Expenditure Control
Key Features, Stages, and Actors

Contributor Notes

Author’s E-Mail Address: SPattanayak@imf.org

This paper defines and explains key stages of the government expenditure chain and describes the controls applied at each stage, including their objectives and key features as well as centralized vs. decentralized approaches in application of those controls. The paper also examines the influence of different administrative traditions on types of expenditure controls, including the authority and responsibility of various institutional actors. Finally, it discusses typical weaknesses/problems associated with different traditions of expenditure control and suggests specific measures for strengthening the control framework. While providing examples of expenditure control practices from more than 32 countries, the paper points out that more than two-thirds of the 85 low and middle income countries covered by the publicly available Public Expenditure and Financial Accountability (PEFA) assessments have weak systems of expenditure control that are also associated with higher levels of expenditure arrears and a lack of budget credibility. This paper will help public financial management practitioners to evaluate budget execution systems and identify priorities for strengthening expenditure controls. It will also usefully guide technical assistance work related to modernization of government budget execution and expenditure control systems, including the design and implementation of IT-based financial management information systems.


This paper defines and explains key stages of the government expenditure chain and describes the controls applied at each stage, including their objectives and key features as well as centralized vs. decentralized approaches in application of those controls. The paper also examines the influence of different administrative traditions on types of expenditure controls, including the authority and responsibility of various institutional actors. Finally, it discusses typical weaknesses/problems associated with different traditions of expenditure control and suggests specific measures for strengthening the control framework. While providing examples of expenditure control practices from more than 32 countries, the paper points out that more than two-thirds of the 85 low and middle income countries covered by the publicly available Public Expenditure and Financial Accountability (PEFA) assessments have weak systems of expenditure control that are also associated with higher levels of expenditure arrears and a lack of budget credibility. This paper will help public financial management practitioners to evaluate budget execution systems and identify priorities for strengthening expenditure controls. It will also usefully guide technical assistance work related to modernization of government budget execution and expenditure control systems, including the design and implementation of IT-based financial management information systems.


Expenditure Control: Key Features, Stages, and Actors

This technical note and manual (TNM) addresses the following issues:

  • Key stages of the government expenditure chain;

  • The controls exercised at each stage;

  • The roles and responsibilities of the key institutional actors in exercising those controls;

  • Approaches to expenditure control in different PFM traditions;

  • Diagnosing weaknesses in expenditure control systems; and

  • Options for strengthening the expenditure control framework.

I. Introduction1

Effective expenditure control is the sine qua non of good public financial management (PFM). Fiscal rules, medium-term budget plans, and annual budgets are meaningless if expenditure cannot be controlled during execution. A lack of effective expenditure controls not only threatens macroeconomic stability and fiscal discipline, but can also call into question the integrity of the public financial management system and undermine trust in a government’s stewardship of public resources. While the institutional arrangements for raising government revenue are typically quite centralized in a national revenue authority, the expenditure of those resources involves a wide array of public entities at various levels of government, even in countries with relatively centralized PFM systems.

Lack of adequate control over government expenditure remains a problem in many countries. As shown in Figures 1–3, Public Expenditure and Financial Accountability (PEFA) assessments covering 85 low and middle income countries revealed that:

  • more than two-thirds of these countries have relatively weak systems of expenditure control as indicated by a score of C or D on the PEFA expenditure control indicator PI-20 (Figure 1);

  • weak expenditure controls are associated with higher levels of expenditure arrears as measured by PEFA indicator PI-4 (Figure 2); 2 and

  • weak expenditure controls are also associated with a lack of budget credibility as measured by PEFA indicator PI-1 (Figure 3).

Figure 1.
Figure 1.

PEFA Scores (2006–14) of 85 Countries for Expenditure Control Indicator

Citation: Technical Notes and Manuals 2016, 002; 10.5089/9781513574639.005.A001

Figure 2.
Figure 2.

Expenditure Control and Arrears

Citation: Technical Notes and Manuals 2016, 002; 10.5089/9781513574639.005.A001

(Source: PEFA PI-20, PI-1 and PI-4 indicators and dimension scores database (www.pefa.org). The ratings “Strong,” “Moderately Strong,” “Moderately Weak,” and “Very Weak” are aligned with the PEFA “A” through “D” scores.
Figure 3.
Figure 3.

Expenditure Control and Budget Credibility

Citation: Technical Notes and Manuals 2016, 002; 10.5089/9781513574639.005.A001

Evaluating the strength of expenditure controls and addressing any weaknesses requires a clear understanding of the key features of an effective expenditure control system as well as the different approaches to putting them into practice. While expenditure control frameworks differ greatly from country to country, it is nonetheless possible to define, in a generic sense:

  • the key stages of the budget execution cycle;

  • the specific control objectives at each of these stages; and

  • the responsibilities of the relevant actors in enforcing these controls.

However, the complexity of the expenditure chain, the precise nature of the controls exercised at each stage, and the degree of centralization varies considerably across countries and is heavily influenced by their respective administrative traditions.

Despite their different administrative origins, there has been some convergence between various expenditure control systems in recent years. This convergence is in the direction of:

  • an increased focus on ex ante controls over expenditure commitments rather than ex post controls only at the payment stage of the expenditure cycle;

  • a shift from controlling only cash expenditures towards controlling the accumulation of accrued liabilities as well;

  • greater devolution of responsibility for routine expenditure controls towards ministries and agencies and a more risk-based approach to the exercise of centralized controls;

  • a stronger reliance on internal and external audit to ensure the integrity of financial control systems in ministries and agencies; and

  • an emphasis on transparency and accountability to the legislature and the public for expenditure overruns.

Strengthening expenditure control in a particular country can, therefore, sometimes require difficult judgments about whether to reinforce traditional administrative arrangements or seek to modernize them.

To help PFM practitioners evaluate a country’s budget execution system and identify priorities for strengthening expenditure controls, this TNM:

  • explains the key stages of the government expenditure chain (Section II);

  • describes the (i) types of controls applied at each stage of the chain, their objectives, and key features; (ii) nature of expenditure limits in cash-based, commitment-based and accrual-based budgeting environments; (iii) centralized vs. decentralized approach to the exercise of those controls; and (iv) authority and responsibility of various institutional actors throughout the expenditure cycle (Section III);

  • examines the influence of different administrative traditions on types of expenditure controls exercised and the allocation of responsibility for their application (Section IV);

  • identifies the typical weaknesses and problems associated with different expenditure control traditions (Section V); and

  • discusses specific measures for strengthening expenditure controls and addressing weaknesses in countries at different levels of administrative capacity (Section VI).

II. Key Stages of Expenditure Control Framework

The objective of expenditure control is to ensure that public resources are spent as intended, within authorized limits, and following sound financial management principles. A distinguishing feature of a government’s budget, unlike the budget of a typical business entity, is that it is funded primarily via compulsory taxation of citizens and authorized via an act of Parliament. The role of an expenditure control system is to ensure that the level and allocation of government expenditure reflect the will of the legislature as voted for in the budget.3 Expenditure controls should also reflect sound financial management principles, ensuring that public resources are utilized efficiently, incurred obligations are cleared in a timely manner, abuse/ misappropriation of public money is prevented, and private actors compete on a level playing field for government contracts.

