united nations development programme comprehensive...

Zimbabwe

Working Paper 2

International Aidand its Management:Some Insights for Zimbabwein the Context of Re-engagement

MARK SIMPSON AND DALE DORÉ

United Nations Development Programme

Comprehensive Economic Recovery in ZimbabweWorking Paper Series

2009

Working Paper 2

International Aid and its Management:Some Insights for Zimbabwe in the Context

of Re-engagement

MARK SIMPSON AND DALE DORÉ

Copyright © UNDP [2009]

All rights reserved

Layout and Design: Fontline International

The views expressed in this publication are those of the authorsand do not necessarily represent those of the United Nations or UNDP

iii

Team Members of UNDPs Working Paper Series iv

Foreword v

Executive Summary vii

Acronyms viii

Section 1: Introduction 11.1 Defining Official Development Assistance 11.2 Trends in Zimbabwe 1

Section 2: From Project Aid to Budget Support 32.1 Project Aid and its Limitations 32.2 Common Funds 42.3 The Rise of SWAPs 52.4 Budget Support and its Advantages 5

Section 3: The Role of Poverty Reduction Strategy Papers (PRSPs) in Aid Delivery 73.1 The Development of a PRSP 73.2 The Role of the Bretton Woods Institutions (BWIs) 83.3 The Impact of PRSPs 8

Section 4: International Debt Relief Efforts and Debt Management 104.1 The Paris Club and Naples Terms 104.2 The Highly Indebted Poor Country (HIPC) Initiative 114.3 The Multilateral Debt Relief Initiative (MDRI) 124.4 Impact, Slippage and Moral Hazard Considerations 12

Section 5: Using the Paris Declaration to Maximize Aid Effectiveness 155.1 Key Features of the Principles 155.2 How Have the Principles Fared? 17

Section 6: Aid and Fragile States 206.1 Applying the Fragile States Principles 216.2 A Toolkit for Fragile States Transitions 23

Section 7: The Macroeconomic Management of Aid Flows 26

Section 8: The Aid-Growth Debate 30

Section 9: Enhancing Aid Predictability 349.1 The Nature of Aid Uncertainties 349.2 The Magnitude of the Problem and its Consequences 349.3 Measures to Reduce Aid Uncertainties 36

Section 10: Aid Dependency as an Impediment to the Rise of Developmental States 39

Section 11: National Aid Policies and Management Systems 4111.1 Efforts to Address the Power Imbalance 4111.2 Some Results in Terms of National Aid Policy Frameworks 4311.3 Aid and Private Capital Flows 44

Section 12: Policy Implications 47

Bibliography 49

Table of Contents

iv

Dr Dale Doré is the Director of Shanduko in Harare, a non-profit research institute on agrarian andenvironmental issues in Zimbabwe. After an undergraduate degree in Business Adminstration, hecompleted his M.Sc in Regional and Urban Planning at the University of Zimbabwe, and his D.Phil inAgricultural Economics at the University of Oxford. Prior to joining the UNDP recovery strategyresearch team, he was the coordinator of a WWF and IUCN community-based natural resourcemanagement programme in five Southern Africa countries, and later a land tenure specialist for afour country study in S.E.Asia on behalf of the Center for International Forestry Research and theFAO.

Professor Tony Hawkins is currently Professor of Economics at the Graduate School of Managementat the University of Zimbabwe, an institution which he founded. Professor Hawkins received hisB.A. in Economics from the University of Zimbabwe, was a Rhodes Scholar at Oxford where heread for a B.Litt in Economics, and a Fulbright Fellow at Harvard. He has written extensively on theZimbabwean economy with a special focus on the private sector, and consulted for a range ofinternational organizations including the World Bank, UNIDO, UNDP, UNCTAD and SADC.

Dr Godfrey Kanyenze is the Director of the Labour and Economic Development Research Institute ofZimbabwe (LEDRIZ). Prior to this he was an economist with the Zimbabwe Congress of TradeUnions (ZCTU) and statistician with the Central Statistical Office (CSO) in Harare. He currentlyalso serves as a member of the Technical Committee of the Tripartite Negotiating Forum (TNF) andthe Tripartite Wages and Salaries Advisory Board. He received his B.Sc in Economics at the Universityof Zimbabwe, his Masters at the University of Kent, and completed his Ph.D in Labour Economics atthe Institute of Development Studies, University of Sussex (UK).

Professor Daniel Makina is currently Professor of Finance and Banking at the University of SouthAfrica in Pretoria. Following his B.Acc at the University of Zimbabwe, Professor Makina completedhis M.Sc in Financial Economics at the School of Oriental and African Studies, University of London,and his Ph.D. at the University of the Witwatersrand in Johannesburg. He has also taught at theGraduate School of Management at the University of Zimbabwe, and conducted training programmesfor public servants from Sudan and South Africa in public financial management.

Dr Daniel Ndlela is a Director of Zimconsult, a consultancy firm in Harare. He completed hisundergraduate degree at the Higher Institute of Economics in Sofia, Bulgaria, and his Ph.D at theUniversity of Lund, Sweden. After serving as Principal Economist at the Ministry of Planning andDevelopment in the early 1980s, he lectured at the Department of Economics at the University ofZimbabwe. He briefly served as Deputy Dean of the Faculty of Social Studies at the University ofZimbabwe before becoming a Senior Regional Advisor at the United Nations Economic Commissionfor Africa (UNECA). He has consulted extensively for, amongst others, the World Bank, COMESA,SADC, UNIDO and UNCTAD.

UNDP Chief Technical Adviser

Dr Mark Simpson completed his B.Sc (Econ) and Ph.D at the London School of Economics, and hisMasters at the School of Oriental and African Studies, University of London. Prior to his posting toUNDP Zimbabwe, Dr. Simpson served with the UN’s Department of Peace-Keeping Operations inmissions in Angola and East Timor, and with UNDP in Mozambique.

Team Members of UNDPs Comprehensive EconomicRecovery in Zimbabwe Working Paper Series

v

This second paper in UNDP’s working paper series draws upon the broad outlines of the internationalaid system, its instruments and processes, contained in the UNDPs Comprehensive Economic Recoveryin Zimbabwe – A Discussion Document launched in 2008. The paper is an attempt to now providereaders with second and third order levels of detail on what can often seem an overwhelmingly complexsystem, while at the same time focusing on what were seen by the authors as the key features of thatsystem, the essential debates that have in the past shaped its development, and that continue to do sopresently.

In a context in which Zimbabwe is currently re-engaging with the international development community,the authors have sought to provide a primarily Zimbabwean readership with insights drawn from theglobal experience that might prove useful in helping the country manage what may possibly be significantinflows of aid, the return of a number of development agencies, and a scaling up in the operations ofthose who have maintained a presence throughout the recent past – with all the demands that this willinevitably place on national systems. The authors have also purposely tried to steer away from detailedcountry specific prescriptions, since the overall objective of this particular working paper is primarily apedagogic one, aimed at helping the target readership take on the role of informed and empoweredinterlocutors as the country re-engages with the donor community.

Dr Agostinho ZacariasUNDP Resident Representative

Mr Lare SisayUNDP Deputy Resident Representative

Dr Mark SimpsonChief Technical Adviser, Recovery StudyUNDP

Foreword

vii

After years of isolation from the mainstream of international development processes, Zimbabwe’s decision-makers need to familiarize themselves with the new aid architecture and mechanisms as the country re-engages with the international community. The purpose of this paper is therefore to discuss the mainprinciples and processes of various aid modalities, international debt relief efforts and the management ofaid flows so as to enable decision-makers to ensure that engagement in these initiatives best meetZimbabwe’s development objectives. In doing so it draws on the international debate on the role andeffectiveness of aid and the experiences of other countries.

Broadly, the paper is structured around three main themes. The first is the gradual shift of aid deliveryaway from project-based approaches which are specific ‘stand alone’ interventions by donors, towardsprogramme-based approaches where donors provide direct support to various sectors and the nationalbudgets of recipients (partner countries). This evolution in approach finds expression in the second theme,the Paris Declaration of 2005, which is a set of principles to improve the efficiency and impact of aid.The third theme is the need for a comprehensive and technically sound national development strategy,known as a Poverty Reduction Strategy Paper (PRSP).

The Paris Declaration groups aid delivery principles into five broad categories. ‘Ownership’ is the firstcategory whereby the partner country undertakes to exercise effective leadership over the developmentpolicies and strategies and coordinate development actions. The second and third categories, ‘alignment’and ‘harmonization’, require donors to base their overall support on the partner countries’ nationaldevelopment strategies, institutions and procedures, and ensure that their actions are more harmonized,transparent and collectively effective. ‘Managing for results’, the fourth category, means managing andimplementing aid in a way that improves decision-making to achieve development objectives. The lastcategory, ‘mutual accountability’, requires both donor and partner countries to be responsible fordevelopment outcomes.

The formulation of a national development strategy, or PRSP, through a participatory process involvingboth government and non-state actors – the private sector, labour and civil society – places ownership ofthe development process firmly with the partner country. The PRSP then becomes the centrepiecearound which donors can align and harmonize their own development assistance and, with improvementsin public financial management and national oversight by Parliament, move towards providing directbudgetary support. Under such a scenario, donors and partner countries are likely to benefit significantlyfrom greater transparency and oversight, a reduction in transaction costs and fiduciary risks, as well asgreater predictability of aid flows.

The paper concludes with measures that Zimbabwe needs to take when requesting debt relief andmanaging aid flows as it re-engages with the international donor community.

Executive Summary

viii

Acronyms

BWIs Bretton Woods InstitutionsCF Common FundDAC Development Assistance CommitteeDSA Debt Sustainability AnalysisGBS General Budget SupportGDP Gross Domestic ProductHIPC Highly Indebted Poor CountryIDA International Development AssociationIFI International Financial InstitutionsIMF International Monetary FundJAM Joint Assessment MissionLICUS Low Income Country Under StressMDB Multilateral Development BankMDTF Multi-Donor Trust FundMTEF Medium-Term Expenditure FrameworkNGO Non-Governmental OrganisationNPV Net Present ValueODA Official Development AssistanceOECD Organization for Economic Co-operation and DevelopmentPFM Public Financial ManagementPRGF Poverty Reduction and Growth FacilityPRSP Poverty Reduction Strategy PaperSBS Sector Budget SupportSIP Sector Investment PlanSWAP Sector Wide ApproachTRM Transitional Results MatricesUN United Nations

1

1.1 DEFINING OFFICIALDEVELOPMENT ASSISTANCE

Official Development Assistance (ODA), asdefined by the Organization for Economic Co-operation and Development – DevelopmentAssistance Committee (OECD-DAC),1

constitutes:

‘Grants or loans to countries and territorieson the DAC List of ODA Recipients andmultilateral agencies that are undertaken bythe official sector at concessional terms (i.e.,with a grant element of at least 25 percent)and that have the promotion of the economicdevelopment and welfare of developingcountries as their main objective. In additionto financial flows, technical co-operation isincluded in aid. Grants, loans or credits formilitary purposes are excluded.’ (OECD, 2007:232)

This definition of ODA thus excludes othercategories of assistance from OECD-DACcountries, such as Other Official Flows, i.e., thosethat do not meet development objectives or becausethey have a grant element of less than 25 percent.

The volumes of ODA flows from DAC membershave grown from US$20 billion in 1980 to a figureof US$104.4 billion in 2006. This represented anestimated 90 percent of total global ODA flows.The Secretariat of DAC has been pushing for sometime for a measure that excludes debt relief,humanitarian assistance, administration costs,student scholarships and the in-donor country costsassociated with refugees. The objective is to arriveat a figure that reflects net programmable aid.Using this measure, the figure for programmableODA from DAC members for 2006 stood at aroundUS$46 billion.

It is noteworthy that as a proportion of totaldeveloping countries inflows, ODA has fallenrelative to remittances, commercial loans andforeign direct investment. In regards to foreigndirect investment, for example, private flows fromDAC countries in 2006 (covering export credits,direct investment and portfolio investments) stoodat just under US$195 billion (OECD, 2007: 138).To this must be added the unknown flows from theso-called ‘new donors’ such as China, Brazil andIndia. Firm figures for these countries are difficultto ascertain with any degree of certainty given thatthey are not members of the OECD and aretherefore under no obligation to reports such flowsto DAC.

International aid architecture is becoming evermore complex, given the multiplication of both thenumber of actors as well as the financingmechanisms. In recent years there has been agrowing recognition that the system is bothexceedingly convoluted as well as imposing hightransactions costs on all parties.

Broadly speaking, attempts to reform theinternational aid system have been rooted in twofundamental concerns, namely the efficiency of aiddelivery systems and the impact of aid. In otherwords, on the one hand what recipients (i.e., partnercountries) and donors can do to improve the mannerin which aid is delivered, received and accountedfor, and on the other how to ensure that it achievesits intended objectives.

1.2 TRENDS IN ZIMBABWE

Multilateral and bilateral ODA to Zimbabwedecreased sharply as the economic and political crisisdeepened after 2000. The World Bank Group

Section 1

Introduction

1 The Organization for Economic Co-operation and Development – Development Assistance Committee (OECD-DAC) bringstogether 22 OECD member countries plus the European Commission. Its objective is to work towards an expansion of resourceflows to developing countries and to improve the effectiveness of the same.

2

International Aid and its Management

imposed restrictive measures in May 2000 due tothe accumulation of payment arrears and loans, andall undisbursed loans and grants were cancelled.Zimbabwe subsequently made substantial paymentsto the International Monetary Fund (IMF) to offsetits arrears in the General Resources Account (GRA),but only made token payments to the World Bankand the African Development Bank in contraventionof the World Bank Group’s non-discriminatory debt-servicing clause. The World Bank put Zimbabweon non-accrual status in October 2000, and the IMFclosed its Resident Representative Office in Hararein October 2004. Similarly, the European Union’sCountry Strategy programming process, which hadbeen agreed in July 2001, was suspended in Februarythe following year.

Due to these setbacks, the World Bank and theIMF’s engagement with Zimbabwe has been limitedmainly to policy advice and dialogue in an effort toimprove relations and explore ways in whichZimbabwe could design a mutually agreeablearrears clearance plan to enable re-engagement.The IMF Executive Board meeting in February2007 felt unable to restore Zimbabwe’s voting rights

or its eligibility to use the general resources of theFund. It also noted that Zimbabwe’s paymentstowards settlement of its Poverty Reduction andGrowth Facility (PRGF) arrears had beenminimal.2

OECD bilateral donors withdrew their developmentaid, but continued their humanitarian assistance.The poorest and most vulnerable elements ofsociety have been provided with vaccines,prophylactics, anti-retroviral drugs to fight HIV andAIDS and, crucially, emergency food assistance.Limited funds have also been channelled towardslocal community-based recovery initiatives, as wellas support in the areas of governance, democracyand human rights. By 2006, this humanitarianand governance ODA as a percentage of GDPis estimated to have stood at six percent.3 Itis noteworthy that in contrast to prevailinginternational trends, and as a result of theestrangement between the Zimbabwe and Westerngovernments, ODA has been channelled outsidegovernment systems directly to beneficiariesthrough United Nations (UN) agencies and Non-Governmental Organisations (NGOs).

2 The IMF’s PRGF was established in 1999 and replaced the Fund’s Economic and Structural Adjustment Facility (ESAF) and wasaimed at making the IMF’s lending operations more focused on poverty reduction and growth. PRGF-supported programmes areanchored on Poverty Reduction Strategy Papers (PRSPs) and assistance is provided through concessional loans to countries withprotracted balance of payments problems. Areas covered by the PRGF include macroeconomic frameworks and financial policies,exchange rate and tax policy, fiscal management, budget execution, etc.

3 It should be noted that given the rapid deterioration in the country’s humanitarian situation between 2006–2008 and significantincreases in international humanitarian assistance, coupled with a shrinking GDP, ODA currently stands at probably around 15-20percent of GDP.

3

The last decade has seen significant shifts in termsof aid delivery mechanisms to improve aideffectiveness. Most significant amongst these hasbeen a gradual policy shift away from project-based approaches, which are characterized bydiscrete, externally financed interventions, toprogramme-based approaches, especially sectorwide approaches and direct budget support. Ingeneral terms, programme-based approaches maybe defined as development cooperation based orcoordinated support for a single, nationally ownedstrategy, be this a national development or povertyreduction strategy, one covering a specific sector(e.g., health or education) or a thematic area suchas Governance. More specifically in terms ofimplementation modalities, these approaches arealso characterized by the fact that they possess asingle budget and monitoring and evaluationframework, as well as formalized partner country-donor coordination mechanisms.

2.1 PROJECT AID AND ITSLIMITATIONS

The shortcomings inherent in project-based aiddelivery include the fact that there is often littlecoordination between donors operating in a givencountry, as well as coordination between the donorsand the partner country government in terms ofprioritization. Both disconnects often lead toextensive duplication, poor targeting and waste.Project-based delivery is also often associated witha brain drain from the public to the donor sector, asthe donor agencies draw away already scarcenational technical capacity in support of their owninitiatives by offering salaries which are often manymultiples of local average wages.

Given the ad hoc manner in which projects are oftenimplemented – and the disconnect between sectorline ministries and their Ministry of Finance,particularly in the areas of planning and budgeting– there is often a growth in the contingent liabilitiesthat governments subsequently found increasinglydifficult to sustain. Once projects have been handed

over to the partner country, recurrent expenditures,such as maintenance costs for infrastructureprojects and personnel costs for social servicedelivery projects, are often extremely onerous forgovernments to sustain. There is also a growingrecognition amongst donor countries that project-based aid also carries with it the risk of underminingboth the ability and incentives for partner countriesto perform a number of key government functions.These range from expanding the tax base andrevenue collection, planning and prioritizing throughthe budgetary allocation process, and to imple-menting, monitoring and evaluating programmes.

Given the pervasiveness of the project-based aidmodality, it is clear that political pressures andinstitutional incentives act on both donors andpartner country governments which play out in itsfavour. Amongst these are the pull factors on theside of partner countries. The parallel funding andmanagement mechanisms generate both materialand non-material incentives for ministers and civilservants of partner countries, such as vehicles,salary top-ups, training, and travel opportunities. Inaddition, the high visibility impact of stand aloneprojects redounds to the benefit of ministers, civilservants and local interest groups, who are able toclaim immediate credit for these, notwithstandingdoubts over their longer-term sustainability. Thisstands in stark contrast to the deeper but moreprotracted benefits that flow from sector-wide orcountry-wide reform processes such as PublicFinancial Management or capacity-building in theareas of planning, monitoring and evaluation.

Other specific advantages for partner countryministries include the fact that aid flows from astand-alone project are relatively predictable andsimple to track. In addition, partner country ministryofficials can avoid having to justify funding for theseprojects within their central government planningand budgetary processes. One not insignificantadvantage is that they avoid parliamentaryoversight when the flow of donor funds is notreported to their Ministries of Finance, therebydiluting accountability for ODA resources.

Section 2

From Project Aid to Budget Support

4


Pull factors in favour of project-based aid alsooperate on donors. As with partner countryministries, the high visibility of stand-alone projectsassists bilateral and multilateral developmentagencies in terms of their own internal budgetaryallocation processes. In addition, the attribution andevaluation of the impact of stand-alone projects iseasier and faster than in the case of longer-term,multi-donor, programmatic aid flows. There alsoneeds to be a greater acceptance of the ‘fiduciaryrisks’4 inherent in the programmatic initiatives thatentail greater reliance on the financial systems ofpartner countries, whereas oversight over ‘ring-fenced’ project-based resources is seen as easier.By using project-based aid, donor agencies andtheir staff are also able to avoid engagement incomplex dialogue processes around the need forreforms in the areas of national policy frameworksand administrative systems in partner countries.Finally, an argument often advanced by both donorand partner countries is that project-based aid –precisely because it bypasses national systems –allows direct assistance to the poor.

The interplay of these forces sometimes leads towhat might be characterized as a vicious cycle ofaid projectization. Donors are often confronted witha difficult operational environment where partner’spolicies are either technically weak or absent.Coupled with weak national capacities, institutionsand accountability systems – and in many placespervasive corruption and patronage – the donors’default position is to prefer project overprogrammatic aid because it offers them greatercontrol over resources and targeting. This in turnresults in multiple projects that often bypass theplanning and financial systems of the centralgovernments of partner countries, and which areoften not aligned to national policies and priorities.Project coordination or implementation units outsidegovernment structures are then established toprovide the necessary support to these projects.

