## _wp0996

## Source details

**Canonical URL:** [_wp0996](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0996.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0996.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0996.pdf.json)

---

### I. INTRODUCTION — scope and framing
- Interest in strengthening budgetary institutions and public financial management (PFM) traced for at least two thousand years; example: Roman planners of the Claudian aqueducts considered eventual O&M costs in selecting routes and designs.
- Modern reforms span from seventeenth-century English “tally sticks” to techniques including:
  - fiscal rules,
  - fiscal risk analysis,
  - expenditure ceilings,
  - medium-term fiscal frameworks,
  - performance-related budgeting,
  - accrual accounting and budgeting,
  - expenditure review.
- Key questions addressed:
  - What determines development of budgetary institutions over time?
  - What lessons can developing countries learn from advanced-country experience?
  - How can IFIs (especially the IMF and World Bank) and other providers facilitate reform—what adjustments to current approaches are required?
- Conclusions are tentative and indicate areas for further research.

### II. Historical development and selected milestones
- Institutional evolution linked to North, Wallis, and Weingast (2006) and North (1991): societies progress through primitive societies, natural states (or “limited access orders”), and “open-access orders.”
  - Natural states: dominated by elites, vulnerable to violence, exist on a continuum from fragile states to mature natural states (e.g., emerging markets); comprise approximately 95 percent of countries (NWW, 2006).
- Pattern: basic accounting, budgeting, and financial reporting systems established in the 19th century followed by roughly 100 years of refinement; modern budgeting framework (unity, comprehensiveness, and control) emerged by end of 19th century in Europe.
- Selected country milestones (chronological highlights):
  - France: 1791 Accounting Office reporting to parliament; 1807 independent “Cour des comptes”; 1862 imperial decree on rules for budgeting and treasury single account; 1959 medium-term budget framework for investments; 1968 “Rationalisation des choix budgetaires” (RCB); 2001–06 program budgeting; from 2006 accrual accounting; 2008 full medium-term expenditure framework (MTEF).
  - United Kingdom: 1787 Consolidated Fund established; 1866 Exchequer and Audit Departments Act and establishment of Comptroller and Auditor General; 1960s Public Expenditure Survey (PES) and Program Assessment Review (PAR); 1980s Next Steps Program; 1990s comprehensive multi-annual budgeting; 1991 Citizen’s Charter; 1998 Public Service Agreements; 2000–04 resource (accrual) budgeting.
  - United States (Federal): 1776 Treasury Office of Accounts established; 1809 Appropriations Act (modified 1870 and 1874); 1887–89 consolidated accounting procedures (Cockrill Commission); 1894 “Dockery Act”; 1921 Budgeting and Accounting Act (Bureau of the Budget and General Accounting Office); 1940 consolidation of uniform standards; 1950 Accounting and Auditing Act; 1982 Federal Managers Financial Integrity Act; 1990 Chief Financial Officers Act; 1993 Government Performance and Results Act; 1994 Government Management Reform Act.

### III. The “new wave” of reforms since the 1970s
- Initiated in New Zealand; followed by Australia, Canada, Denmark, France, the Netherlands, Sweden, the U.K., and the U.S.
- Characteristics (Westcott, 2008):
  - budget consolidation and restructuring;
  - move to multiannual fiscal and budget frameworks;
  - regular use of performance information within the budget process;
  - shift from cost accounting to accrual accounting;
  - development of computerized information systems;
  - consolidation of revenue collection;
  - greater use of devolved budget management.
- Leadership varied: New Zealand led by finance ministers with treasury support; France led by parliamentarians then ministry of economy and national finance.

### IV. Challenges for developing countries in reforming budgetary systems
- Core constraints:
  - poor quality of public institutions;
  - weak centers of government and cabinet systems causing policy coordination and planning inefficiencies;
  - strong patronage systems filling public agency heads with presidential supporters;
  - weak capacity in human resources and information systems;
  - insufficient financial resources for technical systems and capacity building.
- Relevance of NWW “doorstep conditions”:
  - Rule of (budget) law (condition #1) is important; laws adopted but not implemented do not meet this condition.
  - Perpetual organizational forms and empowered ministries of finance (condition #2) needed to replace overriding fiscal power of presidents in many low-income countries.
  - Political control of the military (condition #3) matters where anarchy or civil disruption exists.
- Role of finance ministers:
  - Can coordinate and drive improvements but often lack the powerful status seen in developed countries.
  - Fragmentation of budget responsibility (e.g., capital managed by minister of economy and recurrent by minister of finance) weakens leadership.
- Empirical evidence and pace of reform:
  - Systematic long-term data lacking; reforms are often slow (Gupta and others, 2007; IMF technical assistance).
  - Isolated technical progress (revised budget calendar, commitment control, simple cash accounting) occurs, but comprehensive reform remains slow.
  - IEG World Bank study (2008) tentatively indicates some improvement in Bank-supported countries during 1999–2006, but CPIA data may be problematic and not independently validated.
  - Gupta and others (2007) did not find a general trend toward improvement for low-income countries as a group.
  - Kohnert (2008) documents slow-moving history of UN technical assistance in fiscal areas dating to the 1950s and 1960s.
- Variation across countries:
  - Some emerging markets (Chile, central and eastern Europe, Korea, South Africa, and others) show more rapid improvement, but often remain maturing with sustainability concerns highlighted by the global financial crisis.
  - Post-colonial deterioration: basics present at independence have sometimes declined due to neglect, rise of elites, donor dependence, loss of technical skills, and decline in civil service salaries.
  - Post-conflict countries (Afghanistan, East Timor, Liberia) may have more favorable initial conditions for institution building due to institutional vacuums and strong donor presence, but face donor dependency risks.

