extrabudgetary accounts.

Limited to central government; autonomous agencies of the government ere not Included.

Includes, In several cases, contingency funds maintained as accounting entities.

The data are not complete In that several trading accounts maintained by the departments are not always centrally consolidated.

Separate sinking funds are maintained for each debt issue. For this purpose, however, they are considered as a single entity.

Includes extrabudgetary accounts.

Rationalization necessary

At another level, accounting entities (with no operational funds such as interdepartmental adjustment, suspense, exchange, and settlement accounts) have all added their own complexities to the situation, thus contributing in a substantial way to the long delays in the compilation of accounts. If these are to be avoided, it is necessary that some rationalization of the funds be undertaken. Experience in this regard is somewhat paradoxical; while there were very few concerted efforts at reducing the number of funds, there have been some efforts at increasing their number, particularly in Latin American countries. It should be noted, however, that the latter tendency, which may have been advantageous as a short-term expedient, has already given rise to the types of problems indicated earlier, and recently there have been moves toward recentralization. Thus, this is an area that needs to be reviewed in the specific setting of each country and efforts made to minimize the bottlenecks currently experienced.

A related question often raised in the developing world is—who should maintain the accounts, a central agency or spending departments? The experience of developing countries indicates both these practices. In countries such as Bangladesh, India, and Pakistan, the accounts of the government are centrally maintained by the accounts department. In these cases, certain accounts are also expected to be maintained by the spending departments, but generally these are found to be inadequate. In some cases, the system of “self-accounting departments” under which the spending agencies are responsible for their accounts is in vogue. A set of central accounts are also maintained in these cases by the Accountant General. In both cases, however, considerable difficulty is experienced in having timely data, since frequently a reconciliation of these two sets of accounts is found to be a time-consuming and complicated process.

Increasingly, with the growing complexity of expenditures and related management, it appears advantageous to decentralize certain accounting responsibilities and entrust the spending departments with the maintenance of records in terms of budget activity, objects or categories of expenditure, cost of the activity, status of obligations, liabilities awaiting payment, asset maintenance, and so on, while maintaining a set of central accounts in terms of broad budget categories. Such a distribution of responsibilities is likely to be in greater accord with the pattern of functions of the government departments and would also minimize the current problems of reconciliation. The decentralization of accounting responsibility assumes the existence of accounting cells in the departments and adequate manpower equipped with needed skills, to man these cells. As of now, however, a major part of the developing world still lacks the manpower needed for this purpose. Although major efforts are being made to attract more persons into government accounting, much remains to be done. When such skills become available, it would also be relatively easy to undertake a simplification of the payment procedures.

• Classification

A major purpose of government accounts is to meet the diverse requirements of development planning and budget formulation. This, inter alia, would require that receipts and expenditures be organized on a functional and economic basis that would meet the analytical requirements. While economic classification of receipts and expenditures is usually undertaken on a supplementary basis, it is suggested that functional classification of expenditures be built into the budgetary and accounting systems so that these data are utilized in the annual budgetary process. This has, however, been rendered difficult in many of the developing countries in view of their traditional practice that the organization of accounts should bear a close reference to the department incurring the expenditure. Because of the considerable overlapping of activities among departments, however, the traditional system means that, in practice, outlays on the same function are distributed over a large number of departments (each one controlling its own little pocket of money without being quite certain about its contribution to the overall implementation of the goals of that function).

In order to avoid these problems, reforms have been undertaken during recent years in developed countries, notably in the United Kingdom, to bring the accounting classification in tune with their newly devised functions and programs. Similar efforts are also being undertaken in a few developing countries. The problem, however, persists in a number of developing countries, and, therefore, it is necessary to make efforts to streamline these aspects. But in doing so care needs to be taken to ensure that the revised classifications also facilitate the decentralized financial management responsibilities.

• Program management

The basic objective of accounting systems in the context of program management is to facilitate appraisal and evaluation that would in turn permit the formulation of policies and ensure their proper working. Although the structure of accounts needs to be broadly evolved in terms of the requirements of each managerial level, its functions include some preaudit and periodic reporting with a view to measuring and assessing what is happening. Such reporting requirements include those that are of a purely financial nature and those that are somewhat broader. As for the former, these generally include reports on the status of authorizations and quarterly apportionments, obligations, and disbursements of funds, and concentrate on the assessment of budgetary lags, lapses and rush of expenditure, and other associated phenomena. From the point of the latter, emphasis is laid on physical and financial achievements and a quantitative and qualitative assessment of work. At a project level, emphasis is laid on cost reporting aimed at an early identification of the cost overruns and the factors contributing to them. In addition to these, accounting systems in governments organized on a federal basis have additional responsibilities, particularly in regard to evolving and ensuring proper delivery mechanisms for central grants and observance of conditionality regarding specific and matching grants. Further, accounting reports need to be so structured in these situations as to indicate distributional and fiscal effects of the transfers made from one level to another.

