CHAPTER 5. Strengthening Institutions to Promote Effective Utilization of Aid

Kevin Fletcher, Sanjeev Gupta, Duncan Last, Gerd Schwartz, Shamsuddin Tareq, Richard Allen, and Isabell Adenauer
Published Date:
April 2008
  • ShareShare
Show Summary Details

Ensuring efficient public expenditure calls for strengthening fiscal institutions, including PFM systems.20 Sound and effective PFM systems are important for several reasons: (1) to increase the prospects of achieving key economic and social priorities; (2) to support transparency and accountability—that is, to ensure that the government is held responsible for managing public resources and that donors and taxpayers have access to information about the allocation and use of such funds; and (3) to reduce the transaction costs of aid-related donor requirements. Weaknesses in PFM systems can undermine budgetary planning, execution, and reporting and result in leakage of scarce public resources.

Weaknesses of Existing Public Financial Management Systems in Low-Income Countries

A number of recent diagnostic studies have evaluated the quality of fiscal institutions and PFM systems in low-income countries.21 The results of these assessments are summarized below.

  • The HIPC Assessments and Action Plans (HIPC-AAP), prepared jointly by the IMF and World Bank, covered 23 eligible low-income countries in 2001 and 26 in 2004.22 These assessments provided the first opportunity to undertake periodic PFM assessments in such countries aimed at measuring progress over time. The 2004 assessment concluded that 19 of the 26 countries assessed still required substantial upgrading of their PFM systems. Budget execution and the ability of countries to track poverty-reducing expenditures were especially weak (Figure 1.)23

  • Report on the Observance of Standards and Codes (ROSC) Fiscal Transparency Module for a sample of 28 PRGF-eligible countries suggests that, in certain areas of fiscal transparency, the performance of low-income countries is not out of line with that of more advanced economies. However, their performance is generally weak in many other areas that are important to a well-functioning PFM system, such as a comprehensive and credible budget and effective audit procedures (Figure 2).

  • The Public Expenditure and Financial Accountability (PEFA) assessments, which began only in 2005, cover just 12 PRGF-eligible countries, although the coverage is expected to expand quite rapidly. Early results suggest a pattern of relatively poor performance across the board in key areas of budget preparation and execution, with a median score of about 2.0 against the international good practice standard of 4.0 (Figure 3).24

  • The World Bank’s Country Policy and Institutional Assessment (CPIA) ratings also display a picture of relatively poor performance by many low-income countries. These ratings are based on an assessment of whether the countries concerned have a comprehensive and credible budget, effective financial management of revenues and expenditures, and timely and accurate fiscal reporting. For 27 countries reviewed in OECD (2006b), scores ranged from 2.0 (weak) to 4.5 (moderately strong) (Figure 4).

  • Recent evaluations of the IMF’s technical assistance activities in low-income countries reached a similar conclusion. One evaluation concluded that, in many countries, budget plans were based on unrealistic assumptions, were not comprehensive, and lacked a medium-term focus; accounting and payments systems and other areas of budget execution were weak; budgetary institutions were fragmented; and broader institutional problems such as weak legislative oversight and poor accountability of senior budget officials were common.25 In some countries reviewed, civil conflict had added to these problems.

Figure 1.HIPC-AAP: PFM Performance by Key Categories

(Number of countries meeting benchmarks in percent of total)1

Source: IMF and World Bank estimates.

Note: HIPC-AAP = Heavily Indebted Poor Countries Assessments and Action Plans; PFM = public financial management.

1 Total number of assessed countries in each year is given in parentheses.

Figure 2.Fiscal ROSC Assessments, 1999–2005: Fiscal Transparency Performance in PRGF-Eligible Countries1

(Percent of countries with strong performance in total)

Source: IMF staff estimates.

Note: NFPE = nonfinancial public enterprises; PRGF = Poverty Reduction and Growth Facility; ROSC = Report on the Observance of Standards and Codes.

1 Based on IMF ROSC Fiscal Transparency Module for 28 PRGF-eligible countries between 1999 and 2005.

Figure 3.PEFA Assessments Undertaken During 2005–06: Aggregate PEFA Ratings by Indicator Categories

(In aggregate score; perfect score = 4)

Source: Data provided by the Public Expenditure and Financial Accountability (PEFA) Secretariat.

Figure 4.CPIA: Quality of Partner Country PFM Systems in 20051

Source: World Bank, Country Policy and Institutional Assessment (CPIA subcomponent 13) 2005. Note: The ratings of PFM systems are based on the following criteria: 1, very weak; 2, weak; 3, moderately weak; 4, moderately strong; 5, strong; 6, very strong.

1OECD (2006b).

The broad conclusion is that, despite sustained efforts in many countries—supported by international financial institutions and donors—to undertake PFM reforms, progress has been uneven. Some countries, such as Ghana, Mali, Mozambique, and Uganda, have shown capacity for quite rapid improvement in their PFM systems, especially those that have promoted reforms in budget management and given incentives to civil servants for better performance (World Bank, 2006). However, reform of PFM systems in many other countries has stagnated. The uneven pattern of progress may reflect changes in political leadership or in the management of key departments, such as the budget office or the national treasury, that are critically important to the reform process. Other reasons include weak capacity in many countries and overdependence on foreign advisors. Another important lesson is that countries have not always prepared an action plan for reform that is well structured and realistic about the prioritization and phasing of reform measures, given the countries’ limited capacity.