### V. Areas prone to rent-seeking and comparative ease of reforms
- Budget process areas especially vulnerable to rent-seeking:
  - control of allocation of appropriations by ministry or sector;
  - planning and management of government investment projects, including procurement;
  - control of decisions relating to planning and management of external aid;
  - management and control of government bank accounts (resistance to consolidating into a treasury single account);
  - management of cash disbursed through the budget;
  - management of the external audit process.
- Expenditure-side reforms are generally harder than revenue-side reforms because they close loopholes, enhance controls and transparency, and reduce rent-seeking opportunities.
- Tax reforms generally easier than expenditure reforms in developing countries; examples include introduction of VAT, unified tax and customs authorities, and large taxpayer offices.
- Possible explanation: tax modernization can create new revenue sources and rent-seeking opportunities aligning incentives of elites and policymakers; expenditure reforms are technically more complex and politically sensitive.

### VI. Donor and IFI influence, incentives, and diagnostics
- Donor dynamics:
  - Donors provide substantial funding and technical assistance, creating donor dependency and inhibiting local capacity development.
  - Donor-driven projects can be elaborate (e.g., computerized systems) and inappropriate or poorly timed.
  - Consultants may extract large fees without direct accountability and be complicit in rent-seeking dynamics.
- Aid effectiveness benchmarks:
  - Paris Declaration (2005) and Accra Agenda for Action (2008) set benchmarks including increasing donor aid “on budget” from about 45 percent to 80 percent by 2010; target may be unachievable given slow PFM progress and donors’ fiduciary constraints.
- Differences in World Bank and IMF perspectives:
  - World Bank: broad public sector reform linked to long-term development, governance, anti-corruption, and reducing fiduciary risk; microeconomic perspective using PER, CFAA, CPAR.
  - IMF: budget as instrument for macrofiscal stability and sustainability; priorities include strengthening ministries of finance, fiscal transparency, reducing fiscal risk, broadening tax base, and revenue collection.
- IMF technical assistance trends (sample of 81 low-income and middle-income countries, past five years):
  - Reforms of treasury and cash management: 41 percent of the countries surveyed
  - IFMIS systems: 38 percent
  - Accounting and budget classification: 38 percent
  - Medium-term fiscal and expenditure frameworks: 37 percent
  - Budget legislation: 29 percent
  - Fiscal transparency: 23 percent
  - Performance and program budgeting: 22 percent
  - Commitment controls: 21 percent
- Diagnostic instruments and limitations (PEFA):
  - PEFA includes 28 performance indicators and three indicators describing donor practices; about 80 countries have undertaken some PEFA assessment in the past four years.
  - Limitations: does not account for weak institutions and governance critical to improving budgetary systems; ratings partially subjective; lacks rigorous quality control; variability of ratings can be politically influenced.
  - Risk: donors may use PEFA as a condition for budget support, creating incentives for governments to collude with donors to boost PEFA ratings.
  - Supplementary institutional and governance reviews exist but are resisted and infrequently carried out.

### VII. MTEF, IFMIS, performance budgeting: experience and outcomes
- MTEF and IFMIS programs have been “dependent upon large injections of external funding and technical assistance.”
- Surveys and evaluations:
  - Survey of MTEF and IFMIS projects in Ghana, Tanzania, and Uganda: “mixed success” (Wynne, 2005).
  - Schiavo-Campo (2009) summarizes a decade of MTEF experiments in Africa:
    - positive: “greater awareness of the need to look beyond the annual budget horizon, and to focus on the results of government spending”;
    - negative: “costly failures” from premature implementation, little or no local ownership, distraction from basic PFM improvements, heavy stress on limited budget capacity, and little improvement in macroeconomic balances, financial control and predictability, or spending efficiency.
  - Brumby (2008) reaches similar conclusions.
  - Earlier Bank flagship projects (PIPs in 1980s, cash budgeting) “also largely failed to deliver their expected benefits” (Lienert and Sarraf, 2001).
- Performance-related budgeting:
  - Strongly supported by IFIs and donors; linked to shift toward budget support and development of PRSPs and monitoring systems.
  - In many cases has “degenerated to a ritualistic exercise” producing “vast databases of redundant and unused information” not linked to the budget (“Data cemeteries” / zahlenfriedhof).

