EV667 Developing capacity? An evaluation of DFID-funded ...

VOLUME 2 EV667June 2006

DEVELOPING CAPACITY?

AN EVALUATION OF

DFID- FUNDED

TECHNICAL CO-OPERATION

FOR

ECONOMIC MANAGEMENT

IN SUB-SAHARAN AFRICA

VOLUME 2: CASE STUDIES

Oxford Policy Management

Technical Cooperation for Economic Management Volume 2: Case Studies

i

Preface/Disclaimer

This evaluation was commissioned and funded by the UK Department for International Development. However the report is the responsibility of the consultants alone and neither the conclusions drawn nor the analysis presented should be attributed to DFID.

Both the synthesis report and this second volume of case studies were prepared by a team comprising: Stephen Jones (study director), Val Imber (country team leader, South Africa and Zambia, and leader for pooling study), John Gray (country team leader, Kenya), Andrew Lawson (country team leader, Ghana), Adom Ghartey (organisational/capacity development expert, Ghana), John Kashangaki (economic management expert, Kenya), Donald Maphiri (economic management expert, South Africa), Inyambo Mwanwina (economic management expert, Zambia), Ole Therkildsen (organisational/capacity development expert, Zambia), Andrew Wyatt (organisational/capacity development expert, Kenya and South Africa), and Katarina Kotoglou and Aarti Shah (researchers). Nils Boesen contributed to the development of the study methodology but was forced to withdraw from the Ghana country team as a result of illness. Anne Thomson provided quality assurance.

Thanks are due to the many DFID staff and others who have contributed time and views to the evaluation team.


ii

Table of Contents

Preface/Disclaimer i

Table of Contents ii

List of Tables vii

List of Figures ix

Abbreviations x

Executive Summary xix

Introduction xix

Main Findings Of The Evaluation – Results Of Dfid’s Technical Cooperation xxi

Evidence on capacity development technical cooperation emerging from the evaluation xxv

Implications for DFID policy and practice xxviii

EVALUATION FRAMEWORK 1

1. Introduction 2

Objectives and scope of the evaluation 2

Structure of the report 3

Issues for the evaluation 3

2. Framework and approach for the evaluation 7

Organisational capacity 7

Economic management: framework and performance 10

Approach and methodology for the country studies 11

Overview of the activities covered by the evaluation 13

Issues in applying the framework 17

GHANA CASE STUDY 1

Executive Summary 2

Context for capacity development 2

Economic management in Ghana 3

Results of the technical cooperation: organisational capacity for economic management 4

Results of the technical cooperation: summary of the evaluation results 4

Conclusions and implications 5

1. Introduction 7

2. The context for capacity development 10

The political context 10

The institutional context 13


iii

DFID’s strategy for technical cooperation 17

Summing up 18

3. Economic management in Ghana 19

Overall economic performance 19

Macroeconomic management (1999-2004) 19

The economic management process 23

Has economic management improved? 25

4. Results of the technical cooperation: organisational capacity for economic management 26

National Development Planning Commission (NDPC) 26

The VAT Service 31

Ministry of Finance and Economic Planning 34

5. Results of the technical cooperation: summary of the evaluation results 41

Relevance 41

Effectiveness 41

Efficiency 42

Impact 42

Sustainability 42

6. Conclusions and implications 43

References 46

List of persons consulted 48

Summary of Findings – Evaluation Matrices 50

KENYA CASE STUDY 56

Executive Summary 57


Economic management in Kenya 58




1. Introduction 62

2. The Context for Capacity Development 65

The Political Context 65

The institutional context – public sector administrative capacity 66

DFID’s strategy for TC in Kenya 67

3. Economic management in Kenya 70



iv

Performance of Economic Management: 1999-2004 71

Budget and public financial management 72

Has economic management improved? 74


The Ministry of Finance (MF) and the Ministry of Planning and National Development (MPND) 75

Office of the President (Directorate of Personnel Management) (OP/DPM) and Kenya Institute of Administration (KIA) 78

Ministry of Local Government (MOLG) 80

Kenya Anti-Corruption Agency (KACA) 83

Summary 84


Relevance 85

Effectiveness 86

Efficiency 86

Impact 87

Sustainability 87


References 91



SOUTH AFRICA CASE STUDY 101



Economic management in South Africa 103




1. Introduction 107


3. Economic management in South Africa 117

Overview of economic performance 117

Macroeconomic management: 1999-2004 118

Summing up 121

4. Results of Technical Cooperation: organisational capacity for economic management 122


v

The National Treasury 122

The Department of Public Enterprises 127

5. Results of the Technical Cooperation: summary of the evaluation results 135

Relevance 135

Effectiveness 135

Efficiency 136

Impact 137

Sustainability 137


References 140



ZAMBIA CASE STUDY 147



1. Introduction 153


The political context 157

The institutional context 158

DFID’s strategy for TC in Zambia 162

3. Economic management in Zambia 164


Macroeconomic management: 1999-2004 165

Budget and public finance management 166

Summing up: has economic management improved? 168


Ministry of Finance and National Planning - overview 169

MFNP: Budget and Economic Division 171

The Internal Audit Unit 173

The Tax Policy Unit 175

The Zambian Revenue Authority 177

Department for Human Resources, Information and Planning (HRIP), Public Service Management Division (PSMD), the Office of the President 180

Summary 183



vi

Relevance 185

Effectiveness 186

Efficiency 186

Impact 186

Sustainability 187


References 191



STUDY OF POOLING ARRANGEMENTS FOR TECHNICAL COOPERATION 203


1. Introduction 206

2. Framework and issues for the study 208

The changing international landscape for harmonisation 208

Findings from the Baser and Morgan study 209

Budget, procurement and management issues 211

3. Country experience: progress on donor harmonisation and TC pooling 214

Introduction 214

Government policy on harmonisation, including TC pooling 215

The provision of TC under the arrangements reviewed 217

Summary of progress 226

4. TC pooling in Ethiopia, Mozambique and Tanzania 227

Range of pooling arrangements 227

Strategic management 230

Implementation management 232

5. Conclusions 235

Ownership 235

Alignment 235

Harmonisation 237

Overall assessment 237

References 239

Synthesis References 241

Appendix A: Terms of Reference 244

Appendix B: Evaluation Matrix for Country Case Studies 251


vii

List of Tables

Evaluation Framework

Table 1. Expenditure on the projects reviewed 16

Table 2. Areas of intervention of projects reviewed 16

Ghana Case Study

Table 1. Organisations and projects selected 8

Table 2. DFID TC project expenditure in Ghana – economic and governance sectors 9

Table 3. Key political events and their significance (1992-2000) 12

Table 4. Key political events and their significance (2001-5) 13

Table 5. Public sector reform and intent in Ghana 14

Table 6. Major aid events and their significance 15

Table 7. Macroeconomic indicators (1997 – 2004) 20

Table 8. Average budget deviation index, 2002-4 23

Table 9. NDPC staffing trends 1998- June 2005 29

Table 10. Number of years at NDPC 29

Table 11. VAT collections relative to targets and GDP 33

Table 12. Changes in quality of MOFEP products, 1997-2005 39

Kenya Case Study

Table 1. DFID TC project expenditure in Kenya: Economic and Financial management 64

Table 2. Key macroeconomic indicators 70

Table 3. Recurrent Expenditure (Shares in Total Expenditure) 72

South Africa Case Study

Table 1. DFID TC project expenditure in South Africa: Economic and Financial Management109

Table 2. Unequal participation 112

Table 3. Key macroeconomic indicators 117

Table 4. Unemployment, inequality and poverty indicators 118


viii

Table 5. Fiscal development indicators 118

ZambiaCase Study

Table 1. DFID TC project expenditure in Zambia: Economic and Financial Management 156

Table 2. The Phasing of PSCAP 160

Table 3. Public sector reform intent and results in Zambia under PSCAP 161

Table 4. Progress against HIPC “Triggers” 166

Table 5. Economic management and the structure of the Ministry of Finance and National Planning 169

Table 6. DFID support for PEM 171

Pooling Initiatives Study

Table 1. Budget Support and Sector Support as a Percentage of ODA in 2004 214

Table 2. Selected pooling initiatives in Ethiopia, Mozambique and Tanzania 215

Table 3. Sources of funding for pooling initiatives 219

Table 4. Characteristics of selected pooling arrangements in Ethiopia, Mozambique and Tanzania 229


ix

List of Figures

Evaluation Framework

Figure 1. Framework for analysis of TC impacts 9

Figure 2. Identification of functions within economic management 10

Ghana Case Study

Figure 1. Expenditures and revenues, 1991-2004 (% GDP current prices) 21

South Africa Case Study

Figure 1. Key political events 1994-2005 110

Zambia Case Study

Figure 1. ZRA revenue collection as a percentage of GDP 179

Pooling Initiatives Study

Figure 1. Partnership Commitments under the Paris Declaration on Aid Effectiveness 209

Figure 2. Funding of TC in relation to national planning and budgeting system 213

Figure 3. Funding of TC in relation to Ethiopia’s national planning and budgeting system 220

Figure 4. Funding of TC in relation to Mozambique’s national planning and budgeting system221

Figure 5. Funding of TC in relation to Tanzania’s national planning and budgeting system 222

Annex A: Terms of Reference

Figure 1. Chain of results from TC inputs to capacity outcomes 246


x

Abbreviations

ABB Activity Based Budgeting

ACC Anti-Corruption Authority

ADMU Aid & Debt Management Unit

ANC African National Congress

APR Annual Progress Review

BD Budget Division (of MOFEP)

BDU Budget Development Unit

BED Budget and Economic Division

BEE Black Economic Empowerment

BHC British High Commission

BoG Bank of Ghana (Central Bank)

BPEMS Budgeting & Public Expenditure Management System

CABRI Collaborative African Budget Reform Initiative

CAGD Controller & Accountant General’s Department

CAP Country Assistance Plan

CDF Comprehensive Development Framework

CEPS Customs Excise and Preventive Service

CG Consultative Group

CGD Centre for Governance and Democracy

CIDA Canadian International Development Agency

CIT Central Implementation Team (for the MTEF)

COSATU Congress of South African Trade Unions

CPIA Country Policy and Institutional Assessment

CPIX Target price index used for South African monetary policy

CSP Country Strategy Paper


xi

CSPG Cross-Sectoral Planning Group

CSPIP Civil Service Performance Improvement Programme

CSR Civil Service Reform

CSRP Civil Service Reform Project/Programme

DA Democratic Alliance

DAC Donor Assistance Committee

DAG Development Assistance Group

DBS Direct Budget Support

DFID Department for International Development

DFIDCA DFID Central Africa

DFIDK DFID Kenya

DNPO National Directorate of Planning and Budget in the Ministry of Finance and Planning

DPs Development Partners

DPE Department of Public Enterprises

DPM Department of Personnel Management

DPSA Department for Public Service and Administration

DTI Department for Trade and Industry

EC European Commission

ECDPM European Centre for Development Policy Management

EPCC Economic Policy Coordinating Committee

EPF Education Pooled Fund

ERMD External Resources Management Division (of MOFEP)

ERS Economic Reform Strategy

EU European Union

FoPOS Fortalecimento da Planificação e Orçamentação Sectoriais


xii

GBS General Budget Support

GEAR Growth, Employment and Redistribution Strategy

GNI Gross National Income

GNU Government of National Unity

GoE Government of Ethiopia

GoG Government of Ghana

GoK Government of the Republic of Kenya

GoT Government of Tanzania

GPRS Ghana Poverty Reduction Strategy

GRZ Government of the Republic of Zambia

GSA Government of South Africa

GSS Ghana Statistical Service

HDI Human Development Index

HIP Harmonisation in Progress

HIPC Highly Indebted Poor Country

HIPC-AAP HIPC Public Expenditure Tracking Assessment and Action Plan

HMG Her Majesty’s Government (UK)

HMT Her Majesty’s Treasury

HRIP Human Resources, Information and Planning Department

IAU Internal Audit Unit

ICASA Independent Communications Authority for South Africa

IDSA Institute for Democracy South Africa

IFI International Financial Institutions

IFMIS Integrated Financial Management and Information System

IGF Internally Generated Funds

IMF International Monetary Fund


xiii

IP-ERS Investment Programme of the Economic Recovery Strategy

IP-ERS MED IPERS Monitoring and Evaluation Department

IPPD Integrated Personnel and Payroll Database

IPSP Integrated Provincial Support Programme

IRS Internal Revenue Service

IT Information Technology

JAS Joint Assistance Strategy

JASZ Joint Assistance Strategy for Zambia

JRM Joint Review Mission

KACA Kenya Anti-Corruption Authority

KACC Kenya Anti-Corruption Commission

KANU Kenya African National Union

KEPSA Kenya Private Sector Alliance

KIA Kenya Institute of Administration

KIPPRA Kenya Institute for Public Policy Research and Analysis

KLGRP Kenya Local Government Reform Project

KUPP Kenya Urban Poverty Project

LA Local Authority

LATF Local Authority Transfer Fund

LASDAP Local Authority Service Delivery Action Plan

LGRP Local Government Reform Programme

MDA Ministry, Department or Agency (the entities holding budget authority)

MDBS Multi-Donor Budget Support

MDD Management Development Division, Cabinet Office

MDGs Millennium Development Goals

M&E Monitoring and Evaluation


xiv

MED Monitoring and Evaluation Department

MF Ministry of Finance

MFED Ministry of Finance and Economic Development

MFNP Ministry of Finance and National Planning

MMD Movement for Multiparty Democracy

MOE Ministry of Education

MOF Ministry of Finance

MOFED Ministry of Finance and Education

MOFEP Ministry of Finance and Economic Planning

MOHA Ministry of Home Affairs

MOLG Ministry of Local Government

MOU Memorandum of Understanding

MPER Ministerial Public Expenditure Review

MPF Ministry of Planning and Finance

MPND Ministry of Planning and National Development

MTBPS Mid Term Budget Policy Statement

MTEF Medium Term Expenditure Framework

MTR Medium Term Review

NARC National Alliance Rainbow Coalition

NDC National Democratic Congress (party)

NDPC National Development Planning Commission

NER National Electricity Regulator

NIRP National Institutional Renewal Programme

NNP New National Party

NPP National Patriotic Party

NPV Net Present Value


xv

NT National Treasury

NTRU Non-Tax Revenue Unit

NUPP Nairobi Urban Poverty Project

ODA Official Development Assistance

ODI Overseas Development Institute

OECD Organisation for Economic Cooperation and Development

OP Office of the President

OP/DPM Office of the President, Directorate of Personnel Management

OPM Oxford Policy Management

PAD Policy Analysis Division (of MOFEP)

PAF Performance Assessment Framework

PAMNUP Partnership Approach to Meeting the Needs of the Urban Poor

PARPA Plano de Acção de Redução da Pobreza Absoluta

PCR Project Completion Report

PE Personal Emolument

PEFA Public Expenditure Framework for Assessment

PEM Public Expenditure Management

PEMFA Public Expenditure Management and Financial Accountability

PEMFAR Public Expenditure Management and Financial Accountability Review

PEMU Public Expenditure Monitoring Unit

PER Public Expenditure Review

PFG Project Focus Group

PFM Public Financial Management

PFMA Public Finance Management Act

PFMRP Public Financial Management Reform Programme

PFP Pooled Fund Partner


xvi

PGBS Partnership General Budget Support

PIG Project Implementation Guide, PMECP

PM Project Manager, PMECP

PMA Programme Management Agency

PMECP Payroll Management and Establishment Control Project

Pmem Project Memorandum, PMECP

PMU Project Management Unit

PO-RALG President’s Office – Regional Administration & Local Government

PPP Public Private Partnership

PPPU Public Private Partnership Unit

PRBS Poverty Reduction Budget Support

PRCF Poverty Reduction Co-Financing Fund

PRDC People’s Revolutionary Democratic Congress

PRGF Poverty Reduction and Growth Facility

PROLOGS Poverty Reduction through Optimising Local Governance Systems

PRS Poverty Reduction Strategy

PRSP Poverty Reduction Strategy Paper

PS Permanent Secretary

PSC Public Services Commission

PSCAP Public Service Capacity Building Project

PSMD Public Service Management Division, Office of the President

PSSRP Public Service Reform Programme

PUFMARP Public Finance Management Reform Programme

PUK Partnership UK

PURC Public Utilities Regulatory Commission

RAGB Revenue Authorities Governing Board


xvii

RBM Results Based Management

RDP Reconstruction and Development Programme

RIZES Revenue Institutions Zambia Extended Support

SA South Africa

SADC Southern African Development Community

SASE Selected Accelerated Salary Enhancement

SATI Southern African Tax Institute

SDP Skills Development Programme

SDPRP Sustainable Development and Poverty Reduction Programme

SERP Support for Economic Reform Project

SIDA Swedish International Development Agency

SMP Staff Managed Programme

SOE State-Owned Enterprise

SOER State Owned Enterprise Restructuring

SRPESA Support for Restructuring of Public Enterprises South Africa

SSA Sub Saharan Africa

SWAps Sector Wide Approaches

SWEEP Sector Wide employment, Enterprise and Equity Programme

TA Technical Assistance

TAP Tax Administration Project

TAS Tanzania Assistance Strategy

TC Technical Cooperation

TCO Technical Cooperation Officer

TDP Teacher Development Programme

TNDP Transitional National Development Plan

TOR Terms of Reference


xviii

TPU Tax Policy Unit, Ministry of Finance and National Planning

TST Trans-Sahara Trading

UNDP United Nations Development Programme

UNIP United National Independence Party

USAID US Agency for International Development

UTRAFE Technical Unit for the Reform of the State Financial System

VAT Value Added Tax

VIPS Value Added Tax Information Processing System

WB World Bank

WTO World Trade Organisation

ZESCO Zambian Electricity Supply Company

ZNCB Zambian National Commercial Bank

ZPA Zambian Privatisation authority

ZRA Zambia Revenue Authority


xix

Executive Summary

Introduction

S1. Volume 2 of this evaluation of DFID technical cooperation (TC) for strengthening economic management in Africa presents the four country case studies undertaken for the study (Ghana, Kenya, South Africa and Zambia). It also presents the technical co-operation pooling study, which was a complementary desk-based study examining experience with technical cooperation pooling initiatives as a means of improving the effectiveness of technical co-operation (through increasing ownership, improving transparency and accountability, reducing administration costs, simplifying procedures and improving efficiency) . The pooling study reviews recent initiatives in Ethiopia, Mozambique and Tanzania to examine the extent to which these objectives have been achieved in particular instances, and compares the findings with an earlier review by Baser and Morgan (The pooling of technical assistance: An overview based on field research experience in six African countries, 2001). However, its limitations as a review are emphasised by the authors, as no primary fieldwork was undertaken.

S2. The structure and approach of the evaluation is outlined below, and a summary of the findings is included. Greater analysis of the implications of the evaluation and a presentation of findings and conclusions is contained in Volume 1, the Synthesis Report. The Terms of Reference for the Study are included at Annex A.

S3. Overall, the evaluation focuses mainly on the period 1999-2004. It sought to understand the contribution of technical cooperation to the development of organisational capacity for economic management – that is the ability of the key organisations (Ministries, Departments) involved in the economic management process to discharge their functions, where organisational capacity can be described in terms of three dimensions, which interact with external factors in determining outputs:

• Individual capacities within an organisation (staff numbers, skills, productivity)

• The organisational framework (which determines how individual capacities are deployed and combined). This includes the organisational structure, the quality of management and the modes of working.

• The institutional framework which includes the organisation’s mandate, incentives for performance, structures of accountability, and operating rules, for instance in relation to finance and personnel.

S4. The main information sources used for the evaluation included:

• A review of DFID project documentation.

• Interviews with current and past DFID staff.

• Interviews with providers of TC services.

• Staff of organisations receiving TC.


xx

• Staff of other government organisations with responsibility for capacity development and technical cooperation policies.

• Interviews with other informed observers including representatives of other donors and civil society organisations.

• A wider review of recent literature on capacity development and technical cooperation used to identify hypotheses for the evaluation to examine.

• A desk study of TC pooling initiatives in three African countries (Ethiopia, Mozambique, Tanzania).

S5. The country case studies assessed DFID’s support to the following organisations:

Ghana

Ministry of Finance and Economic Planning –support to public finance reform, financial management reform, Integrated payroll and personnel database.

National Development Planning Commission – support to poverty monitoring, monitoring and analysis project

VAT Service – support to VAT implementation

Kenya

Ministry of Finance – support to public expenditure management and integrated financial management information system

Ministry of Planning and National Development – Monitoring and evaluation

Office of the President, Kenya Institute of Administration – support to civil service reform

Ministry of Local Government – local government reform

Kenya Anti-Corruption Authority – anti-corruption policy

South Africa

National Treasury – strengthening planning, budgeting and financial management

Department of Public Enterprises – support to state-owned enterprise restructuring

Zambia

Ministry of Finance and National Planning – support to public expenditure management and to tax policy

Zambia Revenue Authority – organisational development, computerisation of direct tax, creation of integrated tax administration system


xxi

Office of the President – payroll management and establishment control.

Main Findings Of The Evaluation – Results Of Dfid’s Technical Cooperation

Relevance

Were the objectives of the support and of the organisations supported consistent with DFID’s objectives?

S6. All the DFID activities reviewed addressed important problems of economic management in the countries concerned and were focused on organisations, processes and systems with the potential to have a significant and transformational effect on aspects of public sector performance that were identified as important for poverty reduction. Problems were encountered where there were unresolved issues about the role, mandate or policies of the organisation supported.

Were the activities relevant to the needs of the organisation and to the context?

S7. A general theme through the activities reviewed was the limited extent to which an adequate analysis of the implications of either the wider institutional context or of organisational capacity was undertaken and/or sufficiently informed project design. In general, a clear distinction between transactional and transformational (capacity development) objectives was not made in the specification of the objectives of project activities, and most activities did not clearly articulate how transformational impact was to be brought about. For this reason the relevance of many of the activities reviewed is questionable, despite the appropriately strategic focus of the DFID support.

Was the support consistent with actions being taken by other development partners and by the government?

S8. In Ghana, Kenya and Zambia, the DFID supported activities have generally formed part of a wider multi-donor effort in support of reform agendas articulated by government around Poverty Reduction Strategies. This framework of cooperation has become more effective and formalised in Ghana and Zambia as moves have been made towards HIPC completion and as the commitment (by many but not all of the significant donors) to general budget support as the preferred aid modality has strengthened. In South Africa (unlike any of the other case study countries), the government has articulated a policy framework for the use of technical cooperation for capacity development.

Was there a consultation or management process to ensure the continuing relevance of the support?

S9. The processes of consultation and management around the activities reviewed were generally effective, and in some cases compensated for weaknesses in the original project design and responded well to changes in context.


xxii

Effectiveness

Were the intended outputs produced?

S10. In general planned project or programme outputs have been produced successfully. The problem of failure to produce planned outputs has been greatest for some support to complex processes of systems reform within government ministries and where issues about the role and management of the organisation supported have not been fully resolved.

S11. Long-term technical cooperation has been most effectively provided where the provider is of a high technical calibre and is seen by the organisation being supported as responsive to its needs and as under its direct management control. In the case of short-term technical cooperation, the provision of flexible consultancy resources has in most cases been effective at a transactional level. Cases where outputs have not been achieved have been characterised by:

• Technical weaknesses or lack of appropriate (particularly interpersonal) skills among the TC providers

• A perception from the organisation supported that it had insufficient effective control over the providers

• Insufficient attention to issues about organisational capacity and hence the capacity to use short-term support effectively

• Lack of effective government commitment to the reforms supported (leading to decisions to terminate support to anti-corruption initiatives and civil service reform in Kenya).

Were they generated in a way to maximise ownership by the organisation and government and alignment with government actions?