Seven key stages of the expenditure cycle

To ensure these objectives are met, government expenditures typically go through seven stylized stages4 between authorization by the legislature and payment to the final beneficiary (Figure 4). These stages are:

  • 1. Authorization of expenditure. A fundamental principle of public finance is that expenditure and revenue proposals must be legally authorized to ensure accountability. The authorization for expenditure is usually given through the budget law which defines the time horizon for, limits on,5 purpose of, and administrative unit accountable for government expenditure (Box 1).6 To deal with unanticipated spending pressures, some flexibility in the allocation of expenditure between sectors may be allowed subject to clear rules/criteria (e.g., through virements and/or allocation from a contingency reserve). Budgets are not the only mechanisms that provide the legal authorization to incur expenditure. Certain sums may be spent under permanent rather than annual legislation. For example, there could be standing legislation for entitlement programs,7 servicing of debt, or payment of subscriptions to international organizations, which provides permanent legal authority to incur such expenditure subject to meeting specified parameters or criteria. However, even in such cases and in line with the principles of budget comprehensiveness, transparency, and accountability, such expenditures should be included in the budget documentation, and subjected to the same regularity controls discussed in Section III (see Table 1).

  • 2. Apportionment of authorization for specific periods and spending units. The purpose of apportionment is to prevent spending agencies from incurring obligations at a rate which would require the authorization of additional funds for the fiscal year in progress.8 Once expenditure authorization is in place, it is apportioned for specific periods and/or specific spending units. Apportionment usually follows two steps: (i) apportionment by the ministry of finance, which consists of releasing the appropriation on a quarterly or monthly basis to the line ministries; and (ii) allotment by the line ministries or main spending units of their apportioned appropriations to their subordinate spending units. This authority to spend is released to the spending units through the issue of warrants/allotments/décret de répartition, or other mechanisms.9 Some form of centralized control during this phase of the expenditure cycle is common in almost all countries and is usually enforced by the budget department of the ministry of finance. The apportionment process is critical to ensuring that expenditure totals are respected and any virements or claims on the contingency reserve are reflected in the revised allocation of appropriations. Each request for apportionment or reapportionment should be accompanied by a financial or cash plan from the relevant ministry or spending agency supporting the request for ensuring that apportionment and cash management functions are well integrated.10

  • 3. Reservation. Once the apportionment of expenditure authorization is made and the spending authority has been released, some countries’ PFM systems include a stage at which funds are reserved for a specific known expense but for which no contract has yet been issued. This is known as retenciones de crédito in Spain (and a similar arrangement in Portugal) and “engagement budgétaire” in France which precedes the “engagement juridique” or legal commitment stage. At this stage, there is no commitment, but it is known that the expense will be incurred during the budget year and, therefore, the reserved funds should not be used for other activities. This setting aside of an allotment for a future expenditure should not be confused with a legal commitment as no specific contract is signed at this stage.

  • 4. Commitment. The commitment stage is the point at which a potential future obligation to pay is established. A commitment occurs when a formal action, such as placing an order or awarding a contract, is taken that renders the government liable to pay at sometime in the future when the order or contract is honored by its counterpart. A commitment thus entails an obligation to pay when the third party has complied with the provisions of the contract. In cases where the expenditure is subject to a previous ongoing contract (e.g., wages, utilities, rent, debt service) or statutory obligation (e.g., transfers to subnational governments), an estimate of obligation to pay should be made and treated as a commitment. Since commitments usually mature as payments, their control is an essential part of overall expenditure control and prevention of expenditure arrears.11 A commitment does not mean that a payment will necessarily be made within the same fiscal year. This is especially true for expenditure on multi-annual investment projects (see Section III for multi-year expenditure limits on commitments).

  • 5. Verification (or certification). At this stage, after goods have been delivered and/or services have been rendered by a supplier, an authorized officer within the spending unit concerned verifies their conformity with the contract or order, and that a liability and due date of payment are recognized. Assets and liabilities of the government are increased and recorded in the books, if an accrual accounting system is established. In cases where the expenditure involves a previous ongoing contract (e.g., wages, utilities, rent, debt service) or statutory obligation (e.g., transfers to subnational governments, payments of household benefits, etc.), the verification requires confirmation that the obligation has actually fallen due. Expenditures at the verification stage are sometimes called accrued expenditures (e.g., in the US), accounts payable, or actual expenses. The defining characteristic of an expenditure at the verification stage is that a liability has been incurred. Arrears are the expenditures at the verification stage that have not been paid by the due date of payment specified either in specific contracts or procurement legislation or assumed under general commercial terms.12

6. Payment order. Once checks are made to ensure that all previously stipulated controls have been performed and documented, a payment order is issued. A payment order is an authorization for payment (usually against a bill or invoice) made by officials of line ministries, other spending agencies, or the ministry of finance. Before issuing a payment order, the issuing authority will typically check that sufficient funds are available to make the payment. Following confirmation that sufficient liquidity is available, a designated official approves the payment and issues a payment order. In cases where a centralized payment system has been established, the individual spending units may prepare the payment orders electronically and submit them to the central unit/treasury for payment through a Financial Management Information System (FMIS).

7. Payment. Once a payment order has been issued, payments are made through various instruments including checks, electronic fund transfer (EFT), and sometimes cash, in favor of a supplier or other recipient to discharge the liability. In line with internationally accepted good practice, the payment should be made through a treasury single account (TSA) system.13 Payments by checks are, in most countries, recorded at the point of their issuance. The process of issuing checks should be managed to monitor and minimize check float14 and ensure that sufficient cash is available when they are presented for encashment. Sometimes, a consolidated check is issued to cover multiple payments by the bank to the respective beneficiaries’ accounts (e.g., payroll payments) as per the treasury’s instructions. To ensure bank reconciliation and reliability of expenditure data used for financial reporting, it is important to compare and reconcile the transactions recorded in the cash book (which records the details of checks issued) with those in the bank statements. When the float of unpaid checks is significant, payments should also be reported on the basis of checks encashed/paid. It is not a good practice to net payments against revenue due from the same recipient, as it hinders the transparent reporting of government revenues and expenditures as they pass through the various stages.15

Authorization for Public Expenditure

  • Limit on amount of expenditure. Government’s expenditure must be within the amounts that the budget appropriations have established, with some flexibility allowed through virements and contingency reserve mechanisms. The nature of those expenditure limits depends on the accounting basis (cash, commitment, or accrual) used in the budget (see Section III).

  • Limit on time horizon of expenditure. The expenditure must occur within the time limits applicable to the expenditure authorization. Most countries adopt annual budgets authorizing spending for one year; however, some countries authorize multi-year limits for certain types of expenditure (e.g., autorisation d’engagement for multi-year investment projects in France—see Box 3).

  • Authorized purpose of the expenditure. The authority for expenditure is given for a specific pre-defined purpose. The budget classification (which may be organized by programs, sub-programs, projects, economic categories, or line items) usually specifies the purpose for which the expenditure can be made.1

  • Administrative unit accountable for expenditure. A unit of government, typically a line ministry, department or agency, is assigned the responsibility to ensure that the appropriated resources are spent as intended within the authorized limits.

1 For a further discussion on budget classification, see D. Jacobs, J. Hélis and D. Bouley (2009).
Figure 4.
Figure 4.

Seven Key Stages of the Expenditure Chain

Citation: Technical Notes and Manuals 2016, 002; 10.5089/9781513574639.005.A001

Some countries’ PFM systems explicitly recognize all or most of the above stages and track them through a budgetary accounting system, while others formally track only a few of them.16 For example, the Spanish and Portuguese-speaking countries track all seven stages and the Francophone countries track at least six stages (the reservation stage or “engagement budgétaire” is also sometimes tracked). By contrast, British Commonwealth countries do not formally track controls at every stage and typically track only the first and last two stages of the expenditure chain: authorization, payment order, and payment (see also Section IV). Three of the seven stages (commitment, verification, and payment) involve a third party (a creditor, supplier, beneficiary, etc.) while others are purely internal to the government.