### VIII. Sequencing, “platform approach,” and critiques
- Core constraint: “absence of an enabling institutional environment,” not merely lack of technical skills or IT systems.
- “Basics first” argument (Schick, 1998): focus on effective control of inputs, cash-based accounts before accruals, and effective financial audit before performance audit.
  - Andrews (2006) criticizes “basics first”; paper finds critique unconvincing for low-income countries and highlights risks of weak basics (loss of expenditure control, widening fiscal deficits).
- Platform approach (Brooke, 2003):
  - Package PFM reforms into sequential “platforms”; overall reform strategy may span ten years with four or five platforms, each lasting two or three years.
  - Claimed advantages: structured sequencing, clarity of roles, focus on interconnection among measures, realistic migration paths via small steps.
  - Kenya example (Strategy to Revitalize Public Financial Management, April 2006) illustrates Platforms 1–4 from short-term credibility and payroll/procurement improvements to long-term service delivery and resource allocation.
- Critiques and practical difficulties:
  - Cambodia’s platform experiment revealed major flaws: insufficient emphasis on institutional constraints; overloading of activities; unrealistic time horizon; poor prioritization; micromanagement; skewed incentives toward deliverables over system improvements.
  - Fundamental objection: difficulty of unambiguously defining stages/platforms (e.g., disagreement on what constitutes a “credible” budget).
  - If goals cannot be rigorously defined and measured, platforms become operationally meaningless.
- Suggested refinements:
  - Distinguish functional components into subplatforms addressing budget laws and regulations, business process changes, IT systems, organizational upgrades, and training/capacity building.
  - Caveat: refinements are difficult in practice given platform challenges.

### IX. Box: Cambodia — weaknesses of the “platform approach” (key points)
- Weaknesses identified:
  - Insufficient emphasis on institutional constraints; underestimated resistance on payroll transfer, bank account consolidation, and arrears prevention.
  - Overloading: Platform 1 included 27 activities (14 needed to prepare ground for Platforms 2–4) and more than 250 specific actions; some activities (e.g., piloting performance-based budgeting) were questionable given weak capacity.
  - Unrealistic time horizon: developing countries cannot undertake more than a few reform measures within a three- to four-year period.
  - Poor prioritization: decisions to proceed with IFMIS premature without basic accounting and treasury improvements.
  - Micromanagement: implementation dominated by technicalities and hundreds of deliverables; insufficient attention to change management at ground level.
  - Skewed incentives: implementers focus on performance indicators or deliverables rather than improving underlying PFM systems.
- Suggested refinements to platform design:
  - Create subplatforms by functional component (budget laws, business processes, IT, organization, training) to ensure components advance broadly in step.
  - Streamlined alternative used in South East Europe: divide 10 important areas into basic reforms (minimum capacity) and advanced reforms (move toward international good practice); consistent with IMF fiscal transparency code and PEFA.
  - Limitations: subjective distinction between “basic” and “advanced”; unclear criteria for transition; absence of analysis of underlying institutional conditions.
- Implications and practical principles distilled:
  - Recognize broader goals: tie budget reform to fiscal objectives (macroeconomic stability, protecting funds from theft, spending consistent with approved budget, linking public priorities and the budget, monitoring results and external accountability).
  - Get the basics right for low-capacity countries: comprehensive and credible budget, sound accounting and control to avoid payment arrears, and stronger links between budgeting and poverty-reduction strategy.
  - Favor selection over broad multi-platform sequencing; sequencing still relevant for specific components.
  - Avoid simplistic prioritization based only on worst diagnostic scores; adopt nuanced approach considering regulatory framework, business processes, human resources, IT, external finance, interdependencies, and political drivers.
  - Strengthen coordination and impartial advice; need an honest broker to provide realistic, impartial guidance on expectations, scope, and sequencing.
  - Practical design guidance:
    - Avoid overly complex, rigid, technocratic programs.
    - Link budget modernization to broader public administration reform and merit-based civil service.
    - Focus on changes that impel further changes (Hirschman, 1958).
    - Provide inducements and incentives (financial and nonfinancial) to civil servants to support reform.
    - Raise and institutionalize professional standards in economics, accounting, and financial management; introduce codes of conduct and whistle-blowing legislation where appropriate.
    - Define required outputs/outcomes and tests of success; engage middle- and low-ranked budget officials; hold officials and advisors accountable for implementation failures.