It is therefore critical that countries focus on strategic PFM reforms and sequence them appropriately, consistent with the country’s capacity to undertake them. Continued capacity building in public debt management, to help countries develop their own Medium-Term Debt Strategy (MTDS), is also crucial.26 The PFM measures described below are those judged to be especially important in the short term. Work on implementing many of these measures, however, will need to be sustained over a period of years if they are to become fully effective.

Countries that are emerging from conflicts or have suffered major disasters face special challenges in strengthening their fiscal institutions.27 These countries have generally received large injections of aid resources that their PFM systems often are not able to cope with, either because of the logistics of handling a sudden increase in inflows of aid, or because institutions and PFM capacities have been significantly weakened and cannot respond adequately or meet donor expectations. In the experience of many postconflict or disaster-affected countries, there is often a sudden influx of foreign experts to work alongside national staff or sometimes even run the central fiscal institutions until national staff can take over. In such cases, both the focus of PFM reforms and the perception of where the main weaknesses are concentrated may be different from what has been described above. These issues are highlighted in Box 3.

Overall Strategic Planning

The strategic planning process should be anchored in the budget process to help the government achieve its medium- and long-term policy objectives, as defined in the PRSP and MDGs. This includes strengthening the relationship between the planning and budgeting cycles, strengthening the role of the cabinet in strategic decision making, closer integration of the recurrent and development budgets, consolidating responsibility for preparing and executing the budget within the finance ministry, and increasing the coverage of donor-funded development projects in the budget. Another important issue is the development of a top-down budgeting approach that focuses on the correct sequencing of budget decisions—starting with agreement by the cabinet on the overall resource envelope and broad allocations to sectors, and followed by negotiation between the finance ministry and line ministers on detailed expenditure allocation by chapter or line item. Top-down budgeting is often associated with the development of a medium-term approach to budgeting described in the following section. Both require sufficient capacity and effective coordination of decision making through the cabinet, to achieve good results. Some reforms are likely to take many years to become fully effective and functional. In cases in which the planning and finance functions are managed by separate ministries, strong coordination between the two is needed.

Developing a Medium-Term Approach to Budgeting

Box 3.Strengthening Public Financial Management in Postconflict and Disaster-Affected Countries1

Postconflict countries provide insight into the challenges and opportunities that scaling up aid more widely may pose to their public financial management (PFM) systems. The countries concerned may also display special circumstances that affect PFM issues and priorities, such as the following:

  • There is a greater need to consider PFM reforms within the context of any emerging political and constitutional debates. Postconflict countries are often in a state of political and constitutional flux. Early PFM reforms may favor greater centralization of control to increase efficiency, whereas political forces may pull in the opposite direction.

  • Potential loss of physical and institutional infrastructure can greatly increase the capacity constraint faced by the recipient country. Postconflict countries have often experienced some form of regime change, and key government positions may be filled by international experts, diluting local capacity.

  • Large off-budget donor expenditures can increase the volatility of aid revenues as the emergency phase recedes. Changes in donor priorities and withdrawal of support can lead to large projects suddenly being brought on-budget. In the short run, this can crowd out other budget items and lead to ad hoc changes in appropriations.

  • The right balance must be struck between slow, thorough approaches and fast, pragmatic solutions. In emergency situations, it becomes harder to undertake rigorous reviews to tackle fundamental issues and a strategic approach to planning and sequencing PFM reforms.

  • Postconflict countries often have a larger, concentrated donor presence, which can facilitate coordination of priorities and systems, if well managed.

PFM reforms that are likely to require priority attention in postconflict countries include the following:

  • Establishing/strengthening the central budgetary institution.

  • Preparing a credible annual budget with clear and reliable estimates. The first step in many cases is the formulation of an emergency budget, which is by necessity rapidly constructed and subject to revision.

  • Developing a clear government accounting system that can track the use of budget appropriations, produce clear reports on outturns, and support the government’s financial reporting obligations.

  • Undertaking a review of the legal and regulatory framework for budget preparation and execution (which may be necessary if regime change leads to rejection of the existing legal framework).

  • Taking initial steps to harmonize the medium-term fiscal and strategic planning framework so that it transparently shows both donor commitments and planned transfers to the government’s budget.

1 Countries such as Rwanda (1994), Bosnia and Herzegovina (1995), Kosovo (1999), Timor-Leste (1999/2000), Sierra Leone (2000), the Islamic Republic of Afghanistan (2001), the Democratic Republic of the Congo (2002), and Iraq (2003) experienced sharp increases in the volume of aid flows following the cessation of periods of conflict. In others, such as Honduras (1998) and Sri Lanka (2004), the international community provided support following devastating natural disasters.

In an environment of scaled-up aid, a key challenge for low-income countries is to ensure that additional spending helps to achieve the MDGs and maintain macroeconomic stability. Achieving the MDGs will require undertaking ambitious spending programs spanning several years. Financing requirements for such programs may exceed immediate aid commitments. Consequently, countries need to take a longer-term view of their spending needs and potential resource availability—both domestic and external—in fiscal policy formulation. This can be achieved by anchoring fiscal policy in an MTF.28 Such frameworks can help align the budget with the government’s medium-term planning goals, such as progress toward the MDGs and the Poverty Reduction Strategy. In general, MTFs set the overall spending limit, allocate the annual spending envelope across different sectors, and then allocate sectoral spending across different programs and projects. The instruments corresponding to these three tasks are as follows:

  • Medium-term fiscal frameworks (MTFFs): The MTFF outlines the fiscal framework based on projections for broad fiscal aggregates that are consistent with key macroeconomic variables and fiscal targets. Broad fiscal aggregates (for example, revenues and expenditures) are projected based on the expected evolution of macroeconomic variables (for example, growth and inflation) and policy measures. Projecting a realistic path for scaled-up aid is essential to underpin overall spending growth. As noted earlier, the MTFF should be reexamined and updated at least annually to reflect recent developments and shocks to the macroeconomic outlook.