S12. Ownership has been variable. In Ghana, ownership within the National Development Planning Commission and VAT Service was assessed as strong but it was much weaker in the Ministry of Finance and Economic Planning and at the level of overall effective political commitment to reform. In Kenya, ownership within the organisations supported has generally been strong but has been much more fragile at the political level. In South Africa, ownership was strong. In Zambia ownership within the organisations supported has been strong but progress on the wider reform agenda was limited over much of the period of the evaluation.

S13. Some aspects of project design, procurement and management as noted above have not maximised ownership, particularly where the organisation supported has not had effective control over consultant selection and management.


xxiii

Were they generated in a way to maximise harmonisation with initiatives from other development partners and alignment with government initiatives?

S14. Harmonisation with other donors has generally been strong and has improved to the extent that wider frameworks for coordination have been developed, for instance in Ghana and Zambia.

Were they generated in a gender-sensitive manner, which might promote mainstreaming of gender concerns with the organisation?

S15. Gender issues have not been explicitly addressed in the activities reviewed and have generally not been regarded as relevant to the objectives of the activities supported.

Efficiency

Were the outputs generated efficiently in terms of their costs and timeliness?

S16. The efficiency of the activities reviewed has varied significantly (as has the levels of expenditure involved). Some strategic support that has responded to very clear government needs was judged to have had a significant impact for very modest expenditure (e.g. support to the NT in South Africa in bringing UK experience to bear on the improvement of the budget system, and in helping Zambia achieve HIPC completion). The support to development of the integrated financial management information system in Kenya was also judged highly efficient in comparison to similar initiatives elsewhere in Africa. By contrast the support to payroll system development in Ghana was judged as exceedingly inefficient, while the efficiency of support to the similar initiative in Zambia was assessed as questionable. Quality of design and procurement (particularly in matching skill requirements to needs) seem to be the main determinants of efficiency.

Impact

Did the provision of the outputs lead to results? Was any impact achieved either transactional or transformational?

S17. A consistent picture emerges across the case study examples. DFID’s TC support has contributed in many cases towards the development of improved systems and procedures (e.g. for budgeting, financial management, monitoring and evaluation, audit, payroll management) that if fully implemented could significantly improve aspects of economic management. Failure to achieve this level of contribution is associated either with situations where support has been terminated because of lack of government commitment and progress (the cases in Kenya), uncertainty about the mandate of the organisation supported (Tax Policy Unit in Zambia), and management weaknesses or design flaws. In some other cases, support has been mainly transactional – that is assisting organisations to perform their functions without necessarily strengthening their capacity to perform the functions without continuing support.


xxiv

S18. However, in most cases the realisation of this potential depends on a strengthening of government commitment to the use of these systems and procedures, as well as to measures to address constraints on the capacity of the organisations supported, notably in relation to staffing and incentives but also in some cases in relation to the role and mandate of the organisation.

S19. As a result, it is only in a minority of the cases reviewed that a capacity development impact can be identified. The main examples are support to the National Treasury in South Africa (a notably well-staffed organisation with a clear mandate and well-articulated reform agenda) the Ghana VAT service and the Zambia Revenue Authority. Both of the latter have established management and remuneration systems outside the normal civil service structure in each country, and have been the subject of sustained and long-term external support, which has helped to build a strong organisation.

S20. A significant problem in the assessment of capacity development impact is that the activities reviewed in general did not involve exercises to benchmark capacity and capacity targets were not set 1. As a result, monitoring of capacity impact was either not built into the project monitoring and evaluation system or was not done so using a consistent analytical framework (for instance one that distinguished the wider institutional setting, the elements of organisational capacity, or individual staff capacities). The absence of such a systematic framework or focus on capacity development limits significantly the quality of the information on which judgements about capacity development impact can be based. The lack of such a framework is a reflection of the fact that DFID has not, to the knowledge of the authors when preparing these studies, articulated strategic approaches to capacity development either globally or within the country strategies for the case study countries.

Sustainability

To what extent have transformational impacts been sustained beyond the period during which inputs were provided? Was there a coherent exit and sustainability strategy?

S21. Only in a minority of the activities evaluated has a sustained impact on capacity been achieved. These have been where the mandate of the organisation supported is clear and issues of management, staff retention and incentives have been satisfactorily addressed. The critical issues for sustainability are the extent to which the organisations supported can develop and retain the skills needed to use new systems that the DFID support has helped to develop, and the extent to which there is political and managerial commitment within the organisation to using them. These issues have not been addressed through the activities supported although the wider public sector reform agendas in Ghana and Zambia have included measures focused on these problems. These have however not been effectively implemented. Outside South

1 It is important to note that the activities covered were not conceived or designed as comprehensive

programmes of organisational capacity development for the organisations concerned, though several were implemented within the framework of wider programmes of organisational development support.


xxv

Africa, the resolution of problems about staff incentives and management have only been achieved within the Zambia Revenue Authority and Ghana VAT Service.

Evidence on capacity development technical cooperation emerging from the evaluation

Evidence on the significance of the context

S22. The evaluation evidence suggests that the wider institutional setting has been a major factor in determining the extent to which TC has both been effective and has served to develop capacity. The evidence from the case studies suggests that a favourable institutional setting appears to be necessary but not sufficient to bring about improvements in organisational capacity:

• Influence of political commitment to capacity development: South Africa provided a favourable institutional context, which meant that some TC was able to contribute to a major transformational impact even on the basis of the expenditure of quite limited resources. All of the activities reviewed in South Africa were judged effective.

• Limits to capacity development impact in unfavourable contexts: In situations where the institutional context remained unfavourable it was still possible for TC to be used effectively in a more transactional role.

• Pay and management structures: None of the activities reviewed in Ghana, Kenya or Zambia resulted in clear evidence of improved organisational capacity within government, except in agencies where there was both sufficient clarity about, and government commitment to, their role and objectives. The two cases where this occurred focused on improving revenue performance. Most fundamentally, measures had been taken to address staffing and management constraints, outside the framework of normal civil service rules. .

• Even where the environment is highly constrained, it does appear that appropriate design of a TC activity may enhance the likelihood of capacity impact.

• Quality of leadership: As noted in the Ghana case study in relation to the National Development Planning Commission, good leadership in the organisation supported, political support for key reform measures, and well-structured and well-delivered TC activities may be particularly important factors for success where the institutional environment is unfavourable.

• Policies on technical cooperation and capacity development: Had clearly articulated government policies towards the use of TC and capacity development been in place (and implemented) in Ghana, Kenya, and Zambia, it is possible that the capacity development impact of the activities reviewed would have been enhanced. DFID itself has not implemented any overarching policies to guide the use of TC and capacity development, at least as evidenced by the activities under review, and so has not substituted for this lack of policy guidance from partner countries. The DFID country strategies did not generally provide guidance on how TC should be used and made most effective in the country context.


xxvi

Evidence on identification and design

S23. The evidence from the evaluation supports the view that an understanding of the wider institutional context, as well as the specific features of the organisation, is critical in designing TC activities and for effective capacity development. Generally the design of the activities reviewed did not pay sufficient attention (at least as reflected in the project documentation) to the analysis of the context at either level. More systematic guidance and project appraisal procedures (prompted either by DFID or by the aid receiving government) would probably have improved performance and enhanced impact even in unfavourable circumstances.

S24. The findings also suggest that there has been some tendency to overestimate partner capacity and commitment. This was particularly the case in Ghana. In Kenya, most of the activities reviewed were designed with an awareness that government commitment to reform might be fragile, and included milestones that were related to commitment and evidence of progress, so that exit or suspension was possible in an orderly manner. In the case of South Africa, support to the Department of Public Enterprises was based on assumptions that the Government was clearly committed to privatisation, when in practice the policy position was not fully resolved.

S25. Harmonised and aligned approach to capacity development: The most successful activities reviewed were marked by high levels of government ownership and consensus between DFID and other development partners on priorities and approach, and on close involvement of the organisation receiving support in the selection of TC providers. Examples include the support to the Budget Division in Zambia (where DFID essentially provided a mechanism for enabling the organisation to retain and use effectively a key external consultant), the support to the National Treasury in South Africa and support to the VAT Service and to NDPC in Ghana. By contrast, activities that encountered problems or were less successful tended to be marked by weaknesses in ownership and features of the design including the procurement process.

Evidence on TC provider characteristics and role

S26. The evidence from this evaluation strongly suggests the need for caution in drawing any general conclusions about the relationship between the “type” of technical cooperation provided and its effectiveness and impact. The overall context in which the support is provided and the quality of the design and management of the activity appear to be much more fundamental determinants of success.

Evidence on management and delivery of technical cooperation

S27. The evaluation findings suggest that the style and techniques of TC project management by DFID have a major bearing on the success of the activities. Specific management issues that are highlighted from the case study experience are the following:


xxvii

• Effectiveness is likely to be enhanced where the choice of contractors is a joint decision by DFID and Government and where rigorous interview based selection of consultants is used.

• The quality of certain types of project is likely to be enhanced by the provision of an independent, properly qualified Quality Assurance overseer, who can advise both Government and DFID as an honest broker.

• The use of project management units operating outside the normal civil service structure may be useful and necessary for the task of carrying through a particular task like the management of the development of a new IT system but impact and sustainability is in question unless underlying staffing and capacity constraints are addressed.

• The use of special allowances and performance bonuses was not judged successful in the cases reviewed.

Evidence on TC pooling initiatives

S28. The background study on TC pooling found that there has been significant progress in improving harmonisation and alignment driven in large part by moves towards providing general budget support and sector support compared to the situation in 2001. Joint strategy processes and related organisational mechanisms have been developed to strengthen coordination. Governments are increasingly articulating the objective of moving towards receiving aid through General Budget Support (GBS) and of strengthening capacity to procure and manage TC. Policies towards capacity development and the use of technical cooperation are also being developed.

S29. In terms of the TC funding arrangements used, there has been progress towards extending and deepening pooling arrangements, but in practice different arrangements coexist in each of the countries and sectors examined. While significant progress has been made at the policy level in the approach to TC management, in each of the countries there are constraints to effective implementation, with recognition that capacity development in procurement and financial management in particular is required.

S30. Compared to the earlier assessment by Baser and Morgan in 2001, progress appears to have been made in several areas:

• Progress has been made in addressing the capacity constraints that they identified as obstacles to progress, including through a more realistic assessment of the time and resources required to establish effective mechanisms. This is recognised in the range of arrangements that are being used.

• National ownership of both the overall policy agenda for harmonisation and alignment in relation to TC, and in relation to the management of TC under the arrangements reviewed, does seem to be increasing, though it is not complete particularly in relation to the strategic management role.

• The lesson that incremental and context-sensitive approaches work best appears generally to be reflected in the approaches taken for advancing the initiatives reviewed.


xxviii

• This study has not been able to examine whether the introduction of these arrangements has led to a reduction in the quantity or cost of TC, though the arrangements may have led to an increase in the effectiveness of TC as a result of greater government ownership and more transparent selection procedures.

• Making progress in establishing pooling arrangements remains labour-intensive and time-consuming for both donors and government.

S31. The overall conclusion of the pooling study is that while pooling initiatives have now begun to realise their potential for improving ownership, there remain important constraints on progress. Key strategic decisions tend to remain under joint (effectively donor) control, and there is a conflict between the ability to meet urgent demands (which can be effectively addressed by informal coordination arrangements or individual bilateral donors) and the process of capacity building for effective procurement and management of assistance. The scope for progress (and the priority to be placed on moving forward in this area, compared to lower cost and simpler processes of coordination and cooperation between donors) depends on the national context but the long-run determinants of success are likely to be the strength of government commitment to exercising more effective control over TC, and the level of success in developing organisational capacity for effective procurement and financial management.

Implications for DFID policy and practice

S32. The following implications emerge from the evaluation for DFID’s policy and practice in the provision of support to capacity development and the use of technical cooperation:

S33. The commitments towards strengthening harmonisation and alignment that are set out in the Paris Declaration provide an improved framework for the effectiveness of capacity development and for technical cooperation in many contexts.. All DFID’s practice should continue to emphasise and reinforce these commitments and the principles behind them. This should include the articulation of and commitment towards strengthening the capacity of governments effectively to manage and use technical cooperation resources.

S34. A more rigorous and systematic approach to the assessment of the institutional and organisational context is required if DFID’s activities are to contribute to the sustainable building of capacity. This needs to be reflected in a long-term and strategic approach to engagement in capacity development processes and a greater attention to understanding the process by which change may occur as well as strengthened monitoring of key factors affecting the scope for effective capacity development. Both DFID’s general policy guidance, and the treatment of capacity development issues within Country Assistance Plans may need to be reviewed.

S35. In general achieving transformational impact requires a wider framework of action (that will need to be agreed and implemented with involvement from all donors and clear government leadership) that addresses identified institutional and organisational issues. The implementation of reforms and strengthening of management within the


xxix

public sector of the government supported will be the key steps in this process. In the absence of progress in these areas, there may be a case for providing support with a more explicitly transactional focus as well as for some investment in systems development, but the limitations of what this is likely to achieve in terms of longer-term capacity development and impact need to be recognised.

S36. TC can be highly effective in a range of contexts (including ones where the institutional setting for capacity development is poor – indeed it may be of most value in achieving transactional impact in such conditions) but this effectiveness is not necessarily related to capacity development. Technical cooperation is not fundamentally what builds organisational capacity, since high levels of organisational effectiveness are in general needed for technical skills to be absorbed and used within an organisation. It would be helpful also to move away from the common tendency in donor practice to equate TC with capacity development.

S37. For long-term technical cooperation staff (whether provided as consultants or seconded staff), the quality and appropriateness of skills for the role and organisation are critical, with interpersonal skills being particularly important. Selection processes should be as rigorous and individualised as would apply for the appointment of permanent staff within the organisation. Where technical cooperation is provided to an organisation, the lines of reporting and management control should be extremely clear and the organisation supported should play the key role in the selection and management of technical cooperation staff, against clearly defined objectives and criteria.

S38. Even in relatively unfavourable environments for capacity development, greater attention to the design of particular activities (including the terms of reference for TC providers) could plausibly be expected to increase the capacity development impact. General principles of good project design and management apply in relation to TC as much as in any other type of project and are particularly important in difficult contexts.

S39. Lack of progress in civil service reform emerges as the most significant factor in explaining the limited capacity development impact achieved in three of the four country case studies, and this issue (and the constraints it imposes) does not seem to be sufficiently addressed in the various recommendations on capacity development that emerge from the recent evaluation and policy literature. This suggests that DFID should strongly emphasise the fundamental importance of effective reform in this areas, and support measures that contribute to progress in addressing this issue through coordinated donor and government action, noting the limited success that has been achieved in this area to date and the substantial obstacles to reform that there appear to be in many countries. Exploring the conditions for success for wider public sector reform initiatives (and what role donor agencies might play in bringing about these conditions) emerges as a priority for further investigation.


1

EVALUATION FRAMEWORK


2

1. Introduction

Objectives and scope of the evaluation

1.1 This report presents the findings of an evaluation of DFID technical cooperation2 (TC) for strengthening economic management in Africa. It is based on four country case studies (of Ghana, Kenya, South Africa and Zambia) focusing mainly on the period 1999-2004. Within each country, the evaluation concentrated on DFID activities that were identified as providing TC support to organisations whose functions played an important role in the economic management process.

1.2 The framework for the evaluation (see Annex B and Section 2 below) sought to identify the influence of TC (as one input provided in a wider context) on organisational capacity, and in line with the main purpose and questions specified in the terms of reference (Annex A):

“To map and test the chain of results from DFID-funded TC personnel inputs to enhanced partner country capacity and better policies and service delivery. In doing so, lessons will be drawn on the best ways to provide TC personnel in different contexts in order to maximise effectiveness, efficiency and impact on partner country capacity. The key broad questions that will be answered are: ”

• To what extent does DFID TC aim and achieve to (1) bridge capacity constraints in order to deliver operational outputs (“getting the job done”) and (2) support partner country capacity development?

• What have the most efficient and effective ways of providing long-term TC personnel to Africa been for DFID (focusing on differences between consultancy and other personnel such as TCOs3 and ODI fellows4, short and long-term personnel, local and expatriate consultants)?

• What is the role of recipient country ownership in the delivery of DFID TC? What difference does it make to the effectiveness, efficiency, impact and sustainability of TC?

• How does the context in which DFID TC is delivered influence the relationships mapped above?”

2 The terms Technical Cooperation and Technical Assistance are used interchangeably in this study.

Technical cooperation is defined in the terms of reference (and following OECD DAC guidelines) as “the provision of advice and/or skills, in the form of specialist personnel, training and scholarships, grants for research and associated costs.” This definition was narrowed for the purposes of the evaluation to focus on DFID-funded TC personnel, where TC personnel is defined as assistance provided to recipient countries in the form of specialists, including consultants, distinguishing between long-term (one year or longer) and short-term (less than one year). This broad categorisation includes the situation where long-term contracts may be used to provide short-term inputs delivered in a variety of systematic or ad hoc ways.

3 Technical Cooperation Officers, i.e. staff employed by DFID to provide technical co-operation to supported

organisations.

4 Junior staff selected through a competitive process by the Overseas Development Institute to serve two

year assignments in (usually) line civil service positions in developing countries, being paid on local terms plus a supplementation payment.


3

1.3 The terms of reference for the evaluation did not provide a definition of “capacity development” but stressed that:

• Different types of capacity need to be considered

• Capacities relate to the stage of the policy process (planning, implementation, monitoring, evaluation)

• Capacity development needs to be examined at individual, institutional and societal levels.

• At least six key facets of capacity development need to be addressed – knowledge acquisition, institution building, institutional and environment partnerships, policy environment, country commitment and autonomy, and results and accountability.

1.4 The approach adopted for the evaluation was to focus on organisational capacity for economic management – that is the ability of the key organisations (Ministries, Departments) involved in the economic management process to discharge their functions. This includes strengthening the capacity of these organisations to define policy objectives, reach policy decisions, design policy instruments and implement them effectively.

Structure of the report

1.5 A review was undertaken of the key themes in recent policy debate on capacity development and technical cooperation5. This highlighted the emergence of a “new consensus” on capacity development, and summarised the main elements of the critique of technical cooperation as an aid instrument. On the basis of this review, issues in the debate were identified that the evidence from this evaluation might inform. Section 2 presents the conceptual framework and approach used for the evaluation and reviews issues that arose in applying the framework.

1.6 The findings and conclusions of the evaluation were reviewed in relation to the assessment of the results achieved and the impact on improving capacity for economic management in Volume 1 (the Synthesis Report). Annexes in both volumes contain the terms of reference and the generic evaluation matrix used for the country studies.

Issues for the evaluation

1.7 The review of recent literature on capacity development and technical cooperation undertaken for this study identified several issues as potentially of significance in explaining the performance of TC in bringing about capacity development or achieving other objectives. Four main issue areas were identified (based on the literature and specific questions in the terms of reference6) against which the evidence provided by the evaluation was assessed:

5 See Synthesis Report, section 2

6 Attached as Annex A


4

• The significance of the context

• Design issues

• Characteristics of the technical cooperation provider and the use of technical cooperation

• Management and delivery issues.

1.8 In practice, the extent to which assessments of these issues could be made was variable and in most cases incomplete, because of the limitations in the detail of the review of institutional and organisational context that could be undertaken, the small-scale of the interventions, and the difficulty of making firm causal judgements about the determinants of outcomes.

The significance of context

1.9 The recent literature highlights the importance of both the wider institutional context and the specific characteristics of the organisation (including its role and the nature of its products) in determining the likely success of capacity development initiatives, and the specific impact that technical cooperation may achieve. Key issues related to the wider context which the evaluation sought to examine included:

• The extent of political commitment to capacity development objectives (or the degree of domestic ownership of the capacity reform agenda) both in general and for particular organisations.

• The extent to which governments have articulated or implemented policies to guide capacity development or the use of technical cooperation.

• The influence of the institutional context affecting the performance of the public sector, particularly in relation to civil service pay and management structures, and other factors affecting recruitment, retention and the motivation of staff within organisations supported, but also including the progress of reform initiatives addressing constraints in these areas.

• The quality of leadership, approach to change management, and existence or otherwise of a critical mass of staff within the organisation committed to change.

• The extent to which TC has been used to try to try and encourage the demand for reform, to influence policy, or to tackle wider institutional constraints on capacity development.

1.10 As discussed in Volume 1 (section 5), it was possible to identify the influence of elements of the political and institutional context as determinants of outcomes. The significance of management factors were however much harder to assess. The TC experience reviewed in the evaluation did not seek specifically to influence to the demand for reform or policy, and was not generally focused on addressing wider institutional constraints.7

7 It had originally been intended to include an assessment of support for public service reform in Ghana but

this was not possible as a result of the illness of a team member. The Kenya case study included a review of DFID’s aborted support to civil service reform and anti-corruption initiatives but .in both cases a decision not to proceed with support was made based on a view about lack of government commitment.


5

Design issues

1.11 The review suggests that quality of design is likely to be critically important for achieving capacity development impact, with the following factors specifically identified in the literature and experience:

• The extent to which the design of the activity is based on adequate contextual analysis both in relation to the wider institutional context and the specific features of the organisation.

• The existence of an overall strategic approach to capacity development that is shared with government and other donor agencies.

• The extent to which generic features of international best practice in design are appropriately modified for the particular circumstances.

• The monitoring and evaluation framework used, specifically the extent to which it (a) defines clear targets and a monitoring framework focusing on capacity development or other objectives and (b) the extent to which this framework responds to “endogenous” demands from stakeholders in the performance of the organisation, as opposed to the external accountability requirements of funding agencies.

• The extent to which the organisation receiving support and key domestic stakeholders in the organisation have been able to exercise effective control over each stage of the activity, including the definition of the terms of reference for TC, the selection of TC providers, and the management of the delivery of TC inputs.

• The extent to which it is possible to create a “virtuous circle” by which TC helps create the conditions for further capacity development progress.

• The relationship between TC activities and the mix of aid instruments being used, the other resources that accompany TC, the overall level of TC and its relation to other donor activities.

1.12 The literature around the “new consensus” on capacity development stresses the importance of contextual analysis and having a clear framework to guide capacity development efforts that can apply the results of this analysis to improve the effectiveness of support, as well as the dangers of applying “generic” approaches in an unsuitable context. The evaluation findings tend to suggest that stronger contextual analysis could have led to the more realistic setting of objectives, and that a clear framework for design and monitoring could have led to a more strategic focus on higher- level institutional constraints, but it does not suggest what might in practice have been done to overcome these constraints. The evaluation does lend support to the view that effectiveness is related to the control of the organisation over the activity, and that quality of design is also of critical importance but failures do not appear to be related to an attempt to apply some generic view of good practice but rather to failures to follow consistently good design practice.

Characteristics and uses of technical cooperation

1.13 Issues relating to the characteristics of TC suggested by the literature as potentially important include:

• The relative effectiveness of national and international TC.


6

• The relative merits of short and long term TC.

• The mix of skills (including technical analytical, policy process, or management and organisational development skills).

• The advantages and disadvantages of using externally hired consultants as opposed to Technical Cooperation Offices (TCOs) and secondment arrangements (i.e. using UK government staff as TC providers).

1.14 The findings of the evaluation tend to support the view that first that interpersonal skills are a key determinant of the effectiveness of TC especially in capacity development and second that different forms of support or characteristics of TC providers are effective in different contexts and the challenge is find what is most appropriate (which is largely determined by the demands of the organisation supported) rather than to apply general rules or presumptions about the desirability of e.g. local or regional over international TC provision. Only in one case reviewed (support to the National Treasury in South Africa) was TC provided by seconded UK government staff rather than consultants and there were no projects using TCOs in the sample.