III. Types of Controls and Institutional Actors

Controls applied at different stages of the expenditure cycle

The various controls applied during the expenditure cycle can be grouped into six main categories. These are (i) appropriation control; (ii) commitment control; (iii) aggregate cash control; (iv) control of regularity; (v) accounting control; and (vi) other specific controls. Other specific controls relate to specific types of transactions and are designed to either reinforce macro-fiscal discipline and sustainability (e.g., controls on payroll, pensions, and incurrence or liquidation of liabilities or guarantees) or safeguard the integrity and efficiency of public procurement and payroll systems. Table 1 below summarizes the types of control applied during various stages of the expenditure cycle, their key features and objectives. Box 2 describes the other specific controls that can supplement these general controls.

Table 1.

Types of Control, Their Key Features and Objectives

article image

All uses of public funds should be governed by financial regulations. These regulations, among other things, prescribe the establishment of responsibility for financial decisions, the segregation of duties to ensure appropriate checks and balances, and documentation procedures for maintaining a defined audit trail.

Other Controls Specific to Particular Types of Transactions

Controls on liabilities or guarantees (contrôle de liquidation): This control is applied on the incurrence of a liability or guarantee and again at the payment stage when the liability is extinguished or guarantee is paid. It seeks to verify: (i) the existence of budget cover or space within the authorized limits; and (ii) that the payment is being made to extinguish the liability to a real creditor and for a claim that was not paid earlier.

Payroll controls (a subset of commitment control): The objective of payroll controls is to control personnel expenditures and staffing numbers. This requires that personnel database (where personnel information files are kept) and payroll records be linked, regularly updated, and reconciled. Where the two are not routinely or automatically reconciled, special surveys may be required to identify ghost workers and remove them from the payroll. Payroll audits should also be undertaken regularly to identify weaknesses in the control system.

Pension controls: The liability and associated expense for pensions and other retirement benefits should be recognized at the time the employee's services are rendered. Any part of that cost unpaid at the end of the period is a liability. To be able to exercise upfront control over the future resource requirements related to pensions, countries implementing accrual budgeting (e.g., the UK, Australia and New Zealand) include the accruing cost and any unfunded liabilities of pension schemes17 within budgetary limits for each government department.18

Verification of goods and services (contrôle du service fait): This control involves: (i) verification of the goods and/or services delivered by a supplier to ensure that these conform to the specified quality and quantity; and (ii) a calculation of the liability incurred by the government to the supplier. This control can also apply to the wages/personnel expenditure in the sense that a designated official (e.g., the head of the division/department) certifies that the respective staff have performed their duty during the time period for which wages are to be paid.

Control of procurement: Significant public spending takes place through the public procurement system. The main objective of the government as a purchaser is to obtain high-quality goods and services at a competitive price. Procurement procedures should provide a fair opportunity for all bidders to compete for government contracts, and be designed to get good value for money and to minimize risks of corruption and patronage.

Manual processing controls: Key manual processing controls for purchasing, payment, and confirmation of receipt of goods and services are performed outside the typical information systems (e.g., FMIS) environment and should be subject to periodic internal control checks and audit. They can be more easily circumvented, presenting the potential for error or fraud.

Some of the controls during the expenditure cycle can be automated and applied through an FMIS. These include appropriation control, commitment control, and accounting control. The introduction of an FMIS can thus strengthen expenditure controls as the system can replace several key controls that were previously applied manually and systematically track them. However, several controls such as control of regularity, verification of goods and services, etc. would still require manual intervention.

Nature of expenditure limits

The nature of the expenditure limit enforced at each stage depends on the accounting basis used in the budget appropriation framework.19 Specifically:

  • Cash-based budgeting systems primarily enforce a limit on the accumulation and liquidation of cash obligations incurred during the budget year. A strict legal interpretation of a cash appropriation would mean that the appropriation is utilized when the government makes a cash payment. Therefore, entering into a commitment or incurring a liability in excess of the limit would not, in the absence of other controls, constitute a breach of law. Where countries have cash appropriations and accrual based financial statements, this usually gives rise to differences between budget execution reports and financial statements that require reconciliation.

  • Commitment-based budgeting systems impose limits on both expenditure commitments and cash payments. Commitment limits may be multi-year in nature (usually for capital projects) and carried over from one financial year to the next, while cash expenditure limits are usually set for the budget year. For example, in France, autorisation d’engagement authorizes commitment for an investment operation that may entail payments over a multi-year period, while crédit de paiement limits the actual payments during the budget year subject to the overall limit under the respective autorisation d’engagement (see Box 3).

  • Accrual-based budgeting systems enforce limits on the incurrence of liabilities, expenses or expenditure even when no immediate cash transactions are involved. In addition, they may impose limits on accumulation of cash obligations, multi-year commitments, and long-term obligations (such as pensions) and contingent liabilities (such as guarantees). Under accrual budgeting systems, there can be differences in both the choice of “binding constraint” and the fungibility of other constraints. For example, under an accrual-based budgeting system, the ministry of finance may: (i) delegate to spending agencies all decisions on their cash requirements and the timing of cash payments as long as the agencies conform to the expenses authorized on accrual basis; (ii) enforce a cash limit (either at an aggregated level or at a more detailed level) in addition to the amount of accrual-based expense authorized; and/or (iii) allow spending agencies to move cash appropriation into accrual-based expense but not vice versa.20

Commitment-based Expenditure Limits in France

As part of the comprehensive reform of its budget framework in 2001, France introduced multiannual commitment authorizations as a means of controlling expenditure obligations and associated payments for programs or projects that span more than one year (e.g., investment projects). For multiannual programs/projects, the approved budget includes both the multiannual commitment limits (autorisations d’engagement or AE) against which it also sets annual limits (crédits de paiement or CP) for cash payments during the year. As such, the focus of expenditure control at the commitment phase of budget execution has moved from an annual to a multiannual basis, i.e., the total cost of a legal commitment into which the government is entering into is fully recorded against the available multiannual commitment authorizations/AEs. As such, the AEs are consumed at the commitment stage of the expenditure or the legal act of signing a contract of the State with a third party.1 In the case of a commitment running over several years, its associated CP is spread over several budget years up to the cumulative maximum amount of the initial multiannual AE. As part of the budgetary accounting, commitments and payments made (through the CPs) against each AE are tracked to identify and report on (i) AE approved in budget law; (ii) AE consumed through legal commitments; (iii) AE unused/available for new commitments; (iv) CP authorized in budget law; and (v) payments made or CP consumed (see figure below). This information is then used for preparing the baseline estimates of the detailed medium-term budget forecast.

1 Unlike the program authorizations (autorisation de programme or AP), as defined by the French ordinance of 1959, AEs are no longer valid indefinitely and are extendable to all types of expenditures, including investment expenditure.

The presence of dual appropriations (either commitment/cash or accrual/cash) can complicate control of budget execution by spending agencies. Under commitment-based budgeting systems, there is a need to separately track and account for both expenditure commitments and payments, liquidate the latter against the former during the course of the budget year, and carry unused commitment appropriations between years. Under a dual accrual and cash budgeting/appropriations framework, there is a need for each agency to make projections of both its anticipated cash requirements as well as incurrence of expenses and liabilities and accumulation/realization of long-term obligations and contingent liabilities. In such cases, agencies execute their accrual budgets by incurring expenses and liabilities—even if they had not been settled in cash during the fiscal year—which are reflected in their budget execution reports or financial accounts.21 Therefore, the control of accrual appropriations relies primarily upon the monitoring of spending agencies’ accounts, whereas ex ante control over cash transactions is the primary instrument under cash budgeting.