  • Medium-term budget frameworks (MTBFs): The MTBF goes further than the MTFF by allocating the overall spending envelope across sectors (for example, health and education), based on the country’s development priorities, as set out in the PRSP or the national development plan.

  • Medium-term expenditure frameworks (MTEFs): The MTEF outlines the sectoral allocation of spending across programs and projects according to government priorities, based on detailed costing from sectoral analysis. The MTEF also involves a more detailed indication of performance objectives and measures.

Most low-income countries lack adequate capacity to adopt a comprehensive MTEF, but many have the rudimentary form of an MTFF and MTBF in place. Those countries that lack MTFFs could draw on the macroeconomic scenarios developed in the context of DSAs for a basic MTFF. To develop MTBFs, important steps include strengthening the government’s strategic decision-making procedures regarding the budget (for example, by involving the cabinet), improving the relationship between the national planning and budget processes (and between the ministries of finance and planning), and strengthening capacity to prepare sectoral ceilings and estimates on the basis of an improved economic and functional classification of budgetary revenues and expenditures.

Designing and implementing an MTEF, however, is a complex and challenging process that needs to be approached cautiously and step by step. Successful adoption of an MTEF requires, among other things, (1) a capacity for making realistic forecasts of macroeconomic variables over the medium term; (2) procedures for estimating fiscal developments beyond the current and upcoming fiscal years; (3) a review of the decision-making processes of government and associated institutional arrangements; and (4) enhanced coordination between the finance ministry, the planning ministry, and other central agencies and line ministries through the cabinet or cabinet committees. For these reasons, MTEFs have proved difficult to implement in many developing countries: necessary institutional adjustments have not been made and the MTEF has not been fully integrated into the annual budget process29 (Box 4). A recent IMF review concluded that developing an MTEF can be effective when circumstances and capacities permit; otherwise, it can be a great consumer of resources and may distract attention from immediate needs such as improving the annual budget execution process (IMF, 2006b; and World Bank, 2006).

In Africa, in cases in which the preconditions were not right, the MTEF was introduced prematurely and turned out to be largely a paper exercise, partly because it failed to recognize the essentially gritty and political nature of the budget bargaining process, which cannot be reduced to a technocratic exercise.30 For countries whose detailed MTEFs have failed, a simplification process should be encouraged.31

Strengthening Budget Execution and Reporting

In the short term, countries with low capacity should focus on “getting the basics right”—relatively simple reforms of the kind summarized in Box 5.32 The emphasis should be on building capacity in budget execution and reporting. This will help countries avoid both ad hoc decision making on the allocation of budgetary resources and frequently encountered problems, such as expenditure arrears and low-quality and untimely fiscal reports. This will also facilitate efficient and effective use of resources and better tracking of poverty-reducing expenditures.

Box 4.Current Medium-Term Fiscal Planning Practices

Most low-income countries do not have a medium-term framework (MTF) in place, and even where an MTF exists, it is often not well integrated with the budget and not used for analytical purposes. Out of 31 low-income countries examined,1 only 3 had a full-fledged macrofiscal framework in place, 5 had a relatively comprehensive framework, 13 had a basic framework, and 10 had no MTF (see figure below). Only 10 of the 21 low-income countries that had an MTF systematically aligned their yearly budgets with these frameworks. In addition, (1) medium-term fiscal projections and macroeconomic assumptions were often unrealistic and therefore lacked credibility;2 (2) the underlying macroeconomic assumptions for forecasts were often not explicit, with Reports on the Observance of Standards and Codes Fiscal Transparency Module suggesting that 9 out of the 21 low-income countries with an MTF did not explicitly state key macroeconomic assumptions; (3) the fiscal effects of different macroeconomic scenarios were often not quantified; (4) on occasion, long-term policy scenarios were not prepared; and (5) medium-term expenditure estimates often did not reflect expenditure priorities or changing priorities and were not adjusted in a rolling fashion.

Medium-Term Framework (MTF) Rating by Category

(Number of countries in each category)

Source: IMF, fiscal ROSC reports for 31 countries in Africa, Asia, Europe, and Latin America.

1 Based on information collected from the fiscal module of the Report on the Observation of Standards and Codes (ROSC). The country sample includes Albania, Armenia, Azerbaijan, Bangladesh, Benin, Burkina Faso, Cameroon, El Salvador, Equatorial Guinea, Georgia, Ghana, Guatemala, Guyana, Honduras, India, the Kyrgyz Republic, Malawi, Mali, Mauritania, Moldova, Mongolia, Mozambique, Nicaragua, Pakistan, Papua New Guinea, Rwanda, Samoa, Sri Lanka, Tanzania, Uganda, and Zambia.2 Some low-income countries even invite spending agencies to draw up their own macroeconomic assumptions to determine expenditure levels for the annual budget, which are not in line with the budget document.