The development processes and the requirements of intergovernmental fiscal management have cast additional burdens on the traditional accounting systems in a number of developing countries. Though these responsibilities have been growing, the responses of the systems have remained largely inadequate. (There are, however, some countries, such as Malaysia, where major improvements have been made in the monitoring of projects involving massive outlays.) As of now, the systems in the developing world reveal a number of weaknesses. First, the accounting reports are generally prepared after a considerable time lapse. Second, a number of redundant and superfluous reports are prepared at considerable expense and time and very little effort is made to rationalize these practices. Third, the reports do not generally seek to analyze the physical progress or its relationship to the outlays. Fourth, not much data are available on costs, with the consequence that cost overruns are known only when they have taken place and there is little or no preparedness to meet such critical situations. Thus the very role that accounting is to play in program management remains largely unfulfilled.

The task of accounting is to move in the direction of simplicity, certainty, and measurement in an economy that is acquiring more and more complexity. It is imperative that further progress be made in this area. Toward this end, it is essential to undertake in each country a broad survey of the requirements of information, the periodic collection of such information, the capabilities of the agencies for maintaining such a system, and the manpower needed for the purpose. In pursuance of these objectives, responsibility centers that are entrusted with the management of costs and administrative tasks would have to be organized in each agency. Often, efforts in this direction have been slow owing to the exaggerated fear that costing of operations may need an elaborate cost accounting framework; this is not always true and much can be achieved even within the existing resources.

• Asset accounting

During recent years, following the trend in the developed world, governments in developing countries have been acquiring a vast range of assets; however, few governments have a comprehensive inventory of these assets, or compiled data about their valuation, longevity, and the cost of their replacement. It is often suggested that in spheres where assets are acquired for the purpose of providing some common services to other governmental agencies (for example printing) or where the capital has to be retained intact with a view to maintaining the credit standings of a government, it would be necessary to develop asset accounting. This approach has, however, met with an uneven response. One view is that such asset accounting, with its attendant features of financing and depreciation, can make full sense only in a proper commercial context and not in the government. The other view is that such an asset accounting would improve the direction of resources by contributing to a greater cost consciousness in the government. Indeed, the recent report of the Swedish Budget Commission (Ministry of Finance, Proposal for a Reform of the Swedish Budget System, Stockholm, 1974) suggested that appraisable assets should be recorded by the agencies in a balance sheet and included in the budget, and that they should be appraised on the basis of the replacement value adjusted for depreciation. Such explicit recognition and supporting accounting procedures would undoubtedly go a long way in the acquisition, maintenance, valuation, and replacement of assets and should contribute to greater economies in public expenditure. This is an area where a good deal remains to be done in the developing countries.

A more recent aspect that has cast greater responsibilities on the accounting system is inflation. Given the double digit inflation in many parts of the world, it is necessary that adequate preparatory work is undertaken to furnish regular data so that necessary adjustments in wages and pensions and the impact of the cost escalation factors on construction outlays and procurement contracts can be made. In the absence of such data inputs, budget execution may grind itself to a halt owing to allocations that may find themselves overtaken. It would also be necessary to investigate the basis of government services so that their tariffs, based on historical costs, could be revised to reflect the current costs. As with other areas, this is one where what remains to be done is far more than what has been done to date.

Implications for the future

Admittedly, it is imperative that the inherited systems of accounting should be reformed by the developing world to make them more relevant and responsive to their present needs. At the same time, however, it should be recognized that the task, which involves considerable effort and extra costs, is one for long distance runners and not for sprinters. Any attempt at reform will initially have to aim at removing the short-term bottlenecks in the payment procedures (and in maintaining basic accounts) and the lags in reporting. Over the medium term, with simultaneous efforts at improving technical skills, attention should be devoted to organization of accounting units in the spending agencies and reform of classification formats. After these have been achieved, emphasis should be laid on the development of costing and related purposes of program management. Thus over a period, the systems would have to be developed from the current stage of mere accountability to a stage at which the tasks of program management would be better fulfilled.

Other Resources Citing This Publication