1.15 A further set of issues relates to the relationship between the effectiveness of TC and how it is used in the organisations supported:

• The use of TC in line positions in the organisation supported as opposed to its use in an advisory or challenge role outside the normal line functions and organisational structure.

• The role of specified counterparts in ensuring skill transfer.

• The advantages and disadvantages of using TC to fill gaps in skills needed to produce specific outputs as opposed to having capacity development objectives.

1.16 The findings show that TC in line positions can be highly effective and responsive to an organisation’s needs, but that the capacity development impact of such TC is likely to be limited unless the institutional and organisational context is appropriate. Whether TC is or is not in a line position is likely to be of secondary importance. Counterpart arrangements may help in individual skills transfer but it is hard to draw firm conclusions on this from the evidence of the case studies.

Management and delivery of technical cooperation

1.17 There is a further set of issues around the quality of (or the vast majority of them) management of TC. These include:

• Clarity of accountability and management (particularly in relation to the balance of accountability between the organisation receiving support and the funding agency).

• Effectiveness of the monitoring and evaluation function in ensuring that the TC strategy is implemented effectively, and flexibility and speed of response to emerging problems or opportunities.

• Quality of selection of TC providers and procurement procedures.

1.18 Quality of management and selection emerges strongly from the evaluation as of central importance along with quality of project design.


7

2. Framework and approach for the evaluation8

Organisational capacity

2.1 The key feature of the approach used for this study was a focus on understanding the contribution of technical cooperation to the development of organisational capacity for economic management – that is the ability of the key organisations (Ministries, Departments) involved in the economic management process to discharge their functions.

2.2 Organisational capacity can be described in terms of three dimensions, which interact with external factors in determining outputs:

• Individual capacities within an organisation (staff numbers, skills, productivity)

• The organisational framework (which determines how individual capacities are deployed and combined). This includes the organisational structure, the quality of management and the modes of working.

• The institutional framework (determined outside the organisation, but potentially amenable to influence through capacity development processes) which includes the organisation’s mandate, incentives for performance, structures of accountability, and operating rules, for instance in relation to finance and personnel.

2.3 The process of developing organisational capacity is necessarily a multi-faceted one. It can involve the provision of different types of input, including staff, training, tools (such as computer systems), knowledge production, and policy and institutional advice. The starting point for examining the capacity of organisations is the quality and quantity of “outputs” that they produce – which includes key policies and documents as well as the effective management of processes like the budget cycle. An improvement in organisational capacity is associated with an improvement in the quantity, range, quality, timeliness, accessibility and government or wider ownership of these outputs. The main dimensions on which the evaluation focused in making assessments of organisational outputs were:

• The coherence of decision-making processes, as assessed by examining the consistency of key policy statements, decisions and the management of the policy making process such as the timeliness and quality of preparation for Cabinet meetings.

• The extent to which economic management decisions have been internally driven and ‘bought into’ by key political decision-makers.

• Core macroeconomic outcomes and their consistency with stated policy: actuals and targets with regard to inflation, the fiscal deficit and the level of borrowing over a three year period, and the extent to which there is stability in interest and exchange rates.

• Core budgetary outcomes and their consistency with approved budgets: aggregate revenue and expenditure and spending by ministry/ sector relative to budgets.

8 The framework and approach was set out in more detail in the Inception Report (OPM, 2005).


8

2.4 Figure 1 illustrates the framework for understanding the process by which organisational capacity may be developed, and the potential role of technical cooperation in relation to other factors contributing to capacity development. The top row of Figure 1 represents the baseline capacity of the organisation and the products it produces in contributing to economic management before a specific TC intervention. The bottom row represents capacity for economic management after the intervention. TC may have two types of impact. Transactional impact is defined in accordance with the terminology used in the evaluation terms of reference as “bridging capacity constraints in order to achieve operational goals”, while transformational impact relates to “developing countries’ capacity in the longer term.” Within the framework used for the evaluation, a transactional impact is understood as one that improves the organisation’s products (possibly transiently during the period that support is provided), while a transformational impact improves the organisation’s capacity to produce its key products.

2.5 Factors relevant to determining the impact of a particular piece of TC are therefore:

• The wider institutional, political and external context within which the organisation being supported is operating.

• The level of pre-existing organisational capacity – since strong organisations may be able to use TC more effectively.

• The other non-TC inputs provided in parallel with the technical cooperation (such as equipment and finance)

2.6 The causal relationships involved are potentially complex. Each of these factors can affect whether a particular piece of TC produces the planned output, whether this output has an impact on the organisation, either transactional or transformational, and whether this impact is sustained.

•

Te

ch

nic

al

Co

op

era

tio

n f

or

Ec

on

om

ic M

an

ag

em

en

t V

olu

me

2:

Ca

se S

tud

ies

9

Fig

ure

1.

Fra

mew

ork

fo

r a

na

lys

is o

f T

C i

mp

ac

ts


10

Economic management: framework and performance

2.7 In principle, a wide range of public sector activities can be described as economic management. They can be divided into management of the public sector and the way the way in which the public sector reacts with the rest of the economy, that is public sector management and regulation of the private sector. The boundary between the public and the private sector varies from country to country and TC is often provided to shift that boundary. The nature of economic management also varies across countries reflecting history, tradition and economic philosophy inherent in the culture of the society: TC may be used to drive change in that philosophy as a means of promoting development. One of the main challenges of donor assistance is to reconcile differences between the approaches of the host and the donor country not only in terms of economic philosophy but also in terms of the way society, including public administration, actually operates.

2.8 Figure 2 summarises the approach used in the evaluation for defining the functions included as “economic management.” At the core of the concept is the reconciliation between the national planning system to meet poverty reduction objectives (in most cases in Africa as part of a formal PRSP process) and the allocation of public finance to resource the necessary services. Elements bounded by the thick black line were the main focus of the evaluation, though the precise coverage varied between the case study countries depending on the nature of the DFID programme in each case and in two cases (the Zambia Revenue Authority and the Ghana VAT Service) support to revenue authorities was included in the activities evaluated.

Figure 2. Identification of functions within economic management

Functions within Economic Management

State

Bank

Revenue

Authority

Statistics

Body

Donor

GBS

Donor

projects

Economic PolicyMacro forecast

Planning & Budgeting

MTEF

FinancialManagement

SystemPRSP

Service Policies

Line Ministries


11

Approach and methodology for the country studies

2.9 In order to try to understand the chain of causation from TC inputs to capacity for economic management, the approach to the case studies was therefore based on four analytical steps (A to D below):

A. Understanding the context for capacity development and defining the scope of analysis

1) Describing the essential features of the country context over the period, including the key economic and political developments, the level of aid dependence, approach to aid management and relations with donors, and the country’s institutional and political characteristics (civil service capacity, political economy, legal and judicial framework) so as to identify the key external factors that will have influenced the capacity development and organisational development processes.

2) Assessing the key features of economic management over the evaluation period, and the context in terms of the role of stakeholders and the political economy of economic management. This is important for understanding the political and administrative constraints within which TC operates.

3) Documenting and reviewing national policies towards technical cooperation and capacity development, including the extent to which the national poverty reduction strategy or other key documents identify a role for technical cooperation or attempt to establish clear management principles for it, including the extent to which TC appears to be demand- as opposed to supply-driven.

4) Reviewing the analysis of capacity issues and the strategy towards capacity development for economic management in DFID country strategies.

5) Mapping the key organisations and formal and informal processes in the performance of economic management functions.

6) Defining the scope of the exercise within each country based upon the organisations/ ministries in which DFID-financed TC operations were undertaken.

B. Assessing organisational capacity for economic management

7) Benchmarking the status of the organisational processes and functions of economic management at the beginning of the period under analysis.

8) Benchmarking the status of the organisational processes and functions of economic management at the end of the period under analysis.

9) Identifying the changes at the level of organisational capacity and the products of the organisation over the period covered by the evaluation.


12

C. Identifying contributions to capacity change [for each focal organisation]

10) Identifying the range of capacity development inputs being provided over the period – focusing on those provide by DFID but relating the DFID activities to those of other Development Partners and Government.

11) Documenting and analysing the key DFID-supported activities.

12) Identifying the key external factors influencing the capacity development/ organisational development process.

13) Assessing the contribution of the DFID interventions to the changes identified.

14) Considering the alternative ways in which capacity development inputs might have been provided and the most plausible explanation for the changes identified.

D. Assessing the DFID interventions

15) Based on the impact assessment and the identification of contribution, making an assessment of the relevance, efficiency, effectiveness and sustainability of the TC activity. This was presented within a common evaluation matrix framework (Annex B).

2.10 The main evidence base for the evaluation country case studies derived from reviews of documentation and information obtained during two-week visits to each of the case study countries (carried out in June and July 2005). The main information sources used for the evaluation included:

• A review of DFID project documentation

• Interviews with current and past DFID staff

• Interviews with providers of TC services

• Staff of organisations receiving TC

• Staff of other government organisations with responsibility for capacity development and technical cooperation policies

• Interviews with other informed observers including representatives of other donors and civil society organisations.

2.11 The set of questions in the evaluation matrix (Annex B) was used to structure interviews.This analytical framework is reflected in the structure of each case study:

• Assessment of the context for capacity development (political, institutional including policies towards external assistance and capacity development)

• DFID’s strategy for the use of technical cooperation as reflected in country strategy documents

• Economic management in the case study country (overall economic performance, overall economic management, budget and public financial management)


13

• Results of the technical cooperation: organisational capacity for economic management

• Results of the technical cooperation: summary of the evaluation results against the key evaluation criteria

• Conclusion – summarising the overall results of the DFID support, and what might have been done differently.

Overview of the activities covered by the evaluation

2.12 The case study countries were selected using three criteria specified in the terms of reference: (1) the likely availability of data (proxied by the DFID’s TC expenditure); (2) including a range of diverse policy and institutional contexts; (3) avoiding overlap with other on-going and planned evaluations, including the Joint Evaluation of General Budget Support.

2.13 .The starting point for the selection of activities to be covered was a listing of DFID projects in the economic and governance sectors in each country that recorded technical cooperation expenditure over the period from 1999-2004, along with other identified activities that are not included as DFID projects at the country level (for instance, ODI Fellowships and some staff secondments).

2.14 This list was discussed (through an initial country visit by a DFID Evaluation Department (EVD) staff member) with country teams to provide a better understanding of the perceived significance of activities undertaken in relation to capacity development for economic management. On this basis, the main organisations supported were identified, and a preliminary classification of DFID activities was made in relation to their likely significance in terms of potential impact on economic management. The final selection of activities to be evaluated was based on this list, from those identified as of key importance and potentially significant, taking into account the feasibility of collecting information within the constraints of the resources available for the evaluation.

2.15 The DFID activities reviewed (and the main organisations covered) in each country case study were the following:

Ghana9

Ministry of Finance and Economic Planning –support to public finance reform (PUFMARP), financial management reform (PFM), Integrated payroll and personnel database (IPPD and IPPD2).

National Development Planning Commission – support to poverty monitoring, PRS monitoring and analysis project (NDPC)

9 The illness of one of the Ghana team members meant that it was not possible to cover support to public

sector reform as had been intended.


14

VAT Service – support to VAT implementation (VAT)

Kenya

Ministry of Finance – support to public expenditure management (PEM) and integrated financial management information system (IFMIS)

Ministry of Planning and National Development – Monitoring and evaluation (IP-ERS MED)

Office of the President, Kenya Institute of Administration – support to civil service reform (CSR, KIA)

Ministry of Local Government – local government reform (PROLOGS/KLGRP)

Kenya Anti-Corruption Authority – anti-corruption policy (KACA)

South Africa

National Treasury – strengthening planning, budgeting and financial management (SERP I, SERP II)

Department of Public Enterprises – support to SOE restructuring (SERP II, SRPESA)

Zambia

Ministry of Finance and National Planning – support to public expenditure management (PEM) and to tax policy (RIZES - TPU)

Zambia Revenue Authority – organisational development, computerisation of direct tax, creation of integrated tax administration system (RIZES - ZRA)

Office of the President [management location but with system-wide impact] – payroll management and establishment control (PMECP)

2.16 It is important to note that the activities covered were not conceived or designed as comprehensive programmes of organisational capacity development for the organisations concerned, though several were implemented within the framework of wider programmes of organisational development support. The main exceptions were the support to the Ghana VAT Service and Zambia Revenue Authority, where substantial support was provided to core functions of the organisations. The support to public expenditure management involved a combination of long-term TC and the provision of consultancy resources to assist in the carrying out of studies. The focus was therefore technical (particularly systems development) rather than related to organisational capacity (and hence can be seen as based on an assumption that there was sufficient organisational capacity to absorb technical advice), as was the support to the Department of Public Enterprises and Department of Trade and Industry in South Africa (in the form of the provision of consultancy resources). The


15

support to the implementation of improved government systems (IFMIS in Kenya, payroll management in Zambia and Ghana) was intended to strengthen the effectiveness of management across the whole of government in each case, rather than the organisational capacity of the implementing agency. In Kenya, support to the Kenya Institute of Administration and the Kenya Anti-Corruption Authority was in both cases part of a wider donor effort that had a strong organisational development focus, though in both cases, as described below, lack of government commitment led to the termination of support.

2.17 Table 1 provides information on the size of the projects based on recorded expenditure. It should be noted that significant problems were encountered in compiling information on expenditures, reconciling information on expenditures from different sources, and relating information on planned and actual expenditure.

2.18 Table 2 lists the projects reviewed in each of the four countries and the broad objectives in terms of the roles of the organisations supported in the economic management process. In all four countries, there were DFID-supported activities that focused on core functions of the Ministry of Finance (public expenditure management and tax policy), in two of the countries support to development planning functions, and in three of the four countries there were activities supporting wider reform of government systems (e.g. civil service reform, payroll or integrated financial management) which were of wider significance for economic management. In three of the countries there was support to government agencies of a quasi-autonomous nature. The evaluation also covered some initiatives supporting privatisation or public enterprise reform processes and one activity focused on policy research outside government.


16

Table 1. Expenditure on the projects reviewed

Expenditure Ghana Kenya South Africa Zambia

< £0.5m PFM CSR, KIA, KACA, IP-ERS M&E

PEM

> £0.5 m, < £1.0m

NDPC IFMIS SERP I ZPA

> £1.0 m, < £5.0m

VAT, IPPD, PUFMARP

PROLOGS SERP II, SRPESA

RIZES, PMECP

Table 2. Areas of intervention of projects reviewed

Ghana Kenya South Africa Zambia

Support to improving public expenditure management and tax policy

MTEF PEM SERP I

SERP II

PEM

RIZES - TPU

Support to improvement of other government management systems

IPPD CSR

PROLOGS / KLGRP

IFMIS

PMECP

Support to quasi-autonomous agencies

VAT Service KACA

KIA

RIZES - ZRA

Support to privatisation / public enterprise reform

SERP II

SRPESA

Support to development planning functions

NDPC IP-ERS MED


17

Issues in applying the framework

2.19 The evaluation matrix used for the country case studies included assessments of the relevance, effectiveness, efficiency, impact, and sustainability of the activities evaluated, using the standard OECD Development Assistance Committee evaluation criteria.10

2.20 A number of challenges were faced in applying the evaluation approach in the case studies, which have implications for the interpretation of the results. These can be grouped into three main areas:

• The overall evaluation concept.

• The choice of case study countries.

• Access to information in relation to the resources available to the evaluation.

The overall evaluation concept

2.21 As noted above, the purpose of the evaluation was to “map and test the chain of results from DFID-funded TC personnel inputs to enhanced partner country capacity and better policies and service delivery.” Reflecting the terms of reference, the focus of the evaluation was on technical cooperation as a particular type of instrument to achieve capacity development, and on understanding how this input as provided by DFID can actually lead to improved capacity. The terms of reference also focused on the provision of “economic experts” as an instrument for building capacity. In practice applying this approach encountered a number of problems:

• The main evaluation criterion specified in the terms of reference (capacity development impact) did not relate clearly or consistently to the objectives of the activities that were identified as falling within the scope of the evaluation.

• The activities covered in the evaluation had a range of objectives, not all of which were focused on capacity development, and the activities did not have monitoring frameworks that sought to benchmark or measure capacity development impact as part of an overall strategic approach to capacity development.

• Very few activities involved specifically the provision of “economic experts” and were in fact extremely diverse in the forms of TC support provided, making it difficult to draw comparative conclusions about effectiveness given the small sample size in relation to the range of approaches and instruments.

• In general, the support examined involved a wider set of project activities of which technical cooperation formed a part, but often combined with other inputs such as the provision of equipment or financial resources for other activities such as training. The evaluation of the projects therefore involved a wider assessment than the evaluation just of TC as an instrument.

• It was not possible to carry out comprehensive organisational assessments as originally envisaged because of the relatively diffuse nature of the DFID programmes

10 See www.oecd.org/dac/evaluation


18

across organisations (in some cases), and the relatively marginal nature of DFID’s support to the organisations in others.

• The absence of an overall or a country level strategy for capacity development or for strengthening economic management, or for strengthening particular organisations, meant that it was difficult to formulate a purpose that would provide a point of clear point of comparison in constructing a counterfactual.

2.22 The case studies have therefore involved a compromise between an attempt to apply an approach focused, in line with the evaluation terms of reference, on elucidating the understanding of the causal mechanisms by which TC as a particular instrument can lead to capacity development, and the evaluation of a set of specific activities which differ considerably in terms of objectives, scale, and type of support provided. In practice it was possible only partially at best to examine the chain of causality implied by Figure 1 and identify wider impacts on economic management capacity or to determine the specific causes of success or failure.

Issues in country selection

2.23 As noted above, the choice of country case studies was predefined in the evaluation terms of reference. In interpreting the wider applicability of the findings of the evaluation, the following points should be borne in mind:

• The exclusion of countries in which other related evaluations were being undertaken (notably the joint evaluation of General Budget Support - GBS) may have led to some biasing of the country selection. It is understood that the evaluation was originally designed to be complementary to the GBS evaluation but in practice (partly because of timing) it was not possible to establish effective links between the two evaluations.

• None of the countries covered falls within the classification of “fragile states” as used for instance by the OECD Development Assistance Committee and only one (Kenya) is included in the list of 25 African countries on the DFID (2005b) “Proxy List of Fragile States.” In particular the country case studies did not include examples of any countries emerging from conflict –these may present different constraints and opportunities from those included as case studies. South Africa was selected in part to provide a contrasting example of a middle-income country in which capacity was likely to be relatively strong – conditions not replicated in other African countries in which DFID is active in providing TC.

• As noted above, the process of selecting case study countries used the criterion of the level of technical cooperation expenditure classified within the DFID Management Information System as in the economic and governance sectors as a proxy for TC related to capacity development for economic management. The more detailed review and classification of activities undertaken during the Inception Phase of the evaluation suggested that this criterion might not have been a very good proxy, since the share of this expenditure that was in fact identified as related to the core functions of economic management varied significantly between the case study countries.


19

2.24 Particular caution should therefore be exercised in using the findings of the evaluation as a basis for drawing wider generalisations about the performance of DFID’s TC.

Access to information

2.25 The information base for the case studies was constrained in several respects. First,the process of locating relevant DFID documentation on many of the activities (especially those that were completed in the earlier part of the period covered by the evaluation) was time-consuming and as noted above the interpretation of the project documentation in relation to the evaluation objective was made more difficult by the lack of a consistent focus on, or framework for, capacity development in DFID project monitoring and evaluation information. Second, a number of the activities falling within the scope of the evaluation were (or were parts of) large long-term programmes of organisational support, which could not be satisfactorily evaluated in their entirety. Third, the restriction to a single country visit for primary data collection11

necessarily led to gaps in the coverage of informants who could be interviewed (particularly from partner governments and some of the organisations that were recipients of TC but also from DFID offices in some cases). Locating individuals who were able to provide perspectives from within organisations receiving TC was a particular challenge for activities in the earlier part of the evaluation period (for instance before 2002 or 2003). Fourth, while the terms of reference specified the importance of engaging with a wider stakeholder group including civil society this was in practice difficult to do because there were few people outside government and the organisations supported who had had involvement with, or direct knowledge of, the DFID activities evaluated.

Implications for the interpretation of the case studies

2.26 These observations suggest the need for some caution in the interpretation of the case study findings for each country because of the limitations on the information sources available within the constraints of the exercise. It is however considered that the evidence base is sufficient to support the main conclusions drawn in each case study about the relevance, effectiveness, efficiency and impact of the activities reviewed. Caution is required in drawing any wider conclusions about the results achieved from DFID TC for economic management in Africa, or conclusions about the relative effectiveness of different modes of TC delivery from the case study evidence presented.

2.27 For detailed analysis of the Evaluation findings and conclusions and recommendations, see the Synthesis Report, sections 4 – 6.

11 The one case in which a follow up visit was undertaken (Kenya) showed the value of a second visit for

filling gaps in the evidence it is was possible to collect in the initial visit and the opportunity for a fuller discussion of the findings after report drafting.


1

GHANA CASE STUDY


2

Executive Summary

S1. This report is a case study of DFID-financed Technical Cooperation (TC) for economic management in Ghana focusing on the period 1999 – 2004. It evaluates the support provided to three organisations:

• The Ministry of Finance & Economic Planning (MOFEP),

• The National Development Planning Commission (NDPC), and

• The VAT Service.

Context for capacity development

Political context

S2. The evaluation period covered the end of the period of National Democratic Congress party rule in 2000, and the new National Patriotic Party government. This period can be seen as representing one of consolidation of democratic rule in Ghana, and there is evidence of a deepening of democracy including a strengthened role for Parliament, and a more active civil society and media. Despite these positive features, the political system retains a “neo-patrimonial” legacy from earlier periods. Significant constraints on change include the continuing importance of patronage throughout the political system. In some respects intensified political competition might make the situation worse if it leads to a shorter time horizon for policy making.

Institutional context

S3. Basic weaknesses in public sector functions and management practices themselves create major institutional obstacles to capacity development for economic management. While there are some positive developments, within the period covered by our evaluation it is difficult to identify clear-cut changes, which have served to improve the generally negative institutional environment within which TC projects have had to operate. Key problems continued to include:

• A generally demoralised work force, with a framework of incentives that does not promote initiative.

• A highly politicised civil service leadership.

• A confusion of political and civil service roles at the senior level.

• A weak system of controls, meaning that that there is generally weak accountability and that indiscipline and petty (and sometimes serious) corruption are common.

S4. Public sector reform efforts over the period are judged to have achieved little impact in addressing these problems and development assistance has in many ways tended to reinforce (through attempting to bypass) rather than solve these problems particularly through the payment of project-related salary supplements that have distorted incentives.


3

S5. Of the three organisations studied, MOFEP is the one that suffers most from these constraints, whereas the VAT service has suffered least – both because it has a more effective remuneration structure (using IRS scales) and has enjoyed more consistent leadership. NDPC lies somewhere in the middle. Without doubt, this has influenced the differing outcomes of the TC projects in the three organisations.

DFID’s strategy for TC in Ghana

S6. DFID’s country strategy has focused on improving economic management as an important objective and has noted the general institutional constraints on progress, but has not articulated an approach for addressing them and project design has also failed to capture the complexity and intractability of the context. The strategy has also not articulated a specific role for technical cooperation.

Economic management in Ghana

S7. Over the period covered by the evaluation there have been improvements in economic management but there remain major weaknesses. There has been an increase in coherence in, and ownership of, economic decision-making in the period following the introduction of the Ghana Poverty Reduction Strategy (GPRS). Nevertheless, there remain significant off-budget flows both from internally generated funds and from donor projects. Moreover, the gains appear fragile.

• Inflation has fallen from its peak of 39 % in 2000 to 11.8 % in 2004. This is a notable improvement but until single digit inflation can be sustained over a multi-year period, the inflationary threat cannot be said to be under control.

• Fiscal management has improved due both to a strengthening of revenue collection and to increased aid resources from budget support and from HIPC relief.