Authority and responsibility of various institutional actors

The allocation of authority and responsibility to various actors for enforcing the controls at each stage of the expenditure cycle varies from one country to another, but some common features can be noted (see Table 2). Both the central agency/finance ministry and line agencies are involved in various tasks during the expenditure cycle. The distribution of responsibilities between them is typically organized along the following lines:

  • The budget department of the ministry of finance issues regulations on matters related to the execution of the budget, apportions appropriated funds to spending agencies, monitors their expenditures and performance, authorizes in-year budget revisions, and monitors and reports on budget execution. It may also have a role in appropriation and commitment controls through its representatives posted in spending agencies (as in Francophone systems).

  • The treasury department of the ministry of finance is responsible for maintaining central appropriation and fund accounts, forecasting government cash requirements and raising the necessary finance, supervising government/treasury bank accounts, and monitoring cash balances in these accounts.

  • The accounting department22 of the ministry of finance issues regulations and guidelines on matters related to the preparation of financial accounts by spending agencies, prepares and issues government-wide financial reports, and conducts regular bank reconciliations. Where centralized payment and/or payroll systems exist, they may also be responsible for authorizing payment orders and/or making payments. In some countries, it may conduct its own pre-payment audits prior to executing payments.

  • The line agencies have responsibility for executing their budget and managing the funds/ resources assigned to them. They allocate funds among their subordinate units, make commitments, purchase and procure goods and services, verify the goods and services acquired, prepare requests for payment (and make payments, if the payment system is not centralized), prepare progress reports, and may keep accounts and financial records. They maintain systems of internal control, and regularly report to the ministry of finance and other central agencies on their financial operations.

When several departments in the ministry of finance and other agencies are involved in the supervision of the expenditure cycle, clear business process rules delineating the respective functions of each are required. In particular, it is important to ensure effective coordination between issuance of warrants/allotments by the budget department and cash planning and management by the treasury so that adequate cash is available for payment when the line ministries/agencies’ expenditure commitments materialize.23 Similarly, transaction level bank reconciliation by the accounting department and/or line agencies should be coordinated with the treasury department’s monitoring of cash balances in various bank accounts.

Table 2.

Stage of the Expenditure Cycle, Controls, and Actors

article image

Note: BD-Budget Department; TD-Treasury Department; AD-Accounting Department; SA-Spending Agencies.

Centralized and decentralized controls

As described in detail in the next section, the application of controls at various stages of the expenditure cycle may be centralized within the ministry of finance or decentralized to line ministries and agencies.

  • Centralized systems, particularly with centralized commitment and accounting controls,24 have the advantage of: (i) reducing the scope for variable interpretation and application of control criteria by multiple agencies; (ii) facilitating integration between aggregate cash control and commitment control at the transaction level;25 and (iii) allowing the ministry of finance direct access to a centralized repository of expenditure data for budget execution monitoring/reporting. At the same time, centralization has the disadvantage of: (i) undermining spending responsibilities of managers in line agencies in the day-to-day management of line ministries/agencies’ budgets; (iii) inefficient decision-making (including superimposed prioritization) and rigid controls by the ministry of finance when it lacks the detailed information on the spending requirements of agencies;26 and (iv) presenting opportunities for rent seeking by officials implementing multiple and cumbersome controls.

  • Decentralized frameworks have the advantage of: (i) aligning expenditure decision making with the spending priorities of line agencies; (ii) minimizing/eliminating redundant controls which in turn improves the efficiency and speed of expenditure execution; and (iii) making each line agency directly accountable for its spending programs. At the same time, they have the disadvantage of: (i) potential disparate application of controls by various agencies particularly when the control criteria are not well defined; (ii) increasing the risks of non-compliance and/or collusion (as both the authority to spend and the responsibility to ensure the regularity of transactions is assigned to the same agency) in the absence of strong internal and external audit functions; and (iii) prolonging the preparation of financial reports (as expenditure data has to be collected and complied from multiple sources) required by central agencies for budget execution monitoring.

In both centralized and decentralized systems, there should be regular bank reconciliation of transactions27 and reporting of expenditure against authorized allocations by spending agencies.

Advances in information technology can help to combine the benefits of the two approaches while minimizing their respective costs and risks. In fact, with the functionalities available from a modern IT-based FMIS, information on budget execution can be made quickly available, and it becomes straightforward for the ministry of finance/treasury to track expenditure transactions as they pass through the various stages of the expenditure cycle, even when transactions are fully administered within line agencies. Ideally, the FMIS should have, in addition to the usual transaction processing function, a consolidation feature or module that receives periodic data from relevant entities, carries out consolidation of data in accordance with relevant standards, and generates required management reports for control purposes. In many cases, a transaction processing system may have to be supplemented by specialized consolidation software to generate the necessary reports in a timely manner. In case of relatively simple requirements, a spreadsheet-based application may suffice. The scope for establishing such advanced systems, however, remains challenging in many developing countries.

IV. PFM Traditions Influencing the Expenditure Control Framework

The allocation of responsibility to various actors in the exercise of expenditure controls is heavily influenced by their respective administrative traditions of PFM and level of development. The key difference is in the degree of centralization between countries that follow the British Commonwealth, German-Austrian, and Scandinavian traditions of PFM and those that follow the Napoleonic traditions of PFM (France, Portugal, and Spain). In countries of the British Commonwealth tradition, officials in spending agencies are charged with initiating and authorizing expenditure transactions, from commitment to payment, based on apportionments/allotments/warrants issued by the ministry of finance. In Francophone and Lusophone systems, such wide ranging responsibilities are not provided to spending agencies and various departments of the ministry of finance play a major role at key stages of the expenditure cycle. Countries of the former Soviet Union (FSU) have tended to retain centralized treasury departments to control and process government payments, while controls at the commitment, verification and payment order stages remain with the line ministries/agencies. Table 3 provides a summary of the roles played by central and line agencies at different stages of the expenditure cycle as well as the key strengths and challenges in different traditions.

The descriptions of the various traditions provided below are broad and general, and in practice, there are variations among the countries belonging to each tradition. In some cases, these variations are noted in the text. In general, relatively advanced countries have moved further in the direction of devolving their expenditure control systems than developing countries, but this is not true in all countries.

British Commonwealth tradition

In broad terms, Commonwealth systems28 are characterized by the devolution of the responsibility for financial control and the issue of payment orders to line ministries. The accounting officer in the spending ministry, usually the permanent secretary, is responsible for proper use and control of the ministry appropriations. Therefore, after warrant releases, line ministries have the power to: (i) make commitments against their budget appropriations and authorized cash ceilings without reference to the ministry of finance; (ii) issue payment orders to liquidate those commitments that have materialized; and (iii) prepare accounts of their expenditures. The issuance of payment orders and checks may be decentralized—with spending ministries carrying out these tasks and reporting back to the center—or centralized in a treasury department, typically called the accountant general's department within the ministry of finance, which acts both as paymaster and prepares the final accounts of the government. In the British Commonwealth system, there is no complementary period, and at the beginning of a new fiscal year, in principle, no cash transactions pertaining to the previous year’s budget should take place. Therefore, unlike the old French system, the key principles here are integration and delegation. This model assigns both the authority to spend and the responsibility to ensure the regularity of each transaction to the same agency. To function effectively, this approach requires a high level of capacity in the line agencies with devolved authority, timely and transparent financial reporting, and a strong internal and external audit function.