The pull factors operating on both sides of the donorand partner countries aid relationship then lead to

situations in which neither donors nor partnercountries make an effort to improve nationalpolicies or systems. Sector ministries becameaccountable to multiple donors and not centralgovernment or parliament, and their ability andinterest in performing core functions such asplanning, budgeting and revenue collection isundermined.

2.2 COMMON FUNDS

Moves to address the constraints with stand-aloneprojects and introduce programme-basedapproaches began in the early 1990’s with the riseof basket or pooled funding, also known as donorCommon Funds (CFs). These may be used tosupport activities in a whole sector, e.g., health, athematic area such as rural development, or a sub-theme such as condom or vaccine procurement anddistribution. The key motivation of CFs is to reducetransaction costs on both donors and partnercountries, and to strengthen alignment by ensuringthat CFs are supporting a national priority for asector or sub-sector. One key characteristic is thatdonor CF resources are usually kept separate fromother resources intended for the same purpose, andare channelled in any one of three ways:

• Using parallel systems, with donors takingcontrol of design, appraisal and quantity ofinputs and using their own disbursement andaccounting procedures. Under this modality,CFs stand alone outside sector or nationalbudgets;

• Through NGOs or other non-state serviceproviders; or

• Through government systems and in supportof government policies and priorities, but withspecific earmarking of expenditure for adiscrete set of activities. In such situations,the inputs, outputs and objectives of such aidare defined and evaluated separately.

4 The idea of fiduciary risk means that expenditure may not be properly accounted for, or does not represent value for money, andthat it may not be used for the original intended purpose.

5

Section 2 – From Project Aid to Budget Support

2.3 THE RISE OF SWAPS

If one could characterize the development of aidinstruments in evolutionary terms, the next step upthe ladder one finds Sector-Wide Approaches, morecommonly known by their acronym SWAPs. Thesehave the following defining characteristics:

• All significant funding for a sector supports asingle sector or thematic policy, expenditure,implementation, and monitoring and evaluationframework;

• These frameworks are the result of an iterativeprocess of negotiations between nationalinstitutions (usually line ministries) and donors;

• Sectoral/thematic activities are implementedunder government leadership;

• All donors gradually move towards commonapproaches in terms of participation in theprocess of sector diagnostics, the establishmentof priorities, and the means of achieving them;and

• All donors also move gradually towards relianceon Government procedures for disbursementand the accounting of funds.

Two additional planning tools come into play underSWAPs. Firstly, governments need to develop 3-year Medium-Term Expenditure Frameworks(MTEFs) through donor-ministry negotiations. Theyare meant to show projected partner governmentcontributions to the sector/thematic strategy, andthe shortfall to be covered by donor funds. A SectorInvestment Plan (SIP) is an integral part of theMTEF, and shows the capital investmentrequirements.

2.4 BUDGET SUPPORT AND ITSADVANTAGES

Taking the evolutionary analogy one step further,one arrives at various budget support modalities.Budget support can take two main forms:

• General Budget Support (GBS) which isunearmarked donor aid disbursed through apartner country’s public financial management

(PFM) system via the partner country’sTreasury. Besides financial resources, aprogramme of GBS may also include provisionfor donor-government dialogue processes,technical assistance and the alignment of donorinterventions.

• Based on SWAPs, Sector Budget Support(SBS) involves earmarked donor fundsdelivered through a beneficiary government’sPFM system, but strictly linked to a sector ormulti-sector policy framework. SBS is providedwith sector-conditionality, agreement on andexecution of an agreed policy and expenditureframework through a SWAP process.

GBS and SBS are jointly referred to as DirectBudget Support (DBS), and are normally providedin support of Poverty Reduction Strategies (seebelow) and/or sector strategies. One keydistinguishing feature of both is that donor fundsare mixed with domestic resources.

Advocates of budget support name a number ofadvantages for partner countries. By directlycontributing to the national budget, accountabilityshifts from the government-donor partnership to therelationship between partner governments, theirparliaments and wider citizenry. The direct injectionof budget support resources into national financialsystems allows partner governments to treat theseresources as they would domestic revenues withwhich to undertake both capital and recurrentexpenditures. Direct budget support is thus alsointended to make it easier for partner countries tomanage their own planning and budgetary processes.

A fundamental consideration that has receivedincreased attention since the G-8 promised largeODA increases at Gleneagles in 2005 is the capacityof donor countries to plan, implement and monitora multitude of projects. Partner countries are indanger of being swamped by projects which maynot respond to national priorities, undermine nationalinstitutions, systems and accountability, and havelimited impact and sustainability. This point wasmade forcefully by the European Commissioner forDevelopment and Humanitarian Aid in 2008:

‘At the moment there are 600 projectsinvolving less than ∈1 million in the healthsector in Mozambique alone. This means that

6


the time and energy of the Minister of Healthand his department are taken up in discussionswith various donors involved in the sector. Ifin future, as the volume of aid increases, weall continue to fund our own micro projects,each with our own requirements, staking ourown little claims, we can say goodbye to anyaid efficiency agenda. Budget support andmore of it is the only answer.’5

Changes in the volumes of aid being delivered asbudget support (as a percentage of total aid flows)have been significant in a number of countries, andhave gone hand in hand with the rise of PovertyReduction Strategy Papers (PRSPs).6 In the caseof Uganda, for example, between the launching ofits 2nd PRSP in 2000, and its 3rd PRSP in 2004,general budget support rose from U$176 million toU$409 million, which represented 45 percent of totalaid flows.7 The evidence to date indicates that thebudget support modality does meet its statedobjectives of reducing the demands on the scarcenational capacity of many partner countries andlowers the transaction costs on both sides of thedonor-partner country relationship. While it remains

difficult to establish the precise impact of PRSPson the one hand, and of budget support on the other,in terms of improvements in the systems andprocedures of partner countries, it is neverthelessclear that when both have been present there havebeen significant improvements in terms of nationalplanning, budgeting and accounting systems.

However, while the global shift towards GBS hasbeen significant, it still remains relatively low as apercentage of total OECD-DAC ODA to low-income countries. In 2007, for example, theStrategic Partnership for Africa carried out a surveyof a sample of committed GBS donors in Africa.One of its findings was that GBS still representedonly between 20-25 percent of total ODA flows toAfrica from that particular group of donors.8 Clearlyfiduciary risk considerations, as well as otherfactors, still loom large in the calculations of donorsin terms of transferring funds through the systemsof partner countries, while the pull factors on theside of partner countries that operate in favour ofprojects, as opposed to programme-based aidmentioned above, also still exert strong pressures.

5 Louis Michel, Foreword, ‘Budget Support – A Question of Mutual Trust’, Luxembourg: Office for Official Publications of theEuropean Communities, 2008, p.4.

6 These are similar to national development plans, except that they are drawn up with the support of the Bretton WoodsInstitutions (BWIs) and bilateral donors.

7 International Development Department, School of Public Policy, University of Birmingham, UK, ‘Evaluation of General BudgetSupport, Synthesis Report – Executive Summary’, May 2006, p.S2.

8 Strategic Partnership with Africa – Budget Support Working Group, ‘Survey of Budget Support 2006’, Washington DC, 2007.

7

Though it is difficult to attribute precise causalityin the evolution of aid delivery modalities, it seemsclear that the rise of PRSPs in the late 1990’s playeda key role in driving forward the principles containedin the Paris Declaration, which is the currentinternational normative framework guiding donor-partner country relations (see below).9

At the 1999 annual meetings of the Boards of theWorld Bank and IMF, it was agreed that all futureconcessionary lending by the Bretton WoodsInstitutions (BWIs), as well as debt relief, wouldbe conditional on the development of nationally-owned PRSPs. The comprehensiveness of PRSPs(multi-year, multi-sectoral, linked to medium-termexpenditure plans and monitoring and evaluationframeworks and annual progress reports), andbecause they are meant to be country-owned andresults-oriented, means that they have become formany donors the national development strategiesreferred to and called for in the Paris Declaration.Given their centrality in current international aidarchitecture around which the Paris Principles canbe implemented, it is worth exploring the processof developing PRSPs, and their content and impactin more detail.

3.1 THE DEVELOPMENT OF A PRSP

At the design stage, some basic groundwork needsto be carried out in order to develop an under-standing of the extent and nature of poverty in agiven country and its key determinants. The qualityof this ‘poverty profile’ is heavily dependant on thequality of data. According to the guidelines (World

Bank, 2002), this profile should go beyond incomemetric and asset-holding measurements of povertyand consider aspects such as health and educationstatus (the so-called capabilities approach topoverty) and disempowerment (access toinstitutions and capacity to influence public policyprocesses).

An effort should be made to disaggregate povertyby gender and region, as well as carry out labourmarket diagnostics,10 the impact of macro-economic variables on the national poverty profile,as well as the growth and distributional impact ofpast policies. Additional areas requiring analysisinclude the equity, efficiency and effectiveness ofpast and current patterns of public expenditure, thesoundness of a country’s public financial manage-ment system and service delivery systems. Bringingtogether these findings should then enable theestablishment of linkages between all these variablesand their impact on poverty, and thereby theconstraints which prevent movement out of poverty.

The next step in the design process is to set targetsfor the PRSP, and embed these in both existingand future macroeconomic and sectoral policies.Key public actions need to be designed andprioritized. This design stage should be informedby the targets that are set, what is known aboutinter-sectoral linkages and their impact on poverty,what is known from the poverty diagnostics in orderto ensure proper targeting and sequencing, theestimated costs of the interventions, the availabledomestic resources and a realistic estimate of theexternal resource envelopes required to cover thefunding gap. Institutional capacities should also be

Section 3

The Role of Poverty Reduction Strategy Papers(PRSPs) in Aid Delivery

9 A UNDP report in 2000 also played an essential role in moving the problem of poverty to the front burner. Amongst otherinsights, the report noted that less than one-third of developing countries had targets for eradicating poverty, that many anti-poverty measures were vague statements of intent, and that many poverty reduction programmes were too narrow and could moreproperly be seen as a set of social safety net style interventions. See ‘Overcoming Human Poverty’, UNDP Poverty Report,2000, New York, USA.

10 Given the centrality of employment in poverty reduction strategies, what is required is to arrive at labour force participation rates,unemployment rates, earnings and productivity, formal versus informal sector shares in employment, etc.

8


assessed. Indicators are then developed to measureprogress, which can be for the whole lifespan ofthe PRSP, as well as for monitoring and evaluatingon an annual basis.11

In addition to specific sectoral policies, a PRSPmust contain a macroeconomic framework. Thisis meant to be supportive of the growth and povertyreduction targets, and normally includes inflationtargets and an external position, projected growthrates and a fiscal stance. Recent PRSPs have alsolaid out possible trade-offs between short-term andlong-term poverty reduction targets, as well as thepossible impact of variations in growth rates andtheir likely impact on the targets. Many also containstructural reform measures to be implemented inthe context of the PRSP, such as financial sectorreform, privatization, trade and regulatory reformsas well as improvements in governance and publicfinancial management.

An overarching principle of participatoryformulation is meant to inform both the designand implementation processes. In addition toparliaments, partner country governments wereenjoined to include non-state national actors indiscussions about poverty, diagnostic work, thedesign and implementation phases as well as themonitoring and evaluation of progress. Suchparticipation is seen as a means of building national(not just governmental) commitment to povertyreduction, as well as mutual accountability betweennational partners, and to move away from theprevailing primary focus on government-donoraccountability.

3.2 THE ROLE OF THE BRETTONWOODS INSTITUTIONS (BWIs)

Both the World Bank and IMF staff assist countriesduring the design of PRSPs. Their technical

assistance is most prominent during the phase ofpoverty assessments and diagnostics. They alsoprovide advice on reforms to public financialmanagement systems and on public expendituresurveys and budget management. Guidance on themacroeconomic framework is largely assigned tothe IMF. Once a full PRSP has been developed,the staff of the World Bank and IMF conduct aJoint Staff Assessment (JSA). This is intended toplay two key roles. On the one hand they providethe Boards with a technical assessment of a givenPRSP as a basis for debt relief initiatives and futureconcessionary assistance from the BWIs, thusallowing the Boards to exercise ‘due diligence’.They also provide an opportunity for the BWIs toprovide feedback to the country in question as tohow the PRSP might be improved.

3.3 THE IMPACT OF PRSPS

As the process of designing a PRSP by the partnercountry needs to be comprehensive, rigorous andconsultative, the average time required betweenan interim PRSP12 and a full PRSP is aroundtwenty months.13 The ubiquity of this mechanism,and the fact that they lie at the heart of currentinternational development discourse and assistance,can be seen by the fact that as of March 2008,over 70 full PRSPs had been submitted to theBoards of the BWIs for Washington ‘sign-off’.

As PRSPs have been rolled out in partner countries,both donors and the BWIs have been able to rewritetheir country assistance strategies in support ofthese single, national, overarching strategicdocuments and the priorities contained therein. Theyhave enabled donors to make significant progressin harmonizing their joint aid efforts and reducingtheir transaction costs by gradually moving awayfrom the project-based modalities to budget support,and by being able to accept the structure of

11 Some of the most common indicators used are the poverty headcount ratio, i.e., the percentage of the population below a givennational poverty line, unemployment rates, and capability indicators such as primary school completion rates, illiteracy rates,vaccination coverage, maternal mortality rates and life expectancy at birth.

12 Given the recognition that country-owned, participatory PRSPs take time to develop, and in order not to delay progress on debtrelief, the Boards of the World Bank and IMF allow countries to prepare interim PRSPs. These should contain: 1. a statement ofcommitment to poverty reduction, 2. an outline of the nature of poverty, 3. existing government strategies to tackle poverty and4. a timeline and process for preparing a full PRSP, a 3-year policy matrix and a macroeconomic framework.

13 Experience has shown that the time required depends on the existence and quality of extant poverty data, national planningcapacity, the level of integration of Government structures (in particular between line ministries and ministries of finance andplanning), as well as the status of consultation mechanisms between Government and national non-state actors.

9

Section 3 – The Role of Poverty Reduction Strategy Papers (PRSPs) in Aid Delivery

previously negotiated Annual Progress Reports.The rise of PRSPs as the main form of nationaldevelopment strategies in partner countries has ledto a much greater focus on poverty issues withinthe international development community, by partnergovernments and civil society actors. WherePRPSs have been rolled out, another significantdevelopment has been a greatly improvedengagement between partner country governmentsand national non-state actors in terms of policydebates around poverty reduction.

And on a more technical level and in terms ofnational systems, the very process of designing,implementing, and monitoring and evaluating PRSPshas led to a noticeable improvement in nationalpoverty analysis, enhanced national databasesand statistical skills. National public financialmanagement and planning, and budgeting systemshave also been strengthened. Improved inter-ministerial coordination mechanisms arising fromthe multi-sectoral content of PRSPs, and the basicneed for the various bodies of government tocooperate in the process of designing andimplementing these strategies, might also be addedto these positive developments.

Given the large number of countries involved inthe global PRSP exercise, the picture of theirspecific impact on national poverty levels could notbut be mixed. But one example of what can bedone given a government’s commitment to povertyreduction is that of Mozambique. Mozambiquedeveloped its first interim PRSP in 1999, which wasone of the first to be approved by the Boards ofthe World Bank and IMF. This allowed the countryto access Highly Indebted Poor Country (HIPC)debt relief and additional BWI concessional finance(see section 4 on debt relief efforts). A full PRSP

followed in 2000, covering the period 2001–2005.This contained detailed targets, a timeframe andresource requirements.

Between 1997 and 2003, the poverty headcountfell from 69 percent of the population to 54 percent,an early achievement of the poverty reductiontarget set for 2005. In line with the government’scommitment to rebuild human capital after the civilwar, non-income poverty indicators, based on thecapabilities approach to poverty, also showedmarked improvements. In the area of education,for example, spending increased from US$170million in 2000 to US$250 million by 2004. It isnoteworthy that this increase in spending oneducation took place against the background offalling external financing for the sector, from 42percent of total public expenditure on the sector to38 percent over the 5-year period. This might beinterpreted as a strong commitment to povertyreduction, irrespective of external concessionaryfinancing.

An additional and highly instructive aspect of thecountry’s experience lay in the fact that, startingfrom a very low base – given the residual effectsof the protracted civil war from which it emerged– Mozambique was able to sustain very healthygrowth rates over the period of its 1st PRSP,averaging above 8 percent per annum. Theserobust rates of growth were not only a keycontributor to reducing poverty through enhancedemployment opportunities and improved wages, butalso allowed the state to expand its revenue baseand increase the volume of its revenue stream. Byincreasing its fiscal space, the state was able tochannel resources to both directly productive publicinvestments as well as increasing expenditures onhealth and education.

10

Given that Zimbabwe will, on the basis of its existingdebt profile, be a candidate for international debtrelief (see below), a sound understanding of thearchitecture and mechanisms of international debtrelief efforts will be essential for national decision-makers and the wider public. It is therefore worthoutlining the historical developments of this aspectof donor-partner country financial flows.

4.1 THE PARIS CLUB AND NAPLESTERMS

One long-standing actor in the area of external debtrelief to developing countries is the Paris Club. Thisis an informal grouping of official bilateral creditorsfrom the world’s eighteen richest economies, whichmeet regularly to find coordinated responses topayment difficulties experienced by debtorcountries. In terms of their standard practice, ParisClub members agree to either provide debt reliefthrough:

1. Postponement of payments, i.e., rescheduling;or

2. Concessional rescheduling, i.e., a reduction indebt service obligations through downsizing ofdebt stocks.

As of 2007, Paris Club members had reachedaround 400 agreements with 85 debtor countrieswith the total amount covered in these agreementssince 1983 totalling around US$500 billion.

In 1994 Paris Club members agreed to adopt a newposition on the debt of the world’s poorest countries.

These became known as the Naples Terms, whichwere underpinned by the notion of ‘debt overhang’,i.e., a situation in which a government’s externaldebt exceeded its future capacity to repay itsobligations.14 The eligibility criteria applied werethat candidate countries should:

1. Possess a high level of external indebtedness

2. Only be eligible for IDA financing from theWorld Bank.15

3. Have a low GDP per capita (US$755 or less).

In addition, account would be taken of the trackrecord of the debtor country with both the ParisClub and the IMF in terms of their ability to respectthe debt agreement.

Under Naples terms, two basic routes were laidout for debt relief.

1. The debt reduction option: 67 percent of bilateralnon-ODA (Official Development Assistance)claims16 are cancelled, with the outstandingpart being rescheduled at an ‘appropriatemarket rate’ with a 23-year repayment periodand a 6-year grace period;

2. The debt service reduction option: 67 percentof non-ODA claims are rescheduled at areduced interest rate with a 33-year repaymentperiod.

The specific provisions governing ODA claims arethat they are to be rescheduled at interest rates atleast as favourable as the original interest rateapplying to those loans, over a 40-year period with

Section 4

International Debt Relief Efforts and Debt Management

14 Many Highly Indebted Poor Countries (HIPCs) had to resort to expensive, short-term commercial borrowing in order to fulfilannual debt-servicing obligations, and thus become sucked into a debt trap. Naples Terms were an attempt to offer an exit strategythrough the reduction of Net Present Value (NPV) of debt stock. The NPV applied to debt stock is the sum of all future debt serviceobligations (both the principal and interest). The NPV measures the degree of concessionality of a country’s debt stock. When theinterest rate on a loan is lower than the market rate, the resulting NPV of the debt stock is smaller than the face value, thedifference between the two therefore representing the grant element.

15 The International Development Association (IDA) is the World Bank’s lending window for long-term interest-free loans andgrants.

16 i.e., non-concessional debt from Paris Club members.

11

Section 4 – International Debt Relief Efforts and Debt Management

a 16-year grace period. Such rescheduling isintended to result in a reduction of the net presentvalue (NPV) of the claims, as the originalconcessional rate is lower than the appropriatemarket rate of the debt.

4.2 THE HIGHLY INDEBTED POORCOUNTRY (HIPC) INITIATIVE

The next stage in the development of internationaldebt relief efforts took place in 1996, when theBWIs recognize that even after the application ofNaples Terms by the Paris Club, and despite theprovisioning of concessionary financing and theirpursuit of sound economic policies, the external debtsituation of a number of low-income countrieswould not allow them to achieve sustainable externaldebts levels within the reasonable period of timeand without additional external support.

These considerations then lead to the establishmentof the Highly Indebted Poor Country (HIPC)initiative, the key objective being to provideexceptional assistance to eligible countries followingthe adoption of sound economic policies (soundnessbeing determined by the IMF and World Bank) tohelp them reduce their external debt burden tosustainable levels.17 In terms of eligibility criteriafor HIPC, a country’s debt levels were consideredstill unsustainable if either:

1. the ratio of its debt to exports were above athreshold of 250 percent; or

2. the ratio of its debt to government revenue wasabove 280 percent.