• Unanticipated wage expenditures and quasi-fiscal spending (due mainly to oil price subsidies) have continued to be a feature of budget management, and only once an automatic price adjustment formula is institutionally embedded can the latter problem be considered finally eliminated.

• Management of the public sector wage bill remains a major problem and the key source of deviations between approved budgets and out-turns.

S8. The joint IMF-World Bank Public Expenditure Management Country Assessment and Action Plan give some indication of improvements in public expenditure management. In particular, the assessment shows that Ghana fulfilled seven out of the 16 PEM benchmarks in 2004 as compared to only one out of 15 in the 2001 assessment. However, a closer examination suggests that many core areas of budget management, execution and reporting remain weak.


4

Results of the technical cooperation: organisational capacity for economic management

S9. In relation to the achievements of the TC provided, the evaluation noted significant changes in two organisations – the NDPC and the VAT service. Here, DFID-financed TC appeared to have contributed in a significant manner, although in relation to NDPC there were question marks over the sustainability of its impact. By contrast, within MOFEP, despite considerably higher levels of TC expenditure, rather less change had been achieved. There were some essentially ‘transactional’ improvements but no clear-cut transformational impact.

Results of the technical cooperation: summary of the evaluation results

Relevance

S10. The support to the VAT Service was the most relevant in terms of the understanding of organisational needs and operational context. This was helped by the fact that this was a new organisation and issues relating to staffing and salaries were addressed. There was strong political commitment and commitment from other development partners, and an extensive preparation process involving a number of donor agencies. In the case of NDPC, the capacity development needs were identified but probably underestimated, and the limited status and influence that the NDPC had at the start of the project was not fully appreciated. The process of preparing the Ghana Poverty Reduction Strategy (GPRS) presented an opportunity that was seized by the NDPC Director General, and the provision of TC was responsive to these needs.

S11. MOFEP presented a much more difficult context. The initiatives to establish a medium-term expenditure framework (MTEF) and the Integrated Personnel and Payroll database (IPPD) initiatives were identified in the earlier support to public financial management reform (PUFMARP), but the specific technical form these interventions took did not address basic issues underlying the wage bill problem and this was not fully appreciated until significant sunk costs had been incurred. Neither project appreciated or addressed the problems associated in paying allowances. Disputes over allowances and design problems weakened the commitment of government.

Effectiveness

S12. Support to the NDPC and the VAT service were in general highly effective in generating the intended outputs with strong ownership from the supported organisation and close harmonisation and alignment with the activities of other development partners. Less skills transfer was achieved under the NDPC support than was envisaged, and full functionality of VIPS was not achieved under the VAT Service support.

S13. In relation to MOFEP support, of the ten outputs specified in the original project documents for the MTEF project, five were assessed as fully achieved, three partially


5

and two not at all. However the coverage of MTEF was insufficient and the outputs were not appropriate to achieve the project purposes. IPPD2 remains incompletely implemented at the time of the evaluation.

Efficiency

S14. Both the NDPC and VAT Service support delivered most of their outputs cost-effectively and within the intended time schedule. The delivery of outputs for the MTEF project was largely on schedule if somewhat more expensive than intended. IPPD2 was exceedingly inefficient with many of the norms of IT project management best practice being broken by the contractors, DFID and the Government of Ghana.

Impact

S15. The NDPC project succeeded in bringing a coherence and focus to economic management, in raising the importance of poverty reduction as a government objective, and in enhancing the NDPC’s influence. The VAT Service project played an important role in establishing a viable, self-sustaining VAT collection service. The MTEF project has had a transactional impact in changing the way in which the budget is prepared, but it did not achieve its project purpose of improving the cost effectiveness of government resource allocation. IPPD2 remains incomplete after seven years. It could be argued that it has laid the groundwork for an effective payroll system but this has yet to be proven.

Sustainability

S16. Five years after the end of the main part of the TC project, impacts on the VAT Service have been sustained and VAT performance continues to improve. In the case of NDPC, transformational impacts have been sustained so far, but NDPC does not have the capacity to sustain the GPRS system without continuing TC or organisational reforms that have yet to occur, in order to address problems of staff retention and provide greater management stability. The transactional changes introduced by the MTEF project have been sustained, though had the project been better designed in relation to the context, greater success could have been achieved. In neither the MTEF nor the IPPD2 projects have there been effective exit strategies with both projects being prematurely closed, leaving weak organisational structures for the maintenance of the project outputs and the continued pursuit of their objectives.

Conclusions and implications

S17. The findings show overwhelmingly the importance of the institutional context to the success of Technical Co-operation. In Ghana, very substantial TC resources were expended without an adequate prior analysis of the institutional context and without an adequate understanding of the interactions of organisational, institutional and political economy factors. On the other hand, TC projects can achieve some successes in a poor institutional environment, if there is the right leadership and organisational


6

capacity and if the project design is well focused although the sustainability of such successes is likely to remain in doubt. Ghana has a particularly difficult combination of institutional and political economy factors with which to contend, yet some of DFID’s TC projects did achieve success in this difficult environment.

S18. Good inter-personal and ‘process’ skills by the TC provider are crucially important. Clearly, the right technical skills are also necessary if new systems are to be developed and skills transferred but technical skills alone will not be effective. The assumption that local consultants are likely to be more sensitive to the requirements of the operating environment and/or more easily accepted by the government counterparts than international consultants may also be questioned.

S19. Salary supplementation created serious tensions and distortions within the Government agency concerned, as well as undermining the joint Government of Ghana (GoG)-DFID management of these projects and failing to address effectively the problem of staff retention and incentives.

S20. The findings suggest the need for a more systematic approach by DFID to taking account of the institutional and organisational context in project design, as well as a strengthening of DFID’s approach to project management for such activities.


7

1. Introduction

1.1 This case study is one of four (the others being Kenya, South Africa and Zambia) in a wider study to evaluate the role of DFID TC in Economic Management in sub-Saharan Africa. The purpose of the wider study is:

“To map and test the chain of results from DFID-funded TC personnel inputs to enhanced partner country capacity and better policies and service delivery. In doing so, lessons will be drawn on the best ways to provide TC personnel in different contexts in order to maximise effectiveness, efficiency and impact on partner country capacity.”

1.2 The overall approach and conceptual framework for the country studies is discussed in the Synthesis Report (Volume 1). Four criteria were applied in the selection of the projects (Table 1):

• Each of the projects had to be significant in terms of the relative importance of the economic management functions which they were supporting.

• In keeping with the evaluation methodology, it was considered important to focus on projects which were clearly rooted within specific organisations – rather than providing more general advisory support to the whole of government.

• Each of the projects should embody explicit transformational objectives in relation to the organisations within which they were rooted; and

• The projects should cover a range of organisations so as to provide the possibility of considering the influence of the organisational setting as an independent variable interacting with institutional conditions common to the public sector as a whole.

1.3 The choice of projects was also limited by the need to undertake fieldwork with a two-person rather than a three-person team due to the last minute illness of a team member.12 It was our intention to focus on the five-year period from 1999 – 2004 as the basis for drawing judgements on the extent of capacity development in economic management. There is easily available data on this period and it is sufficiently recent to fall within the ‘recall period’ of the government officials and other key informants who might provide information on trends in capacity development and on the contribution of DFID-financed TC projects.

1.4 For the most part, it was possible to follow this intention. However, three of the chosen projects started before this period - support to VAT implementation, to PUFMARP and to IPPD Replacement. Where relevant the consultants have therefore referred to activities undertaken in the early phases of these projects and to the reasons underlying the initial design of these projects.

1.5 It should be stressed that in the context of a relatively short country visit, it was not possible to interview all of the key stakeholders or to obtain a complete set of relevant information. However the assessment of the consultants is that the

12 The main effect of this was that it was not possible to cover the Public Sector Reform activities as had

originally been intended.


8

information obtained was sufficient for valid empirically-based conclusions to be drawn.

Table 1. Organisations and projects selected

Organisation Projects to be Evaluated Summary of Justification

Ministry of Finance & Economic Planning (MOFEP)

Support to PUFMARP

Integrated Payroll & Personnel Database (replacement & deployment)

Support to MoF in Financial Management Reform

MOFEP is the core organisation responsible for economic management.

The three DFID TC projects embrace budget preparation as well as budget execution functions across a number of the divisions of MOFEP, including the Controller and Accountant General’s Department (CAGD).

National Development Planning Commission (NDPC)

Poverty Monitoring assistance

Ghana PRS monitoring & analysis project

The NDPC is the second core organisation in economic mgt. DFID TC focused on the development of a monitoring capacity for the Ghana PRS, a central element in national policy and in GoG-Donor relations.

VAT Service Value Added Tax implementation The VAT Service operates as a semi-autonomous agency under the RAGB. Introduction of VAT has been crucial to increased domestic revenue & hence to macro stability.

1.6 Table 2 summarises total spending by DFID Ghana on TC projects within the economic and governance sectors. It is based on the listing of TC projects provided by the Evaluation Department and DFID Ghana as the basis from which to select relevant projects for evaluation. As can be seen, it represents a relatively high proportion of total spending on the DFID Ghana programme, underlying its importance within the DFID Country Assistance Plans. Both the 2003 Country Assistance Plan and its predecessor included ‘assisting improvements in public financial management’ among their main objectives.


9

Table 2. DFID TC project expenditure in Ghana – economic and governance sectors

Fiscal Year (April - March); £ 000s

Pre

1999/2000 1999/2000 2000/1 2001/2 2002/3 2003/4 Total 99-04

NDPC

Poverty Monitoring Assistance 96.0 201.3 16.1 313.4

Ghana PRS Monitoring & Analysis Project 191.0 111.5 302.5

VAT Service

Value Added Tax Implementation 1,472.2 347.0 221.6 126.2 49.0 743.8

MOFEP

Support to PUFMARP 1,146.2 660.7 649.0 128.1 238.6 1,676.4

Integrated Personnel & Payroll Database 112.6 457.6 564.9 16.7 13.6 1,052.8

IPPD 2 Deployment Project 0.3 85.9 86.2

Support to MoF in Financial Management Reform 69.4 78.2 147.6

Sub-total TC projects evaluated 2,731.0 1,465.3 1,531.5 472.3 578.0 275.6 4,322.7

(as % of Total Country Programme) 3.1% 2.1% 0.9% 1.0% 0.4% 1.4%

Public Sector Reform

Civil Service Performance Improvement Phase II 2,029.0 1,465.3 1,199.0 459.6 37.0 128.8 3,289.7

Value for Money Assessment Services 16.9 218.3 208.9 292.8 736.9

Subvented Agencies Reform Project 5.1 546.4 698.6 587.1 207.1 2,044.3

Public Sector Reform Review 83.2 83.2

Ghana Statistics Service

Economic Statistics Phase II 656.8 27.6 61.9 89.5

Ghana Census 2000 91.5 1,425.9 9.7 1,527.1

Trade and Investment

Ghana Trade Policy Project 1.6 0.1 220.0 348.4 135.3 705.4

Balance of Payments

Programme Aid (TC) 1999 11.7 11.7

Programme Aid (TC) 2000 26.2 20.3 46.5

Debt & Capital Flows Project 64.4 23.9 43.9 107.2 239.4

Preparation for Multi-Donor Budget Support 72.7 72.7

Sub-total TC projects not evaluated 2,685.8 1,619.7 3,542.2 1,641.0 1,309.2 734.3 8,846.4

Total TC Economic & Governance Sectors 5,416.8 3,085.0 5,073.7 2,113.3 1,887.2 1,009.9 13,169.1

Total Country Programme 48,000 73,000 55,000 56,000 68,000 300,000.0

TC for Economic & Governance as % CP 6.4% 7.0% 3.8% 3.4% 1.5% 4.4%


10

2. The context for capacity development

The political context

2.1 The main socio-political features of the Ghanaian state have been identified elsewhere13 and these provide important elements of the context for economic management and capacity development.

2.2 Booth et al. (2004, pp. 6-10) stress that the Ghanaian state continues to have a ‘neo-patrimonial’ character. In other words, instead of the state being an instrument governed by legal rules and explicit policy objectives, it becomes an apparatus serving the interests of the patronage networks that have captured it. In short, the state loses its policy autonomy, its ability to conduct policy focused on the public good. There are two reasons why neo-patrimonial tendencies have been more deeply embedded:

• First, the ‘neo-mercantilist’ model which Ghana inherited from the colonial period, based on the control of exports (cocoa, gold and timber) and imports placed a huge premium on state capture, simply because there were so few sources of wealth which were outside of the control of the state.

• Second, the heterogeneous nature of the state created by colonial rule created a wide variety of forms of social identity or loyalty that were available to be mobilised politically.

2.3 Thus, there were, on the one hand, avenues for political competition, which did not exist in other more homogenous societies, and, on the other hand, great gains to be had from holding political supremacy. The combination of these factors created a post-Independence State with deep neo-patrimonial tendencies, which were manifested in the competition for control of the state witnessed over the first 20 years of independence.

2.4 How far do these historical tendencies persist? Clearly, the advent of economic and political liberalisation has fundamentally altered the political economy but there are signs that patronage remains very important. The provision of public sector employment is an obvious source of patronage and the failure to introduce tight controls over staffing levels and to reinforce a strictly meritocratic system of appointment suggests that it continues to be so. When first elected, the NPP pledged to reduce the number of ministers and deputy ministers, yet it is now higher than under the NDC government. Similarly, it made a campaign pledge to abolish presidential powers of appointment over District Chief Executives, which to date it has not been possible to fulfil. There may be practical or legal reasons why the

13 This section draws in particular on the Drivers of Change Overview report, commissioned by DFID (Booth

et al., 2004), on the comparative analysis of public sector reform in Ghana, Tanzania and Zambia of Stevens and Teggemann (2004) and on the Government of Ghana’s own policy document, “Towards a New Public Service for Ghana” (2004).


11

Government has retained these powers of appointment. But one possible interpretation is that under pressure from its different supporting factions, the Government has found it difficult to quickly force through implementation of its campaign pledges.

2.5 There have been some fundamentally important reforms put in place in Ghana over the past 20 years and the NPP government has made progress in several areas of importance. There are however important constraints on change:

• Within a system that places a high degree of importance on patronage, close discretionary control over appointments and over budgetary allocations is highly valuable to the political class. Steps to limit the degree of political control or to increase its transparency will be resisted.

• When there is sharp political competition – both between parties and amongst the internal factions of parties, the need for patronage to retain party unity and expand support is still stronger.

• The ineffectiveness of the civil service is a major constraint to the effectiveness of government and, as such, a potential vote loser. On the other hand, reforms are only likely to bring gains in the medium to longer term and yet will entail significant short-term costs.

• The four-year electoral cycle places a premium on visible, quick fix improvements and not on longer-term structural reforms.

• Certain types of reforms may avoid short-term costs but in so far as staff reductions were considered necessary to free up fiscal space for salary improvements, then short-term costs would be inevitable.

• Despite the poor pay and conditions, those within the civil service represent a powerful obstacle to change because many of them have found ways of profiting from existing arrangements and will certainly have doubts about their ability to be successful within the private sector.

2.6 The evaluation period coincided with the last NDC government and with the first two NPP governments. The key political events within this period are summarised in Tables 3 and 4. The tables also try to give an indication of the significance of these events. Most notably, this period has seen the consolidation of democracy in Ghana. 1996 saw the first properly contested election since the restoration of democracy; 2000, the first election in Ghana’s history leading to a democratic change of government; and in 2004, the fourth set of elections under the 1992 Constitution. In other respects too, there are reasons to believe that democracy has deepened: civil-military relations have been democratised, a vibrant independent media and civil society have emerged and Parliament has become more involved in law-making, deliberation over policy, executive oversight and scrutiny over international agreements. (Booth et al., 2004, pp 20-21).


12

Table 3. Key political events and their significance (1992-2000)

DATE KEY POLITICAL EVENT SIGNIFICANCE

Dec 1992 Presidential and Parliamentary elections for the 4

th Republic with the New

Patriotic Party (NPP) boycotting the Parliamentary elections

First democratic elections after over 10 years (1982-1992). Change over from military regime to democratically elected government.

7 Jan 1993 Inauguration of the 4th Republic

1995 Change of Minister of Finance from Kwesi Botchwey to Kwame Pepra

Change from long-established Minister who had led many of the economic reforms.

1995 Political riots - “Kumi Preko” as a result of the introduction of the VAT

VAT suspended. Government hope of meeting economic management requirements shattered.

Dec 1996 2nd

Presidential and Parliamentary elections of the 4

th republic

Heightened confidence of the international community in Ghana’s democratic governance and economic management.

1996 Vision 2020 the First Step approved by Parliament

Provided framework for coordinating development programmes in Ghana.

Jan 1997 Inauguration of the 2nd

session of the 4th

Republic

1997 Establishment of Public Utilities Regulatory Commission (PURC).

Gradual progress in introducing automatic tariff adjustments for electricity, water, etc to cover costs.

Dec 1998 Reintroduction of VAT VAT reintroduced successfully at a rate of 10%. Rate increased to 12.5% in June 2000 and to 15% in August 2004 as a National Health Insurance Levy.

1999 Major terms of trade shock: prices of gold and cocoa plunged, while petroleum prices increased. Delays by government in adjustment of domestic petroleum prices and electricity rates.

Led to sharp deterioration in macro performance including build up in external arrears. Also resulted in accumulation of large bank debt by Tema Oil Refinery.

Dec 2000 3rd

Presidential and Parliamentary elections of the 4

th republic

NPP winning the elections and changing the political landscape thus strengthening donor and investor confidence in Ghana.


13

Table 4. Key political events and their significance (2001-5)

The institutional context

2.7 The above analysis has shown some of the important constraints to reform. Unfortunately, the basic weaknesses in public sector functions and management practices themselves create major institutional obstacles to capacity development for economic management. Stevens and Teggemann (2004) review the progress of public sector reform over the period to 2003 and note the limited progress that was made in major reform areas. Their conclusions are summarised in Table 5. The Government’s recent publication, “Towards A New Public Service for Ghana” (GOG, 2004) provides a frank assessment of the problems. The tone of this document, in particular the honesty of its analysis, and its strongly expressed desire to initiate a Ghanaian response to the problem are positive signs of a renewed political concern to address these problems. The appointment of a well-respected new Minister for

DATE KEY POLITICAL EVENT SIGNIFICANCE

Jan 2001 Handing over by the National Democratic Congress (NDC) to the NPP

Smooth transition from NDC, which had been in power for a cumulative period of 19 years.

2001 Imprisonment of former Deputy Minister of Finance (of NDC govt) and Minister of Youth and Sports for causing financial loss to the state

Demonstration of government commitment to zero tolerance for corruption.

2002 Imprisonment of former Minister of Finance & legal advisor (of NDC govt) for causing financial loss to the state.

Demonstration of government commitment to zero tolerance for corruption.

February 2003

GPRS approved by Parliament Mapped out strategy for promoting growth and reducing poverty in line with MDG targets.

2003 NIRP dissolved. Senior Minister assigned oversight of the Public Sector Reform Programme

Acknowledgement of failure of NIRP.

2003 Ghana accedes to African Peer Review Mechanism under NEPAD.

Ghana first country to volunteer to the peer review mechanism, permitting independent public critique of the government.

2004 HIPC completion point Debt relief amounting to US$2.2 billion in net present value terms.

Dec 2004 4th Presidential and Parliamentary

elections of the 4th republic

Sustained democratic governance and heightened investor and donor confidence.

7 January 2005

2nd

term of NPP government Change in Ministerial portfolios including the Minister of Finance.

April 2005 Energy Commissioner removed over alleged mismanagement of state funds

To pursue government commitment to zero tolerance for corruption.

Feb – June 2005

Street protests – dubbed “Wahala Demonstrations” over fuel price hikes of about 100%.

Demonstration of capacity for social mobilisation and of electoral risks the government needs to take to maintain sound economic management.


14

Public Service is also a positive sign. Together these may suggest that the negative political costs of an ineffective public service are becoming more apparent and may perhaps be changing the underlying political incentives for reform. However, clear-cut improvements are hard to identify over the period covered by the evaluation.

Table 5. Public sector reform and intent in Ghana

Reform Intent Results so far Assessment

Dissolve hidden reward systems

Introduce selective accelerated salary enhancement (SASE)

Studies conducted

Backtracking from SASE

Real pay decreasing (except 1998 and 1999)

No reforms under the Public Sector Management Reform programme

Deteriorating reform record

Perform functional analysis

Reform State Agencies (SA)

Retrench 28,000 civil servants

Decentralize

Functional reviews commissioned for 22 ministries

SA staff still on public payroll. No coherent overall SA strategy

No real reduction in public service because civil service moved to Sas

No progress on decentralisation

Reforms barely went beyond studies

Develop sanctions and reward systems

Establish performance-based management system

None No progress

Link MTEF to performance management

Expand coverage of existing integrated personnel and payroll database

Perform stakeholder analysis and consultation

MTEF operational since 1999 but not linked to performance management

IPPD covers only part of public service

Social assessments and stakeholder analysis conducted

Early progress on expenditure and administrative controls stalled

Some stakeholder participation but not systematic

Retrain 1200 civil servants Some training undertaken Long-term needs not addressed

Source: Stevens and Teggemann (2004)


15

2.8 Key problems that therefore continue to constrain capacity development include the following:

• A generally demoralised work force, with a framework of incentives that does not promote initiative.

• A highly politicised civil service leadership, more concerned to keep political masters happy than to manage and motivate staff towards a common vision.

• A confusion of political and civil service roles at the senior level, such that it is not clear who is in charge and who holds responsibility when things go wrong.

• A weak system of controls, meaning that that there is generally weak accountability and that indiscipline and petty (and sometimes serious) corruption are common.

The role of development cooperation

2.9 Ghana has been a major recipient of debt relief and other sources of aid financing and has generally enjoyed good relations with its Development Partners as the calendar of major ‘aid events’ shown in Table 6 demonstrates.

Table 6. Major aid events and their significance

DATE MAJOR AID EVENT SIGNIFICANCE

1999 Piloting of the Comprehensive Development Framework (CDF)

Ghana among first countries to pilot the CDF. Moved Consultative Groups, traditionally held in Paris, to Accra.

1999 Pioneering of successful SWAp in health.

Providing opportunity for piloting common basket funding arrangements.

Feb. 2002 HIPC Decision Point reached Initiation of debt relief process.

2002 Poverty Reduction Growth Facility (PRGF) declared off-track

Serious fiscal and quasi-fiscal slippages in 2002 prevented the completion of the final review of the PRGF supported programme.

February 2003

GPRS approved by Parliament Mapped out strategy for promoting growth and reducing poverty in line with MDG targets.

12 May 2003

IMF Board Approves a 2nd

Poverty Reduction Growth Facility (PRGF) arrangement

IMF recognised improvements since 2002 and renewed potential for strong prudent fiscal & monetary management.

2003 MDBS Implementation Framework agreed with Development Partners

Development Partnership strengthened. Dev’t Cooperation shifting from projects to sector support and multi-donor budget funding.

2004 Ghana reaches HIPC Completion Point Debt relief of US$2.2 billion NPV to reach targeted debt-fiscal revenue ratio of 250%.

June 2005 Over $40 billion debt relief for 18 poor countries including Ghana by the G8.

Ghana expects to derive about $4.1 billion debt relief, removing some 80% of the debt owed by Ghana, thus freeing funds for the implementation of the GPRS.