*Source: _wp0996 (excerpts and summaries from Sections I–III, MTEF/IFMIS discussion, and Box 1).*

### 1. Selected Dates in the Development of Budget Systems: France, the U.K., and the U.S.......6

### 1. Selected Dates in the Development of Budget Systems: France, the U.K., and the U.S.......6

### I. INTRODUCTION — scope and framing
- Interest in strengthening budgetary institutions and public financial management (PFM) can be traced back for at least two thousand years.
- Historical example: Roman planners of the Claudian aqueducts considered eventual O&M costs in selecting alternative routes and designs.
- Modern evolution: reforms range from the “tally sticks” used to record the budget in seventeenth century England to contemporary techniques including:
  - fiscal rules,
  - fiscal risk analysis,
  - expenditure ceilings,
  - medium-term fiscal frameworks,
  - performance-related budgeting,
  - accrual accounting and budgeting,
  - expenditure review.
- In Europe and the United States, a detailed history of the development of budget systems goes back for two hundred years or more.
- The processes and determinants of this evolution, while critically important, are imperfectly understood.

### Core questions addressed by the paper
- What are the main factors that determine the development of budgetary institutions systems over time?
- What lessons can developing countries learn from the long experience of more advanced countries in improving their budgetary institutions?
- How can the international financial institutions (IFIs), especially the IMF and the World Bank, and other providers of financial and technical support, facilitate the process of reform in developing countries—what adjustments are required to the approaches and models they currently apply?

### Nature of conclusions and research needs
- Conclusions reached by the paper are tentative and to some degree subjective, and will be controversial to some readers.
- The paper identifies several areas where further research would be helpful.

### Structure (as stated)
- Section II provides a conceptual framework for strengthening budgetary institutions and an historical perspective.

*Source: _wp0996 - 1. Selected Dates in the Development of Budget Systems: France, the U.K., and the U.S.......6*

### Section III outlines the challenges for developing countries in reforming their budgetary

### _wp0996 - Section III outlines the challenges for developing countries in reforming their budgetary

### Historical development of budget reforms
- Reform of budgetary institutions is linked to the development of political and economic institutions as described by North, Wallis, and Weingast (2006) and North (1991).
- Societies pass through three stages: primitive societies; natural states (or “limited access orders”); and “open-access orders”.
- Natural states:
  - Are dominated by elites with primary access to power and resources and are vulnerable to violence and political conflict.
  - Tend to perpetuate for very long periods.
  - Exist on a continuum from fragile states to mature natural states (e.g., emerging markets).
  - Comprise approximately 95 percent of countries (NWW, 2006).
- Institutional development is not strictly linear; political institutions generally develop before economic institutions, which in turn precede budgetary institutions, but exceptions and discontinuities exist.

### Key historical milestones (selected)
- France:
  - 1791: Accounting Office reporting to parliament
  - 1807: Independent “Cour des comptes”
  - 1814–1819: First Restoration—Baron Louis’ reforms
  - 1862: Imperial decree on rules for budgeting and treasury single account
  - 1959: Medium-term budget framework for investments
  - 1968: “Rationalisation des choix budgetaires” (RCB)
  - 2001–06: Program budgeting
  - From 2006: Accrual accounting
  - 2008: Full medium-term expenditure framework (MTEF)
- The United Kingdom:
  - 1787: Consolidated Fund established
  - 1866: Exchequer and Audit Departments Act (established modern budgeting and accounting system)
  - 1866: Comptroller and Auditor General established
  - 1960s: Public Expenditure Survey (PES) and Program Assessment Review (PAR)
  - 1980s: Next Steps Program
  - 1990s: Comprehensive multi-annual budgeting
  - 1991: Citizen’s Charter
  - 1998: Public Service Agreements
  - 2000–04: Resource (accrual) budgeting
- United States (Federal):
  - 1776: Treasury Office of Accounts established
  - 1809: Appropriations Act (modified in 1870 and 1874)
  - 1887–89: Consolidated accounting, bookkeeping, reporting procedures (Cockrill Commission)
  - 1894: “Dockery Act” established Comptroller of the Treasury; consolidated annual statement of revenues and expenditures
  - 1921: Budgeting and Accounting Act established Bureau of the Budget and General Accounting Office
  - 1940: Consolidation of uniform standards and procedures for accounting and reporting
  - 1950: Accounting and Auditing Act
  - 1982: Federal Managers Financial Integrity Act
  - 1990: Chief Financial Officers Act
  - 1993: Government Performance and Results Act
  - 1994: Government Management Reform Act
- Pattern observed: establishment of basic accounting, budgeting, and financial reporting systems in the 19th century followed by approximately 100 years of refinement and consolidation; modern budgeting framework (unity, comprehensiveness, and control) emerged by end of 19th century in Europe.