In 1999 both the IMF and the World Bank agreedto strengthen the original HIPC initiative andcommit to even deeper debt relief under theEnhanced HIPC initiative. This initiative is basedon the recognition that the empirical workunderpinning the original HIPC was derived from

prior debt sustainability analysis (DSA) based onmiddle-income countries. The reality was that low-income countries had a much lower capacity tosustain external debt. As a result, under enhancedHIPC, the eligibility threshold was reduced further:

1. The ratio of debt to exports was reduced to150 percent; and

2. The ratio of debt to revenue was reduced to250 percent.

Most significantly, an explicit condition wasestablished in 1999 that linked the granting of debtrelief to the need for HIPC countries to use fundsfreed up from debt relief for poverty reductionefforts in the context of Poverty Reduction StrategyPapers (PRSP).

The process for accessing this relief on the partof HIPCs requires, firstly, confirmation that acountry meets the eligibility criteria; that it adoptsan adjustment and reform programme supportedby the IMF and World Bank, and pursues thesefor a minimum of six months; and demonstrates arecord of macroeconomic stability. It must alsohave either already cleared, or reached anagreement to clear, outstanding arrears to theWorld Bank, IMF and other multilaterals, and havedeveloped an interim PRSP. The staff of the WorldBank and IMF then conduct a debt sustainabilityanalysis in order to determine the level ofindebtedness of the economy through anexamination of the loan portfolio and the level ofdebt relief required to bring down a country’s debtindicators to below HIPC thresholds.18

The Boards of the IMF and World Bank thenformally decide on a country’s eligibility (known asthe decision point). Other creditors (bilateralmembers of the Paris Club, other multilaterals suchas regional development banks, other official non-Paris Club members as well as commercialcreditors) then commit to provide sufficient

17 Sustainability being defined in terms of levels of debt that would comfortably enable those countries to service their debt throughexport earnings, aid and capital inflows.

18 Debt Sustainability Analysis for HIPC have uniform thresholds in order to ensure equity of treatment. They also integrate riskscenarios, and given the vulnerability of low-income countries to external shocks, they use a backward-looking 3-year average ofboth exports and revenue in order to iron out volatility in both export earnings and revenues.

12


assistance by the completion point. Delivery ofdebt relief by the IMF and World Bank will dependon assurances of action by these other creditors.

During the interim period the country in questionwill begin to receive conditional interim debt relief‘based on a country’s immediate needs andcapacity for channelling the funds to poverty-reducing purposes’. A full PRSP is developed afterwhich it is submitted to the IMF and World Bankboards for approval. At the completion point, theHIPC country must have maintained macro-economic stability under an IMF Poverty Reductionand Growth Facility (PRGF) supported programmefor a period of time, and carried out key structuraland social reforms agreed upon at decision point inthe context of a PRSP which has been satisfactorilyimplemented for one year. Full, irrevocable debtrelief is then granted.19

As of 2008, the total cost to creditors of the HIPCdebt relief initiative was estimated at U$71 billionin end 2007 Net Present Value (NPV) terms (IDAand IMF, 2008: 18).

4.3 THE MULTILATERAL DEBTRELIEF INITIATIVE (MDRI)

At their 2005 summit, the G-8 countries proposedthat the IMF, World Bank and the AfricanDevelopment Fund of the African DevelopmentBank cancel 100 percent of their debt claims oncountries that either already had, or wouldeventually reach completion point under the HIPCinitiative. This covered debt accumulated prior toend-2004, and became known as the MDRI.

As of the end January 2009, 23 countries that hadreached HIPC completion point had benefited fromthe MDRI, with an additional 10 HIPC currentlyat decision point (known as Interim Countries) alsoeligible for MDRI once they reached HIPCcompletion point. As of the end of July 2008, thetotal amount of relief granted under the MDRIstood at US$49 billion (IDA and IMF, 2008: 10).

4.4 IMPACT, SLIPPAGE AND MORALHAZARD CONSIDERATIONS

In terms of the number of beneficiaries, as of theend of 2006 there were 22 post-completion pointHIPCs (of which 18 were sub-Saharan Africancountries) that also benefited from the relief offeredby the MDRI, 10 post-decision point HIPCs (ofwhich 8 were sub-Saharan African countries) and9 pre-decision point countries (of which 7 weresub-Saharan African countries).

The total costs as of the end of 2006 under HIPCwere US$45.5 billion in NPV terms and US$61.1billion in nominal terms, while under MDRI thefigures were US$21.1 billion and US$41.7 billionrespectively. The impact on the debt stocks ofHIPCs has been significant, with the total debtstocks of 32 post-decision point HIPCs having beenreduced by over 90 percent by the end of 2006. Interms of total volumes, before the application ofNaples Terms this stood at US$106 billion in NPVterms, after Naples it had been cut to US$86 billion,after the HIPC initiative it had fallen to US$41billion, and after the MDRI and additional bilateraldebt relief it had fallen to US$9 billion.

Another significant measure of the impact of theseinternational debt relief efforts is the average debtservice obligations of post-decision point HIPC asof 2006. The average debt service to exports ratiofell from 16.6 percent in 1999 to 6.4 percent in2006; the debt service to GDP ratio also droppingfrom 4.6 percent to 1.9 percent; and the NPV ofdebt to exports being cut from 440 percent to 132percent over the same period.

Given the intimate linkage between HIPC andMDRI debt relief and PRSPs whereby the formerwere intended to release funds for HIPCs to investin national poverty reduction efforts, it would beinstructive to ascertain the extent to which this hastaken place. The evidence to date does suggestthat there is a robust correlation between areduction in debt servicing requirements and anincrease in poverty-reducing spending through an

19 The average length of time between decision and completion point varies greatly, from 3 months in the case of Uganda, to 5.7years in the case of Malawi. The key variable seems to be domestic capacity to both produce a technically sound PRSP and toimplement the recommended reforms.

13

Section 4 – International Debt Relief Efforts and Debt Management

improvement in the fiscal space for priorityexpenditures of countries benefiting from debtrelief. The impact of debt relief on national povertyreduction efforts may be gauged in terms ofchanges in two variables:

1. World Bank and IMF data shows that suchexpenditures have increased by about 2 percentof GDP in HIPCs since the late 1990s as debtservicing as a percentage of GDP hasdecreased by a similar amount (IDA and IMF,2008: 14).

2. Indicators on the macroeconomic front are alsopositive, with the average inflation rate for post-completion countries having fallen from 13.8percent over the period 1994–1998 to 6.6percent over the period 1996–2005. Real GDPgrowth rates have averaged about 4.3 percentfor those countries that reached completionpoint, as opposed to their average of 1.7 percentbetween 1980–1993.

These positive results must be tempered, however,by the fact that these countries have not registeredsignificant improvements in their domestic resourcemobilization, which has remained largely constant.Export performance has remained at an averageof around 26 percent of GDP. More seriously, andgiven that the various international debt relief effortswere aimed at ensuring a permanent exit fromrescheduling, it cannot but be a source of concernthat some slippage has been evident. In 11 of 13post-completion point countries for which data wasavailable in 2005, external debt sustainability haddeteriorated since completion point, and 8 of thesewere once again above HIPC thresholds. Onepossible lesson to be drawn is that international debtrelief efforts, in and of themselves, are not sufficientto improve export diversification, national debtmanagement capacity or the ability of developingeconomies to cope with external shocks througheither a deterioration of terms of trade orfluctuations in exchange rates.

It is worth recalling that poor debt managementand imprudent borrowing played a major role –together with external shocks and the lack of adiversified export base – in the build-up ofunsustainable debt in the first place. Debt reductioncan only provide temporary relief and an opportunityfor countries to undertake the reforms necessaryto avoid slippage. One key area that contributes toavoiding this danger is national debt managementcapacity, the strengthening of which will help toensure more prudent future borrowing. In manycountries, as noted by a World Bank evaluation ofthe HIPC initiative:

‘HIPC countries have a fair loan-by-loanrecord of their sovereign external debt, butnot of loans taken on by state enterprises andthe private sector... [in addition] Agenciesresponsible for debt management…are not ableto analyse the impact of new borrowing onlong-term debt sustainability and on macro-economic scenarios. Countries debt manage-ment units need to strengthen institutionalframeworks, improve staffs’ analytical skillsand upgrade technical software’.20

In addition to the dangers of slippage, the issue of‘moral hazard’ has also loomed large in theconsiderations of both bilateral donors andInternational Financial Institutions (IFIs), i.e., thatcontinual extensions of international debt reliefefforts might constitute incentives for debtorcountries to increase their borrowing to unsustain-able levels and then avail themselves once againof debt relief. One key Multilateral DevelopmentBank (MDB), the African Development Bank,stated as much in a policy document on the dangersof non-concessional debt accumulation by itsmember states:

‘ADF [African Development Fund] grants anddebt relief may introduce an incentive forcountries to over-borrow from other creditorswhich could cause their debt to be unsustainableand compel ADF and other MDBs to increasethe grant share of their assistance.’ (AfricanDevelopment Bank, 2008: 1)

20 World Bank, Independent Evaluation Group, ‘Debt Relief for the Poorest: An Evaluation Update of the HIPC Initiative’,Washington DC, 2006, pp.27-28.

14


Notwithstanding these concerns, the Boards of theWorld Bank and IMF chose to extend the ‘sunsetclause’ of the HIPC initiative on a number ofoccasions for 2-year periods at a time in order toallow potentially eligible countries to enter theHIPC process. But in December 2006 theseBoards sought to ensure that the HIPC initiativewas brought to finality. They decided that onlycountries that had already been assessed (or areassessed in the future) to have met HIPC incomeand indebtedness criteria based on end-2004criteria, and that met policy performance criteriaand agreed to adopt an IMF-World Bank supervisedprogramme, could enter the HIPC stream.

Zimbabwe entered into arrears with the majorityof its multilateral and Paris Club creditors between2000 and 2002, though it did make some tokenpayments to the IMF, World Bank, the AfricanDevelopment Bank as well as some Paris Clubcreditors. But as of this writing, its debt ratios arewell above the HIPC thresholds. While the HIPCdetermined NPV of debt to exports threshold atthe end of 2004 stood at 150 percent, based on2006 calculations the IMF estimated thatZimbabwe’s ratio already stood at 273 percent. Onthe assumption that the then prevailing policiesremained unchanged, the IMF’s baseline scenarioexpected a strong contraction in exports andnegative annual GDP growth of 4 percent to

continue until 2011. The ratio of debt to exportswas projected to increase to over 440 percent. It isnoteworthy that the IMF observed at the time that:

‘Clearly, if policies improve the debt indicatorswould improve as well, but the very high debtratios in the baseline scenario suggest thatbetter policies alone are unlikely to besufficient to make Zimbabwe’s external debtburden sustainable.’ (IMF, 2007: 30)

One technical constraint to Zimbabwe’s eligibilityfor debt relief is that it has been a notionally IBRD/IDA21 blend country for a number of years. Thismeant that had Zimbabwe not been in arrears itwould have had access to both the World Bank’slong-term interest-free loans and grants through itsIDA lending window, as well as through the Bank’sIBRD lending window, which charges interest onloans. However, a precondition for HIPC debt reliefis that the country should be classified as an IDA-only country. For Zimbabwe, this means it will needto be reclassified as a Low-Income Country andtherefore also as an IDA-only country before it iseligible for debt relief under the HIPC initiative.With Zimbabwe’s re-engagement with theinternational donor community, and with thenecessary political support from the IMF and WorldBank boards, a technical case could be made tomake it eligible for HIPC debt relief.

21 IBRD/IDA stands for International Bank for Reconstruction and Development and the International Development Association,respectively.

15

After years of isolation from the mainstream ofinternational development processes, Zimbabwe willhave to rapidly familiarize itself with what hashappened over the last 10 years. The recent pasthas seen significant changes to the way aid isdelivered and how it is channelled in order to achievedevelopmental objectives. Yet at the same time, likelatecomers to any global process, there will beadvantages in terms of being able to quickly playcatch-up and update its understanding based on thestate of the art, drawing on lessons learned fromthe experiences of other countries.

Zimbabwe will also be able to benefit from theexistence of the Paris Declaration, subject to itbecoming a signatory to the agreement. To the extentthat there is a single international normativeframework guiding relations between OECD donorsand partner countries, it is the Paris Declaration. Thismight be seen as the culmination of numerousevaluations and experiments that took placethroughout the 1990’s within the donor communityas well as partner countries. It was rooted in thewidespread disillusionment with the results ofdevelopment cooperation, particularly in Africa.

In 2005, the Paris Declaration was signed up by115 donor and partners countries, as well as a largenumber of international organizations. It seeks tocodify a number of principles aimed at improvingthe effectiveness of aid through focusing on fivekey commitments. Of these, two may be seen asthe primary responsibility of donors, one places theonus in terms of implementation on partnercountries, and two are shared, though in fact actionsare required from both sides for all five principles.Both donors and partner countries jointly committedthemselves to implement these principles in aspirit of mutual accountability. In a welcomedevelopment, the Declaration also goes beyond thesimple enunciation of general principles, with thesignatories agreeing on the need to both monitorand evaluate progress on the basis of an agreedset of progress indicators.

5.1 KEY FEATURES OF THEPRINCIPLES

I - The principle of ownership

The first principle contained in the Paris Declarationis primarily a responsibility of partner countries.Based on decades of experience with theshortcomings of externally generated and imposeddevelopment strategies, under the Paris Declarationpartner countries undertake to exercise effectiveleadership over their development strategies. Theytake the lead in coordinating development activitiesin consultation with donors, as well as with theirrespective civil societies and the private sector.Donors on the other hand undertake to respect theleadership of partner countries in terms of theirdevelopment strategies, while also committing tohelp them to build their capacity to do so.

The principle makes specific reference to ‘nationaldevelopment strategies’, which are seen asoverarching development strategies that includepoverty reduction, and other sector and thematicstrategies. This is significant given the general moveaway from ODA flows in support of discreteprojects to supporting longer-term inter-sectoral andnationally devised development strategies. Needlessto say, the precise boundaries of what constitutes‘country ownership’ have always been difficult todefine and to translate into practice. Yet the ubiquityof what are supposed to be internally generatedPRSPs is an indication that progress has beenmade. The specific indicator used for measuringprogress in terms of partner country ownership isthat by the target date of 2010 at least 75 percentof partner countries have operational developmentstrategies.

II - The principle of alignment

This principle flows logically from principal 1 onpartner country ownership. The ideas underlying

Section 5

Using the Paris Declaration to Maximize AidEffectiveness

16


the importance of alignment are based on therecognition that parallel donor systems andprocesses are, in the long term, both unsustainableand counter-productive, and that the efficiency andimpact of aid flows will be enhanced if aligned tosound national strategies, systems and processes.Under the principle of alignment, therefore, donorscommit themselves to basing their overall supporton partner countries national developmentstrategies, and to tailor their support to the prioritiesidentified in those documents. One key indicatorfor this is the percentage of aid flows to thegovernment sector that is reported on the nationalbudgets of partner countries, with a target of halvingthe percentage of aid flows not so reported by 2010.In addition, they also undertook to ensure thatresource flows would be linked to a single, mutuallyagreed set of indicators on the basis of whichperiodic joint reviews of progress could be based.

Under the alignment principle, donors alsocommitted themselves to using the systems andprocedures of partner countries. What this entailsis the use of Public Financial Management (PFMs)systems, procurement, auditing and monitoringsystems of partner countries to avoid creatingparallel structures. In the case of PFMs, two targetsare set for 2010, namely, that at least 90 percent ofdonors use these systems, and that aid not goingthrough PFMs is reduced by at least one-third.Similar targets are set for using partner countriesprocurement systems. Donor countries also committhemselves to work together with partner countriesto carry out diagnostic reviews of national capacityin these areas. Where necessary, donors areprepared to assist in strengthening national capacityso that these systems meet generally acceptedstandards. At the very least, partner countriesshould have a reform programme underway toreach those standards.22

Donors also commit themselves to untying aid inorder to increase the effectiveness of these flows,23

as well as making aid flows more predictable byensuring that it is released according to agreed

schedules. A specific progress indicator is to halveaid scheduled but not disbursed within a particularfiscal year by 2010.

III - The principle of harmonization

The key objective of this principle is to ensure thatdonor assistance programmes are more effectivecollectively. The primary responsibility for thechanges necessary to implement harmonizationgoals rests with donors. Progress on theharmonization agenda requires work in three keyareas, namely, the development of commonarrangements, the simplification of procedures, anda more effective division of labour to reduce thedemands on the capacity of partner countries.

More specifically, the harmonization principleenjoins donors to set up common arrangementswithin countries in the areas of planning, funding,disbursement, monitoring and reporting to hostgovernments on their aid activities, and to increasetheir use of ‘programme-based’ aid modalities. Atarget was set for two-thirds of donors’ aid flowsto use programme-based approaches by 2010.

Specific reference is made to the need for donorsto address a common source of resentmentamongst partner countries, namely multiple,duplicative missions. Forty percent of donormissions should therefore be joint missions by 2010.In coordination with partner countries, donors arecalled upon to establish a clear division of labouramongst themselves based on the principle ofcomparative advantage, and to delegate authorityto previously agreed ‘lead donors’ to executeprogrammes.

IV - The principle of managing for results

Flowing from the growing disillusionment with theresults of development cooperation in the 1990’snoted above, the concept of managing for resultsshould be seen as an attempt to ensure that aidflows are managed in such a way that the desired

22 The specific target for monitoring progress in the area of PFMs is that at least half of the partner countries move up at least onemeasure on the World Bank’s Public Financial Management/Country Policy and Institutional Assessment (PFM/CPIA) scale.

23 The point here is that numerous studies have shown that tying aid to home country, i.e., donor country, supplier pre-conditions,greatly raises the costs of these goods and services in contrast to situations in which partner countries are able to source these onthe open market.

17

Section 5 – Using the Paris Declaration to Maximize Aid Effectiveness

results are achieved, and that monitoring andevaluation systems are used to improve impact.Responsibility for achieving this is shared betweendonors and partner countries.

Donors are called upon to ensure that their ownprogramming systems and their funding flows arelinked to specific results. They should also alignthese results to the performance assessmentframeworks of partner countries, both of thesebeing derived from the national developmentstrategies of partner countries. Simultaneously,partner countries are to ensure that their ownplanning processes are properly linked to budgetaryprocesses. Their monitoring, evaluation andreporting frameworks should also be results-oriented, i.e., based on indicators derived from theirnational and sectoral development strategies. Thetarget for 2010 is to reduce the number of partnercountries without transparent and monitorableperformance assessment frameworks by one third.

V - The principle of mutual accountability

This principle is meant to give substance to theidea that both donors and partner countries arejointly accountable for development results. Thismutual accountability should be strengthened andthe transparency of development cooperationincreased in order to trigger greater public supportfor both the national development strategies ofpartner countries and for development assistancein donor countries.

Partner countries are therefore enjoined tostrengthen the role of their national legislatures inthe process of designing national strategies andtheir budgetary processes, and to ensure theparticipation of the general public in theformulation, monitoring and evaluation of theimplementation of these strategies. Donors ontheir part will endeavour to provide comprehensiveinformation on their aid flows in a timely mannerthat will enable partner country governments to,in turn, present comprehensive budget reports totheir parliaments and citizens.

Both donors and partner countries are to movetowards mutual assessment reviews in order tomonitor progress in terms of progress in imple-menting these agreed commitments on aideffectiveness, with all partner countries having suchreview mechanisms by 2010. In practice themechanisms range from annual consultationsaround major national strategies, through to jointreviews of sector strategies, and ConsultativeGroup meetings.

5.2 HOW HAVE THE PRINCIPLESFARED?

One of the key thrusts of the Paris Declarationwas to provide partner countries with greater policyspace for formulating and implementing their ownstrategies. There remains, however, the possiblyirresolvable tension between country ownership andleadership of the development agenda on the onehand, and what bilateral donors and the IFIsconsider to be technically sound and realisticstrategies, that they would be prepared to support,on the other.24

Notwithstanding the fundamental asymmetry thatcharacterizes this relationship, annual surveysconducted by the OECD-DAC do indicate thatprogress has been made on both sides of therelationship, though the picture shows widelydiffering progress in terms of individual donors andpartner countries. In aggregate terms, however, anumber of key findings are significant.