Francophone tradition

In Francophone systems, a guiding principle is that the official who orders payments has to be different from the official who makes the payments. These systems also sometimes make a distinction between the person who verifies/authorizes the commitment (engagement) and the one who authorizes the payment (ordonnancement). The key players are financial controllers (contrôleurs financiers) who are generally under the Budget Department of the ministry of finance and exercise a priori control (in the form of issuance of a “visa”) at the commitment stage, commitment officers (ordonnateurs) who approve the issuance of payment orders to the Treasury, and public accountants or cashiers (comptables publics) who belong to the public accounting directorate and are responsible for ensuring that the verified bills are paid. This principle of separation of responsibility means that no single individual or agency controls all stages of an expenditure transaction.29 The system has also a number of potential redundancies, e.g., the control of regularity is applied both by financial controllers and public accountants.

The expenditure cycle and associated controls in Francophone African countries derive largely from the French system as it existed prior to 1960s in the French provincial governments (préfectures).30 However, many Francophone African countries not only intensified the centralization of controls in the late 1990s,31 but the minister of finance also became the single and principal payment-authorizing officer (Ordonnateur principal unique).32 The responsibility for financial control is typically centralized in the ministry of finance, with its staff out-posted in the line ministries/spending agencies. The payment-authorizing officer, called the Ordonnateur, is separate from the financial controller. The annual budget authority is regulated by the ministry of finance which releases warrants (notifications de crédit) specifying a commitment ceiling for each month or quarter. A complementary period (generally two months)33 is allowed after the close of the fiscal year to process and record payments in respect of commitments that were authorized before the close of the fiscal year, but for which the actual delivery of goods or services has yet to take place.

Lusophone tradition

The expenditure cycle and associated controls in Lusophone African countries34 are quite similar to the control framework in Francophone systems, but there are a few key differences. Lusophone African countries are also characterized by a higher degree of centralization in the sense that the overall budget execution responsibility, including the responsibility for financial control, is usually concentrated in one office: either the budget office itself (e.g., in Guinée Bissau, and Sao Tome and Principe) or the accounting office (e.g., in Angola and Mozambique). Whereas, in Francophone systems, the financial control officer is employed by the Budget Office, in Lusophone systems the payment-authorizing officer is employed by the Accounting Office.

Latin American tradition

In some Latin American countries, e.g., Chile, a powerful accounting organization—contraloría general—often also carries out both ex ante and ex post audit functions, in addition to acting as the accountant to the government, and undertaking the payment function and pre-audit of commitments. The contraloría hence maintains overall control of budget execution. While this was a common practice in most of the Latin American countries several years ago, many countries—e.g., Bolivia, Columbia, Paraguay, and Uruguay—have in recent years separated the accounting and audit functions.

German-Austrian tradition

In the German-Austrian tradition,35 as in Francophone and Lusophone systems, there is a clear division between the roles of “ordering or anordnend” (which covers the apportionment of the budget, together with the reservation, commitment, verification and payment order stages) and “executing or ausfuehrend” (which covers the execution of payments). However, overall responsibility for budget execution is assigned to the respective line minister (or the head of an independent State body) who allocates budgeted resources to the various departments/agencies under his/her authority and delegates responsibility for budget execution up to the stage of the preparation of payment orders. The payment stage, however, is executed by a separate centralized agency (with regional branches)—called “cash offices” in Germany and the “Federal Accounting Agency” in Austria—which processes and keeps records of all payments and is in charge of accounting and preparation of financial reports.

Scandinavian tradition

The Scandinavian countries (Norway, Sweden, Denmark, and Finland) do not have a separate treasury department in the ministry of finance. Most budget execution control operations are delegated to the line ministries. Sweden has a separate debt management agency which also administers the TSA system. In the other countries, the central banks usually handle both debt management and TSA administration on an agency basis. Some of the countries have internal audit agencies under the ministry of finance, but they have no financial control role. The line ministries have substantial authority in executing the budget. In most cases, funds can be moved quite freely between sub-programs and items within the year. None of the Scandinavian countries have formalized apportionment (or in-year release of spending authority) mechanisms. Line ministries and agencies can commit and use their allocated resources whenever they want within the year. The ministry of finance in these countries does not carry out any form of detailed commitment or payment control. Financial accounts are prepared by the ministries and agencies and consolidated by the ministry of finance.

Tradition in the former Soviet Union countries

When the FSU countries were centrally-planned economies (i.e., before their transition to market-based economies), the ministry of finance played a minor role, mainly as the financial administrator of the central plan. Accounting and monitoring of budget execution were carried out by the banking system. Reports from the central bank, based on bank payments data classified by bank code (a compressed form of the budget/accounts classification) provided the only basis for in-year control of budget implementation. The system of payments to government suppliers was highly decentralized. The common practice was for lines of credit to be given to line ministries/agencies to spend against accounts in local banks.

As these countries transitioned to market-based economies in the 1990s and early 2000s, and steps were taken to establish independent central banks, separate monetary policy from management of government liquidity, and reduce fiscal dominance over the banking sector, it was obvious that corresponding changes had to be made in the fiscal institutions responsible for budget execution and control. Banks were privatized and deregulated, central bank laws introduced firm limits on the amount of credit available to the government, centralized treasury departments were established to raise financing from the private sector on commercial terms, control and process government payments, and report on budget execution through the main treasury account. Controls at the commitment, verification and payment order stages remained the responsibility of the line ministries/agencies, with variations, however, in the effectiveness with which such controls are exercised.

Table 3.

Role of Central and Line Agencies in Various Traditions and Lessons Learned

article image

V. Identifying Weaknesses in the Expenditure Control Systems

Weaknesses of different expenditure control systems

Different expenditure control systems bring with them their own advantages, but also their own potential weaknesses. Weaknesses at one stage of the expenditure control cycle can undermine the integrity and credibility of the system as a whole. When looking to strengthen expenditure control, it is therefore important to review the whole expenditure cycle/process instead of focusing on a few stages.36 However, based on experience in different groups of countries, it is possible to identify a set of problems that characterize different expenditure control traditions:

  • British Commonwealth. The lack of systematic tracking and control of commitments leads to over commitment and arrears in several countries of commonwealth tradition, particularly in Africa. Cash plans in these countries are used as tools for rationing expenditure authority but themselves tend to be unrealistically optimistic, as they do not reflect expected cash outflows based on commitments.

  • Francophone and Lusophone. Excessive number of redundant controls leads to payment delays,37 arrears, and proliferation of exceptional procedures that bypass the normal expenditure control framework. These procedures also lead to significant differences between fiscal accounts and financing data (i.e., between above- and below-the-line data).38

  • Austrian. There is potential scope for over commitments and/or manipulation of in-year data on commitments which may not be systematically recorded/tracked in a timely manner at the respective stage of the expenditure cycle.39

  • Scandinavian. There is scope for disparate application of controls by line agencies, particularly when the control criteria are not well defined. Allowing ministries and agencies to commit and use their resources whenever they want complicates cash management.40

  • FSU countries. The lack of effective communication between treasury, which uses a cash-based budget execution system, and line agencies, which maintain their own accounting records (normally on accrual basis), leads to problems in reconciling expenditure data. There is also lack of control over agencies with own revenue outside the treasury system.

Diagnosing weaknesses in a particular country

Identifying gaps and weaknesses in expenditure control in a particular country requires a systematic review of the integrity of the expenditure cycle, looking at:

  • Definition/specification of key stages of the expenditure cycle, including the control criteria. The key questions to be asked are: (i) whether all the key stages of the expenditure cycle and associated business rules and processes have been clearly defined; and (ii) whether the required controls at each stage have been clearly specified and consistently applied.