### The “new wave” of reforms since the 1970s
- Initiated in New Zealand and followed by Anglo-Saxon and Northern European countries including Australia, Canada, Denmark, France, the Netherlands, Sweden, the United Kingdom, and the United States.
- Characteristics and areas covered (Westcott, 2008):
  - Budget consolidation and restructuring
  - Move to multiannual fiscal and budget frameworks
  - Regular use of performance information within the budget process
  - Shift from cost accounting to accrual accounting
  - Development of computerized information systems
  - Consolidation of revenue collection
  - Greater use of devolved budget management
- Leadership and political support varied by country:
  - New Zealand: led by successive finance ministers with treasury support
  - France: led by parliamentarians then taken up by the ministry of economy and national finance

### Challenges facing reformers in developing countries (Section III)
- Core constraints:
  - Poor quality of public institutions
  - Weak centers of government and cabinet systems causing policy coordination and planning inefficiencies
  - Strong patronage systems filling heads of public agencies with friends and followers of the president
  - Weak capacity in human resources and information systems
  - Insufficient financial resources for technical systems and capacity building
- Relevance of NWW “doorstep conditions” to budget reform:
  - Rule of (budget) law (condition #1) is an important precondition for improving PFM; laws adopted but not implemented do not meet this condition.
  - Perpetual forms of organization and empowered ministries of finance (condition #2) are required to replace overriding fiscal power of presidents in many low-income countries.
  - Political control of the military (condition #3) matters primarily where anarchy or civil disruption exists.
- Role of finance ministers:
  - Finance ministers can coordinate and drive budget improvements but often lack the powerful status seen in developed countries.
  - Fragmentation of budget responsibility (e.g., capital investment managed by minister of economy and recurrent expenditures by minister of finance) weakens finance minister leadership.
- Empirical evidence and pace of reform:
  - Systematic long-term data are lacking on time required for budgetary improvements, but experience summarized in Gupta and others (2007) and IMF technical assistance indicates reforms are often slow.
  - Isolated technical progress (e.g., revised budget calendar, commitment control system, simple cash accounting) has occurred but comprehensive reform remains slow.
  - The IEG World Bank study (2008) tentatively indicates some improvement in PFM systems in Bank-supported countries during 1999–2006, but the data (CPIA) may be problematic and not independently validated.
  - Gupta and others (2007) did not find a general trend toward improvement for low-income countries as a group.
  - Kohnert (2008) documents the slow-moving history of UN technical assistance in fiscal areas dating to the 1950s and 1960s.
- Variation across countries:
  - Some emerging markets (Chile, central and eastern Europe, Korea, South Africa, and others) show more rapid improvement, but these are exceptions and their reforms may still be maturing with sustainability concerns highlighted by the global financial crisis.
  - Post-colonial deterioration: basics of budget systems existing at independence (commitment controls, single treasury account, comprehensive budget, regular budget calendar, end-of-year accounts, internal and external audit, in-year reporting) have sometimes declined due to neglect of rules, rise of elites, donor dependence, loss of technical skills, and decline in civil service salaries.
  - Post-conflict countries (Afghanistan, East Timor, Liberia) present special cases where initial conditions for institution building may be more favorable due to institutional vacuums and strong donor presence, albeit with risks of donor dependency.

### Areas of the budget process prone to rent-seeking and reform resistance
- Areas especially vulnerable to rent-seeking that frustrate reform include:
  - Control of the allocation of budgetary appropriations by ministry or sector
  - Planning and management of government investment projects, including procurement operations
  - Control of decisions relating to planning and management of external aid
  - Management and control of government bank accounts (resistance to consolidating into a treasury single account)
  - Management of cash disbursed through the budget
  - Management of the external audit process
- Expenditure-side reforms are generally harder than revenue-side reforms because expenditure reforms tend to close loopholes, introduce controls, enhance transparency, and reduce rent-seeking opportunities.

### Comparative ease of tax versus expenditure reforms
- IMF and World Bank technical assistance experience suggests:
  - Reform of tax policy and tax collection has generally proved easier in developing countries than reform of expenditure processes.
  - Examples of tax-side progress include introduction of VAT systems, unified tax and customs authorities, and large taxpayer offices.
- Possible explanations:
  - Tax modernization can create new revenue sources and opportunities for rent-seeking, possibly aligning incentives of elites and policymakers.
  - Expenditure-side reform is technically more complex, covers a wider set of functions and organizations, and tends to reduce rent-seeking opportunities.

### The dominating influence of donors and multilateral institutions
- Donors often provide substantial funding for reforms and supply technical assistance, which can create donor dependency and inhibit development of local capacity.
- Donor-driven projects may emphasize elaborate initiatives (e.g., computerized information systems) that may be inappropriate or poorly timed for country circumstances.
- Consultants can extract large fees without direct accountability for results and be complicit in rent-seeking dynamics.
- The Paris Declaration on Aid Effectiveness (2005) and Accra Agenda for Action (2008) set benchmarks, including increasing donor aid “on budget” from about 45 percent to 80 percent by 2010; the target may be unachievable given slow PFM progress and donors’ fiduciary constraints.
- Differences in mandates and perspectives of the World Bank and IMF:
  - World Bank: focus on broad public sector reform linked to long-term developmental objectives, governance, anti-corruption, and reducing fiduciary risk of Bank lending; emphasis on microeconomic perspective and tools such as PER, CFAA, CPAR.
  - IMF: regards the budget as an instrument to promote macrofiscal stability and sustainability; priorities include strengthening ministries of finance, fiscal transparency, reducing fiscal risk, broadening tax base, and revenue collection.
- IMF technical assistance trends (sample of 81 low-income and middle-income countries, past five years):
  - Reforms of treasury and cash management: 41 percent of the countries surveyed
  - IFMIS systems: 38 percent
  - Accounting and budget classification: 38 percent
  - Medium-term fiscal and expenditure frameworks: 37 percent
  - Budget legislation: 29 percent
  - Fiscal transparency: 23 percent
  - Performance and program budgeting: 22 percent
  - Commitment controls: 21 percent