On the fundamental first principle of nationalownership of the development agenda, oneconclusion reached in a highly representativeOECD-DAC survey of 2008 – which covered atotal of 55 partner countries and more than 50percent of all aid delivered by OECD members topartner countries in 2007 – was that progress hadbeen made in terms of designing nationaldevelopment strategies since 2005. Weaknesses,however, still remained in linking these to budgetingprocesses and in the prioritization and sequencing

24 Some analysts have argued that the concept of national ownership as defined by the Bretton Woods Institutions (BWIs) translatesinto an acceptance of primary responsibility by countries for the policy reforms put forward by the International FinancialInstitutions. This is contrasted with more robust definitions that include the designing and implementation of strategies andpolicies that national governments freely choose based on their own analysis of what reforms are necessary and how and whenthey should be introduced.

18


of actions. From a baseline of 33 partner countriesused in a previous survey in 2006, the percentageof partner countries possessing operationaldevelopment strategies had increased from 17percent in 2005 to 24 percent in 2007. But this wassignificantly below the progress indicator and targetlaid down in 2005, namely that by 2010 at least 75percent of partner countries should have suchstrategies in place (OECD, 2008: 22).

The highly amorphous and inherently political natureof the concept of ownership helps to throw light onthe mixed progress to date, with questions oftenbeing raised as to the utility of this progress indicatorin capturing the reality of country ownership. Thisis particularly true in the case of highly aiddependant countries, which have to juggle the needfor their development strategies to containobjectives that are both endogenously generated,while at the same time ensuring that they areendorsed by donors. The tension between the twoposes real dilemmas for many partner countries:

‘…highly prescriptive approaches are notconsistent with local ownership… On the otherhand, national strategies that do not givesubstantial weight to internationally sharedobjectives are unlikely to provide an enduringbasis for partnership between partnercountries and donors.’ (OECD, 2009a: 38)

On the alignment principle, the OECD-DACsurvey noted that in general, donors wereincreasingly gearing their activities in support ofnational development strategies, as well as usingthe systems of partner countries. Given therecognition of the key role of PFMs in contributingto both improving the prospects of partner countriesachieving public policy objectives as well asprogress on alignment, significant efforts have beendirected towards improving those systems. Basedon the baseline of 2005, which had 40 percent ofdonors using the PFMs of partner countries, thefigure for 2007 had improved to 45 percent (OECD,2008: 38).

Once again, however, given that the target for thisprinciple is 80 percent by 2010, it is clear thatprogress has been very slow. Interestingly, therelationship which is assumed to underlie theprinciple, i.e., that the improved quality of partnercountry PFMs would logically trigger increased use

of these systems by donors, is shown to be lessthan robust. The survey notes, for example, thatwhile Zambia’s PFM system improved from 3.0 to3.5 on the World Bank’s PFM performance scale,which triggered a 25 percent increase in donoruse of those systems, in the case of Ghana,notwithstanding an improvement from 3.5 to 4.0on the same scale, there was a 10 percent decreasein the use of country systems (OECD, 2008: 42).

While improvements in national systems are anecessary condition for donor alignment to partnercountry systems, they are clearly not a sufficientone. The continued attractions of project deliverymodalities, which largely bypass national planning,budgeting, accounting, auditing and procurementsystems, may contribute to this. In addition, thereis the possibility that donor perceptions of corruptionwithin a partner country, independently of theobjective quality of a partner country’s PFM, willdiscourage the use of those national systems.

On the harmonization agenda – as measured bythe percentage of aid delivered through programme-based approaches in which donors engage arounda single country-owned programme and budgetframework – progress has also been made since2005. The proportion of aid provided in this mannerincreased from 43 percent in 2005 to 47 percent in2007 (OECD, 2008: 52). Other harmonizationindicators relate to the objective of reducingtransaction costs. So, for example, the number ofcoordinated missions between donors has also risenin the partner countries surveyed. The number ofsuch missions increased from 18 percent in 2005to 21 percent in 2007 (OECD, 2008: 54). Jointcountry analytical work by donors, once again aimedat reducing both transaction costs and to avoidduplication, also shows some admittedly limitedprogress, having increased from 42 percent in 2005to 44 percent in 2007 (OECD, 2008: 55).

On the 4th Paris Principle covering the areas ofaccountability on the one hand, and results andassessment orientation on the other, the picture isagain mixed. Accountability and improvements inthe recording of aid flows in national budgets – toenable partner countries to present accurate andrealistic budgets to their national parliaments –show that there has been some progress due to themore comprehensive and timely reporting of aidflows by donors. The percentage of aid recorded

19

Section 5 – Using the Paris Declaration to Maximize Aid Effectiveness

on budget improved between 2005 and 2007 from42 percent to 48 percent amongst the partnercountries surveyed, though this still falls far shortof the target of 85 percent by 2010 (OECD, 2008:58). The complementary component that the ParisDeclaration sees as contributing to greater budgetrealism and therefore accountability, under whichdonors are called upon to provide reliable indicativecommitments over multi-year frameworks, showsthat the gap between aid scheduled and aiddisbursed has also begun to narrow, with animprovement from 41 percent in 2005 to 46 percentin 2007 (OECD, 2008: 61).

The second aspect of the 4th Principle, namely theneed to nurture a ‘performance culture’ in partnercountries through the introduction of appropriatemonitoring, reporting and assessment frameworks,has shown much more modest progress. Many ofthe partner countries surveyed continued to not onlylack such frameworks, but even the capacity todesign them. The 2008 OECD survey concludedthat:

‘As is the case with strategic planning fordevelopment, results monitoring succeedswhen there is high-level political interest, andnot otherwise… Case studies… revealshortcomings concerning the demand for, aswell as the supply of, monitoring information.’(OECD, 2008: 63)

In terms of the 5th Paris Principle governing mutualaccountability for commitments and results, allcountries were to have the necessary mechanismsin place to carry out mutual assessments by 2010.On this count, according to the OECD survey,progress has not been encouraging, with the surveynoting that the ‘expansion of mechanisms forreviewing partnership commitments seems to havecome to a halt…’ The number of countries withsuch mechanisms in place has only increased from12 to 14 between 2005–2007 (OECD, 2008: 64).

This lack of progress may be partly attributable tothe novelty of the concept of mutual accountability,as well as the power imbalance between donorsand partner countries referred to above, whichhinders two-way accountability.

It is important to understand the characteristics ofthe ideal situation that the full implementation ofthe Paris Declaration would logically lead to. Thishelps to throw light on the slow progress to date.In this theoretical end game there would be perfectalignment between donors and partner countriesaround technically sound national developmentstrategies. The national systems of partner countrieswould be considered completely reliable and wouldbe used by donors to channel ODA resources.Sound monitoring and evaluation systems would bein place, as well as comprehensive and transparentmutual accountability mechanisms. General budgetsupport transfers from donor treasuries to partnercountries’ treasuries would be the dominant modeof interaction, with partner countries thereby beingable to procure technical assistance and projectsupport on the open marketplace. Under thisscenario, the intermediation role played by the vastpanoply of bilateral, multilateral and non-statedevelopment agencies would, logically, also fallaway.

Progress to date indicates that certain patterns ofbehaviour on both sides of the donor-partnercountry relationship are deeply entrenched andmilitate against very rapid progress. While the ParisDeclaration currently constitutes the mostimportant, if not the only, overarching inter-national normative framework for improving aideffectiveness, and under which both donors andpartners can be called to account by their externaland domestic constituencies, it is also clear thatthe initial targets set down in 2005 may have beenoverly ambitious and failed to take account of themultiple incentives playing out in favour of slowprogress.

20

A sub-set of considerations underpinning the ParisAgenda concerns the specificities of aid deliveryin ‘fragile state’ contexts. In terms of the intellectualhistory of the concept of ‘fragile states’, in 2000the World Bank developed the concept of Low-Income Countries Under Stress (LICUS) in orderto characterize a specific group of countries that itsaw as meriting special treatment.

LICUS were seen as low-income countries (GrossNational Income of less than US$825 per capita)characterized by a debilitating combination ofweak governance, policies and institutions. Thosedefined as LICUS are ranked amongst the lowest(<3.0) on the World Bank’s Country Policy andInstitutional Assessment (CPIA)25. One key notionunderpinning the concept of a LICUS was thatproblems of weak policies, institutions andgovernance had persisted over time. It was notsimply a country in a brief spell of crisis.

Running in parallel with the World Bank’s workwas the OECD-DACs initiative entitled ‘DACLearning and Advisory Process on DifficultPartnerships’ which subsequently became knownas the ‘Fragile States Initiative’. According to theDAC, ‘States are fragile when state structures lackpolitical will and/or capacity to provide thebasic functions needed for poverty reduction,development and to safeguard the security andhuman rights of their populations’ (OECD-DAC,2007).

These two processes eventually converged atthe ‘Senior Level Forum on DevelopmentEffectiveness in Fragile States’, held in London in2005 where a set of ‘Principles of good internationalengagement in fragile states’ was produced, whichwere subsequently included as a supplement to theParis Declaration following their approval in 2007.

Out of a total of 12 key principles, the mostsignificant might be seen to be:

• The need to take context as a starting point –that analysis and proposed actions by donorsmust be calibrated to specific countrycircumstances, e.g., whether a specific fragilestate is a post-conflict one, or is undergoingprolonged political crisis or characterized by adeclining governance environment. It is alsonecessary, in order to design appropriateinterventions, to distinguish between capacityconstraints and a lack of political will.

• Do no harm – There is a need to ensure thatdevelopment assistance is conflict-sensitivegiven the very real dangers of externalassistance compounding social cleavages orinadvertently contributing to new conflicts.

• Donor countries are called upon to move fromreaction to prevention – where possible, helpfragile states to build resilient institutionswhich can withstand political and economicpressures. There is also a need to look beyondshort-term palliatives to current problems inorder to address the root causes of crises, andto move rapidly where the risk of furtherdeterioration is high.

• An overarching focus on state-building as thecentral objective – The long-term objective ofdonors engaged in fragile states must be toassist them to build capacity to deliver basicpublic goods such as safety, education, healthand good governance for their citizens. Inaddition, there is a need to assist with buildingan enabling environment for economic growthin order to generate employment as well asrevenue streams for the state.

Section 6

Aid and Fragile States

25 The World Bank’s CPIA is used to assess the quality of a country’s existing policy and institutional framework, ‘quality’ beingdefined as how conducive that framework is for fostering poverty reduction, sustainable growth and effective use of ODA. Forpartner countries, an evaluation of where they stand in terms of the CPIA framework is key to accessing the IDA (InternationalDevelopment Association – the soft loan wing of the WB) credits, i.e., long-term interest-free loans and grants. This is expressedas the IRAI (IDA Resource Allocation Index).

21

Section 6 – Aid and Fragile States

• Alignment with local priorities and systems indifferent ways in different contexts– forexample, in fragile states in which the politicalwill on the part of beneficiary countries existsbut capacity is a constraint, donors should seekto not undermine residual capacity by creatingparallel systems.

• Recognize the political-security-developmentnexus – based on the interconnections betweenpolitical, economic, security and social spheres,failure in any one of these has a negative impacton the others. Donors are enjoined to supportnational reformers so that they can developunified planning frameworks that bring togetherall these variables, while also ensuring thatrealistic targets and priorities are set.

• Aid instruments must be mixed and sequencedin order to fix context. Fragile states, particularlythose in promising but high-risk transitions, mayrequire a mix of aid instruments, including supportfor recurrent expenditure. At the same time,attention must be paid to the need to rapidlyimprove service delivery in health, education andother basic services in order to garner supportfor reforms, while at the same time avoiding long-term dependence on unsustainable systems.

• There is a need for donors to stay engaged forthe long haul in order to give success a chance– given the numerous challenges facing fragilestates, in particular their low capacity, donorsmust be prepared to adopt a longer timeframethan normal before expecting returns.

6.1 APPLYING THE FRAGILE STATESPRINCIPLES

Two aid specialists, Leader and Colenso (2005),carrried out pioneering work on aid instruments thatare most appropriate in fragile states. As they note,the standard donor choice of aid instruments infragile state situations has often been based on a

‘default, risk-reducing, starting point’ (2005: 4). Thisposition is characterized by the restriction of funds,dominance of the project delivery modality as a‘state avoidance’ approach, and a shorter timecommitment. Given the lack of strong governmentleadership and capacity in fragile states, the bestdonors often expect, in terms of the Paris Principles,is some degree of donor coordination rather thanharmonization and alignment. Based on theirfindings, they argue that this framework does notprovide a sufficiently strong base for donorsoperating in fragile state environments.

Contrary to conventional wisdom, their case studiesreveal that, as against the standard approach(largely project-based aid, with short time horizons,distributed through NGOs and with a focus on anarrow set of activities), a much wider range ofaid instruments can be used successfully in widelydiffering contexts. One key conclusion is that‘development actors should avoid “one size fitsall” prescriptions such as “budget support isinappropriate in fragile states”…’ given that theyfound that ‘an imaginative and flexible use of variousinstruments can have a significant impact onpoverty reduction’ (2005: 5).

This is hardly surprising given the widely variedcontexts of fragile states, and they recommend thatwork in fragile states requires creative thinking,experimentation and flexibility, as well as a soundknowledge of the local context. While recognizingthe attractions of project-based aid delivery in fragilestate situations26, they also argue that it is importantto think in programmatic terms right from the outsetof engagement. Based on the experience of one oftheir case studies, namely Afghanistan, they pointout that even in the most challenging of operationalenvironments it is possible to work with govern-ments on the development of national plans, withdetailed decisions on the most appropriate aidimplementation modality, i.e., whether throughgovernment systems, NGOs, direct donor imple-mentation, on or off-budget flows, or a mixture ofall, being decided subsequently.

26 The advantages associated with project aid rehearsed above become even more attractive to donors operating in fragile statesituations. Project aid can be very flexible in responding to new situations, is often based on relatively easily identifiable needs, hasshorter lead times between design and implementation, and can be implemented outside government systems in situations wherethe state possesses either very weak capacity or even constitutes an obstruction. Finally, it allows for a high degree of accountabilityto donors and their domestic constituencies.

22


As they note, in the case of post-TalibanAfghanistan, and notwithstanding the extremefragility of state institutions, budget support wasprovided through a World Bank managed Multi-Donor Trust Fund (MDTF). As of 2008, 27 donorswere supporting this fund. The World Bankdisbursed these funds to the Afghan governmenton a reimbursement basis. Donor fiduciary riskconcerns were further mitigated by the World Bankcontracting an ‘external monitoring’ agent thatevaluated the claims submitted by the governmenton the basis of a set of eligibility criteria andfiduciary standards. Further risk mitigation tookplace through the contracting of an internationalauditing firm, which subjected both the Fund andthe monitoring agent to audits. This arrangementhelped to address donor fiduciary risk concerns,but also strengthened government systems sincefunding priorities were based on those identified inthe country’s 12 National Priority Programmes.27

Important work has recently been carried forwardon the application of Paris Principles to fragilestates situations within the OECD-DAC. Mostnotable amongst these is an OECD-DAC reviewcommissioned in 2008 which draws out the lessonslearned from the application of the ParisDeclaration Principles in fragile states (OPM/IDL,2008). One key lesson arising from the casestudies is that it is important for donors not tounderestimate the task at hand:

‘A fundamental challenge to aid effectivenessin many transitional contexts is the sheer scaleand complexity of the task at hand. Thechallenges faced in (re)building the infra-structure and institutions of a state, deliveringservices, securing and sustaining peace aredaunting.’ (OPM/IDL, 2008: 25)

This therefore translates into a heightened needto effectively prioritize and sequence activitiesbased on a sound assessment of what can be donein the context of multiple constraints. The authorsnote that while the Paris Principles assume somedegree of shared understanding between partnergovernments and donors regarding national

development goals, as well as some capacity ofpartner governments to implement policy, theseassumptions may not hold in the case of fragilestates. Even basic operating assumptions such asan effective control of the national territory andtherefore ‘state reach’ may not apply.

Drawing a distinction between situations of conflict,prolonged crises or impasse with restricted scopefor the application of the Paris Principles, and thosestates characterized as ‘hopeful partnerships’(OPM/IDL, 2008: v), the report argues that the‘state-building’ imperative applies throughout thecontinuum, as do the principles of ‘taking contextas a starting point (principle 1) and adopting a ‘dono harm’ approach (principle 2). The most adversecircumstances are characterized by a rapiddeterioration of socio-economic indicators, conflict,prolonged political impasse and an almost completeabsence of dialogue between donors and govern-ment. Under these circumstances, even a minimalconsensus on an appropriate development agendais often absent. Paris Principles on ownership,alignment, harmonization, managing for resultsand mutual accountability become extremelyproblematic to apply. Nevertheless certain fragilestate principles can still be operationalized in suchcircumstances. The principles of ‘aligning with localpriorities in different ways in different contexts’and ‘doing no harm’ have allowed donors to supportbasic social service delivery through alternativemechanisms in a number of fragile state situations.

In the specific case of Zimbabwe, for example,where the government was often seen by donorsas pursuing policies that were contrary to whatwas required in order to achieve internationaldevelopment goals, there was nevertheless awindow for engagement with, and inter-donorharmonization in areas such as HIV and AIDS andOrphans and Vulnerable Children (OVCs).Adopting the principle of ‘shadow alignment’ to aGovernment National Plan of Action in whichpriorities were identified in the area of OVCs,donors were able to both engage at a technical levelwith line ministry staff and channel funds throughUNICEF, which passed these on to NGO care

27 For an overview of the Afghan MDTF experience see Scanteam, ‘Review of Post-Crisis Multi-Donor Trust Funds’, World Bankand Norwegian Ministry of Foreign Affairs and NORAD, 2007.

23


providers and faith-based organizations. While thismay be seen as a state-avoidance approach giventhe emphasis on engaging with sub-state actors andUN agencies, it was a way to ‘shadow align’ to atechnically acceptable National Plan of Action evenduring a stand-off between donors and a ‘partner’government.

In post-conflict and post-crisis situations in whichtransitions are underway, the principles regardingfragile states become easier to apply. When suchtransitions have begun, this provides a basis for thedevelopment of ‘hopeful partnerships’ betweendonors and countries emerging from crises.Notwithstanding the fact that such transitions maybe extremely fragile politically, that there may beenormous demands on a new administration whenits capacity to deliver is severely debilitated, anincipient development partnership may beestablished around a shared development agendaon the basis of the Paris Principles and thosegoverning engagement with fragile states.

6.2 A TOOLKIT FOR FRAGILESTATES TRANSITIONS

A number of instruments have been designed, testedand refined over the course of recent years, whichhave assisted in improving the quality of therelationship between donors and the governmentsof fragile states in transition. These help to lay thefoundations for national ownership, harmonization,alignment, accountability and a results-focus in thepartnership. In such transitional situations, given theenormity of the task at hand, sound planning,prioritization and sequencing of actions are key toboth ensuring maximum impact, as well asmanaging expectations on both sides of the donor-partner country relationship in terms of what canrealistically be achieved.

Joint Assessment Missions (JAMs) and Post-Conflict Needs Assessments (PCNAs) involvingboth donor and national technicians can, given thatthey provide a factual basis for dialogue, play avery important role in building shared understanding

of what is required and are a first step towardsjoint planning and coordination. JAMs and PCNAshave been rolled out in many post-conflict and post-crisis situations, and lessons learned over time havehelped to refine this instrument.

Another instrument that has proved to be useful intransition situations are Transitional ResultsMatrices (TRMs), often designed on the basis ofthe results flowing from JAMs and PCNAs. TRMsare simple planning, coordination and managementframeworks. These are usually developed by bothdonor and partner countries with a view to betterprioritizing actions in order to achieve a successfultransition in fragile states.28

It is worth recalling that transitions in post-conflict/post-crisis situations are fragile because of theinterconnections between four challenges facingtransition states:

• There are strong links between political andsecurity reforms and the delivery of socio-economic programmes;

• There is a need for prioritization in earlyrecovery environments characterized by highneeds and limited capacity;

• The importance of managing expectations ofdonors, fragile state governments and theirpopulations;

• The above considerations need to be balancedwith the imperative of ensuring rapid progressin order to maintain forward momentum at therisk of the transition process either stalling orbeing reversed.

TRMs can help both donors and national leadershipsto meet these challenges by;

• Placing key actions and results in terms ofpolitical, security, economic and social prioritiesin an easily digestible calendar;

• Forcing prioritization and ensuring realistictimelines and therefore managing theexpectations of all interested parties;

28 See United Nations Development Group and World Bank, ‘An Operational Note on Transitional Results Matrices: Using Results-Based Frameworks in Fragile States’, 2005.