Box 5.Short-Term Priorities for Public Financial Management Reform

The following short-term priorities should be considered for public financial management system reform:

  • Ensure that the ministry of finance or central budget authority has adequate control over the fiscal aggregates and that budgetary spending is in line with the approved budget.

  • Ensure that the process of preparing the annual budget is comprehensive and coherent, follows a sensible timetable, and is adequately integrated with the priorities-setting process of the national plan or Poverty Reduction Strategy Paper.

  • Establish a budget classification, for administrative and economic categories initially, that complies with international good practice.

  • Introduce a basic accounting system, with some level of automation, to process receipt and payment transactions, record information, and produce timely fiscal reports reconciled with banking data, at least for the main aggregates.

  • Take steps to ensure reasonable cash flow planning and control of spending at key points of the spending chain and at the commitment stage.

  • Establish a simple system for recording donor aid commitments and disbursements and tracking poverty-reducing expenditures.

  • Ensure that staff manuals on budget preparation and expenditure authorization procedures are in place.

The maintenance of an adequate and coherent accounting framework is essential for tracking spending, enforcing accountability, and meeting donors’ fiduciary requirements. At a minimum, a functioning accounting system should include (1) regular bank account reconciliation with accounting records; (2) double-entry accounting procedures, with both general ledger and supporting subsidiary ledgers; (3) a chart of accounts that facilitates reporting according to the budget presentation; (4) periodic and timely in-year consolidation of accounts, where accounting is decentralized; (5) maintenance of adequate accounting records; and (6) preparation of appropriate manuals and training for accounting staff (Box 5). Accounting information should be comparable across years and between the approved budget and data on realized expenditures. Any changes in the accounting basis and principles should be clearly specified and necessary adjustments made to ensure that the data are consistent.

Government banking arrangements need strengthening to improve cash management and accounting and to reduce fiduciary risk. A consolidated banking arrangement—ideally in the form of a treasury single account—also improves the quality of accounting data through effective and timely bank reconciliation. Care should be taken during scaling up to ensure that the number of bank accounts does not proliferate and further overload the low capacity of systems in low-income countries to keep track of the increased volume of transactions passing through the banking system. Over time, donors’ bank accounts should be integrated with the treasury’s accounts to avoid problems with cash management, and their respective accounting and banking records should be reconciled in a timely manner.

Regular and timely fiscal reporting is necessary to assure donors, policymakers, legislators, and other stakeholders that the government is on track in implementing its annual budget. In countries where subnational governments account for a significant portion of government spending, fiscal reporting should cover these entities also. Where capacity constraints limit the coverage of fiscal reports, the main requirement should be for timely and reliable reporting of the central government budget. Although many countries have some form of monthly reporting, the coverage and quality of such reports continue to be weak. To ensure that these fiscal reports and statements are complete, donors should provide full information to the authorities on their planned and actual aid disbursements, whether in cash, in-kind, or by direct disbursement to suppliers, particularly when these are not reported through the treasury.33

A sound system of internal control is necessary to provide reasonable assurance that public expenditure is executed in accordance with the approved budget and the established regulatory framework. The effectiveness of budget execution does not necessarily increase by adding multiple layers of redundant control.34 What is required is that the control function or task be clearly articulated and assigned, understood by controllers, and consistently applied to all transactions.35 Given that payroll expenditure forms a substantial part of public expenditure in many low-income countries, improving payroll management and control should also be given priority.36

Weak or inadequate public procurement systems are one of the main sources of corruption. Strengthening procurement is essential to maintain donor confidence in continued funding through the budget and is one of the key measures identified in the 2005 Paris Declaration. Developing standard bidding documents, streamlining and computerizing the system for recording bids and contract awards, publishing such information in the official gazette, and, later in the process, developing an e-procurement system are some of the steps countries can take to strengthen public procurement. Recent efforts, spearheaded by the World Bank,37 have resulted in a significant number of low-income countries enacting new procurement legislation and establishing procurement regulatory authorities. Although useful progress has been made, much capacity building remains to be done.

Integrating Donor Aid in the Budget Process

Aid disbursed through extrabudgetary channels should be coordinated with budget priorities. According to a recent survey, on average, only 37 percent of external aid is channeled through country PFM systems (OECD, 2006b). This complicates fiscal management. For example, the “3 by 5 Initiative”38 mobilized substantial extrabudgetary resources to expand the number of HIV/AIDS patients receiving antiretroviral drug treatment from 400,000 in 2003 to about 2 million in 2006 (WHO, 2007), and the U.S. President’s Emergency Program for AIDS Relief has a provision for US$15 billion over five years (beginning in 2004) largely for nongovernmental organizations, with spending increased in line with local implementation capacity. Donor preferences for specific programs affect the composition of health spending, although these programs are targeted to diseases that have high mortality.39 In addition, earmarking of aid could introduce inefficiencies in expenditure allocations as noted earlier, straining the capacity of already weak PFM systems, reducing the flexibility of governments to reallocate resources in response to changing conditions and priorities, and hampering the implementation of expenditure smoothing.40 Extrabudgetary flows hinder comprehensive budget planning and could lead to duplication and waste in the absence of a well-functioning coordinating mechanism. However, extrabudgetary aid and expenditure might be warranted in the short term in fragile and postconflict states, where resources cannot be channeled through institutions in an efficient and effective manner.