16

2.10 From the early 1990s to the present, a substantial volume of external resources – from the World Bank, and DFID in particular - has been dedicated in different ways to the support of public sector reform. Booth et el (2004) conclude quite unequivocally that this had limited impact in the absence of a clear domestic commitment to reform:

“…Public service reform is a good example of the way even quite substantial efforts by a range of agencies can bear little fruit where the domestic political logic produces an unfavourable balance of political costs and benefits.” (Booth et al., 2004, p.33)

2.11 The Government’s own assessment of the impact suggests that part of the problem lay in the dynamic of aid projects:

“…The cardinal lesson had been learnt from past initiatives, that short-term “projects” of the type usually funded by donors for the country, did not have a suitable perspective for the sort of permanent, long-term changes required for the reform of a public service which had touched the depth of weakness.” (GOG, 2004, p. 9)

2.12 Interestingly, this is the only example that was found of a formal government document, which expressed a negative view of the use of aid projects to support reform. Ghana does not have a formal policy on Development Cooperation. Our discussions with the External Resources Management Department of MOFEP did not suggest that there was even an informal policy, regarding, for example, the use of TC or the relative balance of projects, SWAp arrangements and budget support.

2.13 At an individual level, several Ghanaian government officials did, however, express negative views over the sorts of incentives generated by aid projects and of their potential negative effect on the institutional environment within the public service. The views expressed are well substantiated in the wider literature on aid effectiveness as well as in several Ghana-specific reports (Booth et al., 2004; GoG, 2004):

• Virtually all aid projects in Ghana have since the 1990s incorporated some type of ‘salary supplement’ for the government civil servants associated with the project. This might be in the form of direct salary payments to specific staff designated as project staff or as ‘local consultants’, or significant allowances for participation in steering committees or workshops.

• In many parts of the civil service, the number of projects and the significance of the associated allowances has been so great that this has become a dominant motivation for ‘winning’ projects and has often drawn attention away from potentially more important civil service business.

• In many cases, this has generated significant conflict between those government staff deriving pecuniary benefits from projects and their colleagues on normal civil service pay.

• Many civil servants believe the widespread existence of these project ‘top-ups’ has made it possible to defer the introduction of much needed across-the-board improvements in civil service terms and conditions.

• Civil servants did also speak of certain aid projects as sources of innovation and ideas as well as both formal and on-the-job training. These contributions were clearly


17

valued but there is no doubt that for the most part aid projects were seen as part of the ‘institutional problem’ rather than part of the solution.

DFID’s strategy for technical cooperation

2.14 DFID’s Country Strategy Paper (CSP) for the period 1998-2001 was focused on three main objectives, aligned around the national development goals set in the Vision 2020 policy document:

• Concentration on a limited number of sectors to provide better (mainly social) services.

• Supporting the mainstreaming of the poverty strategy in all budgetary planning.

• Supporting sound macroeconomic and fiscal engagement through programme aid.

2.15 A review of DFID’s implementation of the CSP (quoted in DFID, 2003) noted that DFID “failed to anticipate the weak implementation of key reforms by the Government, most notably on decentralisation, public sector reform and public expenditure management.” In response DFID noted that its new Country Assistance Plan (CAP) covering the period from 2003 to 2006 needed a greater emphasis on the “central reform agenda on public financial management and public sector reform” and that this needed to be better integrated with complementary activities at sectoral and decentralised levels, and that there was a need “to engage civil society, parliamentarians and others as change agents in the central reform agenda” with this envisaged as taking place through the consultations on the GPRS and the MDBS.

2.16 The 2003 CAP was focused on aligning to the Ghana Poverty Reduction Strategy with the objective of having an increasing share of resources provided through general budget support, with a continuing focus on assisting improvements in public financial management. Issues of economic management were identified as key risks for the programme (specifically that political priority is given to expenditures that do not benefit the poor, that there could be a failure in macroeconomic management, and that capacity constraints prevented the GPRS being delivered). The CAP does not however identify actions to address capacity constraints either in general or in relation to DFID financed activities. Nor does the CAP discuss issues relating to the use or role of TC as a particular aid instrument.

2.17 A feature of the evaluation period (as shown in Table 1) has been the relative decline in the proportion of TC spending (and an absolute decline in the total) as a share of the total DFID country budget. In part, this represents a deliberate policy to ‘move an increasing proportion of DFID’s resources for the Ghana Country Programme into Multi-Donor Budget Support (MDBS)’ (DFID, 2003). It also reflects the intention to transfer financial support to PFM improvements from the existing bilateral arrangements to a multi-donor technical assistance fund, to which all the MDBS donors are expected to contribute. Our discussions with government officials also suggested that it is in part a reflection of the declining GoG demand for TC projects.


18

Summing up

2.18 The institutional environment for the operation of DFID’s TC support to economic management has clearly not been favourable over the period covered by the evaluation and is likely to have been less favourable than in many Anglophone Sub-Saharan African countries14. Three facets of the ‘institutional problem’ seem particularly pronounced in the Ghanaian case:

• The degree of politicisation of the civil service, which has not only undermined its autonomy but also significantly confused leadership responsibilities.

• The degree of demotivation of the civil service, particularly within the central agencies.

• The extent to which projects have come to be used as a source of salary supplements, to the detriment of the wider incentive environment.

2.19 Of the three organisations studied, MOFEP is clearly the one that suffers most from these factors, whereas the VAT service has suffered least – both because it has a more effective remuneration structure (using IRS scales) and has enjoyed more consistent leadership. NDPC may be said to lie somewhere in the middle. Without doubt, this has influenced the differing outcomes of the TC projects in the three organisations.

2.20 On the positive side, it is clear that Ghana does enjoy higher levels of human capital development than the majority of countries of Anglophone Sub-Saharan Africa. This is due to high levels of enrolment in secondary and tertiary education, as well as due to the positive influence of returning Ghanaian expatriates. If the structure of incentives could be established by which to capitalise on this advantage, Ghana could expect to have a highly efficient, well performing civil service but such changes would need to be domestically led.

2.21 There are some recent promising signs that there may be a more sustained political effort to reform the public sector, but there is little tangible progress to date, beyond the development of a strategy. Moreover, there are reasons to believe that the political economy factors generating these institutional weaknesses remain powerful.

2.22 However, none of the project documentation for the TC projects being evaluated served to capture the complexity and inherent intractability of the institutional context. It is clear that at a personal level there was a much higher awareness of these problems among the DFID staff responsible for these projects and among the few consultants it was possible to interview. This awareness seems to have been gained through the hard lessons of implementation, and certainly the rate of learning has increased with the establishment of the devolved DFID Ghana office. Nevertheless, the consultants found limited evidence of such an awareness being deployed during the diagnostic or design phases of these projects, when it might have helped to shape their direction and mode of delivery.

14 This assertion is based essentially on the personal experiences of the authors in Botswana, Namibia,

Tanzania, Uganda and Zambia but it is supported by comparative analysis of Ghana with Tanzania and Zambia presented in Stevens and Teggerman (2004).


19

3. Economic management in Ghana

Overall economic performance

3.1 Since independence Ghana has had a mixed record of growth and economic management. During the 1950s and early 1960s Ghana experienced reasonably high GDP growth, followed by two decades of turbulence and unsustainable economic management practices. Following this period of instability, a series of structural reform and stabilisation policies were consistently pursued in the 1980s and 1990s. Ghana’s economic performance began improving in the mid-1980s, as result of better macroeconomic management and a favourable external environment. By the end of the 1980s, Ghana was being declared an economic success story and held up as a model reformer.

3.2 In contrast, following the re-establishment of democracy, the 1990s was a decade of high inflation, substantial and regular currency depreciation and unstable growth. Fiscal performance was marked by a persistent and substantial fiscal deficit. Revenues and grants stagnated in the mid 1990s, whereas public expenditure as percentage of GDP continued to rise. During the 1991-95 period, the deficit was largely financed externally or through government divesture proceeds. After 1995 they were largely financed domestically, creating new and highly burdensome recurrent cost commitments, which the NPP government inherited on assuming power in January 2001.

3.3 Historically, economic mismanagement, most particularly an inability to maintain a controlled rate of inflation, as well as cyclical and external factors, and slow progress on structural issues, have constrained economic growth. The growth performance improved after 1983, averaging 5.4% p.a. over 1983-91 and 4.4% in the decade to 2001; as a result there were significant improvements in the poverty indices. (Booth et al., 2004) However, if one considers that Ghana’s per capita income at Independence was similar to that of South Korea, Malaysia and Thailand, the performance looks rather poorer. Fundamentally, the economy has continued to rely on three export commodities – timber, gold and cocoa – and on a small formal sector, in which the public sector has accounted for over half of wage employment.

Macroeconomic management (1999-2004)

3.4 In 1999, Ghana suffered a major terms-of-trade shock when prices for its two main exports, gold and cocoa, plunged, while petroleum prices increased. The effects were exacerbated by the delay in adjusting domestic petroleum and electricity prices, as well as fiscal and monetary policy. The medium-term strategy set out in 1999 envisaged real GDP growth rising to 6 percent, inflation falling to 5 percent, and gross international reserves reaching 3 months of import cover. Instead the next two years saw erratic fiscal and monetary policies and persistent delays in structural reforms, so that many of these targets were not reached.


20

3.5 Fiscal and monetary discipline faltered in the run up to the 2000 elections, presenting the new government with a difficult post-electoral economic situation. Serious fiscal and quasi-fiscal slippages also prevented the completion of the IMF’s final review of the Poverty Reduction and Growth Facility (PRGF). In January 2001, a new government took office in Ghana, the first democratic transfer of power in the nation’s history. The incoming administration inherited a rapidly depreciating currency, sharply rising inflation, burgeoning public debt, and a substantial depletion of foreign exchange reserves. In late 2002 and early 2003, government redoubled reform efforts by undertaking a number of measures to improve their fiscal situation. On the strengths of the reforms undertaken, the IMF approved a second PRGF arrangement on May 12, 2003.

3.6 The Ghana Poverty Reduction Strategy, approved by Parliament in February 2003, mapped out a medium-term strategy for promoting growth and reducing poverty. Ghana reached HIPC completion point in July 2004 allowing it to access US$2.2 billion of irrevocable debt relief in net present value terms.

3.7 The Ghanaian economy appears to have turned the corner in 2001 with real GDP growth taking an upward turn from the declining trend seen since 1998. There was an upsurge in growth in 2001 due to a recovery in agricultural production aided by improvement in macroeconomic management. This also led to a reduction in inflation, interest rates and to some stability in the foreign exchange rate of the cedi. Over 2001 and 2002, government followed a policy of building up a ‘cushion’ of international reserves in response to the large reserve drawdown in the wake of the external trade shock in 2000. In 2003 and 2004, improved fiscal management coupled with a record cocoa harvest and higher world prices for cocoa and gold, resulted in a more stable macroeconomic environment and sustained growth of over 5% p.a.

Table 7. Macroeconomic indicators (1997 – 2004)

1997 1998 1999 2000 2001 2002 2003 (e) 2004

(p)

Gross Domestic Product (in billions of

Cedis)

14,113 17,296 20,580 27,153 38,071 48,862 66,158 78,650

Real GDP growth rate 4.2% 4.7% 4.4% 3.7% 4.2% 4.5% 5.2% 5.2%

Inflation (% change) 18.4% 16.3% 13.1% 39.3% 23.5% 14.1% 24.0% 10.8%

Exchange rate (Cedi-US dollar) 2050 2314 2669 5455 7170 7932 8677 9004

Percentage depreciation -20.0% -11.4% -13.3% -51.1% -23.9% -9.6% -8.6% -3.6%

Stock of domestic debt (% of GDP) 28.2% 28.9% 26.8% 26.2% 19.6% 17.2%

Trade Balance (% of GDP) -17.9% -10.8% -16.0% -16.5% -18.2% -10.8% -12.0%

Current Account Balance -8.4% -5.3% 0.5% 1.7% 0.3%

(e) - Estimated, (p) - Projected

Source: IMF


21

3.8 The newly elected government made considerable progress in stabilising the economy in 2001. Monetary policy in 2001 focused on reducing the rate of inflation and the rate of depreciation of the cedi. This followed the turbulence in 2000, which saw the rate of inflation reach 40 percent and the cedi depreciate by around 50 per cent.

3.9 The improved fiscal and monetary mix allowed an easing in inflation, following a spike in February 2003 of 13 percentage points in the aftermath of the 95 percent hike in prices of petroleum retail products. Lower levels of domestic public debt and declining inflation allowed interest rates to fall to an annualised rate of 16 per cent, by early January 2004.

Fiscal sustainability

3.10 Aggregate fiscal discipline has improved since 2002. Including grants, the overall fiscal deficit declined to 4.5 percent of GDP in 2003, down from 6.8 percent in 2002.

3.11 Total revenue performance has improved since 1999 and more so after 2002, especially with regards to indirect taxes and trade taxes. Revenue increased by 6 percentage points to 23 per cent of GDP between 2002 and 2004. Donors financed close to a third of the budget and fluctuations of aid flows are a source of fiscal risk. Net aid jumped in 2001, fell in 2002, and jumped again in 2003. Domestic debt, which rose from one per cent of GDP in 1990 to 21 percent in 1995, reached 29.6 percent in 2002. There was no net domestic financing of the budget in 2003, implying a sharp reduction of domestic debt relative to GDP to an estimated 19.6 per cent by end 2003.

Figure 1. Expenditures and revenues, 1991-2004 (% GDP current prices)

Source: IMF

0%

5%

10%

15%

20%

25%

30%

35%

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Total Revenue and Grants Total Expenditure and Net Lending


22

Expenditure management

3.12 Aligning resource allocation with the country’s strategic priorities has been a key challenge in managing the budget in Ghana, especially in the run up to an election. This was the case in the periods leading up to both the 1996 and the 2000 elections. Current expenditure increased by 3 percent of GDP from 1999 to 2001, after which it stabilised to around 20 percent of GDP. HIPC debt relief enabled a declining share of interest payments after 2001. The fiscal space this created permitted a further pre-election boost in spending through the growth in capital expenditure over 2003 and 2004.

3.13 In 2001 and 2002 weak commitment controls had resulted in the build-up of domestic arrears and the payment of unbudgeted expenditures. The wage bill and other spending overruns significantly distorted the expenditure allocations relative to the 2002 budget. To offset wage overruns, capital spending was compressed, and arrears to statutory funds built up, further distorting budgetary execution. At the same time, large quasi-fiscal losses estimated at 2 per cent of GDP for 2002 emerged as a result of the failure to raise petroleum prices in line with movements in world market prices and the depreciation of the cedi.

3.14 In 2004, a wage overspend and the decision by Government to increase its subsidy to the state-owned oil refinery rather than allow prices to adjust to rising global prices contributed towards domestic financing of the deficit significantly exceeding the target. The fact of an impending election clearly influenced this decision, although price increases of over 100 % were introduced in early 2005 when the new NPP government took power. These price hikes were made in the face of heavy social pressure from the ‘wahala’ fuel protests and the government has declared its commitment to move to an automatic price adjustment formula to take account of changes in exchange rates and world oil prices.

3.15 A closer analysis of the consistency of approved budgets with budget out-turns suggests this remains a significant problem in Ghana. As part of the World Bank-led External Review of Public Finance Management (World Bank, 2005), a detailed analysis of deviations was made. The results, shown in Table 8, show that while there has been a significant reduction in deviations at the aggregate level, at the ministerial level and at the item level deviations continue to be well in excess of the norms recommended under the HIPC-AAP process. This is driven mainly by a continued inability to budget accurately for personnel emoluments and to retain PE spending within budget limits. Aggregate balance is generally maintained by under-spending on services and investments, which are exactly the areas one would wish to protect to enhance the quality and efficiency of spending. The analysis in World Bank (2005) shows no clear trend towards improvements at the ministerial and item levels. Booth et al. reach a similar conclusion for the ministries of education and health over 1999 to 2003.

3.16 The joint IMF-World Bank Public Expenditure Management Country Assessment and Action Plan (IMF-World Bank, 2004) give some indication of improvements in public expenditure management. In particular, the assessment shows that Ghana fulfilled seven out of the 16 PEM benchmarks in 2004 as compared to only one out of 15 in


23

the 2001 assessment. However, a closer examination of which benchmarks have been fulfilled suggests that many core areas of budget management, execution and reporting remain weak.

Table 8. Average budget deviation index, 2002-4

( in absolute percentage)

2002 2003 2004

BDI 1/ BDI BDI

Total domestically financed discretionary expenditure

32 6 2

Major MDAs 2 / 22 20 18

Item 3/ 31 25 29

Personnel Emoluments 15 20 14

Administration 39 38 29

Services 27 14 27

Investment 42 27 44

Source: Tables 5 & 6 of the Statistical Annex

1/ The Budget Deviation Index (BDI) is the average variance of the actual out-turn against the budget

2/ Includes all MDAs whose domestic discretionary expenditure represents 0.5 percent or above of total in 2004, and excludes contingency

3/ Includes all MDAs (Table 7) but excludes contingency

Source: World Bank (2004)

The economic management process

3.17 Ghana has had at times coherent economic planning and decision-making, having had a well-structured process of five year planning in the 1960s and 1970s and then having managed a successful process of economic stabilization and structural adjustment in the 1980s and early 1990s. The quality and coherence of economic decision-making clearly declined in the period following the restoration of democracy in 1992 compared to the structural adjustment period. This may be attributed to four key factors, which have continued to be relevant up to the present day:

• First, the reform agenda itself became more complex – requiring sustained institutional reforms, such as privatisation and public sector reform, rather than bold (but discrete) policy decisions such as devaluation or de-regulation of price controls.

• Second, the democratic framework brought more interest groups into the policy process, introducing an element of bargaining and deal making which had not been necessary under the PRDC government.


24

• Third, there was an important change of personnel, with the 1995 retirement of Kwesi Botchwey as Minister of Finance, who had held responsibility for economic management for the previous 10 years.

• Finally, the need to address the wider framework for economic planning and service delivery became increasingly apparent both because the existing national planning structure had become obsolete and because the level of donor involvement at sectoral and district levels was presenting new challenges to coherent planning.

3.18 The Vision 2020 initiative of 1996 and the launching of Sector Wide Approaches (SWAps) in agriculture, roads and health (1998-99) provided the main thrust of the NDC response to the fourth of these problems. The introduction of the MTEF in 1998 (covering the 1999 -2001 period) along with the Medium Term Development Plan helped to create a formal, if rather loose, link between national policy (Vision 2020) and the budget.

3.19 The preparation of the GPRS and its formal approval by Parliament in February 2003 created a more robust national planning framework, with stronger links to resource allocation and a wider base of participation. Although many argued at the time of its production that this was a largely donor-driven process, there is evidence of an increasing government control over the process in recent years and of an increasing government concern to link it more tightly to resource allocation. The advent of MDBS has made it easier to bring aid resources within the budget framework and has also facilitated coherent decision-making.

3.20 Overall then, one can point to both increasing coherence and increasing ownership over economic decision-making in the period following the introduction of the GPRS. Nevertheless, there remain significant off-budget flows both from internally generated funds and from donor projects. Moreover, as discussed below, there continue to be inconsistencies in macro-economic decision making, most notably in response to popular pressure for utility subsidies.

3.21 Booth et al. (2004) note that in the mid-1990s political liberalisation led to the emergence of significant popular protests – against privatisation, against the removal of utility subsidies and most notably against the introduction of VAT in 1995. Notwithstanding this, State shares in Ashanti Goldfields were substantially divested and the move even proved popular due to the reservation of shares for small shareholders. Later VAT was successfully re-introduced, reflecting ‘among other things, the NDC government’s enhanced credibility arising from its conduct of the 1996 elections, the strong presence of opposition parties in Parliament and the reduced incentive for obstructive extra-parliamentary opposition.’ Booth et al. (2004) also suggest that ‘an even stronger “democracy dividend” favouring more rational economic policies, emerged from the alternation of power in 2001’. Thanks to the goodwill and popular legitimacy that the NPP initially enjoyed, the government was able to raise the price of water and electricity substantially in 2001 and to double the price of petroleum products in 2003. Afrobarometer 2 survey data showed a doubling of the proportion of the adult population, “willing to endure hardships now” (an indication of patience with economic reforms) from 36 % in 1999 to 72 % in 2002.


25

Has economic management improved?

3.22 There have been some significant improvements over the period reviewed, but there remain major weaknesses. Moreover, the gains appear fragile, especially because important elements of political mismanagement seem to have persisted.

• Annual GDP growth rates have risen from below 4 % p.a. to above 5 % p.a. However, this has been aided by improved gold and cocoa prices, and it is not clear how robust growth will be in the face of continuing increases in world oil prices.

• The structure of growth has improved modestly, with a continued expansion in services and some diversification of exports into tourism and non-traditional agricultural exports, in particular horticulture.

• Nevertheless, formal wage employment has not grown significantly which has maintained pressure on government to protect public sector employment.

• Inflation has fallen from its peak of 39 % in 2000 to 10.8 % in 2004. This is a notable improvement but until single digit inflation can be sustained over a multi-year period, the inflationary threat cannot be said to be under control.

• Fiscal management has improved due both to a strengthening of revenue collection and to increased aid resources from budget support and from HIPC relief.

• Unanticipated wage expenditures and quasi-fiscal spending (due mainly to oil price subsidies) have continued to be a feature of budget management, and only once an automatic price adjustment formula is institutionally embedded can the latter problem be considered finally eliminated.

• Management of the public sector wage bill remains a major problem and the key source of deviations between approved budgets and out-turns.

• There have been improvements in certain aspects of public finance management – notably through the updating of public finance and audit legislation, but fundamental weaknesses remain in budget formulation, execution and reporting.


26

4. Results of the technical cooperation: organisational capacity for economic management

4.1 This section first examines the changes in organisational capacity in the organisations that have formed the focus of the evaluation and then assesses the contribution of DFID funded TC in bringing about these changes.

National Development Planning Commission (NDPC)

Role in Ghana’s economic management process

4.2 The National Development Planning Commission (NDPC) is a public sector organisation provided for by Articles 86 and 87 of the 1992 Constitution of the Republic of Ghana and established by the National Development Planning Commission Act, 1994 (Act 479).

4.3 Its mission is to advise the President (and Parliament, upon request) on development policies and strategies that will ensure accelerated and sustainable development of the country. NDPC thus performs an advisory and supervisory role in Ghana’s economic management. It advises the Government on (a) Development Planning Policy and Strategy (b) Macro-economic issues and Structural Reform Options (c) National Multi-year Rolling Development Plans (d) Environmental Protection and (e) Even Development of all the districts in Ghana. It also supervises the Sectoral and District Planning Authorities in the formulation and implementation of development plans and programmes.

4.4 To enable the Commission to carry out its functions, a secretariat (which is the technocratic organ of the Commission) exists to draft and formulate the framework for the design of the national development plan. A Director General heads the Secretariat. It is expected to provide the technical inputs for the Commission’s formulation of development planning policy and strategy. The Secretariat under the guidance of the Commission technically coordinated the preparation of the Ghana Poverty Reduction Strategy (GPRS) 2003 – 2005.

The TC inputs provided

4.5 DFID, the World Bank, UNDP and the GTZ have provided the key TC inputs to NDPC for the development and management of the GPRS. The TC input by GTZ focused on IT provision, working mainly with the M&E Division. The World Bank support focused on participatory monitoring of the GPRS, including the use of the Citizens’ Score Card. The DFID support was for the formulation of the GPRS and the design and implementation of the Annual Progress Review (APR) framework.

4.6 TC was also provided for a poverty and social impact assessment under the GPRS. DFID funded the agriculture, vulnerability and exclusivity component, the World Bank


27

funded the electricity component, the UNDP the petroleum component and the GTZ/KfW funded the decentralisation component.

4.7 Development Partners have also provided close guidance to the GPRS process in terms of support for the organisation of workshops and seminars to elicit the participation of key stakeholders. They have further provided comments on documents and reports produced. In addition, support has been provided to enable the staff of NDPC and key stakeholders to participate in capacity building workshops and seminars.