24


• Helping to identify in advance possiblebottlenecks and periods of systems ‘overload’;

• Allowing for the monitoring of progress andthe addressing of areas in which progress issub-optimal; and

• Serving as a basis of dialogue between donorsand fragile states on resource allocationbetween priorities.

Experience with TRMs indicates that the keycriteria used in identifying priority actions must be‘crucial areas’ where lack of progress wouldheighten the risk of a reversal in the stabilizationand recovery process.

The standard format which has evolved over timeis based on four clusters which are key to recoveryin transitional situations, namely: 1) political, 2)security, 3) restoration of minimally functioningpublic financial systems, and 4) socio-economicrecovery. For each of these clusters, a TRM shouldcontain:

A – A Strategic Objective or Goal;

B – A baseline – even though it is often extremelychallenging in such situations to obtain reliabledata, what is available must be incorporated,without which it will be impossible for bothdonors and partner countries to monitorprogress;

C – Properly sequenced and time-bound actions.In situations where the capacity for reform isoften limited, the need for effective sequencingis heightened;

D – Targets and monitoring indicators, i.e.,observable targets that are subject toverification;

E – Clear responsibilities assigned for each action.

Based on assessments of TRMs in widely varyingsituations, one key lesson concerns the need tobalance comprehensiveness with selectivity. Giventhe multitude of constraints that characterize fragilestates in transition (capacity, dilapidatedinfrastructure, lack of financial resources) there isa clear need to prioritize. On a related note, themultiple demands on limited resources also entail a

need to strike a balance between addressingimmediate emergency needs at the same time asplanning for the longer-term. Strategic visioningmust therefore run alongside the process ofdeveloping a TRM to ensure that the inclusion ofshort-term actions in a TRM does not compromiselonger-term, comprehensive and sustainablerecovery. Finally, the simplicity of a TRM is centralto its utility for all users.

It has also become clear that donor support to agovernment-owned TRM should be based onprogress across the overall programme rather thanspecific and detailed targets. Given that progressin fragile states is never a linear one, even whenthey are not undergoing transitions, it is importantfor donors to base the conditionality of their supporton the strategic ‘big picture’ given that fragile statescan simultaneously perform both well and poorlyin different areas of the four clusters. Theclarification of mutual expectations at the outset isan essential ingredient in the process of buildingunderstanding between donors and governmentsof fragile states.

Reviews of TRMs have also brought to light thefact that the security sector variable has not beengiven the necessary attention in a number of TRMs.This may be due to the distinct cultures prevailingamongst humanitarian and development actors whooften have no previous experience in working withsecurity forces. To this should be added an additionalobstacle to deeper analysis and understanding ofthe security cluster and the designing of appropriateactions in TRMs, namely that work in this areaoften falls outside the scope of donor’s ODA funds.Nevertheless, the maintenance of basic security isa key enabler of the objectives in other clusters,and the longer-term goal of security sector reformcan help to ensure the sustainability of the overalltransition.

Pooled funding mechanisms such as Multi-DonorTrust Funds (MDTFs) offer an opportunity fordonors to begin the process of harmonizing andaligning to national priorities contained in recoveryframeworks such as TRMs. World Bank and UNadministered MDTFs have been implemented in arange of transitional situations, from Timor Lesteto the Sudan, Iraq and Afghanistan. Featurescommon to situations where they have been set upinclude large un-met needs in the area of basic

25


service provision in the short-term, as well assignificant medium-term needs in the area of socialand productive infrastructure. The public sector’scapacity to deliver is also usually very weak, if notnon-existent.

The advantages of MDTFs for donors are that theyreduce transactions costs through joint planning,prioritization and the sharing of information.Transferring resources to multilateral bodies to beadministered also means that the administrativecosts of managing those funds are reduced forindividual contributing donors. Single monitoring,evaluating and reporting mechanisms constituteadditional attractions. MDTFs also help to addressthe fiduciary risk concerns of donors in situationsthat are often characterized by weak public financialmanagement systems. In such situations, donorsare unwilling to engage in direct budget supportuntil their concerns have been met, and these canbe allayed by channelling funds through the UNand World Bank. Where MDTFs are structuredaround national priorities (as contained in partnercountry strategies), which are supported by donors,they help to ensure that donor assistance reflectthese budget priorities, even if funds do not initiallyflow directly through national systems.

There are a number of benefits that accrue topartner countries, especially where their capacityis severely constrained. With MDTFs, they are ableto interact with a single, coordinated and harmonizedsource of external assistance. MDTFs thus providea forum in which donors and partner countries areable to discuss priorities and implementationmodalities. As one survey of a mix of 18 UN andWorld Bank MDTFs has noted ‘many actors lookto MDTFs as the key venue for coming together

and exchanging information and views. In a numberof cases, they are in fact the only structured meetingplace…’29

Notwithstanding their promise and positiveperformance overall, and the fact that they representcurrent donor best practice in post-crisis situations,MDTFs have had their limitations. This may beattributed to the sheer scale of the tasks they face,and the time lags that are inevitably part of theresponse of all external actors in a situation in whichthere is an ever expanding picture of needs as anygiven transition develops. This ‘moving target’ hasto be reconciled with the need to maintain strategicfocus and selectivity while at the same timeensuring flexibility. One overview of the experienceof MDTFs notes that they:

‘…can often be overambitious in terms ofwhat they can deliver, and cannot be expectedto simultaneously build state capacity anddeliver public goods and services in a timelymanner. Start up time and costs are oftenunderestimated, and most MDTFs havefailed to provide adequate management andtechnical personnel on the ground’. (OPM/IDL, 2008: 43)

These shortcomings, which are gradually beingovercome with each newly designed MDTF,nevertheless help to explain why in many situationsMDTFs, while commanding substantial resources,still only represent a small percentage of total donoraid in each individual fragile state situation. To thismust be added the simple fact that there is a needon the part of each bilateral donor to be seen to be‘flying the flag’ separately, rather than being oneof many within an MDTF, thus contributing to themkeeping one foot outside the MDTF structures.

29 Scanteam, ‘Review of Post-Crisis Multi-Donor Trust Funds – Country Study Annexes’, World Bank, Norwegian Ministry ofForeign Affairs and NORAD, 2007. p.67.

26

A sub-set of concerns on aid effectivenessconcerns the impact of large flows of ODA on themacroeconomy of partner countries. Theseconcerns must be seen against the backdrop of thecommitment by G-8 leaders at the GleneaglesSummit in 2005 to double aid to Africa by 2010.The prospect of a ‘scaling-up’ of aid of thismagnitude has led a number of developmenteconomists to update earlier work on themacroeconomic impact of ODA flows, and to bringto the fore the macroeconomic challenges facedby many African economies that are alreadyheavily reliant on aid.

Since 2000 there has been a significant increase inthe number of countries receiving ODA that is over5 percent of GDP, with some countries such asMozambique and Sierra Leone receiving ODAflows over 20 percent of GDP for extended periods.In a post-crisis scenario in which there may besignificant net inflows of ODA, Zimbabwe shouldbe aware of the possible effects on its balance ofpayments and the demands placed on monetary,exchange rate and fiscal management.

In examining the possible macroeconomicimplications of aid flows, a distinction needs to bemade between the absorption and spending ofaid by partner countries (IMF, 2005). According tothe IMF’s definition, aid absorption is the extent towhich a country’s balance of payments non-aidcurrent account deficit widens in response toincreased aid flows. It reflects the increase in netimports resulting from increased aid that should betreated essentially as extra foreign exchange. Theextent of an economy’s absorption of aid dependson domestic demand for imports and the exchange-rate policy. Assuming there is no spare capacity inthe economy, aid only enables an economy to bothinvest and consume more by financing an increasein imports relative to exports.

Under the assumption of full capacity utilization,and if aid is simply spent on domestically producedgoods and services, it does nothing to increaseaggregate output. Rather, this increased demandwill fuel inflationary pressures. Under theseconditions, it could be said that real resourcetransfers can only take place where there is anincrease in net imports, i.e., increased absorption,or else it will only be equivalent to an expansionarydomestic fiscal and monetary policy, i.e., deficitfinancing.

Aid spending is defined as the widening in thegovernment’s fiscal deficit excluding aid thataccompanies an increase in aid. It indicates theextent to which the government uses aid in fiscalmanagement, i.e., either an increase in expenditureor a reduction in taxation. The Government fiscaldeficit excluding aid should therefore be seen asequal to total expenditures less domestic revenueand is financed by a combination of net aid anddomestic borrowing.

Spending is determined by the government throughfiscal management and absorption by the centralbank through its monetary management. Increaseaid flows can result in four possible combinationsof absorption and spending, each one with differentmacroeconomic implications.

1. Aid is both absorbed and spent

Under this scenario, the government spends aidresources – the central bank providing it withdomestic currency counterpart funds – whilethe central bank sells the foreign currency tofinance imports. The latter are absorbed by theeconomy through a widening non-aid currentaccount deficit.30 Though the non-aid fiscaldeficit becomes larger, it is neverthelessfinanced by increased aid.

Section 7

The Macroeconomic Management of Aid Flows

30 The non-aid current account balance is the current account balance excluding official grants and interest on public external debt.

27

Section 7 – The Macroeconomic Management of Aid Flows

However, sustained high levels of aid inflowscan result in real exchange-rate appreciation.The phenomenon is linked to the ‘DutchDisease’ common to many economiesexperiencing an export boom, in whichcurrency appreciation makes tradable goods31

less competitive and leads to an increase inimports.32 The Central Bank sells the extraforeign-exchange reserves derived from aidflows, which results in foreign exchangebecoming cheaper relative to the domesticcurrency. This rise in the exchange rate of thedomestic currency impacts negatively on bothdomestic traders of exports (and therefore thegains from international trade) as well as theimport-substituting sectors.

However, such an exchange-rate appreciationcan be mitigated if aid is used to increaseproductivity. This can happen when there isspare capacity or significant unemployment inthe economy. If a quick supply-side responsecan be obtained, there will be limited exchange-rate appreciation if the additional aid-induceddemand for non-tradable goods results inincreased employment and production. Inprinciple using aid to fund infrastructure toboost capacity and productivity, in particular,should also mitigate the macroeconomic effectsof aid inflows as it increases production in thelong run.

2. Aid is neither absorbed nor spent

This is a situation whereby aid is used forbuilding up international reserves. This mightbe an appropriate short-run strategy if aidinflows are volatile or if the country’sinternational reserves are low. Under thisscenario, government expenditures are notincreased, hence there is no expansionaryimpact on aggregate demand and no pressureon either the exchange rate or price levels.However this scenario might not be compatible

with donor requirements in terms of povertyreduction, and should therefore be seen as alargely academic alternative unless it is pursuedas a very short-term measure to build sufficientreserves of aid money before absorbing andspending it.

3. Aid is absorbed but not spent

This is a scenario whereby increased aidinflows are used to reduce inflation where thelevels are excessively high. The governmentsaves the domestic currency counterpart of aidwhile the central bank sells the foreignexchange to finance increased imports. Thishas the effect of releasing more resources tothe private sector and results in reducedmonetary growth.

The strategy of absorbing and not spending aidis an appropriate strategy as a tool for reducinghigh inflation, because it allows for adeterioration of the balance of trade to befinanced by aid inflows. However, in the longer-term the no-spend strategy is once againincompatible with donor requirements forpoverty reduction expenditure. Therefore, oncestabilization is achieved, the government isexpected to pursue an ‘absorb and spend’strategy.

4. Aid is spent but not absorbed

This is a situation whereby the fiscal deficit(net of aid) increases as a result of increasedaid inflows. There is no attempt to sell theforeign exchange required to finance additionalnet imports in order for a real transfer ofresources to the partner country to occur. TheGovernment increases expenditures, but keepsaid dollars as reserves in the Central Bank. Theresulting fiscal expansion is the same asincreasing government expenditures in theabsence of aid, since the fiscal stimulus must

31 i.e., items that are readily imported or exported such as consumer durables, as opposed to non-tradables which are items that arenot readily imported or exported such as housing.

32 The theoretical underpinnings of Dutch Disease as an economic concept is used to explain the phenomenon whereby an increasein revenues from natural resources will deindustrialize an economy by raising the exchange rate which makes the manufacturingsector less competitive. More broadly, it can refer to any development that results in a large inflow of foreign currency, includinga sharp surge in natural resource prices and Foreign Direct Investment (FDI). The term was first coined to describe the decline ofthe manufacturing sector in the Netherlands after the discovery of natural gas in the 1960s.

28


be financed by either government borrowingfrom domestic money markets (thus crowdingout the private sector through higher interestrates) or by printing money and thus generatinginflation. Increased aid inflows would thereforeraise money supply and domestic demand alsothereby fuelling inflation. The only positiveeffect of this scenario is that once againinternational reserves would rise but then theywould only contribute to increased liquidity.

This strategy is the least desirable one and notfavoured by donors as it defeats the centralpurpose of aid in effecting a real transfer ofresources to a recipient country, while from amacroeconomic perspective it shows fiscalindiscipline.

The key point about the above scenarios is thathighly aid-dependant economies need to adapt andclosely coordinate their monetary, exchange rateand fiscal policies in the context of aid surges so asnot to penalize their domestic private and exportsectors. The point is made forcefully by Killick andFoster (2007), who believe that both donors andpartner countries must be aware of the policyimplications of aid surges and the possibility ofunintended consequences. Partner countries shouldrecognize the need for close coordination ofmonetary and fiscal policies between central banksand their Ministries of Finance – something thatcannot be taken as given in many developingcountries. In particular, they need to have soundpublic policy with a focus on the development ofthe tradable goods sectors so that aid surges donot disadvantage domestic producers. Withouttimely countervailing policy measures, aid surgesmay well contribute to rising unemployment andpoverty levels. Amongst the various solutions theyrehearse is aid being both absorbed and spent onmore directly productive uses such as infrastructuraldevelopment so as to unblock supply-sideconstraints operating on the tradable goods sector(2007: 167-192).

It is important to note that the relation between aidinflows and their differentiated impact on varioussectors of a recipient economy, depending on howit is managed, has been subject to a large numberof empirical studies. Overall, it would be fair to saythat the jury is still out on a verdict, with some studiesconfirming the existence of Dutch Disease effectsand others refuting it. For example, an IMF surveyof 12 such studies on the relation between aidinflows and exchange-rate appreciation confirmsthe overall inconclusive picture, with six country-specific studies confirming the relationship, whileanother three found no evidence for it. Three wereinconclusive.33

An indication of the difficulties facing analysts inestablishing the robustness of the relationship canbe seen from another study produced from withinthe same institution, namely the IMF, that foundthat aid inflows did in fact have ‘systematic adverseeffects on a country’s competitiveness, as reflectedin a decline in the share of labour-intensive andtradable industries in the manufacturing sector’(Raghuran and Subramanian, 2005: 1). Furthermorethe researchers found evidence that,

‘in countries that receive more aid, labour-intensive and exportable sectors grow slowerrelative to capital-intensive and non-exportable sectors respectively. As a result ofthe reduced competitiveness, employmentgrowth in these sectors is slower, and thesesectors account for a lower share of theeconomy than in countries that do not get asmuch aid.’ (2005: 6)

The significance of these findings on the possiblenegative effects of aid flows on a country’scompetitiveness is obvious. Labour-intensiveindustries are a source of employment generation,especially for low wage economies. By renderingthem uncompetitive, aid inflows could end upconstricting a channel through which surplus labouris absorbed and poverty levels reduced. The

33 Gupta, S., Powell, R., and Yang, Y., ‘Macroeconomic Challenges for Scaling Up Aid to Africa: A Checklist for Practicioners’, IMF,Washington, 2006. Another IMF study based on the experience of five countries that at the time it was conducted had recentlyexperienced aid surges, namely Ghana, Ethiopia, Mozambique, Tanzania and Uganda, found that contrary to the assumptions ofthe theory of Dutch Disease, there was an absence of any Dutch disease effects on exports via appreciation of the exchange rate.On these findings see IMF, ‘The Macroeconomics of Managing Increased Aid Inflows: Experiences of Low-Income Countries andPolicy Implications’, Policy Development and Review Department, Washington DC, 2005.

29

Section 7 – The Macroeconomic Management of Aid Flows

structural transformation of an economy throughthe gradual shift of surplus agricultural labourengaged in subsistence agriculture to higherproductivity and wages in the manufacturing sectoris also jeopardized. Economic history demonstratesthat manufacturing exports constitute an essentialvehicle for ‘take-off’ for developing countries.Participation in international trade helps economiessuch as those of many Low Income Countries toovercome their restricted domestic markets in orderto achieve economies of scale, as well as acting asan important channel for technology transferswhich contribute to productivity gains. The possibleshort-term negative effects of aid flows on thecompetitiveness of the tradable goods sector musttherefore be factored into the design of a longer-term growth strategy, as well as how fiscal,monetary and exchange-rate policy should react.While any aid surge may be temporary, its longer-term effects may be more lasting and damagingfor domestic producers of tradable goods.

As argued earlier, partner country governmentsshould therefore take a longer-term perspective andensure that a significant portion of aid is used tohelp raise the productivity and lower the costs ofagents in the tradable goods sector. There may bea need to offset the disincentives that flow from anappreciating exchange rate by tackling non-priceconstraints on both exporters and import-substituters. Investments of aid funds in infra-structure and other services may offer the highestlong-term returns to aid spending.

Given the recent dollarization of the Zimbabweaneconomy, the dangers of Dutch Disease forZimbabwe through the transmission channel ofexchange-rate appreciation in the context of largeaid inflows may remain purely of academic interestfor the meantime.34 But in a post-dollarisationscenario, characterized by continued large inflowsof aid, the need for careful management ofmonetary, fiscal and exchange-rate policy willremain strong.

34 One of Zimbabwe’s major challenges in the short-term will be to reconcile the need to maintain levels of employment, throughwhich poverty reduction is mediated, with the imperative of improving the country’s competitiveness. Under dollarization onestandard means of improving competitiveness, namely devaluation, is no longer an option. In the absence of a domestic currency,increased aid cannot give rise to Dutch Disease through the exchange rate, but similar effects may come to be felt through acountry’s price level and cost structure. If significant volumes of aid are injected into what is already a skills-scarce and input-constrained economy such as Zimbabwe’s, aggregate demand and domestic price levels may rise and make businesses, especiallyexporters, even more uncompetitive. To forestall job layoffs, increased aid needs to be accompanied by policy measures thatincrease competitiveness.

30

Given the current volumes of Official DevelopmentAssistance (ODA) flowing from donors to partnercountries, it might surprise the reader to know thatthere is an ongoing and heated debate amongst aideconomists revolving around what might best becharacterized as an ambiguous association betweenaid and demonstrable development outcomes. Thevery effectiveness of ODA in helping partnercountries to achieve economic growth and broaderdevelopment targets has been subjected to intensescrutiny. Available evidence has been used todetermine the impact of aid in relation to intendedoutcomes at both sectoral and aggregate levels.

Early views on how aid was supposed to relate togrowth in developing countries were based on theassumption that the main constraint to growth inthese economies was their poor rates of capitalaccumulation. Helping these economies toovercome trade gaps and their low domestic savingsrates was therefore the role of aid. This would inturn help them to undertake a ‘big push’.35 Thelarger the volumes of aid, the sooner these ‘two-gaps’ could be closed and developing economieswould reach ‘take-off’, and thereafter embark ona path of robust and sustained growth.

It is noteworthy that this ‘big push’ thesis on aidreappears in the 21st century through the advocacywork of Jeffrey Sachs (2005). He argues that amajor international effort is required in terms ofincreasing aid volumes if the Millennium Develop-ment Goals (MDGs) are to be achieved, preciselybecause the accumulated physical, institutional andhuman capital deficits in low-income countries areso large that these countries find themselves in apoverty trap, and therefore unable to embark on

sustainable growth and poverty reduction (Sachs,2005). The key underlying assumption is, onceagain, that of the primacy of additional capital inunlocking better growth performance.

As early as the 1970’s, however, the aboveassumption was being queried through a numberof so-called ‘first generation’ studies on aideffectiveness.36 These focused specifically on theassociation between aid and national savings, thisfocus on savings being an indirect way of evaluatingthe aid-growth relationship. Many reported zero oreven negative correlations between aid receipts andgrowth. A common conclusion was that aidresources were being spent on consumption (largelyimports) rather than investment. In addition, aid wasalso seen to be acting as a disincentive to recipientgovernments in terms of their revenue collectionefforts, the expansion of their tax base, and thesound targeting of government expenditures ongrowth-enhancing investments.