Strengthening PFM Systems, Including Their Capacity to Track Poverty-Reducing Spending

The emphasis should be on building capacity in budget execution and reporting to ensure efficient and effective use of resources. This capacity includes budget classification, accounting, public procurement, payroll management, and internal control systems. Such reforms will help countries avoid ad hoc decision making on the allocation of budgetary resources and frequently encountered problems, such as expenditure arrears and low-quality and untimely fiscal reports. Efforts to strengthen PFM systems, promote greater transparency, improve budget procedures and reporting, and prepare MTFs should also reassure donors that aid will be used effectively and encourage them to channel more aid through the budget.

Special attention needs to be paid to tracking poverty-reducing public spending to ensure that it reaches the intended recipients (Box 6).41 The key PFM areas identified above are critical to achieving desired expenditure allocations. Techniques such as the use of Public Expenditure Tracking Surveys (PETSs) and audit reports can help identify persistent weaknesses in the expenditure chain by estimating the amount of public funds that actually reaches frontline service delivery units through the budget system.42 These techniques can be helpful in assessing the quality of cash management and internal control systems, can focus attention on the need for reform in these areas, and, eventually, can strengthen mechanisms of internal and external audit. Uganda initiated a PETS to assess potential leakage of funds following a dramatic increase in primary education spending that failed to boost enrollments. The survey found that 87 percent of the nonwage resources were diverted to other uses (World Bank, 2004). Several lessons emerge from similar tracking surveys in Ghana, Honduras, Madagascar, Mozambique, Papua New Guinea, Rwanda, Senegal, Tanzania, and Zambia. In particular, poor resource management can result from excessive discretion in budgetary procedures in an environment with weak internal controls, imperfect information, and vested interests.

Box 6.Illustrative Expenditure Tracking Mechanism

The essential component of an effective mechanism for tracking poverty-reducing expenditures is a budget classification system that conforms with the international standard:

  • At a minimum, a basic economic and administrative classification of expenditure should be put in place. This would allow administrative units to be “aggregated” into sectors that allow poverty-reducing expenditures to be tracked at a broad level;

  • A functional classification of expenditures would provide a better indication of the allocation of expenditures within a sector or subsector, and hence allow improved tracking against the government’s poverty-reducing priorities; and

  • A simple program classification would provide additional information for more detailed policy analysis and evaluation of aid effectiveness.

In addition, for aid that is delivered directly by donors outside the budget, donors should provide timely reports to government on disbursements:

  • At a minimum, such reports should include information on aggregate disbursements by sector (aligned with government definition) and geographical location;

  • If possible, data should be provided on expenditures according to their economic classification. The most basic requirement would be a distinction between recurrent and capital expenditures, which would allow the government to plan for the medium term adequately; and

  • Eventually, donors should report against the functional or program classification of expenditures adopted by the government.

All poverty-reducing spending should be monitored because of fungibility in resource use. If resources are fungible, then a recipient country can offset the scaled-up aid that is intended for additional poverty-reducing programs by lowering its own spending in those areas.43 This problem is compounded by the difficulty of determining a counterfactual, that is, what a government would have spent on poverty reduction in the absence of higher aid flows. Moreover, substantial earmarking of resources for certain poverty programs by vertical funds would require their sustained support to ensure that these programs are maintained over time.

A gradual move toward a results-oriented budget would help strengthen accountability and help assess the effectiveness of different programs. Advanced forms of results-oriented budgeting are not appropriate for low-income countries. Still, some initial steps in this direction can be useful, such as identifying key program objectives and associated “responsibility centers,” defining the intended program outputs, strengthening the link between performance and rewards/sanctions, and implementing pilots in performance evaluation and results-oriented budgeting in selected ministries, such as health and education, where much donor aid has been focused. Burkina Faso, Ethiopia, Ghana, and Mali are examples of low-income countries that have made progress in implementing program and performance budgeting.

Formulating and Implementing PFM Action Plans

Low-income countries should prepare an action plan for strengthening PFM systems based on a comprehensive diagnostic study such as the PEFA framework. Reforms should be sequenced in line with government capacity, for example, by following a properly defined “platform approach” so that core building blocks precede more advanced reforms. The platform approach entails identifying indicators of success for each stage of the overall reform process to determine when to move on to the next stage (Table 1). This approach strengthens the linkage between the functional components of the reform.44 The action plan may also need to be tailored to the unique circumstances of postconflict countries, which usually face weak legal and regulatory frameworks (for example, tax and budget laws), an ill-defined fiscal authority (for example, the ministry of finance), and inappropriate PFM systems (Gupta and others, 2005).