  • Clarity of the legal and regulatory framework, including the roles of the key actors. The key questions to be asked are: (i) whether there are clear laws and financial regulations regarding the controls and the authority and responsibility of relevant actors who should apply them; and (ii) whether the relevant actors understand and apply them in practice.

  • Broader PFM and enforcement issues affecting budget execution and expenditure control. PFM weaknesses such as lack of a comprehensive and credible budget,41 poor cash planning or shallow markets for government debt, reporting delays, and accumulation of liabilities/arrears also undermine the effectiveness of expenditure control. While the controls may be well specified and the roles and responsibilities of the key actors clarified in a country’s legal/regulatory framework, problems may still arise due to lack of enforcement. This includes, but not limited to: failure to check the availability of funding before authorizing expenditure; failure to record and maintain data on commitments; delays in processing of payments; circumvention of controls at key stages, including through collusion; and poor record keeping, including of verification documents.

While particular expenditure control weaknesses may manifest themselves at a particular stage of the expenditure cycle, addressing them may require reforms at various points in the process. Table 4 lists the typical problems that may arise at different stages of the expenditure cycle. Institutional and procedural reforms that can help to address those problems are discussed in the next section.

VI. Strengthening Expenditure Control

Specific measures to address expenditure control weaknesses

Once the specific problems and weaknesses in expenditure control have been identified, the government needs to develop tools and measures to address them. Reforms to address budget execution issues might require clarifying budget execution procedures, introducing or upgrading an FMIS, improving budget warrant/allotment system, establishing commitment control, strengthening cash management, introducing accrual accounting, etc. Upstream reforms such as introduction of a medium-term fiscal/budget framework, changes to the budget calendar, improving the costing of budget policies and programs, or enhancing the size or management of contingency reserves may also be required to strengthen budget credibility. Table 4 lists some specific tools and measures that can address weaknesses at different stages of the expenditure cycle. It also proposes some indicators—mainly based on the PEFA framework—that could be used to assess reform progress.

The main reforms required to address the weaknesses at different stages of the expenditure cycle are as follows:

  • Authorization. The main reforms include enhancing the coverage of the budget, improving the methodology of costing budget policies, introducing medium-term fiscal and budget frameworks, and ensuring timely submission and approval of the budget by the legislature.

  • Apportionment. The key reforms include development of expenditure plans by line agencies and submission of these plans to the ministry of finance for decision on apportionment, and preparation of reliable cash flow forecasts to serve as the basis for apportionment.

  • Reservation/pre-commitment. The reforms should focus on streamlining the procurement process to reduce the time of withholding the reserved funds before they are committed, and ensuring that these reserved funds are in the TSA.

  • Commitment. The main reform at this stage is the introduction of a comprehensive commitment control system, including for multi-year commitments and standing commitments (e.g., salaries, utilities, subsidies, transfers, etc.).

  • Verification. The key reforms include clarifying responsibilities for verification of delivery of goods and/or services, ensuring documentary proof of delivery, and eliminating any undue delay between verification and issuance of payment orders.

  • Payment order. The main thrust of reforms is to ensure that payments are made within the due date to prevent accumulation of payables/arrears, extend the horizon of the cash plan which also reflects expected payments, and eliminate exceptional procedures for payment.

  • Payment. The reforms should focus on modernizing the mode of payment (e.g., through electronic fund transfer), streamlining check floats, introducing active cash management to ensure cash availability for payments, and ensuring regular bank reconciliation.

Table 4.

Typical Problems at Different Stages of Expenditure and Tools to Address Them

article image
article image

The PEFA indicators are based on the new PEFA 2016 framework.

Phasing the reform process

The reforms in the expenditure control framework should be tailored to the specific country context and capacity, and be scaled up as the capacity improves. A key question is whether reforms should focus on strengthening the traditional system or leapfrogging to the modern devolved approach. While the answer to this question will depend on specific country context and weaknesses, in general caution should be exercised in the devolution of controls in countries that are at the initial stage of development of PFM systems (e.g., fragile states), or with weak capacity at line agencies to implement the required controls, and/or without strong institutions of ex post oversight (internal and external audit). Reforms could usefully be implemented in phases as follows:

  • In the first phase, the focus should be on establishing basic control functions such as centralized control of apportionments and simplified/streamlined but effective controls42 at other stages of the expenditure cycle, particularly commitment control backed by cash planning43 linked to timely release of funds to spending agencies. This should be supported by adequate monitoring at each stage of the expenditure cycle and ex post audit—particularly external audit to start with—to ensure effective compliance. In addition to reviewing and streamlining the formal procedures defined by the financial regulations, informal and any “special procedures” and practices that bypass the normal expenditure cycle should be thoroughly reviewed and (to the extent possible) eliminated.44 Business rules and processes should be established to enable faster spending execution to address priority needs rather than relying on informal/special procedures that undermine the effectiveness of the expenditure control framework.

  • In the second phase, the focus should be on progressive devolution of controls to spending agencies in parallel with a reinforcement of procedures for auditing and reporting. This devolution/decentralization of financial and expenditure control should be gradual and based on predefined competency criteria for line ministries/agencies who should demonstrate that they can operate with higher levels of delegated authority. This will be dependent on sustained improvements in financial management standards and management information, and assurance of a control-conscious culture in each agency. Box 4 provides a few examples. A key point to note here is that without enhancing the capacity of the line agencies to meet the specified competency criteria, such devolution of controls are unlikely to take hold (e.g., the case of Thailand discussed in Box 4).

  • In the third phase, a risk-based approach to control (control modulé de la dépense) could be introduced, which in essence comprises the replacement of systematic ex ante control at the line item/transaction level by ex post audit and strengthened oversight. The central idea here is that not all expenditure transactions need to pass through an identical control process. A distinction can be made between the treatment of high value and risk-prone transactions vis-à-vis low value transactions. However, the design and implementation of such a differentiated control arrangement would depend on several factors, including the effectiveness of the internal control and assurance system to identify and alert management to control risks.45

Devolution of Expenditure Control in France, Morocco, and Thailand

France, following the introduction of a new organic budget law in 2001, overhauled the budget execution system by decentralizing it somewhat toward line ministries. In particular, the new law: (i) divests the financial controllers of the responsibility for regularity control (contrôle de régularité) of budget managers (administrateurs de credit), which was transferred/decentralized to the line ministries; (ii) requires the financial controllers to assess the fiscal sustainability of decisions taken by budget managers to reinforce macro-fiscal discipline; and (iii) also requires the controllers of central agencies, such as the General Inspectorate of Finances (Inspection générale des finances), to apply a risk-based approach to control.

Morocco has been implementing a reform of financial and expenditure control since 2006 that seeks to gradually transfer this responsibility from the ministry of finance to line agencies while ensuring adequate safeguards (through strengthened oversight) against the resulting risks of abuse. In this context, the function of financial and expenditure control has been merged with the treasury and verification of the regularity of certain current expenditures (such as salaries, leases and certain procurement contracts) has been delegated to line agencies at the commitment level. For other expenditure items, the devolution is based on the assessed effectiveness (through formal “capacity audits”) of the internal control system of the line agency and its risk management capacity. For this purpose, the spending units were grouped into two categories and the devolution of financial and expenditure control started with the best-performing line agencies at the superior level. This is what constitutes the hierarchical and risk-based control (control modulé de la dépense) that Morocco has started to implement since 2008 (based on Decree n° 2–07-1235 of November 4, 2008).