In these first generation studies, and in line withthe two-gap model of their predecessors, aid wastreated as an exogenous net increment to the capitalstock of beneficiary countries. In other words, eachdollar of aid would supposedly result in an increaseof one dollar in total savings and thereforeinvestment. Aid was therefore treated as exactlyadditive to domestic savings, with a causal chainrunning from aid to savings to investment and thengrowth. But the possibility of aid being used forends other than investment was not controlled. Inother words, the problem of the fungibility of aidwhereby all types of aid can be used for purposesother than investment broadly defined was notcatered for.37

Section 8

The Aid-Growth Debate

35 The intellectual foundations of this approach were laid down by W.W. Rostow in his ‘The Stages of Economic Growth’, CambridgeUniversity Press, 1960, as well as Chenery and Strout. For the latter, see Chenery, Hollis. B. and Strout, Alan, ‘Foreign Assistanceand Economic Development’. American Economic Review, Vol.56, No.4, 1966, pp.679-733.

36 For a representative sample of such work see Griffin, K.B and Enos, T.L, ‘Foreign Assistance: Objectives and Consequences’,Economic Development and Change, Vol.18, No.3, 1970, pp.313-327.

37 The point is made succinctly by one aid analyst who noted that ‘…by reallocating domestic resources in the budget, the recipientgovernment can re-direct aid in accordance with its own preferences. For instance, aid intended by donors for investments inhuman capital (health and education) can be diverted into higher salaries or other perks for ministers and politically favouredgroups by reducing domestic budget allocations to the health and education sectors. The money thus freed can be used to financewhatever the government prefers.’ Peter Svedberg, ‘Comment on Finn Tarp: Aid and economic growth: An alternative interpretationof the evidence’. Swedish Economic Policy Review, 13, 2006, p.17.

31

Section 8 – The Aid-growth Debate

At the same time as these negative conclusionsfrom the earlier generation of studies were beingreached, there were others that were more sanguinein their conclusions. Some argued that while it maybe true that aid did not increase recipient countrysavings and investments on a strict one for onebasis, there was nevertheless evidence that aid didhave a net positive effect in terms of increasingtotal savings, i.e., that aid did lead to an increase intotal savings, although not by as much as the aidflow (Newlyn, 1973: 867-869).

One landmark study characterizes the most recentgeneration of studies, namely the Burnside andDollar work of 2000 (2000: 947-968). Givenadvances in computing and econometrics, Burnsideand Dollar were able to work with panel data thatcontained a much larger sample of countries overa longer timeframe than previous efforts. In a noveldeparture from previous efforts, they factored ineconomic policy and institutional environmentvariables in their regressions. These economicpolicies were grouped into a single policy indexconsisting of measures on inflation, trade opennessand budget deficits. Their hypothesis was that theimpact of aid on growth is likely to be greater whenthere are fewer policy distortions affecting theincentives of economic agents. This assumption ispredicated on the existence of transmissionchannels whereby an improved policy environmentwould result either in the increased productivity ofcapital, and therefore of invested aid, and/orbecause a larger percentage of aid received wasactually invested rather than consumed or diverted.The marginal productivity of capital, and thereforeof invested aid, increases as distortions in theincentive system decrease.

Burnside and Dollar were able to include 56countries in their sample over an extendedtimeframe covering the period 1970–1993.Analysing the aid-policy environment-growthnexus, they concluded that aid does contributepositively to growth, but only in good policyenvironments defined in terms of sound fiscal,monetary and trade policies. The impact of thesefindings should not be underestimated in terms oftheir influence on how international developmentagencies sought to interact with partner countries.Having originally circulated as a World Bankdiscussion paper in the late 1990’s, the Burnsideand Dollar findings came to buttress a landmark

World Bank Report on aid that was published in1998 (World Bank, 1998). Selectivity was the orderof the day in terms of ensuring that aid was to bemade conditional on sound policies on the part ofbeneficiary countries. Aid would therefore promoteeconomic growth in those countries that possessedsuch policy frameworks.

Just when it seemed that the issue had been put torest, however, critiques of the Burnside and Dollarfindings soon appeared. One group of economiststested the robustness of the model used by Burnsideand Dollar by expanding the number of samplecountries and extending the panel data from 1993to 1997, and showed that the positive relationshipbetween aid, the policy environment and growthdetected by Burnside and Dollar disappeared whenthe new data was introduced (Easterly, Levin andRoodman, 2004: 774-780).

A similar fate awaited another research initiative,the results of which were published in 2001 andwhich also seemed to offer advocates of aid strongevidence on the positive links between aid andgrowth. The Hansen and Tarp study was evenmore ambitious in terms of the representativenessof its sample, covering over 100 cross-countryregressions on the impact of aid on growth fromthe late 1960’s to the late 1990s. What they foundwas that while the link between aid and growthwas a robust one, and that aid did increaseaggregate savings and investment, the relationshipwas not on a one-to-one basis. But their findingsalso contradicted those of Burnside and Dollar tothe extent that they did not find a robust positiveinteraction between sound policies, aid and growth.Aid was shown to have a positive effect on growthrates wherever and whenever it was driven bycapital accumulation, i.e., aid worked, even incountries hampered by an unfavourable policyenvironment as defined by Burnside and Dollar(Hansen and Tarp, 2001: 103-128).

The layperson would be well entitled to ask whythe results are so inconclusive, especially given thefact that at last count over 130 studies had beenconducted around the world seeking to bring someclarity to the issue. The title of a relatively recentreview of the literature, namely ‘Macro AidEffectiveness: A Guide for the Perplexed’ isapposite (Roodman, 2007). For partner countries,this bafflement should be all the greater given the

32


often highly prescriptive content of the aid policesof donors, as well as global commitments tosignificantly increase aid flows.

To the extent that light can be thrown on the reasonsfor the inconclusiveness of the findings, one shouldseparate methodological issues from the possibilitythat the very definition of what constitutes aid mayimpact on results. In regards to the first set ofconsiderations, it can be said that multivariableregressions of the sort used in the most recentstudies are always sensitive to model specifications,the time period covered, the quality of panel dataused and the choice of control variables. Given thecomplex interactions between the multitude ofdeterminants of economic performance, theconnection between aid and growth cannot be asimple linear one, thus making it extremely difficultto isolate the effects of aid in any economic systemsince:

‘as the consequences of an aid project rippleout and diffuse, they become harder to detect.It is not hard to tell if a road is paved. But itmay be impossible to discern whether all aidraises total output per capita from a nationaleconomy in the long run.’ (Roodman, 2007: 3)

To this might be added the problem of establishinga counter-factual, i.e., what might have happenedin the absence of aid? These complexities aresuccinctly expressed by one participant in thedebate who noted that ‘simply, if one tortures thenumbers for a sufficiently long time, they willconfess to anything’ (Svedberg, 2006: 13).

However, a recent and more profitable line ofinvestigation seems to be that based on thedisaggregation of aid. In contrast to earlier work,Clements, Radelet and Bhavani (2004) start outfrom the premise that much of the earlier work onthe relationship between aid and growth has beenflawed due to the simple failure to divide aid intotheir distinct purposes as declared ex ante by donors.Economic growth is not, and never has been, thesole objective of aggregate foreign aid.

Humanitarian assistance for example is not directlytargeted at building productive capacity but ratherhas a ‘saving lives’ metric, and aid directed toeducation and health programmes are unlikely tohave a measurable impact on growth rates in theshort-term. To this list might be added aid in supportof the global ‘good governance’ agenda, which mayhave growth only as a secondary and longer-termobjective.

Having separated these components of aggregateaid, they arrive at a sub-set of aid which theydesignate as ‘short-impact aid’, i.e., aid that couldplausibly stimulate growth over the short-termwhich they define as a 4-year period, such as aiddirected to investments in productive sectors suchas agriculture and industry, balance of paymentssupport, and investments in infrastructure,communications and energy generation. When thisapproach is adopted, and the regressions are carriedout, the results are encouraging, with a very robustpositive relationship between ‘short-impact’ aid andgrowth. More specifically they found high returnson such aid, with US$1 in short-impact aid leadingto a gross economy-wide cumulative increase inaggregate output of US$1.64 (Clements, Radeletand Bhavani, 2004: 34).38

As the authors themselves caution, however, giventhat their study sets out to only quantify the impactof one sub-set of aid, the results should not beinterpreted as meaning that the best use of aid is todirect all ODA resources to short-term impact aidof the ‘directly productive investment’ kind, letalone that social-sector investments or humanitarianaid may not have longer-term growth effects. Butthe results do, nevertheless, suggest that partnercountries should be aware of possible trade-offsgiven a finite aid resource envelope. These trade-offs include decisions regarding how to directresources in order to strike the right balancebetween short-term growth imperatives and theneed to enhance longer-term growth potential,which is partially dependant on the building of humancapital over the longer-term.

38 The figure is arrived at using a Keynesian multiplier. As the authors note, this is higher by a factor of 2-3 times when comparedto previous studies that used non-disaggregated aid. It is noteworthy that the robustness of the sample can be gauged by the fact thatthe researchers classified a total of 275,590 donor-partner country transfers contained in the OECD’s database for the period1973–2001, dividing them into short impact (4 year epochs), long impact and humanitarian transactions.

33

Section 8 – The Aid-growth Debate

In the case of Zimbabwe, and given the situation itcurrently faces with multiple demands on anextremely limited resource envelope, the need tomeet short, medium and longer-term objectives isall the greater. Given finite domestic and externalresources, key decisions have to be made in termsof the amounts to be channelled to developmentand humanitarian assistance. While the imperatives

of humanitarian aid may, at first sight, seem to easilytrump the longer-term objectives of developmentassistance, there is evidence to suggest that thearguments in favour of solely focusing on immediatehumanitarian needs are not clear cut, and thatdevelopment assistance plays an essential role inmitigating vulnerability to external shocks.39

39 A study conducted in 1998 specifically used Zimbabwean household panel data from 1994–1997 and therefore catered for the1994/95 drought. The researchers sought to throw light on the opportunity cost in terms of foregone poverty reduction of a shiftfrom development to humanitarian assistance. The authors found that under a scenario in which a certain amount of humanitarianassistance was redirected to development assistance (increasing the agricultural capital stock of households, such as throughimproved extension services and the provision of inputs and equipment), the incidence of food poverty was significantly reducedeven in drought years. See Trudy Owens and John Hoddinott, ‘Investing in development or investing in relief: quantifying thepoverty trade-offs using Zimbabwe household panel data.’ Centre for the Study of African Economies, University of Oxford,Working Paper Series 98, 1999.

34

9.1 THE NATURE OF AIDUNCERTAINTIES

In addition to having to coordinate fiscal, monetaryand exchange-rate policies in order to counter thepotential negative macro- and micro-economicconsequences of large aid inflows, partnercountries also have to deal with the problems ofaid predictability and volatility. Widespreadrecognition of this problem helps to explain thecommitment undertaken by donors in the ParisDeclaration to ‘… provide more predictable andmulti-year commitments on aid flows to committedpartner countries (Principle 26)’. This is translatedinto a specific progress indicator to measure donorperformance on this issue. The measure is thepercentage of donor aid disbursed according toagreed schedules in annual or multi-annualframeworks. The target is to halve the proportionof aid not disbursed within the fiscal year for whichit was scheduled by 2010 (indicator number 7).

The problem can be defined by looking at both sidesof the donor-partner country equation. According tothe OECD-DACs guidelines on harmonization, aidis predictable if ‘partner countries can be confidentabout the amount and timing of aid disbursements’.In other words, predictability is defined by howreliable commitments made by donors are. It is‘volatile when fluctuations in aid flows are largerelative to the volume involved’ (OECD, 2005: 22).Volatility may also be defined in terms of fluctuationsrelative to domestic revenue. A sub-set of concernsrevolving around volatility has to do with the impactof conditions attached to aid and their impact onvolatility. Recognition of this relationship betweenaid volatility and conditionality, though not the onlyconsideration, underpins principle 16 of the ParisDeclaration which calls upon donors to ‘Link fundingto a single framework of conditions and/or amanageable set of indicators derived from thenational development strategy’.

An additional problem for partner countries is thataid has been shown to be pro-cyclical, i.e., there is aseemingly strong correlation between aid flows and

movements in partner countries’ GDPs and theirrevenue base. The evidence to date indicates that,contrary to the seemingly logical assumption thatdonors would be most willing to increase and speedup the disbursement of aid to partner countries thatneed it most urgently, work on time series dataindicates that aid is not performing this compensatory,counter-cyclical function. When partner countrieshave suffered external shocks over which they hadno control (e.g., a rapid deterioration in their termsof trade or drought which has negatively impactedon their agricultural exports) and which then leadsto sharp contractions in GDP, counter-cyclical aidhas not been forthcoming.

The challenges on this front are manifold for partnercountries. Given the requirement to developoverarching national and sectoral strategies, thesecan only sensibly be designed on the basis ofrealistic resource envelopes in which ODA oftenfigures as a large component both in terms of capitaland recurrent expenditure. However, much aid isonly committed in the short-term and on the basisof conditionalities by donors, which can makedesigning long-term spending programmes anextremely precarious exercise. To this must beadded the common problem of donor commitmentsnot translating fully into disbursements. When asignificant part of public resources is unpredictableit makes the already difficult exercise of deployingresources effectively even more challenging. Theconsequence is that ‘national authorities responsiblefor executing macroeconomic policy thus have todo so in ignorance of actual and forthcomingmovements in a major source of balance ofpayments and monetary changes’ (Killick andFoster, 2007: 172).

9.2 THE MAGNITUDE OF THEPROBLEM AND ITSCONSEQUENCES

Surveys of sample countries show large year-by-year swings in ODA flows. The magnitude of theproblem can be gauged from a survey conducted

Section 9

Enhancing Aid Predictability

35

Section 9 – Enhancing Aid Predictability

by the Strategic Partnership for Africa (SPA) usinga sample of donor and partner governmentbehaviour in 15 African countries. The survey foundthat on average, 81 percent of 2003 commitmentswere disbursed during 2003, 10 percent weredisbursed the following year, and 9 percent ofcommitments were never translated intodisbursements (SPA, 2005).

A more recent IMF survey, covering 76 countriesover the period 1975–2003, makes even moresober reading. It demonstrates that notwithstandingcertain parallel initiatives anchored in the roll outof PRSPs – Medium-Term Expenditure Frame-works, improved public financial managementsystems, strengthened accounting and reportingsystems, and donor coordination mechanisms – aidremained both unpredictable and volatile. Indeedtheir findings indicate that contrary to expectations‘aid has become more volatile in the late 1990’sand 2000’s –the post-PRSP period – than in themid 1990s –the pre-PRSP period…’ (Bulir andHamman, 2006: 11). Using as a proxy long-termODA loans from OECD countries for the period2000-2003, they demonstrate that disbursementsfell short of commitments by about one-third.An even more startling finding is that thisunpredictability is negatively correlated with GDPper capita, with countries at the lower end of theincome scale receiving on average about one halfof commitments (Bulir and Hamman, 2006: 16).

The above results help to validate the earlier workof the same IMF researchers. For countries underIMF-supported programmes, IMF staff have tocarry out projections for both project andprogramme aid. The authors found that thediscrepancies between projections prepared priorto the start of the programme (derived from pastdisbursement patterns as well as donor commit-ments) and the actual disbursement as reportedby the partner country authorities, ranged from 8percent in the case of project aid, while programmeaid was overestimated by an average of more than30 percent (Bulir and Hamman, 2001). Theconclusion must be that for many partner countries,donor commitments are poor predictors of actualdisbursements.

The consequences are particularly pernicious ineconomies that have come to depend on budgetsupport disbursements. It is incongruous that in a

situation in which partner countries are enjoinedto move to comprehensive national developmentstrategies such as PRSPs in order to strengthennational ownership and facilitate donor alignmentand harmonization, budget support predictabilityis more vulnerable to fluctuations than stand-aloneproject delivery. Such volatility cannot butcompromise the quality of budget planningprocesses ‘by rendering original allocationsobsolete and forcing expenditure adjustmentsduring budget execution’ (Celasun and Walliser,2005: 2). The unpredictability and volatility of aidflows compound the vulnerability of many partnercountries, exposed as they already are to externalshocks of various kinds such as a suddendeterioration in their terms of trade, unpredictableprivate capital flows and exchange-ratefluctuations.

The problem of the pro-cyclicality of aid flowscompounds these difficulties. In a seminal studybased on time series data for 76 countries, thefindings showed,

‘disbursements to be procyclical on averageand, worse, we found strong evidence that aidhas failed to play any meaningful role inassisting countries to cope with large negativeincome shocks…of all countries hit bynegative GDP shocks of 5 percent, only onebenefited from a concomitant increase in aid.’(Bulir and Hamman, 2006: 20-21)

Explanations for this phenomenon vary. On the onehand the argument has been advanced that sincethe business cycles of both donors and partnercountries are linked, downturns in donor economiesresult in shrinking aid budgets, thus reducing thepotential for aid flows to play a counter-cyclicalrole in the economic downturn of partner countries.Another attempt to explain the phenomenon isbased on the problem of imperfect information abouta partner country’s policy efforts and its relationshipto conditionality. In light of the difficulties inassessing ‘effort’ on the part of partner countries,donors tend to use easily observable, though in alllikelihood imperfect, macroeconomic indicators, toassess performance (results) and use this as aproxy for effort. Given the amorphous nature ofthe concept of ‘effort’ in terms of the imple-mentation of sound policy, donors may withholdcommitments through invoking conditionalities,

36


ignoring the possibility that factors exogenous to apartner country’s commitment to agreed policiesmay explain weak performance. The net result isa reinforcement of the pro-cyclical nature of aidflows, with good performers receiving extra aid andpoor performers being penalized for reasons oftenoutside their control.

How do partner countries cope with the unpre-dictability, volatility and pro-cyclicality of aid flows?In other words how do they deal with ‘aiduncertainties’? Amongst the questions they areforced to address is the degree of dependence oftheir budgets on external assistance. The decisionof whether to design and scale up nationalprogrammes, such as employing more public sectorworkers in health and education services, may bebased on overly optimistic projections of future aiddisbursements. Given the restricted room formanoeuvre (fiscal space) and structural rigiditiesof many low-income economies, as well as liquidityconstraints that hinder their ability to adoptcompensatory measures, coping with possible aidshortfalls has to be taken seriously in terms of theirplanning and budgeting processes.

The aid uncertainties listed above are anothercontributing factor to the continued attractivenessof project-based aid which is less subject tounpredictability and volatility since it is oftencommitted for the lifespan of the project, with fewerconditionalities attached. Any shortfall on a projectis also likely to be less disruptive to national planningand budgeting than unexpected shortfalls in budgetsupport inflows. A rational response on the part ofpartner countries to these uncertainties is thereforeto spread the risks of unpredictability, volatility andpro-cyclicality of aid flows by negotiating withdonors an appropriate mix of stand-alone projects,technical assistance, sector-wide programmeswhich may be either on or off their national budgets,and direct budget support.

A more damaging consequence of these aiduncertainties is that partner countries have alsocome to ‘discount’ aid. Given that they have beenencouraged to move from uncoordinated project

support aid delivery modalities to more coordinatedprogramme support, the benefits of lowertransaction costs associated with the latter may beoffset by increased risks linked to budget supportaid uncertainties. As a result, based on pastexperience and given the need to budget prudently,some countries have come to discount pledges ofassistance.40 While this may have the benefit ofintroducing greater realism to the financing plansof government, there are a number of less welcomeconsequences. One is that discounting can lead toa situation whereby, notwithstanding the objectiveneeds of a partner country, donors find few fundinggaps to which their funds could be applied. This inturn may trigger a downward spiral of decreasingcommitments and disbursements followed byfurther aid discounting.

Given the structural rigidities of their economies,discounting acts as a brake on the partner countries’ability to react rapidly and make up the shortfallthrough an increased export effort. The ‘thinness’of their domestic capital markets, and their ownpoor international credit ratings, means that low-income and highly aid dependant economies cannotoff-set the non-disbursement of aid by borrowingeither domestically or internationally. Excessivelysharp fiscal adjustments, or an inflationaryexpansion of the money supply, are two possiblecounter-measures that may end up being adopted,the latter with possibly negative consequences formacroeconomic stability.