Table 1.Illustrative Platforms for Strengthening Budget Formulation in a Typical Low-Income Country
PlatformsPriority-Setting ProcessMacrofiscal Forecasting CapacityPlanning Instruments and Budget ProcessCapital and Recurrent BudgetingPerformance BudgetingBudget Classification and Chart of AccountsDonor Aid CoordinationCapacity of Government to Manage Scaled-Up Aid
Initial systemTop-down approach, dominated by presidency and MoP with MoF in a supporting role.None.NDP, PRSP, and annual budget—largely disconnected.Dual budget system—MoP responsible for capital investment and MoF for recurrent expenditures.None.Basic system of administrative and economic classification, not compliant with GFSM 2001.Donor aid flows uncoordinated and largely managed outside the budget.Low
First platform: basic system (1–2 years)As in initial system but improved coordination within government and with donors.Small unit in MoP.Simple MTFF. Improved coordination among NDP, PRSP, and annual budget.As in the initial system.Rudimentary classification introduced for pilot sectors.Economic classification compliant with GFSM 2001.Better coordination through government/donor group, but aid largely executed outside the budget.Low
Second platform: intermediate system (3–5 years)Top-down but with involvement of key sector ministries and a stronger role for MoF.Small units in MoP and MoF.MTBF in place. Estimates incorporated in budget—NDP focuses on longer-term priorities.Integrated budget managed by MoF but some role for MoP.Rudimentary performance information for pilot sectors.Functional classification compliant with GFSM 2001. Basic program classification.Strong donor coordination; most aid flows executed by the treasury.Medium
Third platform: final system (6–10 years)Mixed top-down and bottom-up. Prominent role for MoF. Some decentralized decision making to sectors.Strengthened unit in MoF.Functioning MTFF and MTEF in place.Integrated budget managed by MoF.Introduction of program budgeting linked to MTEF.Classification fully compatible with international good practice.Donor aid estimates incorporated in MTFF and budget; aid flows executed by the treasury.High
Source: IMF staff.Note: GFSM 2001 = Government Finance Statistics Manual 2001; MTBF = medium-term budget framework; MTEF = medium-term expenditure framework; MTFF = medium-term fiscal framework; MoF = ministry of finance, MoP = ministry of planning; NDP = National Development Plan; PRSP = Poverty Reduction Strategy Paper.
Source: IMF staff.Note: GFSM 2001 = Government Finance Statistics Manual 2001; MTBF = medium-term budget framework; MTEF = medium-term expenditure framework; MTFF = medium-term fiscal framework; MoF = ministry of finance, MoP = ministry of planning; NDP = National Development Plan; PRSP = Poverty Reduction Strategy Paper.

Key requirements of a successful PFM reform program are as follows:

First, studies have demonstrated that political economy factors are important for the successful implementation of PFM reforms. This helps explain why such reforms are so difficult to achieve and often slow to implement.45 Recent studies of PFM reforms in Ghana, Malawi, and Mozambique demonstrate that, although sound budgetary rules and procedures may be in place, they are frequently distorted by informal practices that determine the actual distribution of budgetary resources (DfID, 2007). The studies reveal that dysfunctions and distortions can occur at all stages of the budgetary cycle. They underscore that demand for better governance and greater accountability is a key driver of change in budgetary systems.46

They also suggest that incremental reforms are more likely to be successful than “big-bang” approaches, especially those that seek to incorporate models of reform imported from other countries.

Second, PFM reforms need to be linked to the annual budget cycle. There are two main reasons for this: (1) budget offices are especially busy at certain points of the year and have little time to focus on new reform initiatives at such times; and (2) certain reforms, for example, a change in the timetable for budget preparation or the budget classification, can be introduced only at the beginning of a fiscal year. In some countries (for example, Tanzania), formal consideration of new reform initiatives has been institutionalized as part of the regular budget calendar and involves consultation with technical assistance providers and other stakeholders.

Third, the main constraints to reform should be identified early on, and means should be found to eliminate or reduce their impact. Such constraints may include the absence of a suitable legal framework, technical issues (for example, the absence of a good budget classification), institutional weaknesses (for example, rivalries among different ministries, agencies, or departments over managing the budget process), or political factors. It is often useful to include an analysis of constraints within the terms of reference of diagnostic studies or technical assistance missions, and to discuss methods of dealing with them with the recipient government.47

Fourth, mechanisms should be put in place to promote effective leadership and coordination of the reform program among the authorities, donors, and other stakeholders. Some countries have established formal mechanisms for coordination, such as the donor partnership developed in Mozambique. Key roles and responsibilities of donors are indicated in Box 7.

The action plan for short-term measures should

  • Focus on key PFM areas for aid utilization. These include the capacity to prepare sectoral ceilings and estimates based on improved functional and budget classifications and strengthened internal controls on budget execution, accounting, and reporting at the central level.

  • Decompose reforms into a core set of functional components. Where appropriate, PFM reforms should be broken down into functional blocks, such as the core design elements, required changes to legislation, information technology procedures, and training. Monitoring progress in each area is essential so that problems can be addressed before they become obstacles to the overall reform process.

  • Take initial steps to give the budget a results orientation. This would allow the governments to get a sense of whether scaled-up spending is having the desired effect on economic and social outcomes.

In addition, the action plan should recognize medium-term reforms where an early start is needed but change will occur more gradually, such as the following:

  • Developing capacity in treasury systems, cash management, and debt management to strengthen budget execution and help countries build their own MTDS.

  • Strengthening the capacity of subnational governments: As noted earlier, delivery of services such as education, health, and sanitation is increasingly being delegated to subnational governments, whose PFM systems are typically weaker than those of the national government. Effective use of aid for such services would require strengthening PFM systems at the subnational level as well.

  • Linking PFM reforms to broader public sector reforms: The reforms of PFM systems can be strengthened if they are part of a broader public sector reform of the civil service, governance and transparency, and the legal framework.

  • Gradually increase the role and capacity of accountability institutions, such as the national audit authority, whose mandate should include undertaking value-for-money audits of key expenditure programs.

The Role of Technical Assistance in Supporting the Reform Process

The international community should support implementation of the action plan through technical assistance that targets the priority areas noted in Box 6. Recent evaluations of that provided by the IMF and World Bank have highlighted common problems in reform programs supported by donor technical assistance. These problems include inadequate diagnostic assessments, overloaded reform agendas, improper sequencing of reforms, political-economy constraints, poorly selected experts and inadequate quality control of their work, and insufficient coordination among the different providers of the assistance.48 In some cases, there have been significant deficiencies and gaps in the design and implementation of the programs.49 In addition, technical assistance recommendations require sustained follow-up by donors to monitor progress on achieving short- and medium-term goals.