Changes identified in organisational capacity

4.8 In the 1990s, NDPC was a relatively ineffective organisation that held a formal status but had no real influence over resource allocation and policy-making. Its role as the coordinator of the GPRS has dramatically increased its level of influence and importance and its organisational framework has now been re-moulded to this role. A Cross-Sectoral Planning Group (CSPG), made up of representatives of MDAs, Research Institutions, the Private Sector and Civil Society, was set up to lead the GPRS 1 process and three years later, the same group is leading the GPRS II process with support from consultants. Thus, the practice of using the CSPG is well established, co-opting more MDA representatives as necessary to support the groups’ activities. Thematic Groups have also been set up for the preparation of the GPRS. The Commission forms the Steering Committee that provides oversight, policy guidance and support to the CSPG and the Thematic Groups.

4.9 The knowledge and skills of some NDPC professional staff have improved to the extent that they are able to provide guidance and support to the GPRS design, monitoring and evaluation as well as reporting and dissemination processes. TC support to the GPRS process has also strengthened linkage and coordination arrangements between MOFEP and NDPC. Until the inception of the GPRS process, each MDA had its own vision and mission that informed the budgetary process using the MTEF. The NDPC is now responding to the provision in the National Development Planning Systems Act 1994 (Act 480) that gives it the mandate to review plans submitted to MOFEP to ensure that they are consistent with the GPRS and other GoG policy documents. To strengthen the coordination arrangements between the NDPC and MOFEP, a Planning Unit has just been established in the MOFEP by the Deputy Minister of MOFEP in charge of Economic Planning to ensure more intense interactions between the NDPC and MOFEP in the planning and budgeting process as well as in the monitoring, evaluation and dissemination of the GPRS. 15 The Commission has also submitted a proposal to the government to turn the NDPC into a Think Tank of professionals.

4.10 Nevertheless, there remain serious questions over the sustainability of these structures and their ability to continue the process of GPRS monitoring, updating and development in the absence of the extensive consultancy support which NDPC

15 The current Deputy Minister in charge of Economic Planning was the former Director General of the

NDPC and thus has direct experience of the inadequate linkage and coordination arrangements.


28

continues to receive. Although the TC provided has built some level of internal capacity in the NDPC to steer the GPRS process, key institutional and personnel issues have precluded a more substantial impact. The NDPC is still not well resourced. The establishment requirement of the NDPC is 54 staff out of which 42 should be professional staff and the rest ancillary staff. Currently, there are 35 staff and only 12 are professional staff. The attrition rate is very high due largely to the unattractive conditions of service. The perception of many NDPC staff interviewed during the field visit was that NDPC had become a fertile ground for development partners and the NGOs to ‘poach’ professional staff.

4.11 Tables 9 and 10 show the staffing trend in the organisation between 1998 and June 2005. NDPC has never had the full complement of the 42 professional staff in any of the years under review, ranging between 40 and 59 per cent of the complement. Currently, there are no Directors for any of the Divisions. Recruiting highly qualified staff is a big challenge for the NDPC. Coupled with the inadequate professional staff capacity is the perception that NDPC has over the years been a dumping ground for the government. A number of military personnel and government functionaries, without the requisite professional and academic backgrounds, have been redeployed to the NDPC, heading technical departments and units.

4.12 Leadership continues to be a challenge. There have been at least five changes in the Director General (DG) position over the last decade. The current DG is the sixth. All the DGs have been in acting positions. The general perception among key stakeholders consulted was that most of the DGs that are appointed were more concerned about delivering the tasks they had been assigned. They therefore generally failed to adopt a holistic approach to addressing the institutional and staffing issues together with delivery of their contractual obligations. In general, the task-oriented focus of TC support has tended to reinforce this short-term focus without addressing the recruitment and retention challenges.

4.13 Owing to the inadequate technical capacity of the NDPC, the Commission has become highly dependent on consultants to execute key functions, including the design, implementation, monitoring and evaluation of the GPRS. When tasks come up, the few professional staff available are mobilised and consultants are used to supplement the staff input. Most of the consultants are national consultants, especially for the GPRS. They are mostly from the universities with some consulting firms as associates.


29

Table 9. NDPC staffing trends 1998- June 2005

Number at Post Category of Professional Staff

1998 1999 2000 2001 2002 2003 2004 2005

Director General - - 1* 1* - 1 1 1*

Director 1 1 1 1 - 1 2 1**

Deputy Director 3 3 3 3 3 6 6 6***

Principal Planning Analyst 5 5 5 5 2 - 3 3

Senior Planning Analyst 4 4 4 4 6 5 4 4

Planning Analyst I & II 9 9 11 10 8 4 1 1

Total 22 22 25 24 19 17 17 16

* In Acting Position; ** On contract; ***2 Deputy Directors on Contract

Table 10. Number of years at NDPC

Number of Years in the Organisation Category of Staff

1 2 3 4 5 6 7

Director General 1 - - - - - -

Director - - - - - - 1

Deputy Director - - 3 - - - 3

Principal Planning Analyst - - 1 1 - 2 -

Senior Planning Analyst 2 1 3 4 1 - 4

Planning Analyst I & II 3 - 4 3 1 - 4

Source: Information supplied by NDPC

4.14 In terms of the Ministry’s “products”, based on experience with GPRS1, a more participatory approach is being adopted for the GPRS 2 process. During the design of GPRS1, there were limited consultations with stakeholders, especially at the district and community level. Dialogue sessions were held with MDAs and four regions selected for consultations. Civil society was not involved in the consultations. In GPRS I, themes were used with MDAs working on the relevant themes.

4.15 The GPRS II process was initiated in September 2004 and has been highly participatory from the drafting stage. The level of consultations and dialogue with stakeholders has been strengthened at all levels. In addition to consultations with MDAs, intense consultations were held at the regional, district and community levels. The involvement of NGOs, Cross Sectoral Working Groups and other civil society groups has addressed the gaps identified in the level of participation in GPRS I. It has ensured ownership and commitment by Government, MDAs and civil society, including the private sector.

4.16 In terms of content, GPRS I did not cover vulnerability and exclusion, but GPRS II will. A poverty and social impact assessment has also been done to inform the GPRS II. The Commissioners of NDPC have been steering the process. They meet


30

monthly to present and discuss the outputs. The outputs are critiqued and further work is done to ensure that the GPRS addresses all requirements.

4.17 An Annual Performance Review (APR) of the GPRS has been instituted. It provides evidence on the impact of government programmes and projects in the five thematic areas of the GPRS. It also describes the policies, programmes and projects of the GPRS within each thematic area, and presents evidence on their progress. Linkage has been established between the GPRS, the Annual Budgets and the MTEF. For example, the 2004 APR analysed the correlation between medium budget trends and GPRS spending objectives. Plans are underway to produce a half year APR to ensure that the outcomes are fed into activities of the next half-year.

4.18 A key challenge of the APR is the continued difficulty in collecting comprehensive, reliable and timely data on GPRS implementation. Data inconsistency and availability from different sources was viewed as a major challenge in the monitoring exercise (NDPC, 2004a). The MDAs are now becoming conversant with the APR and this may lead to enhanced data quality. The APRs are disseminated to the Executive, Parliament, the donors, and civil society.

Contribution of DFID TC

4.19 There is no doubt that DFID TC support has played a key role in the set up and functioning of the GPRS and its monitoring process. In particular, it played a key role in addressing the technical capacity gap of the NDPC in pursuing its roles and functions. NDPC did not have an effective structure to lead the GPRS and had limited qualified personnel for monitoring & evaluation. It is the TC support that bridged the technical resource gap and helped establish an organisational framework to manage the GPRS. The TC support also helped to prepare the M&E plan for the GPRS and facilitated the identification and establishment of core indicators for monitoring GPRS implementation.

4.20 The GPRS has made a significant contribution to the coherence of economic policymaking and is beginning to have an impact on resource allocation. Over 2002 to 2004 actual spending on poverty related areas increased from 28.3 % of total discretional spending (5.5 % GDP) to 33.5 % of discretionary spending or 7.7 % of GDP (World Bank, 2005). Perhaps more significantly, it has led to a change in attitudes: MDAs now know that if their programmes are not linked to the GPRS they will not be funded. Government priorities are fashioned along the GPRS and they are monitored. The GPRS has in turn been the basis for the Multi Donor Budget Support (MDBS) by the Development Partners.

4.21 Of course, there remain doubts over the level of political support to the GPRS and over how to improve the linkage to the MTEF and the budget. Nevertheless, the PRS process is clearly proceeding in the right direction.

4.22 There are more fundamental doubts over the sustainability of the organisational structures – specifically over the ability of NDPC to continue to manage and develop the GPRS without consultancy support. It is understood that NDPC have developed internal proposals to address this issue, based in part on an organisational re-


31

structure and in part on the development of new salary scales for NDPC staff. This is clearly a positive initiative, which hopefully will receive an appropriate form of support from MOFEP and from Cabinet. It will be important for any future TC to be provided in a low-key way that supports rather than hinders the development of a sustainable organisational structure for the GPRS.

The VAT Service


4.23 Value Added Tax (VAT) was established by the Value Added Tax Act, 1998 (Act 546) to provide for the imposition of value added tax and purposes related to the tax. The VAT Service (VATS) was accordingly established to administer the tax. It is governed by the Revenue Agencies Governing Board. Its role in Ghana’s economic management is to contribute to raising revenue through tax collection alongside the other two tax collection agencies namely the Customs Excise and Preventive Service (CEPS) and the Internal Revenue Service (IRS).


4.24 In addition to Government of Ghana funding of personnel input, the World Bank provided initial TC support for the design of the VAT system. DFID funded TC support for further design of the VAT and for the development of management capacity with respect to a) the policy and regulatory requirements, b) IT support, c) staff training and d) public education campaigns. This included the installation of the necessary IT systems and other procedural requirements in each of these areas.


4.25 An effective and efficient VAT Service has been established with adequate trained staff and a computer based system for administering the tax. Hardware and software for the administration of VAT have been installed and operationalised.

4.26 The VAT Service has a responsive organisational framework, staffed with suitably qualified personnel that are motivated to deliver effective and efficient services. An important aspect of the motivation is that employees are on IRS scales, allowing them a considerably better remuneration package than that available in MoFEP or NDPC. The Service has a fairly stable staff most of whom (over 70%) have been with the organisation since it was established in 1998. It also has a fairly stable leadership. The current Commissioner has been in place since 2001 and is the second Commissioner since the service was established in 1998. Staff strength has


32

increased progressively to respond to the increasing demands of revenue collection.16

4.27 The VAT Service has a well-equipped training unit that provides continuous capacity building to its staff. For example, staff have been trained in ‘Key Accounts Management’ to enable them to manage key clients effectively. It provides good customer service by addressing the educational and sensitisation needs of the taxpayer and attending to their needs promptly. Regular announcements and publications are made to remind taxpayers of their VAT obligations. A system is in place for continuous education of taxpayers to minimise the potentially antagonistic relationship between the taxpayers and the tax collector.

4.28 Establishment of the Revenue Agencies Governing Board in 2001 has further strengthened the performance of the VAT Service. The Board provides policy direction and guidance to the VAT Service and facilitates the determination of revenue and performance targets that are responsive to the economic management requirements of Ghana.

4.29 Long-term relationships have been developed with Crown Agents, the consulting firm that provided the TC input to provide trouble-shooting support. Crown Agents continues to provide support to the VAT Service in the maintenance of the Value Added Tax Information Processing System (VIPS) and to address implementation challenges.

4.30 The VAT Service continues to deliver its revenue targets and is generally able to take swift action to tackle non-compliance to maintain the credibility of the VAT Service and the tax system. It has successfully increased the VAT rate from an initial 10% on its re-introduction in December 1998 to 12.5% in June 2000 and 15% in August 2004.

4.31 The VAT Service has continuously exceeded its revenue targets as illustrated in Table 11 and has collected a steadily increasing percentage of GDP, rising from 3.85% in 1999 to 5.55% in 2004. Collections of Import VAT have increased significantly faster than Domestic VAT, which to a large extent reflects the relative ease of collection of VAT on imported goods. Extending the coverage of VAT over domestic goods and services is the most immediate challenge the service faces but overall trends in collection have performed well relative to other countries in Sub-Saharan Africa which have recently introduced VAT.

16 The establishment requirement of the Service as of 31 May 2005 is 839 and the number of staff at post is

952, an excess of 113, but a new establishment structure is awaiting approval.


33

Table 11. VAT collections relative to targets and GDP

Domestic VAT Import VAT Total VAT VAT as % GDP

1999 Actuals 274,528 518,687 793,215 3.85%

Target 265,780 434,220 700,001% Deviation 3.29% 19.45% 13.32%

2000 Actuals 385,158 886,936 1,272,095 4.69%

Target 437,124 825,876 1,263,000% Deviation -11.89% 7.39% 0.72%

2001 Actuals 574,042 1,352,986 1,927,027 5.07%

Target 545,300 1,194,400 1,739,700% Deviation 5.27% 13.28% 10.77%

2002 Actuals 724,426 1,675,293 2,399,719 5.17%

Target 716,000 1,492,000 2,208,000% Deviation 1.18% 12.29% 8.68%

2003 Actuals 1,025,749 2,505,117 3,530,867 5.34%

Target 1,024,400 2,119,200 3,143,600% Deviation 0.13% 18.21% 12.32%

2004 Actuals 1,435,665 2,989,798 4,425,463 5.55%

Target 1,353,927 2,720,000 4,073,927% Deviation 6.04% 9.92% 8.63%


4.32 DFID TC support has focused on the design and implementation of the VAT system. TC support was provided from 1993 to 1995 for the design of the policy and regulatory framework and again from 1997 to 2000 to facilitate reintroduction of an effective and sustainable VAT system after implementation was initially stopped and the law was repealed as a result of violent riots against introduction of the VAT in 1995.

4.33 Crown Agents provided the TC services from inception to the end of DFID support. The TC was provided in close collaboration with the local counterparts. There were conscious attempts at capacity building, which have clearly been successful given that the service has continued to run effectively without TC since 2001.

4.34 The IT support enabled the development of a computer-based system, the Value Added Tax Information Processing System (VIPS), to capture information on traders and records on collection. It involved the procurement and installation of the hardware and the software. A resident trainer was provided to train the staff in the use of the hardware and the VIPS.

4.35 The TC support also facilitated the design and implementation of a publicity and education campaign to ensure smooth introduction of the VAT. It involved development of the education and publicity strategy materials and training of the staff in their use.

4.36 Although introduction of the VAT in 1995 failed, the commitment of the Government of Ghana to implement the system enabled the identification of the factors leading to the false take off and measures adopted to address them appropriately. There was a


34

political misjudgement, which the TC service provider together with the local counterpart did not anticipate. The rate was fixed at 17.5% and this was found to be too high and difficult to calculate by the traders. The timing was also poor. It was introduced at a time when the ‘lean season’ was at its peak and there was general economic hardship among the populace. It was therefore easier for the opposition parties to mobilise the populace to demonstrate against the introduction of the VAT.

4.37 Despite the false take off, the second phase of the TC support from 1997 to 2000, involving the redesign and reintroduction of the VAT, proved successful. Although the initial attempt failed and the consultants left, the project team was kept together. The hardware, the vehicles and other equipment provided by DFID were retrieved and kept by DFID until the VAT was reintroduced before they were made available to the VAT Service. This strategy ensured that equipment was readily available for use at the time of reintroduction of the VAT. At the passage of the law in 1998, and subsequent implementation of the VAT, the rate was pegged at 10%. This rate was more acceptable to the public and it was easier to calculate.

4.38 The introduction of VAT has represented a major improvement over the previous retail sales tax, being both more progressive (Lawson and Salpietro, 2000) and more buoyant. It has been an important part of an ongoing, and largely successful, process of upgrading of revenue policy and administration.

4.39 The one aspect in which the TC was not fully successful was in relation to the installation of VIPS, which was not completed and fully operational before the TC support ended. A modest amount of additional TC has been provided by DFID (through the Crown Agents) since 2001 in order to complete operationalisation but it still remains incomplete. Currently, the VAT service is seeking internal funding to complete installation of VIPS.

Ministry of Finance and Economic Planning


4.40 The Ministry of Finance/ MOFEP has been at the heart of the economic management process in Ghana since Independence. Its primary role has always been to administer the public finance acts, and hence to lead formulation of the budget, control expenditures relative to budgets, manage treasury operations and payments and prepare the consolidated accounts of government. As a result of this core role, it has also enjoyed considerable influence over economic planning and policy-making, although functions in this area have been shared with the Office of the President and the NDPC.

4.41 Notwithstanding its obvious importance, it has never had the pre-eminence, which, for example, the UK Treasury currently enjoys and which is typical of most Anglophone Sub-Saharan African countries. Two areas where its powers are notably less extensive are in relation to the regulation and supervision of State-owned enterprises (SOEs) and in relation to managing public sector recruitments,


35

promotions and terms & conditions. The continuing problems of control of the public sector wage bill, particularly for subvented agencies, and of quasi-fiscal losses through SOE subsidies reflect in large part the diffusion of responsibilities in these areas.

4.42 On the other hand, in those aspects of budget management and expenditure control where its responsibility is more clear-cut, MOFEP has tended to exercise control to a relatively high level of detail, prompting charges of ‘micro-management’ from the MDAs. As a result, its relations with MDAs have often been antagonistic and, consequently, MDAs have a tendency to hide information from MOFEP, which is perhaps more pronounced than it would be in other countries. This has in turn undermined the type of constructive dialogue that would be necessary to bring MDAs’ internally generated funds and off-budget donor projects properly within the budget.

4.43 To a large extent, the reforms that have been pursued in Ghana under PUFMARP and, more recently with the updating of public finance and audit legislation, have aimed to broaden and clarify MOFEP’s responsibilities for overall public sector spending, whilst introducing greater operational flexibility in the way those controls are exercised. Although progress has been made, there remains much unfinished business, as is revealed by the scope and nature of deviations of budgetary out-turns from approved budgets.

4.44 Unravelling the source of the budget deviations problem is crucial to securing sustainable improvements in public finance management. In particular, it is necessary to have a realistic sense of the relative importance of a) legislative errors and omissions in the clarification of roles, b) non-compliance with defined roles and rules and c) informational problems which undermine the exercise of control and oversight. The reform process and the associated TC projects have tended to emphasise solutions focused on improved information and control. It is clear why the incentive framework would favour such solutions, in particular where pursuit of an ‘information and control’ solution involves TC projects with the related benefits that they might bring for staff. However, a more objective diagnosis of the problems would almost certainly place more emphasis on non-compliance as the principal cause of budget deviations.


4.45 A substantial level of technical cooperation and associated computer hardware and software support has been provided to MOFEP since 1997 in support of the public finance reforms. The PUFMARP programme has provided the framework for this, the principal elements of which have been:

• Support by the World Bank to the development of a computerised Budget & Public Expenditure Management System (BPEMS) – an integrated financial management system for managing commitments, expenditures and payments and for the associated accounting and financial reporting functions.


36

• Complementary support by the IMF to develop short-term improvements in methods of commitment control, treasury management and expenditure tracking.

• Support by the World Bank to the revision of legislation on public finance, auditing and procurement.

• Support by DFID to the revision of procurement laws and regulations and to their implementation.

• Support by the European Commission to the strengthening of audit services.

• Support by the Canadian government (CIDA) and others to the strengthening of parliamentary oversight and committee work.

• Support by DFID to the development of a Medium Term Expenditure Framework (MTEF) and, as a part of this, to the strengthening of macro-economic forecasting.

• Support by DFID to the development of a new Integrated Personnel and Payroll Data-base (IPPD2).

4.46 The analysis has focused predominantly on the latter two projects. In recent years, (under the Support to MoF in Financial Management Reform project) DFID has also provided inputs to advise on the sequencing and management of financial management reform17 and to assess the readiness of the BPEMS and IPPD2 systems.

4.47 Simultaneous with these projects, the Government has also attempted to undertake an organisational restructuring of MOFEP and CAGD. A comprehensive organisational review was initiated under the auspices of the NIRP but this review was never properly finished, largely because the review team failed to establish a productive working relationship with MOFEP. 18 Various offers of TC support to this organisational restructuring have been made by DFID and the World Bank but have not been taken up by GoG.


4.48 Regarding individual capacities, with the exception of the CAGD, the team were unable to obtain documented information on staff numbers and grades but did discuss the question both with the Budget Division and the Policy Analysis Division. Within CAGD, once account is taken of the staff transferred to the Ministry of Health (and assuming that this involved the transfer of equivalent accounting functions), the overall level of staffing is remarkably stable. There is evidence of recent difficulties in recruiting junior accounting officers. However, at more senior levels most posts have been filled, presumably from promotions. Overall staffing data does not suggest a capacity crisis.

17 This involved the provision of senior advisory support to the PUFMARP Central Implementation Team

(CIT) and to the CAGD, through Mr. Peter Murphy, who had previously assisted the successful introduction of integrated financial management systems in Lesotho and Tanzania.

18 It is understood that in part this was the result of the working style of the team and, in part, as a result of

its emphasis on the potential need for staff retrenchments.


37

4.49 A similar impression of stability in staffing numbers is obtained from the Budget Division and the Policy Analysis Division. Although the PAD has several well qualified and enthusiastic, young economists, they pointed to the lack of senior, qualified staff as a significant constraint. On the other hand, reports from the MTEF consultants suggest this was also a problem in the late 1990s. In contrast to the CAGD, both the BD and PAD suggested that staff skills had been upgraded from their involvement in MTEF workshops and in economic forecasting workshops. The CAGD did not believe there had been any significant skills transfer from either the IPPD projects or from subsequent advisory inputs to BPEMS financed by DFID.

4.50 Regarding organisational issues, as noted above, the intended organisational restructuring of the ministry has never been completed. Certain minor organisational changes have, nevertheless, been made – such as the establishment of the Budget Development Unit within the Budget Division, and the more recent creation of the Economic Policy Coordinating Committee. Although minor changes, these are not without significance as the creation of the BDU has brought the full set of responsibilities for the development of the budget and MTEF within the BD, whereas previously MTEF work had been managed from a separate PUFMARP project office. The EPCC has introduced a formalised structure for monitoring budget out-turns and managing the cash budget system.

4.51 At the leadership level, a lack of continuity at the most senior levels (Minister/ Deputy Ministers and Chief Director) has been a persistent problem since the late 1990s. The impact of this has been exacerbated by the tendency of the Minister and Deputy Ministers to use special advisers for what would generally be civil service functions.

4.52 With regard to the role and mandate of MOFEP, the recent approval of new public finance and audit legislation should in principle have served to strengthen its role as the guardian of public finances. However, the legislation and its associated regulations are not fully in place and it is perhaps too early to judge.

4.53 What is clear is that the ministry still operates under a framework of unsatisfactory civil service terms and conditions, as it did throughout the period of the evaluation. Aggregate staff numbers are comparable with ministries of finance in Anglophone Eastern and Southern Africa but there are clearly problems of recruitment and retention of young qualified staff. Whilst older staff tend to stay, they bear a strong resentment over differentials in pay between themselves and Central Bank or IRS employees. Moreover, the fact that the various TC projects in the ministry (including the DFID-funded ones) have generally provided salary supplementation of different kinds has made this resentment still deeper. This supplementation is generally short-term and always limited to a small number of individuals; as such it seems to serve not so much as a motivation but as a reminder of how poor the base conditions are. With some exceptions (such as PAD, where the learning opportunities seem to act as a counter-balancing factor), the impression within MOFEP is of a poorly motivated workforce. Most of those interviewed suggested there had been a slow but steady decline in morale since the late 1990s.

4.54 Table 12 provides a summary of the key changes in MOFEP’s “products” over the period in question, where “products” are taken as referring to the analytical,


38

management and reporting systems necessary for MOFEP to execute those stages of the budget cycle for which it is responsible. Overall, the Table suggests that there has been some improvement at this level but many of the improvements are partial and remain incomplete. Moreover, these improvements have largely failed to address the core problems in budgetary and economic management which existed at the outset of the period, most of which have persisted up to 2005.