9.3 MEASURES TO REDUCE AIDUNCERTAINTIES

From the vantage point of donors, therefore, theirresponsibilities require them to enhance aidpredictability, to deliver on their commitments andto programme aid over multi-year frameworks.This would in turn require donor agencies to protectODA as a slice of their own national budgets, andto ensure that such information is transmitted topartner countries so as to allow the latter toincorporate these projections of external resourceinflows into their own medium-term plans.

40 The Ugandan experience is instructive. Based on years of experience with donors, the Ugandan government in the course of itsplanning and budgeting introduced a discount factor for budget support inflows based on the average shortfall between commitmentsand disbursements over the previous five years.

37

Section 9 – Enhancing Aid Predictability

Corporate constraints militate, however, against anyrapid solution on the donor side since:

‘…aid volatility reflects deeply rootedproblems with the way donor budgets areapproved and administered… donor develop-ment agencies that make aid commitmentsare different from those that approve funding(parliament) and disburse aid (ministries offinance). While the magnitude of thisdisconnect differs form country to country,it seems widespread.’ (Bulir and Hamman,2006: 4)

The specific issue of the impact of conditionalityon aid volatility needs to be addressed by bothdonors and partner countries. On the side ofpartner countries, there is a need to pursue policiespreviously agreed with donors in a consistentmanner, thereby reducing the dangers of shortfalls,or in extreme cases, cut-offs in aid flows. On theside of donors, conditionality needs to bestreamlined and designed in such a way thatdisbursement is conditional on a restricted numberof clear and specific performance criteria,mutually agreed to by both donors and partnercountry governments. Disbursement shouldpreferably be over periods longer than one yearin order to allow for the measurable results ofpolicy reform programmes and poverty reductionefforts to make themselves felt. It would also belogical for targets and indicators on whichconditionality is based to be reflective of variablesover which partner governments have a substantialdegree of control, and for which they can then becalled to account for.

It should be noted that there has been significantprogress by the BWIs in terms of streamliningtheir policy conditionalities. In 2002, for example,the IMF issued a set of new Guidelines onConditionality to reduce the negative effects oftheir conditionality. The guidelines tried to ensurethat conditions were negotiated through a processled by partner governments, and that they were

focused on the overall objectives of a PRGF-supported, but nationally designed, programme.41

The same process was undertaken by the WorldBank following an extensive review of its ownpractices (2005), which led in 2006 to the adoptionof a set of ‘Good Practice Principles’, wherebyconditions were to be agreed upfront withgovernments, adapted to country circumstances,selective in terms of the number of actions requiredas preconditions for disbursement of World Bankfunds, and based on regular and transparentprogress reviews.

Supplementary measures by donors should includeensuring that performance reviews, upon whichdisbursements depend, are scheduled to fit into theplanning and budgetary cycles of partner countriesto ensure aid flows are as predictable as possiblefor national authorities, and to help smooth theirbudget implementation. In recognition of thenumerous challenges facing partner countries whichmay result in performance slippage, the OECD hasproposed that mechanisms be set up that allow forpartial disbursements in the case of partial fulfilmentof conditions: ‘an intermediate option betweenwithholding all funds and releasing them’ (2005: 34).Historically, conditionality has been of the ‘all ornothing’ variety. Given the numerous domestic andexternal constraints facing partner countries, thereare compelling reasons for steering a middle course,with the amount of aid provided reflecting the extentto which targets are met. This diminishes the riskof partner countries having their support suddenlysuspended with potentially very serious fiscal anddevelopmental consequences.

A notable example of such a mechanism is foundin the European Commission’s budget supportprogramme. This is based on ‘fixed tranches’ and‘variable tranches’. Fixed tranches are linked tothe fulfilment of basic macroeconomic con-ditionalities such as macroeconomic stability asevaluated by an IMF review, and improvements innational PFM. Variable tranches are additional

41 IMF, ‘Guidelines on Conditionality’, Legal and Policy Development and Review Department, Washington DC, 2002. It isnoteworthy that an IMF Staff Statement in 2006 on the guidelines clarified that ‘Fund staff will endeavor to reach understandingswith the authorities on a mutually acceptable means of achieving program goals, while paying due regard to the domestic social andpolitical objectives, the economic priorities, and the circumstances of the member including… the member’s capacity to implementreforms in the necessary time frame’. See IMF, ‘Statement of the IMF Staff – Principles Underlying the Guidelines onConditionalities’, Washington DC, 2006, pp.1-2.

38


resources disbursed in accordance with thefulfilment of specific, mutually agreed targets inthe framework of PRSPs, normally centred onimprovements in service delivery. The idea is tocombine a reasonable degree of predictability forpartner countries with performance-basedincentives that answer the needs of donors ascustodians of public funds to show that they areexercising ‘due diligence’, as well as monitoringand assessing the results of their developmentassistance programmes. The mechanism thusallows for a graduated response on the part ofthe EC to shortcomings in a partner country’soverall performance.42

Additional means of helping to overcome ‘aiduncertainties’ are to ensure that conditionalities arenot excessively detailed, that they are negotiatedwith partner countries and rooted in a soundunderstanding of the ‘carrying capacity’ of a partnercountry to implement a given set of reforms, andthat reviews of progress upon which disbursementsare based reflect medium-term considerations.Donors extending their programming horizons byproviding their partner countries with multi-yearindicative resource envelopes will also assist partnercountries with their own medium-term financialplanning. Arrangements such as the IMF’s PovertyReduction and Growth Facility (PRGF), whichcovers a three-year period, and the CountryAssistance Strategies (CAS) of the World Bank,which run for four years, are cases in point.

Greater understanding on the part of donors to the‘defensive’ posture adopted by partner countriesin the face of the aid uncertainties is also essential.In addition to recognizing the reasons behind aiddiscounting in national planning and budgetaryprocesses in partner countries, there is a case tobe made for greater donor flexibility in terms ofagreeing to donor funds being used to build up

buffers of international reserves, notwithstandingtheir preference for partner countries to fully absorband spend aid inflows (see above). In high aid years,there is a compelling logic to saving some aid (i.e.,neither absorbing nor spending) so as to cope withfuture aid shortfalls.

This idea has been fleshed out by Eifert and Gelb,who propose a mechanism based on a reservetranche of between 50 and 100 percent aid fundedspending which translates into between 2-4 monthsof import cover in a representative aid-dependantlow-income country. After subjecting theirmechanism to a simulation exercise, they find thatit was:

‘effective in smoothing expenditure in mostperiods under a range of levels of aidinstability…[and that] while our simulatedstabilization fund does in some cases go‘bankrupt’, this usually requires 3-5 years oflarge negative shocks to aid flows… [which]would allow countries and donors… plenty oflead time…. to organize an emergencyresponse.’ 43 (Eifert and Gelb, 2005: 10)

An alternative to using donor funds as a buffer toaid flow volatility would be to use revenue generatedby a Sovereign Wealth Funds (SWFs) – but onlyunder the strictest conditions. As the SWFrepresents the capital accumulation of a depletingnational natural asset (mainly minerals), its fundsare ‘ring-fenced’ and earmarked for reinvestmentin specific productive assets and intangible capital.As such, SWF resources could only be applied tocertain expenditures mandated by its enablinglegislation. The judicious use of SWF funds could,however, reduce fiduciary risks and enable donorsto be more flexible in allowing a portion of aidfunding to be disbursed as a temporary reserveagainst volatility.

42 Another suggestion, advanced by a team of World Bank analysts aimed at improving the predictability and volatility of budgetsupport, is to differentiate between changes in performance and levels of performance. They advance the argument thatcore budget support flows should remain stable over the medium-term, and that sharp changes to such flows should only takeplace when ‘catastrophic’ declines in the quality of partner countries PFM, budget discipline or macroeconomic managementoccur. On this proposal see Been Eifert and Alan Gelb, ‘Improving the Dynamics of Aid – Towards More Predictable BudgetSupport’, World Bank Policy Research Working Paper 3732, Washington DC, 2005.

43 The authors note, however, that one of the difficulties in implementing the idea would be that donors, while possibly accepting theneed for such a fund as an insurance scheme for partner countries to cater for shortfalls in disbursements that are not performance-related, will also require a mutually-agreed and clear performance framework without which they will come to see such a fund asa means for partner countries to access funds when the funding gaps are the result of performance shortfalls that are not due toexogenous aid volatility.

39

The phenomenon of aid dependency has beenextensively treated in the literature, with somearguing that aid, especially large volumes of aid,undermine endogenously generated and sustaineddevelopment efforts by partner countries. Oneparticular target of criticism is technical assistance.The effectiveness of this particular component ofODA has been questioned largely on the groundsthat the current aid system provides incentives forits continuation, rather than building in obsolescencefor technical assistance based on the transfer ofskills and the building of local capacity.

Another concern arising from high levels of aiddependence is that it can constitute an obstacleto a partner state transforming itself into adevelopmental state, i.e., a state that shows ‘a clearcommitment to a national development agenda, thathas solid capacity and reach, and that seeks toprovide growth as well as poverty reduction andthe provision of public services’ (Fritz and Menocal,2006: 4)

Ghani attributes a set of ten functions, which anystate must be able to perform in order to beconsidered effective. It must: (1) possess amonopoly over the means of coercion, (2) enjoyadministrative control over the national territory, (3)exercise sound management of public finances, (4)invest in human capital, (5) foster citizenship rightsand duties, (6) provide sound infrastructure, (7)engage in market formation, (8) exercise soundmanagement of state assets, (9) engage in effectivepublic borrowing, and (10) uphold the rule of law(Ghani, Lockhart and Callaghan, 2005).

The prominence of sound economic managementcriteria in the above set of functions raises thequestion of the effect of aid flows on stateeffectiveness, and more specifically whetherchanges from project aid to budget supportmechanisms through PRSPs has affected incentivemechanisms in terms of state performance. Somelight has been thrown, in particular, on the questionof the possible disincentive effects in terms of theexpansion of the tax base of developing economies.Moss and Subramanian (2005) have argued thataid lowers the incentives for governments tobroaden their tax base and reduce their dependencyon aid. They argue that aid dependence locks bothdonors and partner countries into a permanentsituation of high aid and low institutional capacity,and show that aid is negatively associated with taxefforts.44

They also note that the composition of aid exertsdifferentiated pressures on revenue efforts. Theneed to repay loans leads to increased domesticrevenue generation, while grants have the oppositeeffect since they are free resources that cansubstitute for domestic revenues, and hence aremore likely to reduce domestic efforts to collectmore revenue.

Other studies, however, do not detect a negativefiscal effect of aid where governments implementsound macroeconomic policies and make wisedecisions in terms of the use of aid in the fiscaldomain. McGillivray and Morrisey (2001) show amixed picture from a survey of various fiscalresponse models. In some cases they show that

Section 10

Aid Dependency as an Impediment to the Rise ofDevelopmental States

44 While foreign aid may be necessary at the early stages of growth (or in the case of Zimbabwe, recovery) in order to finance the‘two gaps’, i.e., the savings gap and the trade (foreign exchange) gap, over the longer-term and given finite aid, there is also a needfor longer-term strategies aimed at the efficient reallocation and creation of factors of production, a process in which a state’staxation and fiscal policy plays a key role. The point is hardly a novel one, and has a long and eminent history. As long ago as 1963Nicholas Kaldor observed that incentives for collecting taxes, i.e., one of the basic functions of the state, might be undermined byan over-reliance on external resources. See his ‘Will Underdeveloped Countries Learn to Tax?’ Foreign Affairs, 41, WashingtonDC, 1963, pp.410-419.

40


aid may discourage tax efforts and encourageincrease public borrowing, but that in other casesaid may increase tax efforts, encourage increasedspending on productive investments and supportimproved fiscal management, thereby reducing overthe longer-term, borrowing requirements.

To these concerns might be added anotherdistortion that impacts on partner countries thatare highly dependant on aid, namely the negativeconsequences for domestic accountabilitymechanisms. The argument is a simple one: largeaid inflows trigger a shift in the accountabilitynexus from the primary contract between stateand citizen, to one between governments and

external agencies. The hypothesis, borne out by anumber of cross national studies, is that:

‘large sustained aid flows fundamentally alterthe relationship between government elites andlocal citizens… If donors are providing themajority of public finance and governmentsare primarily accountable to those externalagencies, then it may simply not be possibleto also expect a credible social contract todevelop between the state and its citizens…aid may undercut the very principles the aidindustry intends to promote: ownership,accountability and participation.’ (Moss,Petterson and van de Walle, 2006: 14)

41

11.1 EFFORTS TO ADDRESS THEPOWER IMBALANCE

National aid policies and management systems goto the heart of the power imbalance between donorsand partner countries. Notwithstanding the notionof partnerships embedded in frameworks such asthe Paris Declaration and the MillenniumDevelopment Goals, it is hardly surprising that thereis a clear asymmetry of power in aid relationships,both in terms of who provides the resources anddifferences in national capacity. It is worth recallingthat it is bilateral agencies and donors that determinethe quantity of ODA available globally, how it isdisbursed and to whom. Conditionalities are appliedby these bilateral donors, and the IFIs they largelycontrol. Partner countries have limited influenceover donor policies, except ultimately the right torefuse aid.

Yet the asymmetrical relationship is gradually beingcorrected. Principles such as that of mutualaccountability, part and parcel of the more rules-based approach underpinning the Paris Principles,have played a useful role in levelling the field andare a means through which partner countries canhold their donors to account. Mechanisms such asthe OECD-DAC peer reviews45 and countryspecific Performance Assessments have played auseful ‘name and shame’ function where specificdonors have failed to live up to the Paris Principles.

In situations of high dependence on aid – and wherethe partner country may not possess a strategicnatural resource which could provide it someleverage in negotiations with donors, or does notoccupy a strategically important geographicalposition – the need to develop a sound aid policymanagement framework and accompanyingsystems is all the greater. The list of weaknesses

and uncertainties detailed in this working paper interms of current international aid architecture, itssystems, processes and instruments, means that thedevelopment of effective aid management tools bypartner countries increases their chances ofovercoming these problems (or at least minimizingtheir deleterious effects). A sound aid policy wouldhelp partner countries ensure that aid flows meetnational priorities, that donors are more harmonized,and that they themselves are able to at least partiallydetermine the pace of reforms and exerciseeffective leadership in the aid relationship.

International experience shows that there are anumber of enabling pre-conditions for partnercountries intent on exercising leadership of theirown national development agendas and in theirrelations with donors. Firstly, a stable andpredictable economic environment greatlyfacilitates forward planning by any nationalgovernment, and in the case of highly aid dependanteconomies this becomes all the more importantsince it allows them to make sounder projectionsof their needs in terms of external resourceenvelopes.

Secondly, open and frank engagement with donorsis key to building the necessary mutual trust on thebasis of which donors will be prepared to commitresources to support a partner country’sdevelopment efforts. Thirdly, a partner country’sclear commitment to undertake reforms of publicinstitutions, particularly in the area of public financialmanagement systems, is an important trigger formany donors, reducing their fiduciary riskconcerns. And finally, the technical quality of apartner country’s development strategy, i.e., thedegree to which it is properly prioritized, sequenced,based on sound data, consistent in terms of inter-sectoral linkages, and backed up by verifiable

Section 11

National Aid Policies and Management Systems

45 This OECD-DAC mechanism allows for the monitoring of the efforts and performance of individual DAC members in terms oftheir development cooperation. Individual members are evaluated every four years by a team composed of DAC Secretariattechnicians as well as aid officials from two other member countries of DAC. The task requires visits to both the capital city ofthe member state under review, as well as to partner countries in order to assess the extent of compliance with DAC principles andpolicies.

42


progress indicators and sound monitoring, evaluationand reporting systems – makes it easier for donorsto align to and support such a strategy document.The above enabling conditions, which should bereflected in the long-term objectives of any nationalaid policy framework, might usefully be seen asconstituting the ‘external orientation’ aspect of sucha framework, geared to the donor-partner countryrelationship so that both aid flows and donors arealigned to national priorities and strategies.

To this might be added an ‘internal disciplining’feature of a national aid management policy whichaddresses problems that arise from highlyprojectized aid delivery, and the ‘pull factors’operating on the side of partner countries ministriesmentioned earlier. In the absence of a clearlydelineated approach to national aid management,and assuming that various kinds of project supportand technical assistance continue to loom large forsome time to come, responsibilities for dealing withdonors are usually diffuse and dispersed betweengovernment entities. This results in a problem ofmultiple, disconnected and uncoordinated develop-ment interventions based on donor inputs, be thesetechnical or financial. The problem of Ministries ofFinance being saddled with fiscal responsibilities inthe form of either unbudgeted recurrent and evencapital expenditure of which they may not even beaware, because of commitments entered into byline ministries, is a very real one.

The process of developing an aid managementpolicy provides a means to clearly demarcateresponsibilities. It may help to counter thecontinuing attractions of uncoordinated aid flowsfor line ministries, who understandably arereluctant to give up the freedom to negotiate theirown deals with donors, by reassuring them thatthe development of greater coherence andcentralization will not impact negatively on theirsectoral and corporate interests.

In sum, the advantages of national aid managementpolicies are seen to be that they can constitute:

• a means for partner countries to assertleadership in the local implementation of theParis Declaration;

• a framework for improving the channelling ofaid inflows to priority areas;

• to serve as a filter that will allow governmentsto be more strategic and discerning in terms ofthe volume and types of aid sought;

• over time, assist in raising the proportion of aidpassing through budgets and help governmentsgain a better understanding of the total volumesof aid entering a country;

• help to counter the fragmentation of authorityamongst government agencies dealing with aidmatters; and

• acting as an important vehicle for confidence-building between partner countries and donors.(Killick, 2008)

A clear strategic vision reflecting how any givenpartner country would like to see its relations withdonors develop over time is essential, and shouldbe reflected in any national aid policy document.An interesting example of this may be found inYemen’s Aid Policy Paper which clearlyincorporates both external and domestic aidcoherence considerations. The stated objectivesof the policy paper are described as:

‘…a framework for coordination across theGovernment of Yemen to improve aideffectiveness, since this requires action bymany government ministries and agencies…amanagement framework for developmentpartners to align assistance to Yemeni prioritiesand… contribute to aid effectiveness [andsets] out Yemen’s commitments under the ParisDeclaration… how the Government of Yemenintends to honour them, and what it expectsfrom development partners in this regard.’46

46 Republic of Yemen, Ministry of Planning and International Cooperation, ‘Aid Policy Paper’, Draft for Discussion, 2006, p.1.

43

Section 11 – National Aid Policies and Management Systems

11.2 SOME RESULTS IN TERMS OFNATIONAL AID POLICYFRAMEWORKS

The experiences of some highly aid dependantcountries in terms of the utility of a national aidpolicy are both encouraging and instructive.Mozambique, for example, has made significantprogress since the end of its civil war in terms ofadapting the Paris Principle of mutual accountabilityto its own situation, and developed means ofextending accountability to donors through a rules-based system that contains clear rules andprocedures for both itself and donors.

Donor coordination in Mozambique is structuredaround a single overarching national strategy,namely its PRSP, to which a number of donorsprovide budget support (i.e., Mozambique’sProgramme Partners). In 2004 these budget supportdonors updated and signed a Memorandum ofUnderstanding (MOU) dating from 2000 with theGovernment. The MOU laid out the commitmentsundertaken by both sides in order to improve theeffectiveness of budget support and to lay thefoundations of a partnership-based approach tosupport of the PRSP. Drawing on many of the ParisPrinciples, the group of 18 bilateral and multilateraldonors involved in providing budget supportcommitted themselves to a series of measuresaimed at improving the predictability of their aiddisbursements, to eliminate individual bilateralconditionality and reduce transaction costs, enhancethe transparency of aid flows through improvedreporting, as well as to strengthen the Government’scapacity to lead on the national developmentagenda.

A Performance Assessment Framework (PAF),which draws on the priority targets and indicatorscontained in the PRSP was designed by both theGovernment and donors. As a single conditionalityframework for those donors providing budgetsupport, and relying on the information generatedby Mozambique’s own monitoring systems (anapplication of the Paris Principle of donor alignmentto country systems), the PAF constitutes the basison which joint annual reviews of progress are

carried out, thus allowing donors to make aidcommitments in a coordinated manner.

It is also noteworthy that the principle of mutualaccountability was strengthened when this groupof donors also agreed to the establishment of aProgramme Aid Partners - Performance Assess-ment Framework (PAP-PAF). This allows both theGovernment and donors to monitor donor behaviourand progress in terms of the areas referred to inthe MOU. The assessments of the PAPs-PAF arecarried out by an independent team of aid effective-ness specialists, and the results are made public.