Action is required on several fronts to address these concerns. First, effective measurement systems for monitoring both delivery of technical assistance and improvements in PFM systems need to be established. Tools such as the PEFA framework can be useful in this regard. Although there are some current concerns about the quality of the diagnostic assessments carried out under the PEFA framework, this initiative should be encouraged.

Second, effective coordination between technical assistance providers and the authorities is critical to the success of PFM reforms. There are often numerous donors providing such assistance in low-income countries, underscoring the need for effective coordination. The country authorities should ultimately be responsible for coordinating donor activities, but limited capacity or scarce human resources can require that a major donor assume this role. One promising initiative in this field is the enhanced collaboration between the IMF and the World Bank to support PFM reform in a group of African countries—including Burkina Faso, Ghana, Malawi, Mozambique, Rwanda, and Tanzania—using a range of lending instruments and technical assistance operations.50

Box 7.Role of Donors in Promoting Effective Public Financial Management Reform

  • Donors should encourage the national authorities to establish institutional arrangements that facilitate effective communication between the ministry of finance, the ministry of planning, and other agencies (for example, the prime minister’s office) involved in the process of planning and disbursing official aid.

  • Donors should play their part in producing realistic projections of donor disbursements of aid. Although project aid often suffers most from unrealistic projections, IMF-supported programs in low-income countries are often affected by late disbursements of general budget support by the multilateral donors.

  • Donors should resist earmarking aid for specific purposes. Donors often give less attention to the budget priorities of the recipient country, and instead disburse money on their own schedule for projects they conceive to be useful. In other cases, national producers are rewarded by tying aid to national products and services.

  • Donors should ensure that full information is provided on actual aid disbursements, whether in cash, in-kind, or by direct reimbursement to suppliers, and to the extent possible, attempt to satisfy their national authorities’ requests for such information by using the recipient countries’ own reporting and accounting system.

  • Donors should ensure that, once funds are disbursed, the accounting and banking records of donors’ own bank accounts are reconciled in a timely manner. Ideally, donor accounts should be integrated with the government’s treasury system.

  • Donors can do much to encourage transparency, participation, and accountability in public budgeting by supporting meaningful and regular reporting, timely disclosure of financial information, and external oversight of the budget process. They need to be more aware of the political economy factors that influence the behavior of partner governments, including the potential impact of their own behavior on domestic processes.

Third, the IMF is playing an important role in supporting countries as they develop and implement medium-term action plans for PFM reform, in collaboration with other development partners. The IMF has expertise in many of the areas that are critical to support the short-term capacity-building effort in low-income countries—for example, budget classification, accounting, internal control, and fiscal reporting. Such assistance should emphasize country ownership of the reforms; apply lessons learned to use technical assistance more effectively; make effective use of external finance and partnership arrangements with the World Bank and other technical assistance providers, where appropriate; and leverage the resources of staff from headquarters and the Regional Technical Assistance Centers. Good progress has been made: about 60 percent of the 78 PRGF-eligible countries either have a PFM reform plan in place or are in the process of preparing such a plan. The IMF has provided technical support in more than half of these countries.

PFM comprises the institutional framework, systems, and procedures that govern the preparation, execution, and reporting of the budget. The goals of good PFM systems are threefold: (1) overall control and sustainability of public finances; (2) allocation of budgetary resources to sectors and programs in line with government priorities (for example, as set out in the PRSPs and MDGs); and (3) the efficient delivery of public goods and services to final users.

Other studies include the IMF’s Report on the Observance of Standards and Codes (ROSC) Fiscal Transparency Module; World Bank assessments, including Public Expenditure Reviews, Public Expenditure Management and Financial Accountability Reports, Country Financial Accountability Assessments, and Country Procurement Assessment Reports (CPARs); and reports prepared by the multipartner Public Expenditure and Financial Accountability (PEFA) program. For a useful assessment of the available data, see Chapter 6 of World Bank (2006).

The results of both assessment exercises are summarized in IMF and World Bank (2005).

Dorotinsky and Floyd (2004) concluded that, although budget formulation has improved in a number of countries, budget execution and accountability are still weak in the majority. Thus, fewer than one-third of the countries surveyed had budget outturns that were close to the budget as adopted, and 90 percent of African countries surveyed failed to complete final audited accounts within 12 months of the end of the fiscal year, thus rendering meaningful parliamentary oversight impossible.

PEFA is a partnership whose members include the World Bank, the IMF, the European Commission, and several bilateral donors. The PEFA framework incorporates a four-point rating system (A to D, with intermediate ratings also permitted) compared with the three-point scale (A, B, C) used for the HIPC-AAP assessments. See PEFA Secretariat (2005) and its website,, for further information.

See IMF (2004), which reviewed the technical assistance operations in sub-Saharan Africa of the IMF’s Fiscal Affairs Department.

For a detailed discussion see IMF and World Bank (2006).

See, for example, Heller and others (2006).

In one of the more successful African MTEFs, that of Uganda, the authorities spent five years establishing a stable MTFF before gradually starting to develop projections for individual sectors. Other countries, such as Rwanda and Tanzania, spent a few years testing the new approach in a few ministries before applying the MTEF to the whole budget. See also case studies of Malawi, South Africa, and Uganda in Simwaka (2007), Fölscher and Cole (2007), and Kuteesa and others (2007).