The contribution of DFID-funded technical co-operation

4.55 Table 12 shows that some modest improvements have been made within those aspects of the budget cycle, with which DFID TC was associated. However, these improvements are partial and incomplete, some such as IPPD2 despite substantial project expenditure. Moreover, in terms of eventual impact on economic management, one cannot point to any clear-cut contributions from the DFID TC projects within MOFEP to the modest improvements in economic management identified.

Te

ch

nic

al

Co

op

era

tio

n f

or

Ec

on

om

ic M

an

ag

em

en

t V

olu

me

2:

Ca

se S

tud

ies

39

Ta

ble

12

. Ch

an

ge

s i

n q

uali

ty o

f M

OF

EP

pro

du

cts

, 1

99

7-2

00

5

Sta

ge o

f th

e B

ud

get

Cycle

S

tatu

s in

1997

Sta

tus in

2005

As

so

cia

ted

TC

p

roje

cts

Macro

- econ

om

ic

fore

castin

g a

nd f

iscal

pla

nnin

g

Macro

fore

castin

g w

as r

ud

imenta

ry, re

lyin

g

pri

mari

ly o

n th

e IM

F a

nd th

e B

oG

.

Se

vera

l big

exp

en

diture

ite

ms (

SO

E s

ubsid

ies,

Sala

ries &

dom

estic

debt serv

icin

g)

were

re

gula

rly

un

der-

estim

ate

d.

Tw

o s

imple

spre

adsh

eet m

ode

ls w

ere

de

velo

pe

d

under

the P

UF

MA

RP

pro

ject. O

ne h

as b

ee

n

furt

her

develo

pe

d a

nd is u

sed b

y M

OF

EP

. H

ow

eve

r, r

elia

nce o

n I

MF

& B

oG

rem

ain

s.

With

SO

E s

ubsid

ies a

nd d

om

estic

debt

no

w

better

co

ntr

olle

d, costs

are

als

o b

etter

estim

ate

d.

No im

pro

vem

ents

in f

ore

ca

stin

g s

ala

ry

pa

ymen

ts.

Ste

ad

y im

pro

vem

ents

in d

om

estic

reve

nue

fore

castin

g o

ffset b

y p

oor

fore

castin

g o

f M

DB

S.

Macro

com

ponent

of

DF

ID-s

upport

to

P

UF

MA

RP

.

IMF

short

-cours

es

in M

acro

modelli

ng

.

Bud

get

Pre

para

tion

No m

ediu

m term

pers

pectiv

e &

no c

lear

polic

y lin

k (

because 5

-Ye

ar

Pla

n &

PS

IP n

ot lin

ked to

resourc

e a

vaila

bili

ties).

Bud

get pre

se

nta

tio

n lin

e-

item

and o

rganis

ation-b

ase

d w

ith n

o lin

k to

activ

itie

s, o

utp

uts

or

obje

ctive

s.

Recurr

ent b

udg

et pre

pare

d o

n incre

menta

l basis

w

ith n

o r

evie

w o

f activ

ities.

Investm

ent &

recurr

ent

bud

gets

separa

tely

pre

pare

d.

Covera

ge

of

bud

get

incom

ple

te,

with

man

y donor

pro

jects

and I

GF

s o

ff-b

udg

et a

nd S

OE

lo

sses/ subsid

ies r

are

ly b

udgete

d.

A r

ud

imenta

ry M

TE

F is

in p

lace,

based o

n 3

ye

ar

secto

r polic

y su

bm

issio

ns, lin

ked to G

PR

S.

This

is

in t

urn

con

vert

ed into

3 y

ear

activi

ty b

ase

d

budg

ets

, com

bin

ing In

vestm

ent &

Recurr

ent.

Pre

para

tion p

rocess f

acili

tate

d b

y use o

f bud

get

pre

para

tio

n s

oft

ware

(A

CT

IVA

TE

).

Util

ity

of

MT

EF

un

derm

ined

because a

) sala

ries

& f

ixed o

pera

ting

costs

linked o

nly

to

org

an

isationa

l le

vel, b

) a

pp

ropri

ations f

or

item

s 3

&

4 (

whic

h a

re lin

ked to a

ctivi

ties)

diffe

r substa

ntially

fro

m r

ele

ases a

nd c

) pro

jections o

f oute

r ye

ars

poorl

y pre

pare

d.

Covera

ge

of

SO

E s

ubsid

ies is

bett

er

but m

an

y donor

pro

jects

& IG

Fs s

till

off

budget.

MT

EF

com

ponent

of

DF

ID S

upport

to

PU

FM

AR

P

Bud

get

Execution

Manu

al s

yste

ms p

rovi

din

g ineff

ectiv

e

com

mitm

ent contr

ol and lack o

f fisc

al d

iscip

line

at a

ll le

vels

.

As a

resu

lt, c

ash b

ud

get sys

tem

had b

een

in

trod

uced b

ut

this

still

did

not pre

vent re

gula

r ove

r-spe

ndin

g a

nd

un

pla

nn

ed b

orr

ow

ing.

(Ove

rspend

ing m

ain

ly o

n s

ala

ries &

SO

E lo

sses/

subsid

ies b

ut

als

o s

om

etim

es f

or

unbud

gete

d

pro

jects

).

Com

pute

rised B

PE

MS

sys

tem

cove

ring 8

MD

As

& 4

2 p

ilot

site

s d

ue t

o g

o liv

e f

rom

FY

2006

.

Ne

w p

ublic

fin

ance legis

lation h

as in

tro

duce

d

fram

ew

ork

for

more

fle

xib

le, decen

tralis

ed

expend

iture

managem

ent.

Nevert

he

less, pre

-exis

ting

manual s

yste

ms h

ave

la

rge

ly lost eff

ectiv

en

ess a

nd f

iscal d

iscip

line

relie

s o

n a

pp

licatio

n o

f cash b

udget

sys

tem

. Im

pro

ved c

ash m

onito

ring h

as m

ade this

more

World B

ank-

financed B

PE

MS

Som

e a

dvis

ory

support

fro

m D

FID

T

C (

Under

“Sup

port

to

MoF

mgt of

PF

M

refo

rms”)

DF

ID-f

unded T

C to

IPP

D2

Te

ch

nic

al

Co

op

era

tio

n f

or

Ec

on

om

ic M

an

ag

em

en

t V

olu

me

2:

Ca

se S

tud

ies

40

Sta

ge o

f th

e B

ud

get

Cycle

S

tatu

s in

1997

Sta

tus in

2005

As

so

cia

ted

TC

p

roje

cts

Cash b

udge

t sys

tem

lacked tra

nspare

ncy

in

decis

ions f

or

rele

ase o

f fu

nds.

Pa

yroll

sys

tem

based o

n I

PP

D1,

an o

ld

com

pute

r sys

tem

wh

ose f

unctio

nalit

y lim

ited to

pa

yro

ll on

ly a

nd w

ith v

ery

slo

w o

pera

ting t

ime.

eff

icie

nt and

EP

CC

has m

ade th

is m

ore

tr

anspare

nt.

Ove

r-spend

ing n

ow

larg

ely

lim

ited to

sala

ries.

Pa

yroll

sys

tem

stil

l base

d o

n IP

PD

1. D

esp

ite

substa

ntial in

vestm

ent in

IP

PD

2 (

Ora

cle

-based

inte

gra

ted p

ers

onne

l/ pa

yro

ll sys

tem

), this

is s

till

not fu

nctiona

l and t

here

are

conflic

ting v

iew

s

abou

t ho

w c

lose it

is to c

om

ple

tion.

Accounting &

R

eport

ing

Accounting s

yste

ms h

ad la

rgely

bro

ken d

ow

n,

lea

din

g to

de

velo

pm

ent of

para

llel s

yste

ms o

f expend

iture

monito

rin

g.

Bank r

econcili

atio

ns c

om

ple

ted a

t a

ggre

gate

le

vel of

Nationa

l Tre

asury

account &

gen

era

lly

not a

vaila

ble

at m

inis

teri

al le

vel.

Com

pute

rised B

PE

MS

sys

tem

cove

ring 8

MD

As

& 4

2 p

ilot

site

s d

ue t

o g

o liv

e f

rom

FY

2006

.

Accounting s

yste

ms r

em

ain

larg

ely

d

ysfu

nctio

nal, a

ltho

ug

h s

om

e in

dic

ations o

f im

pro

vem

ents

in H

ea

lth s

ecto

r.

Bank r

econcili

atio

ns c

om

ple

ted a

t a

ggre

gate

le

vel of

Nationa

l Tre

asury

account bu

t ge

nera

lly

not a

vaila

ble

at m

inis

teri

al le

vel.

World B

ank-

financed B

PE

MS

Som

e a

dvis

ory

support

fro

m D

FID

T

C (

Under

“Sup

port

to

MoF

mgt of

PF

M

refo

rms”)


41

5. Results of the technical cooperation: summary of the evaluation results

5.1 This section summarises the key conclusions of the evaluation in relation to the overall evaluation criteria.

Relevance

5.2 The support to the VAT Service was the most relevant in terms of the understanding of organisational needs and operational context. This was helped by the fact that this was a new organisation and issues relating to staffing and salaries were addressed. There was strong political commitment and commitment from other development partners, and an extensive preparation process involving a number of donor agencies. In the case of NDPC, the capacity development needs were identified but probably underestimated, and the limited status and influence that the NDPC had at the start of the project was not fully appreciated. The GPRS process presented an opportunity that was seized by the NDPC DG, and the provision of TC was responsive to these needs.

5.3 MOFEP presented a much more difficult context. The need for the MTEF and IPPD2 initiatives was identified in the GoG’s design of PUFMARP but the specific technical form these interventions took did not address basic issues underlying the wage bill problem. This was not fully appreciated until significant sunk costs had been incurred. This led to significant weaknesses in the design of the projects, which also suffered from insufficient input from government and from other development partners. The MTEF project (unlike IPPD2) was deliberately structured to engage MOFEP leadership, but neither project was based on a strong understanding of the political economy context that drove poor budgetary outcomes (seeing this more as a technical problem). Neither project appreciated or addressed the problems associated in paying allowances. Disputes over allowances and design problems weakened the commitment of government.

Effectiveness

5.4 Support to the NDPC and the VAT service were in general highly effective in generating the intended outputs with strong ownership from the supported organisation and close harmonisation and alignment with the activities of other development partners. Less skills transfer was achieved under the NDPC support than was envisaged, and full functionality of VIPS was not achieved under the VAT Service support.

5.5 In relation to MOFEP support, of the ten outputs specified in the original project documents for the MTEF project, five were assessed as fully achieved, three partially and two not at all. However the coverage of MTEF, in terms of expenditure items, was insufficient and the outputs were not appropriate to achieve the project purposes. IPPD2 remains incomplete largely as a result of the selection of


42

inappropriate contractors and weak project management. Ownership of the MTEF project by government was initially strong but was eroded by the problems identified above, while even initial ownership of IPPD2 appeared weak. PUFMARP provided an initial framework for harmonisation but few development partners were involved in this and those that were did not remain wholeheartedly engaged.

Efficiency

5.6 Both the NDPC and VAT Service support mostly delivered outputs cost-effectively and within the intended time schedule. The delivery of outputs for the MTEF project was largely on schedule if somewhat more expensive than intended. IPPD2 was exceedingly inefficient with many norms of IT project management best practice broken by the contractors, DFID and the Government of Ghana.

Impact

5.7 The NDPC project succeeded in bringing a coherence and focus to economic management, in raising the importance of poverty reduction as a government objective, and in enhancing the NDPC’s influence. The VAT Service project played an important role in establishing a viable, self-sustaining VAT collection service. The MTEF project has had a transactional impact in changing the way in which the budget is prepared, but it did not achieve its project purpose of improving the cost effectiveness of government resource allocation. IPPD2 remains incomplete after seven years. It could be argued that it has laid the groundwork for an effective payroll system but this has yet to be proven.

5.8 In summary, in the VAT service and NDPC, DFID-financed TC appeared to have contributed in a significant manner, although in relation to NDPC there were question marks over the sustainability of its impact. By contrast, within MOFEP, despite considerably higher levels of TC expenditure, rather less change had been achieved. There were some essentially ‘transactional’ improvements but no clear-cut transformational impact.

Sustainability

5.9 Five years after the end of the main part of the TC project, impacts on the VAT Service have been sustained and VAT performance continues to improve. In the case of NDPC, transformational impacts have been sustained so far, but NDPC does not have the capacity to sustain the GPRS system without continuing TC or organisational reforms that have yet to occur, in order to address problems of staff retention and provide greater management stability. An exit strategy to address this problem (and trigger the organisational reforms required) has not been sufficiently developed. The transactional changes introduced by the MTEF project have been sustained, though had the project been better designed in relation to the context, greater success could have been achieved. In neither of the MTEF nor the IPPD2 project have there been effective exit strategies with both projects prematurely closed. In the former case, a change in the technical approach could have reduced the need for a premature exit, while in the latter the project should have been suspended earlier to address its design and management problems.


43

6. Conclusions and implications

6.1 The experiences reviewed in this case study show contrasting experiences. Support to the VAT service was the clear success story of well-structured, well delivered TC operating inside a receptive environment, with strong leadership and motivated staff, in pursuit of an objective, for which there was good political support. Support to the development and monitoring of the GPRS within NDPC was also broadly a success story – again well structured and generally well-delivered TC focused on an objective to which there was political commitment. Yet there remains a question mark over the sustainability of the processes and management systems established, given the capacity weaknesses of the NDPC within the context of a civil service salary structure.

6.2 The TC projects supporting MOFEP have generally been less successful. A large part of the explanation lies in the very difficult operating environment for TC that MOFEP presents. Yet this is only part of the story; there have also been mistakes in project design and project management, which have contributed to project shortcomings. Moreover, it should be recalled that DFID’s support to the MTEF was an acclaimed success in the first two years of implementation before some of the weaknesses in project design started to become evident – this was largely due to the very effective participatory, ‘process-oriented’ approach to consulting which the project team adopted - an approach whose merits served, for a time, to overcome the disadvantages of the difficult operating environment.

6.3 In the cases reviewed, substantial TC resources were expended without an adequate prior analysis of the institutional context and without an adequate understanding of the interactions of organisational, institutional and political economy factors.19

Political factors have been given little attention in the project memoranda and the preparatory documents for the TC projects that form the basis of this evaluation. It is clear that at a personal level there was a much higher awareness of these problems among the DFID staff responsible for these projects and among the few consultants it was possible to interview. This awareness seems to have been gained through the hard lessons of implementation, and certainly the rate of learning has increased with the establishment of the devolved DFID Ghana office. Nevertheless, limited evidence was found of such an awareness being deployed during the diagnostic or design phases of these projects, when it might have helped to shape their direction and mode of delivery.

6.4 The experience reviewed also suggests that TC projects can be successful in a poor institutional environment, if there is the right leadership and organisational capacity and if the project design is well focused.20 Good interpersonal and ‘process’ skills by

19 This does not refer only refer to the TC projects in MOFEP. It could be said that the GPRS project “got

lucky” in the sense that NDPC grew to assume a sufficient political influence to make the project work. It certainly did not have this influence at the outset and without the leadership of Professor Gyan Baffour might never have assumed it.

20 The EC’s support to the Audit Service has also been rated a success. Again, many of the same factors

emerge: good leadership within the recipient organisation, political support for improvements, and well-


44

the TC provider are especially important in a difficult context. Clearly, the right technical skills are also necessary if new systems are to be developed and skills transferred but technical skills alone will not be effective. The TC projects examined in Ghana provided some instructive examples of the right and the wrong combinations of these skills:

• The team working within the VAT Service very deliberately played low key roles in the introduction of VAT: they allowed the Ghanaian managers to take the lead and were sensitive to underlying cultural issues.

• The team responsible for the design of the GPRS monitoring system interacted very effectively both with the local consultants who were their colleagues and with the NDPC staff.

• By contrast, the consultant who undertook the original system design for the second phase of the integrated Personnel and Payroll Database (IPPD2) interacted poorly with the computer services department and the CAGD.

6.5 In this context, the assumption that local consultants are likely to be more sensitive to the requirements of the operating environment and/or more easily accepted by the government counterparts than international consultants may also be questioned based on the experience reviewed. The same issues will determine the acceptability of a local or an international consultant, namely their technical competence and their inter-personal skills and sensitivity to the operational environment.

6.6 A further finding is that in every case where salary supplementation was used to try to address problems of low civil service salaries it created serious tensions and distortions within the Government agency concerned, as well as undermining the joint GoG-DFID management of these projects. Such approaches have not been effective in addressing problems of staff retention and incentives.

6.7 These findings suggest that DFID needs a more systematic approach both to the assessment of the institutional and organisational context in project design and to effective project management. The style and techniques of TC project management by DFID have a major bearing on the success of the project. Knowing how to deal with problems is a crucial aspect of this but there are several other important elements:

• The choice of contractors needs always to be a joint decision by DFID and Government. There will inevitably be costs if this process is shortcut in the interests of speed.21

structured, well-delivered TC. In the case of the AS, improvements were introduced despite the disadvantages of normal civil service pay and conditions.

21 With the IPPD2 contract, the initial contractor for system design was chosen by DFID alone in the

interests of getting the project started and proved to have major shortcomings in interacting effectively with the Government clients and users. The second contractor for software development was chosen directly by GoG, in the interests of ownership, and DFID thereafter never found an appropriate way to voice their misgivings over the technical competence of the chosen contractor. The fact that this was a local company made the issue still more sensitive.


45

• All significant TC projects need to have provision for an independent, properly qualified QA input who can advise both Government and DFID as an honest broker. This is especially important for IT systems projects but also for complicated PFM interventions – a timely intervention by a well informed and impartial third party could have engineered an early re-design of DFID’s support to the MTEF.

• Regular consultations with the Government departments managing the TC are essential: these should be not only with the senior management but also with the immediate project counterparts.

• When things are going wrong, one should be resolute in insisting on corrective action by Government and in ensuring that agreed remedial measures are implemented.

• Project closure – either temporary (‘putting things on ice’) or permanent if the situation calls for it – may be an appropriate response.


46

References

Boesen, N. and O. Therkildsen (2004), Between Naivety and Cynicism: A Pragmatic Approach to Donor Support for Public Sector Capacity Development. May 2004

Booth D. et al. (2004), Drivers of Change in Ghana: Overview Report. London: ODI and Accra: CDD, May 2004

Consulting Africa Ltd. (2000), MTEF Review and Work Plan Workshop (2000), Draft Workshop Report. MTEF Consultants/Consulting Africa Ltd, March 2000

Consulting Africa Ltd. (2002), Medium Term Expenditure Framework, Consultancy Support to the Ministry of Finance. Government of Ghana

DFID (1999), Integration of Donor Flows into the MTEF process, List of funding types. DFID, September 1999

DFID (2000), MTEF Extension Document, Issues to discuss. DFID, May 2000

DFID (2002a), Project Memorandum, Support to the Ministry of Finance Ghana (Draft). DFID, February 2002

DFID (2002b), Project Memorandum, Integrated Personnel and Payroll database Deployment project (IPPD). DFID, July 2002

DFID (2003), Ghana Country Assistance Plan, DFID, April 2003

DFID (2005), Project Completion Report, Integrated Personnel and Payroll database Deployment (IPPD) project (IPPD2). DFID, May 2005

Government of Ghana (2004), Towards a New Public Service for Ghana. Government of Ghana, June 2004

IMF-World Bank (2004), Public Expenditure Management Country Assessment and Action Plan (AAP). World Bank and IMF, May 2004

Lawson A. and G. Salpietro (2000), Government Revenues, User charges and their Incidence in Ghana, Oxford Policy Management: Report to Danish Ministry of Foreign Affairs and Development Co-operation, November 2000

Levy, B. and S. Kpundeh eds (2004), Building State Capacity in Africa: New Approaches, Emerging Lessons, World Bank Institute.

MOF (2003), DFID Public Financial Manager Adviser, Final Report to DFID. Ministry of Finance, Government of Ghana. October 2003

Muggeridge E. (2001) Links between Public Sector Reform and PRSP in Ghana. Consulting Africa Ltd ,April 2001

NDPC (2004a), Implementation of the GPRS 2003 Annual Progress Review, National Development Planning Commission. Government of Ghana, March 2004


47

NDPC (2004b), Ghana Poverty Reduction Strategy 2004 Annual Progress Report. National Development Planning Commission Government of Ghana, March 2004

PUFMARP/MTEF, Ghana Ministry of Finance/DFID discussions. DFID, April 2000

PUFMARP News, Vol.1:No.15, August 1999

PUFMARP News, Vol.1:No.20, September 1999

PUFMARP News, Vol.1:No.21, October 1999

PUFMARP News, Vol.1:No.22, November 1999

PUFMARP News, Vol.1:No.23, December 1999

Stevens M., and S. Teggeman (2004), Comparative Experience with Public Service Reform in Ghana, Tanzania and Zambia in Levy and Kpundeh (2004)

Tihana R. (2000), Developing the Capacity for Macroeconomic Management within the Government of Ghana, Discussion Document. Consulting Africa Ltd, January 2000

Wernham, B., and L. Sutton (2001), Integrated Personnel and Payroll database Deployment project (IPPD), Independent QA Review Report. EmPower Consultancy, April 2001

World Bank (2005), Ghana External Review of Public Finance Management (ERPFM), World Bank


48

List of persons consulted

NAME TITLE ORGANISATION

Emma Spicer Acting Country Director DFID Ghana

Paul Walters

(via Video link)

Senior Economist DFID Ghana

Denise Hill Deputy Programme Manager, Governance Team

DFID Ghana

Ruby Adjaidoo Bentsi Economist DFID Ghana

Jake Tetteh Assistant Programme Manager, Governance Team

DFID Ghana

Roberto Tibana Senior Adviser, MDBS Public Financial Management

DFID Ghana

George Otoo Procurement Adviser National Procurement B

Christian T. Sottie Controller & Accountant General Controller & AccountanGeneral’s Department (CAGD)

Daniel Y. Domilu Director, IPPD CAGD

Cephas Dosu Head, Administration CAGD

Ben Doku Director, Pay Roll Processing Division CAGD

S. B. Nyantakyi Director, Budget MOFEP

Dickson Tamakloe Budget Division MOFEP

Mr. Asamoah Enin Budget Division MOFEP

Osei Gyamera ADMU MOFEP

George Afriyie Gyamfi Information Technology Manager, Team Leader, PUFMARP

MOFEP

Michael Gyamfi Project Implementation Leader, PUFMARP

MOFEP

Mrs. Agatha Gaisie-Nketsiah

Special Assistant to the Minister (Implementation)

MOFEP

Mr. Prempeh Ag Director, Bilateral Relations MOFEP


49

Adu K. Augustus External Resource Mobilisation, Bilateral/DFID Desk Officer

MOFEP

Sammy Arkhurst Policy Analysis Division MOFEP

George Blankson Deputy Commissioner, Operations VAT

Dr. Bentsi Enchil Director, Research VAT

Mr. Doe Head, Research and Statistics VAT

Mrs. Nanemeh Head, Administration VAT

Dr. Regina O. Adutwum Ag. Director General NDPC

Capt Patrick Isaac Donkor (Rtd)

Acting Director, Monitoring and Evaluation

NDPC

Mr. Eric N. Nortey Head of Administration NDPC

Mr. Asiamah Administration NDPC

Ms Angela Brown Farhart Programme Coordinator, Wealth Creation and Social Development Programme

NDPC

Mr. Nelson Planning NDPC

Mr. Wim Olthof First Secretary/Head of Section Micro Economics and Trade

EU Delegation of the European Commission Ghana

Prof Ato Ghartey Former Controller & Accountant General/CEO

Ato Ghartey & Associat

Mr. Ike Duker Managing Director Crown Agents Ghana L

Leonard T. Tingbani Director, Basic Education Division (BED) Ghana Education Servi(GES)

Ms Victoria Donkoh Deputy Director, BED GES

Moses Mabenba Assistant Director, BED GES

Bertinus Gbaben Assistant Director, BED GES

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

50

Su

mm

ary

of

Fin

din

gs –

Evalu

ati

on

Ma

tric

es

Rele

van

ce

ND

PC

VA

T S

erv

ice

MO

FE

P

Rele

vance

to

Org

anis

ational

needs

&

Opera

tional

conte

xt

Sup

port

to G

PR

S p

rocess c

ert

ain

ly

need

ed b

ut th

e e

xte

nt

of

capacity

build

ing r

equ

irem

ents

pro

bably

undere

stim

ate

d. L

imite

d in

fluence/

sta

tus o

f N

DP

C a

t o

uts

et

of

pro

ject not

fully

re

alis

ed.