Thailand introduced a hurdle approach in the late 1990s to devolve budget execution control, moving this function from the finance ministry’s Bureau of the Budget to line agencies. This approach comprised two main components: (i) a set of core financial and performance management competencies (called the hurdles) to be met by each line agency to qualify for delegation of financial management and control; and (ii) semi-contractual arrangements between the Bureau of the Budget and line agencies formally linking the reduction in central control to the achievement of the specified competencies. The hurdles were based on the following indicators of performance by the line agencies: (i) budget planning; (ii) output costing; (iii) financial and performance reporting; (iv) financial control arrangements; (v) procurement management; (vi) asset management; and (vii) internal audit. However, these hurdles were set at such a high level that hardly any agency cleared them. In addition, the approach did not stipulate any time frame for agencies to upgrade their PFM standards, and there was an underestimation of their capacity-building needs. As a result, progress stalled and the reform was not pursued further.

The strengthening of the expenditure control framework should not be viewed as an independent activity and should be integrated with other PFM reforms, including changes to budget execution processes. If an FMIS is planned or under implementation, some measures towards revamping the expenditure control framework will have to be implemented in tandem with the FMIS. For example, a treasury-based centralized payment system with decentralized responsibility for appropriation and commitment controls at the line agency level can be progressively introduced as the FMIS is rolled out (subject to adequate connectivity between the line agencies and the treasury).

Monitoring reform progress and managing the change

The reform strategy and action plan should provide for monitoring the progress. Table 4 lists suggested indicators that could be used to assess progress at different stages of the expenditure cycle. If an FMIS is envisaged, its configuration and rollout should explicitly be linked to planned expenditure control reforms.

A change management strategy should also be developed and implemented, taking into consideration the implications of the reform strategy for diverse stakeholders, from senior officials to agency heads, and the personnel who will support the new systems. Managing the change process would involve communicating effectively to all relevant stakeholders a broad understanding of why the changes are necessary and what objectives are sought to be achieved. Any perceived risks and/or uncertainties should also be adequately addressed.


  • Allen, R., and D. Radev, 2009, “Extrabudgetary Funds,” Technical Notes and Manuals (Washington: International Monetary Fund).

  • Allen, R., and others, 2015, “The Evolving Functions and Organization of Finance Ministries,” Working Paper WP/15/232 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Bouley, D., and others, 2003, “How Do Treasury Systems Operate in sub-Saharan Francophone Africa?OECD Journal on Budgeting, OECD, Vol. 2, No. 4, Paris.

    • Search Google Scholar
    • Export Citation
  • Browne, E., 2010, “Reforming Budget Systems – A Practical Guide,” United States Agency for International Development (USAID).

  • Doe, L., S. Pattanayak, 2008, Financial Control in African Countries, Public Financial Management Technical Guidance Note, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Flynn, S., and M. Pessoa, 2014, “Prevention and Management of Government Expenditure Arrears,” Technical Notes and Manuals, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • HM Treasury, 2013, “Review of Financial Management in Government”, (www.gov.uk). International Organization of Supreme Audit Institutions (INTOSAI), “Guidelines for Internal Control in the Public Sector”, (http://www.issai.org/media/13329/intosai_gov_9100_e.pdf).

    • Search Google Scholar
    • Export Citation
  • Jacobs, D., and others, 2009, “Budget Classification,” Technical Notes and Manuals, (Washington: International Monetary Fund).

  • Khan, A., and M. Pessoa, 2010, “Conceptual Design: A Critical Element of a Government Financial Management Information System Project”, Technical Notes and Manuals (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Khan, A., and M. Pessoa, 2013, “Accrual Budgeting; Opportunities and Challenges,” Chapter 11, PFM and its Emerging Architecture (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Lienert, I., 2010, “Role of the Legislature in the Budget Process,” Technical Notes and Manuals, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Lienert, I., 2009, “Modernizing Cash Management,” Technical Notes and Manuals, (Washington: International Monetary Fund).

  • Lienert, I., 2003, “A Comparison Between Two Public Expenditure Management Systems in Africa,” Working Paper WP/03/2, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Lienert, I., and F. Sarraf, 2001, “Systemic Weaknesses of Budget Management in Anglophone Africa,” Working Paper WP/01/211, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Ministry of Budget and Public Accounts, 2009, Référentiel de comptabilité budgétaire, (France).

  • Moussa, Y., 2004, “Public Expenditure Management in Francophone Africa: A Cross-Country Analysis,” Working Paper WP/04/42, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Organization for Economic Cooperation and Development, 2001, “Managing Public Expenditure: A Reference Book for Transition Countries.” Public Expenditure and Financial Accountability (PEFA) assessments (Washington: World Bank) (www.pefa.org)

    • Search Google Scholar
    • Export Citation
  • Pattanayak, S., and I. Fainboim, 2011, “Treasury Single Account: An Essential Tool for Government Cash Management,” Technical Notes and Manuals (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Pattanayak, S., and J. Cooper, 2011, “Chart of Accounts: A Critical Element of the Public Financial Management Framework,” Technical Notes and Manuals (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Radev, D., and P. Khemani, 2009, “Commitment Controls,” Technical Notes and Manuals, (Washington: International Monetary Fund).

  • World Bank, 2007, “Budgeting and Budgetary Institutions,” Public Sector Governance and Accountability Series.


This TNM has benefited from review and comments of M. Cangiano, M. Fouad, R. Hughes, R. Allen, R. Boukezia, B. Chevauchez, S. Flynn, D. Gentry, T. Hansen, R. Hurtado Arcos, C. Iles, D. Last, J. Menkulasi, D. Moretti, M. Nozaki, B. Olden, M. Pessoa, J Seiwald, H. van Eden, A. Veloz, B. Wiest, and several other colleagues from both the PFM divisions of FAD. The author is also grateful to K. Douglass for her assistance with data analysis.


For a further discussion on expenditure arrears, see S. Flynn and M. Pessoa (2014).


The budget should be implemented as formulated and authorized with as little deviation as possible, but there should be room to adjust to changing circumstances (e.g., genuinely unexpected events) by modifying the budget as necessary during the year. Budget modifications during the year are done according to legally prescribed processes (e.g., virements, contingency reserves, and supplementary/revised budgets), transparently, and in a way that promotes government’s chosen objectives.


Some countries’ PFM systems may not formally track all the seven stages (see discussion in page 9).


Some types of budget appropriations—debt service, for example—may not be subject to a strict spending limit and may be revised according to developments in interest rates and exchange rates.


To provide accountability, the budget proposals should be sub-divided by entity/purpose. This objective is achieved by appropriations. An appropriation is defined as a sub-division of a government budget established for accountability purposes, which shows the amounts legally authorized to be spent for specific purposes in a specific time period. In many countries each appropriation is the subject of a separate vote by the legislature.


For example, in the United States “mandatory” or “entitlement” programs, such as Social Security, Medicare, Medicaid, and certain other programs are not controlled by annual appropriations, except for the requirement to show a corresponding increase or decrease in the costs of these programs due to any envisaged changes.


In some countries, the ministry of finance uses “sequestering” to prevent such risks. Sequestering (or gel/régulation budgétaire in Francophone tradition) is the blocking of appropriations by the ministry of finance. When sequestering appropriations, ongoing commitments should be taken into account. Although sequestering may sometimes be necessary, it diminishes the predictability of budgeted/authorized expenditure and undermines the credibility of the budget, and therefore should be used only in exceptional circumstances.