Where MTEFs with full activity-based budgeting systems have been introduced, for example, in Ghana and Malawi, these have generated large volumes of detailed information in which the main objective of the MTEF, namely, linking policy to budgetary allocations and introducing reliable forward estimates, has been lost.

For example, governments could pilot major new spending initiatives before moving to full-scale implementation and develop expenditure tracking surveys. To help channel funds into the most cost-effective interventions, governments could assess the costs and benefits of new programs on a pilot basis, if possible using randomized trials. If such assessments strain the capacity of low-income countries, donors, nongovernmental organizations, and academics (for example, MIT’s Poverty Action Lab) could be approached to carry out these evaluations. Finally, governments should develop expenditure tracking surveys to assess and improve the administrative efficiency of service delivery structures. Such surveys have been used successfully to identify and reduce expenditure leakages in Uganda and other countries (Reinikka and Svensson, 2006).

Donors could also encourage recipient countries to participate in the General Data Dissemination System (GDDS) and finance a scheme for data improvement developed under the GDDS.

This is the case in some Francophone African countries that exhibit characteristics (for example, emergence of arrears) of a poorly functioning control system in spite of multiple controls being applied at various stages of the expenditure chain. The IMF Fiscal Affairs Department’s technical assistance advice, in such cases, has been to simplify and rationalize the complex structure of control.

All expenditure transactions need not pass through the same control/verification process, and a case could be made for a differential treatment of high-value and risk-prone transactions versus low-value transactions. The design of such a control system would depend on several factors, including the effectiveness of an ex post internal audit system to identify irregular transactions.

Ghana is a good example of a low-income country that has made encouraging progress in strengthening its personnel and payroll system. It adopted a new computerized Integrated Payroll and Personnel Database in 2006 and is finalizing the integration of these data into the budget.

World Bank policy is to build sufficient capacity in countries so that loans can be channeled through their national procurement systems rather than being ring-fenced as donor operations. Diagnostic analysis of these systems, and measures recommended to bring them in line with international good practice, are provided through the World Bank’s CPAR.

In 2003, the World Health Organization (WHO) and Joint United Nations Programme on HIV/AIDS (UNAIDS) launched the “3 by 5 Initiative” to treat 3 million people living with HIV/AIDS by 2005.

The share of donor support for certain infectious diseases doubled during 1999–2004 compared with 1990–98, while allocations for basic health remained broadly unchanged during the same period (OECD, 2006a).

Earmarking aid for specific uses preferred by individual donors has not been uncommon, and comes in different guises, such as providing direct financing for specific projects or tying aid to purchases from vendors in the donor country. For the latter, the United Nations (2005) recently estimated that it reduces the value of aid by 11 to 30 percent. Although official donors and civil society organizations have reaffirmed their intention to reduce earmarking to increase aid effectiveness by aligning aid better with country strategies, it is widely recognized that progress in this area has been limited (OECD, 2005). Going forward, a significant portion of scaled-up aid might continue to be earmarked for specific purposes. An example is the new International Drug Purchase Facility, which raises funds that are earmarked for drug purchases, financed by a levy on air travel.

Countries typically define “poverty-reducing” expenditure in their PRSPs. This implies that some variation in the definition is likely across countries. However, most countries regard spending on basic health, primary education, agriculture, infrastructure, housing, basic sanitation, and HIV/AIDS as poverty-reducing expenditures.

It is not possible to track all types of spending in this way, such as certain large-scale infrastructure spending.

Feyzioglu, Swaroop, and Zhu (1998) have shown that earmarked project aid to developing countries is largely fungible.

The platform approach—described in Brooke (2003)—has been used to help avoid initiating specific reforms for which the preconditions for success are not yet in place. For example, the implementation of a government financial management information system has often been undermined by an exclusive focus on information technology systems without sufficient consideration of other functional components of the reform, such as the review and redesign of business processes, as needed, and the identification of user requirements. Similarly, MTEFs have often been established in low-income countries before essential preconditions are in place, thus undermining their purpose and effectiveness. The platform approach, however, has been applied in only two countries (Cambodia and Kenya) to date, and it is too early to evaluate its effectiveness.

For example, the development of a new budget classification system is a basic building block for other reforms, such as an efficient accounting and reporting system, a treasury system, and a simple performance monitoring system. Such a system can take two to three years to implement, and another four to five years to computerize.

A recent study of budget reform in South Africa highlights the importance of political commitment, a simple framework and transparent norms, comprehensive implementation, convincing stakeholders of the importance of reform, strong leadership by a central agency, building capacity, and demonstrating quick wins (see Fölscher and Cole, 2007).

This approach could build on interesting work done by Hausmann, Rodrik, and Velasco (2006) in relation to alleviating the most binding constraints on economic growth.

One recent evaluation study found that some recommendations were not sufficiently detailed in regard to strengthening the budget office (Chad), reviewing the budgetary process (Madagascar), developing analytical tools for reviewing the budget (Niger), and elaborating a macroeconomic framework for budgetary projections (Côte d’Ivoire and Senegal). See IMF (2006b).

IMF support for PFM reform is primarily through the PRGF program and technical assistance (for example, placing resident PFM advisers in the ministry of finance). Key World Bank instruments include its development policy lending and investment lending operations, as well as related economic and sector work such as public expenditure reviews.

    Other Resources Citing This Publication