*Source: _wp0996 - Section III outlines the challenges for developing countries in reforming their budgetary*

### introduction of a medium-term expenditure framework (MTEF) and integrated financial

### _wp0996 - introduction of a medium-term expenditure framework (MTEF) and integrated financial management information systems (IFMIS)

### Experience and outcomes of MTEF/IFMIS and performance budgeting
- MTEF and IFMIS programs have been “dependent upon large injections of external funding and technical assistance.”
- A recent survey of MTEF and IFMIS projects in Ghana, Tanzania, and Uganda suggests the approach has been a “mixed success” (Wynne, 2005).
- Schiavo-Campo (2009) summarizes a decade of MTEF experiments in Africa as producing:
  - positive lessons: “greater awareness of the need to look beyond the annual budget horizon, and to focus on the results of government spending”;
  - negative outcomes described as “costly failures” arising from:
    - premature implementation of MTEFs;
    - little or no local ownership;
    - damaging distraction from basic PFM improvements;
    - heavy stress on limited budget capacity;
    - little improvement in macroeconomic balances, financial control and predictability, or spending efficiency.
- Brumby (2008) reaches a similar conclusion.
- Earlier flagship projects promoted by the Bank—such as Public Investment Programs (PIPs) in the 1980s and cash budgeting—“also largely failed to deliver their expected benefits” (Lienert and Sarraf, 2001).
- Performance-related (program) budgeting is strongly supported by IFIs and donors and is linked to:
  - the shift from project financing toward “budget support”;
  - emphasis on performance criteria and indicators to measure spending impact;
  - development of poverty-reducing strategy plans (PRSPs) and elaborate monitoring and performance evaluation systems.
- In many cases these initiatives have “degenerated to a ritualistic exercise” producing “vast databases of redundant and unused information, which are not linked to the budget itself.” (Peter Kohnert’s term: “Data cemeteries” / zahlenfriedhof.)

### Donor and IFI roles, incentives, and planning shortcomings
- Governments in many countries “may have accepted too uncritically advice received from the IFIs and donors” as a condition for assistance and debt relief.
- Donors and IFIs often lead drafting of action plans that underestimate time needed for planning and implementation, with little account of institutional barriers and constraints.
- Reasons for compressed time horizons include:
  - electoral cycle and short time horizon of finance ministers;
  - donors’ fiduciary concerns associated with budget support.
- Many reform plans are “overloaded with hundreds of activities and actions.”
  - Example: Cambodia’s reform plan “Strengthening Governance through Enhanced Public Financial Management” (December 2004) included:
    - four platforms covering an 11 year period;
    - under stage 1 (2004–06), 27 activities under 11 broad areas, and 254 separate actions.
  - A revised plan was issued in May 2008; few of the initial activities had been completed as planned two years after the intended completion date.
- Packaging many measures enables donors to assemble a “public sector reform” grant or loan that gains approval and increases donor leverage, but often results in unmanageable, ineffective packages that raise unrealistic expectations and fall into disarray.

### Diagnostic instruments and their limitations (PEFA example)
- The best known diagnostic tool is the PEFA diagnostic tool.
  - The PEFA instrument includes 28 performance indicators covering various aspects of PFM and three indicators describing donor practices in providing aid.
  - About 80 countries have undertaken some form of PEFA assessment in the past four years, some on several occasions.
- Limitations of PEFA in practice:
  - Not designed to take account of weak institutions and governance critical to improving budgetary systems.
  - Diagnostic information from PEFA assessments “does not provide a robust basis for preparing an action plan” for national authorities.
  - Usefulness diminished by:
    - partially subjective nature of the rating system;
    - lack of a rigorous quality-control mechanism;
    - variability of ratings from assessment to assessment, which can be subject to political influence.
  - Risk: donors may use PEFA assessments as a condition for granting budget support, creating incentives for governments to collude with donors to boost PEFA ratings.
- Supplementary institutional and governance reviews exist in principle but tend to be resisted by national authorities and have been carried out relatively seldom.

### Core constraint on PFM reform and the “basics first” debate
- The fundamental constraint on improvement is identified as “the absence of an enabling institutional environment,” not merely lack of technical understanding, skills, or IT systems.
- Budgets today are larger in relation to GDP and more structurally complex than 100 years ago, complicating reform despite technological advances.
- Schick’s argument for priorities in low-capacity countries: aim to “get the basics right” first, e.g.:
  - ensure effective control of inputs before controlling outputs;
  - provide good cash-based accounts before developing accrual-based accounts;
  - have effective financial audit before moving to performance audit.
- Andrews (2006) criticizes the “basics first” approach; the paper notes Andrews’ critique but finds it unconvincing for low-income countries and offers counterpoints highlighting risks of weak basics (loss of expenditure control, widening fiscal deficits).