Progress has been significant. As a result of thePRSP, which has facilitated donor harmonizationaround, and alignment to, a single overarchingnational strategy, as well as the various under-standings and mechanisms reached between theGovernment and donors, there have been significantimprovements in terms of aid predictability. As of2006, 14 of the 18 donors had some form of multi-year aid commitment and disbursement mechanismin place, and there were significant improvementsin terms of disbursements taking place accordingto schedule. In addition, the number of donorswho were providing more than two-thirds of theiraid to Mozambique as budget support had alsoincreased, as had the number of donors aligned toMozambique’s financial planning and budgetcycles.47

Afghanistan provides another interesting exampleof what a highly aid dependant country can do toimprove its relations with donors. It would bedifficult to imagine a country in a weaker positionin terms of its bargaining power vis-à-vis donorsthan post-Taliban Afghanistan. Not only was ithighly dependant on external military forces in orderto safeguard the state, but in 2006, for example,aid accounted for approximately 36 percent ofAfghanistan’s gross national income, while for2006–2007 the volume of aid was almost doublethat of government expenditure (OECD-DAC,2008: 1-2).

The new Karzai government moved quickly to takecontrol of the country’s development agenda and

47 For details see Alina Rocha Menocal and Sarah Mulley, ‘Learning from experience? A review of recipient-government efforts tomanage donor relation and improve the quality of aid’, ODI Working Paper 268, Overseas Development Institute, London, 2006.

44


relations with donors. It designed its own NationalDevelopment Framework in 2002, which thendeveloped into a more detailed AfghanistanNational Development Strategy which clearly setout the government’s priorities. The government’sprinciples and expectations in terms of engagingwith donors and managing aid inflows werelaid out in an annex to the Afghanistan Compactof 2006. This clearly sets out the responsibilitiesof both donors and the government in terms ofjoint efforts to improve the effectiveness of aidto Afghanistan, and draws heavily on ParisDeclaration principles.48

Recognizing the need to take steps to strengthengovernment systems in order to increase donorconfidence in them, as well as to make progress interms of national ownership, the Governmentquickly set up an Assistance Coordination Authorityin the new Ministry of Finance. It also set aboutimproving its PFM systems in order to addressdonors’ fiduciary risk concerns.

Simultaneously, while this capacity-building effortwas underway – and though it early on expresseda preference for budget support from donors – itagreed to the establishment of a small number ofmulti-donor trust funds in order to encourage earlydonor harmonization and alignment to nationalpriorities and thereby reduced transaction costs foritself. And in recognition of the importance that aidwas likely to play for the foreseeable future, thegovernment set up a computerized system tocapture and track aid flows, and stronglyencouraged donors to report to this Donors’Assistance Database.

The results, though seemingly modest, must be seenagainst an extremely poor baseline. By 2007, thepercentage of aid disbursed by donors to theGovernment sector through Afghan PFM systemshad increased from 44 percent in 2005 to 48 percentby 2007.49 By 2006, more than 90 percent of aidflows were being recorded in the Donors’Assistance Database.

The Government was also pro-active in a numberof other areas. Recognizing the dangers of donorfragmentation, and its own weak capacity tonegotiate and monitor a high number of externaldevelopment actors, it sought to strengthen itsposition by limiting the number of sectors that anysingle donor could operate in, thus avoiding thedanger of any single large donor exercising undueinfluence on government policy. Establishingminimum ‘entry fees’ for sectors, i.e., a floor interms of aid resources required from a donor beforeit became involved in a sector, encouraged a divisionof labour amongst donors that developed over time.The effect was to gradually reduce pressures onlimited central government capacities in terms ofmanaging and responding to donor demands. It alsotook a strict line on foreign technical assistance,exercising its right to turn down offers for externalTechnical Assistance (TA) where it felt thatcompetent Afghan nationals were available.

The experience of other countries in terms of theirdevelopment of national aid management policiesand frameworks should be taken on board byZimbabwe as it gradually reengages with theinternational development community. Given thesignificant volumes of ODA that may flow into thecountry over the next few years, there is animperative for Zimbabwe to develop and then reachagreement with its development partners on themost appropriate frameworks, systems andprocesses for managing these resources so thatthey achieve maximum impact in terms of thecountry’s recovery and longer-term developmentobjectives.

11.3 AID AND PRIVATE CAPITALFLOWS

While there are currently heightened expectationsof significant inflows of ODA amongstZimbabweans as the country reengages with theinternational donor community, it should be bornein mind that sub-Saharan Africa has experienced

48 The London Conference on Afghanistan – The Afghanistan Compact, Annex II – Improving the Effectiveness of Aid toAfghanistan, London, 2006.

49 Ibid., p.1-8.

45

Section 11 – National Aid Policies and Management Systems

large swings in both aid and private capital inflows.After 2002, for example, the global commodityboom saw a five-fold upswing in private capitalflows. While small in global terms, Africa reliedincreasingly on these inflows (Figure 1).

Since the onset of the global economic crisis,however, these private capital inflows have virtuallydried up as Western investors rebuild their balancesheets. According to the Institute of InternationalFinance, net private capital flows to poor countrieshave slumped from almost $1 trillion in 2007 to $467billion in 2008; and to just $165 billion in 2009.50

The World Bank estimates that in 2009 mostdeveloping countries will not have sufficientreserves to cover private debt falling due (2009).For these countries, total financing needs areexpected to amount to over $1.4 trillion during theyear, implying a financing gap of about $268 billion.If rollover rates are much lower than expected, orif capital flight increases significantly, this figurecould rise to almost $700 billion.

This swing in terms of the relative importance ofODA versus private sector flows during thecommodity boom saw the share of aid in total flows

to emerging markets fall from 55 percent to just 23percent (OECD, 2009b). This trend will be reversedduring the current global downturn, but given themagnitude of FDI during the commodity boomyears, the level of aid funding is unlikely to fill thegap. Even before the onset of the current financialcrisis, donor countries – with some exceptions –were already falling short (by around $39 billion ayear) of their Gleneagles commitments tosignificantly increase their aid and double aid toAfrica (World Bank, 2009: 9).

Despite pleas from the IMF, and the setting up ofa global Vulnerability Fund by the World Bank,official aid flows are likely to decline in the wakeof the global recession. While Britain hasreaffirmed its Gleneagles’ commitment, Italy,Ireland and possibly France have signalled thatthey intend to cut their aid. Others may followsuit as the recession bites. The OverseasDevelopment Institute estimates that official aidmay fall by about a fifth, or $20 billion, in 2009.51

Others are ‘front-loading’ their aid commitments(borrowing from future years to keep it steadynow), so aid could fall further after 2009.

Figure 1: Sub-Saharan Africa capital inflows

Source: World Bank (2009)

Loans

Portfolio

FDI

Total

60

50

40

30

20

10

0

-102000 2001 2002 2003 2004 2005 2006 2007

Private capital inflows(billions of $US)

50 Institute for International Finance, Capital Flows to Emerging Market Economies, 27 January 2009.51 The Economist, 12 March 2009.

46


When the global economy recovers, aid flows areunlikely to rise to the level of the Gleneaglescommitments, or even return to the pre-recessionlevels. Having infused enormous sums intostimulating their economies to avoid a globaleconomic collapse, the ability of Western govern-ments to maintain aid levels, let alone increase them,will be sorely tested. Not only will rich countries needto cut fiscal spending to re-balance their budgets,but demographic factors – notably the dependencyratio – will put pressure on donor governments to

reduce aid budgets to shore up funds for their ownsocial welfare programmes, especially pensions andmedical insurance.

Given these realities, it may be wise for Zimbabweto lower its expectations of the volumes of aid it islikely to receive. In addition, in a context in whichthere are likely to continue to be multiple demandson a restricted global aid purse, the imperative forZimbabwe to properly utilize such aid as it mayreceive will be all the greater.

47

Zimbabwe is intent on re-engaging with theinternational donor community by addressing theoverarching concerns that have proved to bestumbling blocks to a normalisation of relationswith its development partners. The inclusivegovernment’s Short Term Emergency RecoveryProgramme (STERP) recognizes the need toimplement a sound macroeconomic stabilizationand recovery programme and to address issuesaround the rule of law, the rights to freedom ofexpression and association, and property rights. Theinternational community have indicated they willrespond as soon as there is clear evidence ofreform.

This working paper has sought to provide aprimarily Zimbabwean audience with an overviewof the current international aid system, its systems,processes and instruments as well as its historicaldevelopment, the key debates that have takenplace between aid specialists as to how to improveaid effectiveness, as well as insights into thenational experiences of some developing countriesthat were seen to have some relevance to theZimbabwean situation. As Zimbabwe reengageswith the mainstream of international developmentdiscourse, and if it is to derive maximum benefitfrom future aid flows, it is imperative that it deepenits understanding of this aid system and itsrequirements.

While over the last decade, joint donor-partnercountry efforts have sought to rationalize inter-national aid architecture in order to make it more‘customer friendly’, the paper has sought to conveyto readers the continued complexity of the system.Multiple demands continue to be placed on partnercountries, many of them with limited national capacity.In the specific case of Zimbabwe, this challenge islikely to be all the greater given the need to playrapid ‘catch-up’ in light of years of isolation fromthe mainstream of international developmentprocesses. Yet at the same time, like all latecomersto any global process, it will have the advantage ofbeing able to learn from the experiences of other

countries in terms of macro-level initiatives such asHIPC, PRSPs, and the Paris process. As a result,there is scope for the country to be able to get to theforefront of these processes, drawing on what isalready a significant global track record in terms oflessons learned and best practices.

The need to therefore build national capacity in themanagement of aid flows will determine, to a largeextent, Zimbabwe’s ability to take a lead in termsof relations with the donor community. Morespecifically, such national capacity should includea sound understanding of the technical requirementsfor meaningful engagement in processes such asinternational debt relief efforts, the designing ofoverarching national development strategies suchas PRSPs, the development of appropriateframeworks such as TRMs for those states thatfind themselves in transitional situations, and thedesigning of appropriate national aid and debtmanagement policies and frameworks.

The level of national effort required in terms ofengagement in these initiatives should not beunderestimated, particularly given the fact that thedemands placed on national systems to engage inmultiple processes will take place simultaneouslyrather than sequentially. Irrespective of the precisevolumes and kinds of aid that Zimbabwe mayreceive in coming years, it would be safe to assumethat both aid and donor agencies will loom large innational life for the foreseeable future. Under suchconditions, and in order to avoid many of thepitfalls associated with uncoordinated aid flowsraised in this paper, the need for strong inter-ministerial coordination in terms of the establishmentof technically sound national recovery anddevelopment priorities and strategies cannot beoverstated. Only under such conditions will it bepossible for Zimbabwe to make use of the ParisPrinciples in order to ensure that harmonization,alignment, a results focus and mutual accountabilityof donors and their aid takes place, and for thecountry to exercise leadership in terms of its owndevelopment agenda.

Section 12

Policy Implications

48


In addition, the dangers of aid dependence shouldalso be factored into national dialogues anddecision-making on the role of aid in the country’srecovery and longer-term development. As thispaper has sought to highlight, aid inflows, unlessproperly managed, can lead to both macro- andmicro-economic distortions, as well as impactnegatively on national institutions in terms of both

a country’s need to properly manage and maximizeits internally generated resources, as well as at thelevel of national accountability systems. In the viewof the authors, aid should be seen as comple-mentary to, rather than a substitute for, soundeconomic management, which sustains robustgrowth rates, and the associated political andinstitutional reforms seen as necessary by society.

49

African Development Bank (2007), Country Dialogue Paper 2007.

African Development Bank (2008), ‘Bank Group Policy on Non-Concessional Debt Accumulation’. Tunis.

Bulir Ales and A. Javier Hamman (2001), ‘How Volatile and Unpredictable are Aid Flows, and Whatare the Policy Implications?’. IMF Working Paper, Washington DC.

—— (2006), ‘Volatility of Development Aid: From the Frying Pan into the Fire’. IMF Working Paper,IMF, Washington DC.

Burnside, C. and D. Dollar (2000), ‘Aid, politics and growth’. American Economic Review, 90.

Celasun, Oya and Jan Walliser (2005), ‘Predictability of Budget Aid: Experiences in Eight AfricanCountries’. Paper prepared for World Bank practitioners’ forum on budget support. IMF and WorldBank, Cape Town.

Chenery, Hollis. B. and Alan Strout (1966), ‘Foreign Assistance and Economic Development’. AmericanEconomic Review, Vol.56, No.4.

Clements, Michael, Steven Radelet and Rikhil Bhavani (2004), ‘Counting chickens when they hatch:The short term effect of aid on growth’. Center for Global Development, Working Paper No.44,2004, Washington DC.

Easterly, W., R. Levin and D. Roodman (2004), ‘Aid policies and growth: Comment’. American EconomicReview, Vol.94.

Eifert, Been and Alan Gelb (2005), ‘Improving the Dynamics of Aid – Towards More PredictableBudget Support’. World Bank Policy Research Working Paper No. 3732, Washington DC.

Fritz. V. and A. Rocha Menocal (2006), ‘Rebuilding Developmental States’. ODI Working Paper No.274.London.

Ghani, A., C. Lockhart and M. Callaghan (2005), ‘Closing the Sovereignty Gap: How to Turn FailedStates into Capable Ones’. ODI Opinion No.44. Overseas Development Institute, London.

Griffin, K.B. and T.L. Enos (1970), ‘Foreign Assistance: Objectives and Consequences’. EconomicDevelopment and Change, Vol.18, No.3.

Gupta, S., R. Powell and Y. Yang (2006), ‘Macroeconomic Challenges for Scaling Up Aid to Africa:A Checklist for Practicioners’. IMF, Washington DC.

Hansen, Henrik and Finn Tarp (2001), ‘Aid Effectiveness Disputed’, in F. Tarp (ed). ‘Foreign Aid andDevelopment’. Routledge, London and New York.

IMF (2002), ‘Guidelines on Conditionality’, Legal and Policy Development and Review Department,Washington DC.

—— (2005), ‘The Macroeconomics of Managing Increased Aid Inflows: Experiences of Low-IncomeCountries and Policy Implications’. Policy Development and Review Department, WashingtonDC.

—— (2006), ‘Statement of the IMF Staff – Principles Underlying the Guidelines on Conditionalities’.Washington DC.

—— (2007), ‘Staff Report on Zimbabwe’. 9 February 2007.

Bibliography

50


Institute for International Finance (2009), Capital Flows to Emerging Market Economies, 27 January.

International Development Association and International Monetary Fund (2008), ‘Heavily Indebted PoorCountries (HIPC) Initiative and Multilateral Debt Relief Initiative (MDRI) – Status of Implementation’.IDA and IMF, Washington DC.

Kaldor, Nicholas (1963), ‘Will Underdeveloped Countries Learn to Tax?’ Foreign Affairs, 41. WashingtonDC.

Killick, Tony (2008), ‘Taking control: Aid management policies in least developed countries’. Backgroundpaper for UNCTAD, ‘The Least Developed Countries Report 2008: Growth, Poverty and theTerms of Development Partnership’. Geneva.

Killick, Tony and Mick Foster (2007), ‘The Macro-economics of Doubling Aid to Africa and theCentrality of the Supply Side’. Development Policy Review, 2007, 25(2), Blackwell Publishing,Oxford.

Leader, Nicholas and Peter Colenso (2005), ‘Aid Instruments in Fragile States’. PRDE Working PaperNo.5, March 2005. DFID.

McGillivray, Mark and Oliver Morrisey (2001), ‘A Review of Evidence on the Fiscal Effects of Aid’,Credit Research Paper No.01/13, Centre for Research in Economic Development and InternationalTrade, University of Nottingham, UK.

Menocal, Alina Rocha and Sarah Mulley (2006), ‘Learning from experience? A review of recipient-government efforts to manage donor relations and improve the quality of aid’. ODI WorkingPaper No.268. Overseas Development Institute, London, UK.

Michel, Louis (2008), ‘Budget Support – A Question of Mutual Trust’. Luxembourg: Office for OfficialPublications of the European Communities.

Moss, Todd and Arvind Subramanian (2005), ‘After the Big Push? Fiscal and Institutional Implicationsof Large Aid Increases’. Working Paper No.71, October 2005. Center For Global Development,Washington DC.

Moss, Todd, Gunilla Petterson and Nicholas van de Walle (2006), ‘An Aid Institutions Paradox? AReview Essay on Aid Dependency and State Building in Sub-Saharan Africa’. Working PaperNo.74, January 2006. Center for Global Development, Washington DC.

Newlyn, W. (1973), ‘The effect of aid and other resource transfers on savings and growth in lessdeveloped countries: A comment’. Economic Journal, No.83(331), 1973.

OECD (2005), ‘Harmonising Donor Practices for Effective Aid Delivery, Volume 2: Budget Support,Sector-Wide Approaches and Capacity Development in Public Financial Management’. Paris.

—— (2007), Development Co-operation Report 2007. OECD, Paris.

—— (2008), ‘2008 Survey on Monitoring the Paris Declaration’. Paris.

—— (2009a), ‘Aid Effectiveness – A Progress Report on Implementing the Paris Declaration’.Paris.

—— (2009b), Rebalancing the World Economy: Development Co-operation Report 2009. Paris.

OECD-DAC (2007), ‘Principles for Good International Engagement in Fragile States and Situations’.Paris.

—— (2008), ‘Survey on Monitoring the Paris Declaration: Making Aid More Effective by 2010 –Country Section Afghanistan’. Paris.

OPM/IDL, (2008), ‘Evaluation of the Implementation of the Paris Declaration: Thematic Study –The Applicability of the Paris Declaration in fragile and conflict-affected situations’.

51

Bibliography

Owens, Trudy and John Hoddinott (1999), ‘Investing in development or investing in relief: quantifyingthe poverty trade-offs using Zimbabwe household panel data’. Working Paper Series 98. Centrefor the Study of African Economies, University of Oxford, UK.

Raghuran, G. Rajan and Arvind Subramanian (2005), ‘What Undermines Aid’s Impact on Growth?’.IMF Working Paper 05/126. Washington DC.

Republic of Yemen (2006), ‘Aid Policy Paper’, Draft for Discussion. Ministry of Planning and InternationalCooperation.

Roodman, David (2007), ‘Macro Aid Effectiveness: A Guide for the Perplexed’. Working Paper No.135. Center for Global Development, Washington DC. December.

Rostow, W.W. (1960), ‘The Stages of Economic Growth’. Cambridge University Press, UK.

Sachs, Jeffrey D. (2005), ‘The End of Poverty: Economic Possibilities for Our Time’. Penguin, NewYork.

Scanteam (2007), ‘Review of Post-Crisis Multi-Donor Trust Funds – Country Study Annexes’, WorldBank, Norwegian Ministry of Foreign Affairs and NORAD.

Strategic Partnership for Africa (SPA) (2005), ‘Survey of the Alignment of Budget Support and Balanceof Payments Support with National PRS Processes’. Budget Support Working Group. London.

—— (2007), ‘Survey of Budget Support 2006’. Budget Support Working Group. Washington DC.

Svedberg, Peter (2006), ‘Comment on Finn Tarp: Aid and economic growth: An alternativeinterpretation of the evidence’. Swedish Economic Policy Review No.13.

UNDP (2000), ‘Overcoming Human Poverty’. UNDP Poverty Report 2000. New York.

United Nations Development Group and World Bank (2005), ‘An Operational Note on TransitionalResults Matrices: Using Results-Based Frameworks in Fragile States’.

University of Birmingham (2006), ‘Evaluation of General Budget Support, Synthesis Report – ExecutiveSummary’. International Development Department, School of Public Policy, University of Birmingham,UK. May.

World Bank (1998), ‘Assessing Aid. What Works, What Doesn’t and Why’. World Bank Policy ResearchReport.

—— (2002), ‘A Sourcebook for Poverty Reduction Strategies – Volume 1: Core Techniques andCross-Cutting Issues, Volume 2: Macroeconomic and Sectoral Approaches’. Washington DC.

—— (2005), ‘Review of World Bank Conditionality’. Operations Policy and Country Services, WorldBank, Washington DC.

—— (2006), ‘Debt Relief for the Poorest: An Evaluation Update of the HIPC Initiative’. IndependentEvaluation Group, Washington DC.

—— (2009), Swimming against the tide: How developing countries are coping with the Globalcrisis. Background Paper for the G20 Finance Ministers and Central Bank Governors Meeting,Horsham, United Kingdom, March 13-14.

52


united nations development programme comprehensive...

Documents