In some countries, ministries of finance regard expenditure as having taken place when funds are transferred from the ministry of finance or treasury bank accounts to the line ministries (or first-tier spending units). But these funds may take some time to be further transferred to subsidiary spending units under the line ministries and then be spent on the salaries or goods and services that constitute final expenditure. It is, therefore, necessary to distinguish between such final payments by spending units and the apportionment of spending authorization to them, including the associated transfer of funds.


Following the approval of the budget/appropriation bill, spending agencies are usually asked to submit a proposed plan for apportionment/allotment. This plan indicates the funds required for operations, typically on a quarterly or monthly basis.


For further background information and discussion on specific features of commitment control, see D. Radev and P. Khemani (2009).


For a further discussion on expenditure arrears, see S. Flynn and M. Pessoa (2014).


To centralize cash management, all government cash transactions should go through a TSA system (with a set of accounts linked to a top account). Spending units may have bank accounts which are a subsidiary of the TSA; such accounts could be zero-balance accounts with commercial banks, with money transferred into them as the payments are approved, and with their balances swept daily into the TSA’s top account for cash consolidation. For a detailed discussion on TSA, see S. Pattanayak and I. Fainboim (2011).


Check float has two dimensions: (i) check float time; and (ii) check float amount. Check float time is the time between when a check is written and issued as a payment, and when the check is presented by the beneficiary to the bank for encashment. Check float amount is the total amount of outstanding checks that have been issued, but have yet to be encashed. A long check float time is not a good practice as it not only complicates cash management but can also be misused to write and issue checks despite not having enough liquidity in the bank account to cover the value of the check.


For example, payments after the deduction of taxes are frequent in some countries, with negative consequences not only on transparency, but also on both tax collection and competition among suppliers.


In any case, understanding the seven key stages of the expenditure cycle and associated control systems is also important to effectively design and implement an FMIS.


These valuations measure the full costs of paying pension benefits. The valuations also usefully inform the employer and employee contribution rates to make the pension scheme sustainable.


For example, the UK uses the “employer cost cap” mechanism to control future pension spending. The first and each subsequent actuarial valuation report includes valuation results for the purposes of measuring changes in the cost of the pension scheme against the employer cost cap, expressed as a percentage of pensionable pay. If a future valuation shows that the costs of a scheme have risen or have fallen, action needs to be taken (via adjustments to member benefits or member contributions) to return costs to the level of the cap.


These expenditure limits may not be strict limits for all types of expenditure. For example, most countries with a Francophone tradition of budgeting enforce two types of expenditure limits: (i) the crédit limitatif, which enforces a strict limit; and the crédit évaluatif, which is an indicative limit for certain categories of expenditure, e.g., debt service.


For example, the British budgeting system sets appropriations for both the expenses incurred (the “net resource requirement”) and the cash payments to be made (the “net cash requirement”) by each ministry.


If a liability has been incurred by the end of the fiscal year, this would be enough to report the expense or expenditure against the accrual appropriation. Therefore, there would be no under-execution of the budget under these circumstances. However, this may not be true when only a commitment has been incurred but the government does yet have a liability because, for example, the goods and services have not yet been supplied.


In some countries, one single department may be responsible for both treasury management and accounting functions (which may be discharged by different divisions/units within the same department).


In the absence of this coordination, spending ministries/agencies can potentially frontload their commitments even if adequate cash is not likely to be available for payment when the commitments materialize, or the treasury may not raise the necessary finance to meet cash requirements. This may lead to potential arrears.


This includes countries where the ministry of finance assigns a financial controller or a budget offcer to each line ministry in order to control expenditure commitments. For similar examples, see also R. Allen et al, The Evolving Functions and Organization of Finance Ministries, IMF Working Paper WP/15/232 (2015).


Even when payment processing and expenditure controls are decentralized, a central aggregate control on cash is still required.


Overly rigid and controlled spending procedures in several Francophone African countries have resulted in the proliferation and misuse of exceptional spending procedures, e.g., the issuance by the minister of finance to the treasury (bypassing the normal chain of expenditure) of an immediate payment order subject to regularization later. Originally designed for exceptional circumstances, the use of this procedure expanded in several Francophone African countries to settle most of the expenditures. Similarly, the procedure of using an imprest (dépenses par régies d’avance), which does not follow all the control stages and should be used only for urgent minor expenditures, is sometimes used to speed up the expenditure process.


Bank reconciliations, among other things, are critical to identifying potential misappropriation of public money.


Sometimes called the “Westminster” system of PFM. Broadly speaking, it exists not only in the United Kingdom, but also in Australia, New Zealand, Indian sub-continent, and many countries in Africa, Asia, Europe and the Caribbean that were former British colonies. Canada also shares some of these features.


The public accountant assumes personal financial responsibility for compliance with regulations and is accountable to the Court of Accounts (Cours des comptes), the supreme audit institution.


For example, prior to the 1990s, financial control was under the Presidency (Côte d’Ivoire), the Prime Minister’s Office (Senegal), or was the responsibility of a separate ministry. It was centralized by a 1997 directive that applies to all the West African Economic and Monetary Union (WAEMU) countries.


This used to be the case in all but three countries (Cameroon, Mali and Burkina Faso) of the 12 countries that comprise Sub-Saharan Francophone Africa. The system is unlike that in France where each line minister has always been an Ordonnateur Principal. The new 2009 WAEMU/CEMAC directives, however, call for a shift from centralized to decentralized ordonnancement in these countries.


Several Francophone African countries are taking measures recently to reduce the complementary period to one month. Article 78 of the WAEMU Directive No. 7/2009 sets a maximum one month complementary period.


For example, Angola, Cape Verde, Guinea Bissau, Mozambique, and Sao Tome and Principe.


This is mainly prevalent in Austria, Germany, and partially in some neighboring countries.


For example, the commitment stage of the expenditure cycle has received a lot of attention of late, but addressing issues just at the commitment stage may not result in a robust expenditure control framework (and prevention of expenditure arrears) as evidenced in several countries.


Often the same type of verification/control is applied at multiple stages of the expenditure cycle, thereby rendering the control redundant and tedious.


Transactions undertaken using exceptional procedures often end up being registered in suspense accounts that are rarely cleared due to lack of budget cover and are neither properly tracked nor reported.


For example in Austria, as highlighted by external audit a few years ago, there were instances of either delayed or no recording of commitments until their respective payments materialized.


The TSA systems in some of these countries give financial incentives for smoothing expenditure profiles.


The lack of a comprehensive and credible budget particularly affects the authorization (as the expenditure authority is not realistic), commitment (as ongoing/outstanding commitments are not adequately allocated for) and verification (as the accumulated liabilities are not fully reflected in the budget) stages.


In fact, excessive duplication of controls is a common disease in many countries, particularly in Francophone Africa, which not only results in ineffciency in executing expenditure but also leads to rent seeking by officials who intervene at different stages of the expenditure transaction.


The cash plans should be prepared in conformity with budget authorization and systematically take account of ongoing commitments.


There has been a proliferation of “special procedures” in a number of countries (particularly in Africa) that are designed for the benefit of powerful vested interests (who want a faster spending process for specific transactions, closer tracking of certain resources, and/or the accommodation of special institutional interests/arrangements). Special procedures are also sometimes a symptom of the inefficiency of the normal procedures to respond adequately to priority needs.


See Guidelines for Internal Control in the Public Sector at http://www.issai.org/media/13329/intosai_gov_9100_e.pdf.

Expenditure Control: Key Features, Stages, and Actors
Author: Sailendra Pattanayak