### Sequencing and the “Platform Approach”
- Natural reform trajectories exist across components (e.g., accounting improvements moving from single-entry to double-entry to uniform accounting to asset/liability accounting to full accrual).
- A comprehensive framework for reform hierarchies has not yet been developed.
- Brooke (2003) proposes the “platform approach”:
  - package PFM reforms into groups of activities or measures (“platforms”) forming a logical sequence.
  - an overall reform strategy might stretch over a period of, say, ten years, comprising four or five platforms.
  - each platform would last for a period of two or three years and establish a basis for moving to the next stage.
  - claimed advantages:
    - more structured sequencing and clarity about roles and responsibilities for governments, IFIs, and donors;
    - focus on interconnection and mutual support among measures;
    - encourage realistic migration paths toward desirable technical improvements via small steps that create momentum.
- Illustration (as used in Government of Kenya, Strategy to Revitalize Public Financial Management, April 2006):
  - Platform 1 (Short Term Perspective): Improved quality of financial records and credibility in budget execution for central ministries; fast improvement of service delivery; competitive and open procurement; improved payroll management, reliability, and control; improved collection of revenue; improved effectiveness of internal and external audit.
  - Platform 2 (Medium Term): Improved quality of financial records and budget execution for remaining entities at central, regional, and local levels; improved budget preparation and allocations.
  - Platform 3 (Medium Term): Accountability and result-based management introduced; improved control of payroll, fixed assets, and pensions; improved accuracy of forecast and projections; reduced tax evasion and increases in revenue; reduced costs of debt financing.
  - Platform 4 (Long Term): Substantial improvements in service delivery, increases in allocations in accordance with political priorities; improved effectiveness and efficiency in the public service.

### Critiques and practical difficulties of the platform approach
- Cambodia’s experiment with the platform approach revealed serious practical flaws despite strong government leadership; lessons should be considered by other countries experimenting with platforms (Kenya, Uganda) or considering them (Indonesia, Kyrgyz Republic).
- A fundamental objection: difficulty of reaching agreement on an unambiguous definition of each stage or platform.
  - Example: Cambodia defined stage 1 as achieving a “credible” budget.
  - Disagreement over what “credible” means: statistical definition, economic definition, or political economy definition.
- If goals of a reform stage cannot be rigorously defined and measured, the concept of platforms becomes operationally meaningless.
  - Cambodia authorities declared stage 1 “completed” unilaterally though many advisors and experts would not endorse this decision.

### Policy implications and design considerations implied by the text
- Prioritize realistic sequencing that recognizes institutional dependencies among reform measures (e.g., reliable financial reports needed before effective audits).
- Avoid overloading action plans with hundreds of activities and unrealistic timeframes; ensure plans account for institutional constraints and realistic implementation capacity.
- Ensure local ownership and relevance to country development needs before importing complex practices from developed countries.
- Use diagnostic tools like PEFA with caution: complement PEFA with assessments of institutional and governance constraints and apply quality control to ratings.
- Consider simplified or “reduced form” models of performance budgeting adapted to low-income country capacity rather than full-scale imported systems.
- Define clear, measurable, and agreed criteria for completion of each reform stage to preserve operational meaning in staged approaches (platforms).

*Source: _wp0996 - introduction of a medium-term expenditure framework (MTEF) and integrated financial management information systems (IFMIS).*

### Box 1. Cambodia: Weaknesses of the “Platform Approach”

### Box 1. Cambodia: Weaknesses of the “Platform Approach”

### Key weaknesses identified in the Cambodian platform approach
- Insufficient emphasis on institutional constraints. The reform program underestimated the institutional and political resistance to improvement in areas such as the transfer of responsibility for managing government payroll to the finance ministry, the consolidation of government bank accounts, and measures to prevent the accumulation of new spending arrears.
- Overloading of activities. Platform 1 included 27 activities (of which 14 activities were considered necessary to prepare the ground for Platforms 2–4) and more than 250 specific actions. Such a wide range of activities would constitute a challenging task even in a country with a much higher capacity than Cambodia. Moreover, some of the activities selected for the first platform (e.g., piloting performance-based budgeting) seem questionable in relation to the country’s weak capacity and low level of development.
- An unrealistic time horizon. As argued in this paper, developing countries do not have the institutional or human capacity to undertake more than a few reform measures within a three- to four-year time period. The proposed platforms vastly overestimated this capacity.
- Poor prioritization. Certain decisions, especially to move ahead with the implementation of an IFMIS project, seem premature before basic improvements such as upgrading the accounting system and treasury functions have been implemented.
- Micromanagement. The complexity and technicalities of the exercise have come to dominate the implementation process. Completion of the many hundreds of deliverables and milestones, and important constraints and bottlenecks were disregarded. The crucially important issue of managing the change process was given insufficient attention. The reforms were managed at the strategic level but not sufficiently at ground level.
- Skewed incentives. Managers responsible for implementing the platform approach within the donor community and the government focus more on achieving specific performance indicators or deliverables than with improving the underlying PFM system. This issue of biased incentives―a focus on individual trees not the overall health of the forest―is familiar from literature on performance budgeting.

### Suggested refinements to the platform design
- Distinguish functional components of the improvement process into subplatforms addressing:
  - budget laws and regulations,
  - changes in business processes,
  - the introduction of new technology and IT systems,
  - upgrading organizational structures,
  - training and capacity building measures to strengthen human resources.
- Rationale: Recognizes that reforming institutions is a necessary but insufficient condition for improvement; other important considerations are people, skills, organization, and information (Schiavo-Campo and McFerson, 2008).
- Purpose of subplatforms: develop work plans to ensure improvements in each component (for example, modernizing business processes and IT systems) take place broadly in step with one another to achieve overall reform objectives.
- Caveat: Such refinements are hardly to be recommended in practice, given the formidable existing challenges to the platform approach noted above.

### Streamlined alternative and its limitations
- A streamlined version used in South East Europe divides 10 important areas into:
  - basic reforms, necessary to achieve a minimum level of financial management capacity, and
  - advanced reforms, which will enable countries to move toward international good practice.
- The objectives and benchmarks are consistent with broader benchmarking frameworks for PFM systems, such as the IMF’s fiscal transparency code, and the PEFA framework.
- Benefits: Allows progress to be monitored and correlated with measures of fiscal performance such as aggregate fiscal control, financial discipline, efficient resource allocation, and the cost-effectiveness of public spending.
- Potential difficulties:
  - The logical basis for the distinction between “basic” and “advanced” reforms is largely subjective and to some degree arbitrary (example: many practitioners would regard establishing a unified accounting framework as a precondition for developing a robust system of budget execution and reporting).
  - Unclear criteria for determining when a country has achieved sufficient progress in basic systems to move to advanced measures.
  - Absence of analysis of underlying institutional conditions that may affect the reform process.

### Implications and principles for developing-country reform strategies
- Recognize broader goals: The ultimate goal of budget reform is broader than improving systems and processes; strategies should be tied to fiscal objectives such as:
  - Ensuring macroeconomic stability and fiscal sustainability;
  - Protecting money from theft and misappropriation;
  - Ensuring that public funds are spent consistently with the approved budget;
  - Strengthening the link between public priorities (for example, as reflected in the MDGs or the poverty-reduction strategy) and the budget, and improved multiannual programming;
  - Monitoring results of spending in terms of access to, and quality of public services, and injecting external accountability for effectiveness (“social accountability”).
- Get the basics right (Schick, 1998): Low-capacity countries should focus on basic needs first. Core foundational areas likely include:
  - establishing a comprehensive and credible budget that eliminates (or reduces) extrabudgetary activities and accounts;
  - building a sound system of accounting and control that avoids payment arrears and enables fiscal reports to be produced on a timely and regular basis;
  - strengthening links between budgeting and a country’s poverty-reduction strategy.
- Sequencing vs. selection:
  - Broadly sequenced, multi-platform programs with hundreds of measures are unlikely to be successful; selection rather than sequencing is the recommended organizing principle.
  - Sequencing remains relevant for specific components (for example, reorganizing the ministry of finance, establishing an integrated revenue authority, or a debt management office).
  - Practical politics may require initiating actions out of strict hierarchical order (for example, establishing an external audit authority before basic accounting systems are upgraded).
- Avoid simplistic prioritization based only on poorest diagnostic scores. A nuanced approach must consider regulatory framework, business processes, human resource issues, IT systems, availability of external finance and technical assistance, interdependencies with other public sector reforms, and institutional and political drivers.
- Strengthen coordination and impartial advice:
  - Strong coordination between countries and providers of technical assistance is essential but often ineffective.
  - There is an urgent need for an honest broker to provide national authorities with realistic, impartial guidance on expectations, scope, and sequencing.
- Practical design guidance:
  - Avoid overly complex, rigid, technocratic programs.
  - Link budget modernization to broader public administration reforms and the establishment of a professional, merit-based civil service.
  - Focus on changes that impel or facilitate further changes (Hirschman, 1958).
  - Provide appropriate inducements and incentives (financial and nonfinancial) to civil servants to support reform.
  - Raise and institutionalize professional standards in economics, accounting, and financial management; introduce codes of conduct and whistle-blowing legislation where appropriate.
  - Define required outputs and outcomes and tests of success or failure; ask the right questions of the right people, including middle- and low-ranked government officials in the budget “trenches”; hold officials and advisors accountable for failures to implement required measures.

*Source: Box 1. Cambodia: Weaknesses of the “Platform Approach” (excerpt).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp0996.pdf_