Sup

port

to intr

od

uctio

n o

f V

AT

was

explic

itly

desig

ned

for

a “

sta

rt-u

p”

org

an

isation &

managed

in a

wa

y to

guara

nte

e s

kill

s tra

nsfe

r &

long-t

erm

susta

ina

bili

ty.

En

hance

d IR

S s

ala

ry

scale

s e

nsure

d a

ttra

ctio

n &

rete

ntion

of

appro

priate

sta

ff.

MT

EF

& IP

PD

2 identified in

PU

FM

AR

P a

s

necessary

but specific

tech

nic

al fo

rm b

oth

in

terv

en

tions to

ok d

id n

ot a

ddre

ss b

asic

is

sues o

f non-c

om

plia

nce

underl

ying w

ag

e

bill

pro

ble

m. D

FID

mgt/ r

evie

w p

rocess w

as

not a

ble

to r

ealis

e t

his

until hig

h levels

of

‘sunken c

osts

’ incurr

ed. In

sharp

contr

ast to

IP

PD

2,

MT

EF

pro

ject

was d

elib

era

tely

str

uctu

red to e

ng

age

MO

FE

P le

aders

hip

+

genera

te G

oG

part

icip

atio

n. H

ow

eve

r, n

eith

er

pro

ject based

on p

rop

er

un

ders

tand

ing o

f th

e

polit

ica

l eco

nom

y fa

cto

rs d

rivin

g p

oor

budg

eta

ry o

utc

om

es. N

eith

er

pro

ject

appre

cia

ted s

ensitiv

ity

of

“allo

wances issue”.

Consi

stency

of

DF

ID’s

and

org

anis

ation’s

obje

ctiv

es

ND

PC

DG

im

media

tely

re

alis

ed

import

ance o

f G

PR

S p

roce

ss a

nd

expre

ssed s

trong d

em

and f

or

DF

ID

support

.

The f

irst V

AT

Com

mis

sio

ner

was f

ully

com

mitt

ed to t

he p

rocess. H

e &

his

sta

ff

expre

ssed s

trong d

em

and f

or

DF

ID

support

.

Initia

l dem

and f

or

DF

ID s

up

port

expre

ssed b

y G

oG

but com

mitm

ent lo

st

when a

) d

ispute

s

occurr

ed o

ver

allo

wances;

and b

) desig

n

pro

ble

ms b

ecam

e c

lear.

DF

ID r

eaction t

o

these p

roble

ms s

erv

ed to f

urt

her

dim

inis

h

com

mitm

ent.

Consi

stency

w

ith a

ctio

ns

by

GoG

and o

ther

DP

s

NP

P g

ovt

ga

ve incre

asin

g im

port

ance to

GP

RS

and

its d

evpt, a

s d

id a

ll D

Ps o

nce

decis

ion to

seek H

IPC

re

lief

taken.

Sup

port

by

oth

er

DP

s c

om

ple

menta

ry.

Str

on

g p

olit

ica

l com

mitm

ent to

VA

T

despite initia

l pro

test ri

ots

. D

FID

work

w

ell

syn

chro

nis

ed w

ith I

MF

, W

B a

nd

GT

Z in

itiative

s.

Initia

l PU

FM

AR

P d

iagn

osis

and D

FID

pro

ject

desig

ns w

ere

incom

ple

te a

nd d

id n

ot

pro

perl

y id

entif

y assum

ptio

ns f

or

pro

ject purp

oses to

be a

ch

ieve

d. T

hus, re

qu

ire

d G

oG

actions n

ot

ide

ntif

ied u

ntil l

ate

and

arg

uably

eve

n w

ithin

M

DB

S d

ialo

gue h

ave

not b

een p

roperl

y defined

. O

ther

DP

s (

+IM

F)

did

not

active

ly

enga

ge u

ntil ad

ven

t of

MD

BS

.

Yes –

GP

RS

fra

mew

ork

perm

itted

Exte

nsiv

e p

rep

ara

tion p

rocess b

etw

een

In p

rincip

le,

PU

FM

AR

P p

rovi

ded t

he

consulta

tion p

rocess b

ut aft

er

initi

al desig

n

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

51

ND

PC

VA

T S

erv

ice

MO

FE

P

Consu

ltation

pro

cess

to

ensu

re

rele

vance

?

consulta

tion o

ver

TC

ne

eds.

GoG

, D

FID

, W

B &

IM

F.

Rela

tionship

betw

een V

AT

serv

ice a

nd

Cro

wn A

ge

nts

Ghan

a o

ffic

e h

elp

ed

ensure

appro

pri

ate

desig

n &

mgt of T

C.

work

, G

oG

input

was lim

ited a

nd P

UF

MA

RP

ste

eri

ng

com

mitt

ee n

eve

r w

ork

ed e

ffectiv

ely

. M

ore

ove

r, P

UF

MA

RP

did

not pro

vide c

lear

fram

ew

ork

for

DP

s’ i

nput

on P

FM

issues.

This

has o

nly

re

ally

em

erg

ed w

ith M

DB

S.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

52

Eff

ecti

ven

ess

ND

PC

VA

T S

erv

ice

MO

FE

P

Were

outp

uts

g

enera

ted? I

f not,

wh

y not?

Yes -

Core

outp

uts

genera

ted

and s

yste

ms &

pro

cesses for

GP

RS

mgt cre

ate

d. E

xte

nt

of

skill

s tra

nsfe

r le

ss th

an G

oG

envis

ag

ed.

Aft

er

initi

al ‘fals

e s

tart

’ du

e t

o V

AT

rio

ts

(and h

igh s

tart

ing r

ate

), p

roje

ct outp

uts

w

ere

ve

ry larg

ely

co

mple

ted

with

exceptio

n o

f V

IPS

insta

llation w

hic

h

was 9

0 %

function

al.

Out of

10 o

utp

uts

specifie

d in o

rig

ina

l pro

ject

docum

ent, M

TE

F p

roje

ct a

chie

ved 5

/10 f

ully

, 3 p

art

ially

and 2

no

t at

all.

The p

roble

m w

as that th

ese w

ere

not

the r

ight

outp

uts

to a

chie

ve

the p

roje

ct purp

ose!

More

ove

r, c

ove

rage o

f M

TE

F w

as insuff

icie

nt. (

See

table

12.)

IPP

D2 is s

till i

ncom

ple

te d

ue la

rgely

to s

ele

ctio

n o

f in

appro

pri

ate

contr

acto

rs a

nd p

oor

pro

ject m

gt b

y D

FID

& G

oG

.

Was

ow

ners

hip

m

axi

mis

ed?

Yes –

str

ong o

wners

hip

at

org

an

isationa

l le

vel an

d o

ver

time a

t G

oG

leve

l.

Yes –

str

uctu

re o

f pro

ject m

anag

em

ent

delib

era

tely

lo

w k

ey

with le

ad v

ery

firm

ly w

ith V

AT

serv

ice.

Ow

ners

hip

of

MT

EF

lim

ited

by

use o

f cle

arl

y separa

ted

pro

ject te

am

and b

y d

ispute

s o

ver

allo

wances to

no

n-

pro

ject te

am

mem

bers

but th

ere

is n

o d

ou

bting t

he

initia

l enth

usia

sm

for

the p

roje

ct w

ithin

MO

FE

P a

nd

line m

inis

trie

s.

For

IPP

D2, th

ere

must be q

uestio

ns a

bout

initi

al

com

mitm

ent of

GoG

(especia

lly C

AG

D)

but

DF

ID’s

poor

mgt of

initia

l contr

acto

r sele

ctio

n a

nd c

onflic

tive

appro

ach to r

eso

lvin

g p

roble

ms d

estr

oye

d a

ny

ch

ance

of

real o

wners

hip

.

Was

alig

nm

ent

maxi

mis

ed?

Lin

k to s

ecto

r p

lan

nin

g

pro

cesses a

nd t

o b

udg

et in

itia

lly

weak b

ut has s

tre

ngth

ene

d o

ver

time.

Yes,

VA

T in

tegra

ted to

exis

ting

IR

S

fram

ew

ork

and in

tro

duce

d t

hro

ug

h

GoG

polic

y decis

ions.

In p

rincip

le,

link to P

UF

MA

RP

sho

uld

ha

ve a

ddre

ssed

alig

nm

ent concern

s b

ut th

ere

is a

question m

ark

ove

r depth

of

GoG

com

mitm

ent to

th

is.

Was

harm

onis

ation

maxi

mis

ed?

Yes –

inp

uts

to G

PR

S w

ell

harm

onis

ed a

cro

ss D

Ps.

Yes –

clo

se h

arm

onis

atio

n w

ith W

B

and I

MF

. P

UF

MA

RP

pro

vided in

itia

l fr

am

ew

ork

for

harm

onis

atio

n b

ut

few

DP

s in

volv

ed in th

is a

nd th

ose

who w

ere

(D

FID

, W

B)

did

not re

main

who

leheart

edly

enga

ged

.

Were

ou

tpu

ts

Gen

der

issu

es

were

In

su

ffic

ien

t in

form

ati

on

on

th

is

Ins

uff

icie

nt

info

rma

tio

n o

n t

his

is

su

e.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

53

ND

PC

VA

T S

erv

ice

MO

FE

P

gen

era

ted

in

g

en

der-

sen

sit

ive

way?

reaso

nab

ly w

ell in

co

rpo

rate

d

in G

PR

S &

in

mo

nit

ori

ng

p

roces

s.

issu

e.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

54

Eff

icie

nc

y

ND

PC

VA

T S

erv

ice

MO

FE

P

Were

outp

uts

genera

ted

eff

icie

ntly

in

term

s o

f cost and

timelin

ess?

Pro

ject outp

uts

very

larg

ely

deliv

ere

d a

t cost &

within

dead

lines, b

ut

as n

ote

d

exte

nt

of

skill

s tra

nsfe

r to

ND

PC

less

than G

oG

ha

d e

nvis

age

d.

Takin

g a

ccount

of

“fals

e s

tart

” and n

eed

fo

r re

vised s

che

dule

, o

utp

uts

were

deliv

ere

d c

ost-

eff

ectiv

ely

and w

ithin

re

vise

d s

ched

ule

, w

ith m

inor

exception

of

VIP

S s

yste

m.

Deliv

ery

of

MT

EF

outp

uts

until

pre

matu

re p

roje

ct clo

sure

was la

rgely

to

schedule

, if s

lightly

expe

nsiv

e d

ue t

o

hig

h w

ork

shop e

lem

ent and

rela

ted

allo

wa

nces. P

art

icip

ato

ry m

ode o

f deliv

ery

was a

str

on

g p

oin

t.

IPP

D2

was e

xceed

ing

ly in

eff

icie

nt

with

man

y b

asic

rule

s o

f th

e d

eliv

ery

of

IT

sys

tem

s r

epeate

dly

bro

ken b

y th

e

contr

acto

rs, D

FID

and G

oG

.

Would

there

have b

een

alte

rnative &

m

ore

eff

icie

nt

modes o

f deliv

eri

ng s

am

e

outp

uts

?

No. A

diffe

rent pro

ject

desig

n c

ou

ld h

ave

ensure

d m

ore

skill

s tra

nsfe

r but th

is

would

have c

om

pro

mis

ed tim

elin

ess

whic

h w

as c

rucia

l in

lig

ht

of

HIP

C

accessio

n targ

et.

No. T

he d

esig

n a

nd

de

livery

were

clo

se

to o

ptim

al.

No

n-c

om

ple

tion o

f V

IPS

is a

m

inor

issue w

hic

h is b

ein

g a

ddre

ssed

under

GoG

fun

din

g.

Yes –

for

IPP

D2, contr

acto

rs s

hould

have b

een r

equ

ired t

o f

ollo

w U

K G

ovt

C

CT

A g

ood

pra

ctic

e; an

d D

FID

/ G

oG

should

ha

ve f

ollo

wed

go

od p

ractic

e in

contr

acto

r sele

ctio

n &

mgt.

For

MT

EF

, m

ode o

f deliv

ery

was s

ound,

except th

at d

ang

ers

of

‘allo

wances/

pro

ject encla

ve c

ulture

’ not

fully

re

cognis

ed.

Imp

act

ND

PC

VA

T S

erv

ice

MO

FE

P

Did

pro

jects

achie

ve e

xpecte

d

transactio

na

l/ tr

ansfo

rmatio

na

l im

pact?

Pro

ject succeede

d in b

ring

ing a

cohere

nce a

nd

focus to e

conom

ic

managem

ent, in

rais

ing im

port

ance o

f povert

y re

duction a

nd

in e

nhancin

g

influe

nce o

f N

DP

C. In

short

, a m

ajo

r im

pact

Pro

ject succeede

d in e

sta

blis

hin

g a

vi

able

, se

lf-s

usta

inin

g V

AT

colle

ction

serv

ice.

Ag

ain

, a m

ajo

r im

pact.

MT

EF

pro

ject has h

ad a

tra

nsactio

nal im

pact

in c

ha

ng

ing t

he w

ay

the b

udget

is p

rep

are

d

but

it d

id n

ot

achie

ve it

s p

roje

ct purp

ose o

f im

pro

ving t

he c

ost

eff

ectiv

eness o

f G

oG

re

sourc

e a

llocatio

n a

nd d

id n

ot tr

ansfo

rm the

pro

cess o

f econ m

gt.

IPP

D2 r

em

ain

s in

com

ple

te in 2

005

, seven

years

aft

er

sta

rt. It

ma

y ha

ve la

id t

he

gro

un

dw

ork

for

an e

ffectiv

e p

ayr

oll

sys

tem

but

this

is y

et

to b

e p

rove

n.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

55

Su

sta

inab

ilit

y ND

PC

VA

T S

erv

ice

MO

FE

P

Have

transfo

rmatio

na

l im

pacts

been

susta

ine

d?

So f

ar,

yes, tr

ansfo

rmatio

nal im

pacts

have b

een s

usta

ined b

ut

it is

cle

ar

tha

t N

DP

C d

o n

ot

ha

ve t

he c

ap

acity

to

susta

in t

he G

PR

S s

yste

ms w

ithout

either

TC

or,

pre

fera

bly

, org

anis

ation

al

refo

rms.

Yes. 5 y

ears

aft

er

the c

lose

of

bulk

of T

C

pro

ject, im

pacts

have

be

en s

usta

ine

d

and V

AT

perf

orm

ance c

ontin

ues to

im

pro

ve.

Neither

pro

ject h

as h

ad a

tr

ansfo

rmatio

na

l im

pact but

it is

im

pre

ssiv

e to n

ote

ho

w f

ar

the m

ore

tr

ansactio

na

l chan

ges intr

oduced b

y th

e

MT

EF

pro

ject ha

ve b

ee

n s

usta

ined.

If

the f

ocus h

ad

be

en m

ore

appro

pri

ate

/ re

leva

nt,

sig

nific

ant

success c

ould

ha

ve

been

achie

ved.

Was there

a

cohere

nt

exit

str

ate

gy

pre

pare

d in

advance?

Exit

str

ate

gy

insuff

icie

ntly

deve

loped –

specific

ally

, lit

tle tho

ug

ht h

as b

een g

ive

n

to h

ow

to t

rigg

er

the o

rganis

atio

nal

refo

rm w

hic

h w

ill b

e n

ecessary

to

susta

in a

chie

vem

ents

witho

ut fu

rther

TC

.

Yes –

in t

hat

the w

ho

le p

roje

ct desig

n

was b

ased

on G

oG

lead

ers

hip

fro

m the

outs

et

and p

roje

ct in

puts

were

ta

ilore

d t

o

be s

uff

icie

nt

with

out e

ncro

achin

g o

n

lea

ders

hip

/ re

spo

nsib

ility

of

GoG

.

No –

both

pro

jects

ha

ve h

ad m

essy/

acrim

onio

us e

xits

aft

er

pre

matu

re

clo

sure

. In

the

case o

f M

TE

F, if the

contr

acto

r cou

ld h

ave b

een

forc

efu

lly

redir

ecte

d to c

hang

e th

e te

chnic

al

appro

ach th

en t

here

wou

ld h

ave b

een

no n

eed f

or

this

pre

matu

re e

xit

and

much m

ore

could

ha

ve b

ee

n a

chie

ved.

With

IP

PD

2, a

n e

arlie

r exit

was n

ee

ded

, fo

llow

ed b

y a

pa

use f

or

a G

oG

reactio

n

to th

e p

rob

lem

s a

nd o

nly

th

en a

re-

enga

gem

ent b

y D

FID

, if d

em

anded.

Technical Co-operation for Economic Management Volume 2: Country Case Studies

56

KENYA CASE STUDY


57

Executive Summary

S1. This report is a case study of DFID-financed technical cooperation (TC) for economic management in Kenya focusing on the period 1999-2004. Specifically it evaluates support to four organisations or clusters of organisations:

• The Ministries of Finance and of Planning and National Development (public expenditure management, financial management information system, monitoring and evaluation).

• The Office of the President and Kenya Institute of Administration (civil service reform).

• Ministry of Local Government (local government reform).

• Kenya Anti-Corruption Authority (anti-corruption).

S2. The bulk of the relatively small TC expenditure over the evaluation period (about £4 million in total) was spent in support of local government reform. Other than that, the activities reviewed were limited in scale and reflected a DFID approach based on cautious engagement to support promising reform initiatives many of which did not yield results. Three of the eight projects covered were cancelled or discontinued and others faced significant delays.


Political context

S3. Kenya’s political system has been characterised by centralisation of power on the presidency and associated with a deep-rooted culture of patronage, nepotism and corruption. In December 2002 elections led to the defeat of the Kenyan African National Union (KANU) which had held power since independence, and its replacement by the National Rainbow Coalition (NARC) with an agenda of boosting economic growth, reducing poverty and improving governance. Despite some promising early initiatives, the NARC government has not brought about fundamental political change and has not provided an effective environment for consistent policy-making or budgeting


S4. The concentration of power in the Presidency has been accompanied by an undermining of other public institutions, including the judiciary and the civil service over a period of many years. The capacity of the civil services to play an effective part in designing and implementing policy, including in the field of economic management, is very weak. Lack of capacity has been compounded by endemic corruption at all levels of the state. A number of reforms were initiated by the former government but these achieved little fundamental impact. Whether initiatives introduced by the NARC government will prove more effective remains uncertain.


58


S5. Within this context, DFID along with other donors has faced a difficult environment for effective engagement with government or support to capacity development activities. DFID’s strategy as articulated in 1998 envisaged that technical assistance might be used for capacity development in the event that there was clear progress at the policy level and evidence of government commitment to reform, while in the absence of reform, technical assistance would be directed at building commitment to better governance and pro-poor policies. DFID’s update of its strategy in 2002 envisaged the provision of capacity development support in several areas where there was judged to be progress including local government reform. The 2004 Country Assistance Plan noted explicitly that technical advice and ideas could be used to back reforms that are genuinely owned in country, but in contrast to the 1998 strategy took the view that in the absence of such commitment donors had little capacity to influence policy through TC. In practice the activities reviewed can all be seen as directed towards building capacity to strengthen key government functions. Judgements about progress and government commitment have guided decisions about the level of engagement.

Economic management in Kenya

S6. Overall during the period Kenya has had relatively sound macroeconomic management, which has halted and partially reversed the earlier trend of declining real economic growth, but also a continuing poor record in terms of budgetary management. The result of this has been that budgetary allocations and expenditure have borne little relation to stated policies, a large gap between budget and expenditure out-turns has persisted, and there has been little or no progress in the improvement of service delivery through the budget. The high level of corruption has continued to be a major cause of concern.

S7. Some progress has however been made in laying the groundwork for improved budgetary performance including the development of the PRSP in 2001 and the subsequent Economic Recovery Strategy (ERS), and of the public expenditure reform process since 1997. Since this period and especially under the NARC government there has been progress in the implementation of measures aimed at improving public expenditure management and public service reform. It is too early to judge whether these reforms are going to be sustained and have a greater impact than earlier reform initiatives. However, these measures can be seen as laying the groundwork for improved budgetary management.


S8. There is evidence that significant changes to which DFID TC has contributed have occurred during the period under review in terms of the capacity of the Ministry of Finance and the Ministry of Planning and National Development to provide effective management of the planning and budgetary cycle. However, the full benefits of this increased capacity have not yet been realized. The change in capacity has primarily


59

taken the form of the creation of new processes and procedures. These have the potential to provide greatly enhanced transparency and accountability in public expenditure management. Whether or when the full benefits will be achieved and how sustainable the new processes will prove to be (in the absence to date of measures addressing the wider organisational capacity constraints faced) remains to be demonstrated.

S9. DFID had been substantially involved in support to civil service reform between 1993 and 1998 and contributed to the design of a follow up phase (envisaged as running from 1998 to 2001). However the CSR programme was not consistently implemented. DFID envisaged providing support to the Kenya Institute of Administration (KIA) as part of this process, but this support was terminated after a first phase and DFID has subsequently withdrawn from direct involvement in supporting civil service reform, although there have been a number of subsequent government initiatives in this area.

S10. DFID support to the Ministry of Local Government has contributed to an improvement in the systems for the oversight of local authorities and the strengthening of management systems despite some problems in the design and implementation of the DFID project. However, this progress has been critically dependent on project staffing and finance and so its sustainability remains in doubt. DFID provided some initial support to the Kenya Anti-Corruption Agency but this ceased when the government failed to secure the organisation’s survival in the face of legal challenges to the constitutionality of its powers.


Relevance

S11. The approach and role used for TC was judged to be relevant and appropriate and was focused on areas of reform that did have a potentially major impact on the quality of economic management. The support provided was also consistent with DFID’s strategy and the approaches of other donors. The limited and provisional success achieved suggests that the commitment of additional TC resources would not have led to greater impact but the appropriateness of the particular judgements made about continuing or stopping support is in some cases contested.

Effectiveness

S12. In general the projects reviewed in this study were effective in terms of the delivery of anticipated project outputs, with the exception of the KACA project which was terminated when the beneficiary organisation was wound up. Indeed even in the case of the KIA project, which was discontinued in the face of lack of government commitment, the contribution was effective in the delivery of outputs. With respect to the IFMIS project the judgement on effectiveness is still pending as not all outputs of the project have been delivered. Ownership within the organisations supported (and alignment with government policies) has been variable. For the support to the Ministry of Finance and MPND, there is evidence that ownership has increased over the period.


60

In the case of KACA the project was necessarily terminated when lack of political commitment led to the organisation’s demise. In MOLG, while a level of local ownership was achieved, the projects tended to become relatively isolated islands of competence within a generally weak ministry.

Efficiency

S13. The support to developing the Integrated Financial Management and Information System (IFMIS) at the Ministry of Finance appears to have been highly cost effective but delivery of the outputs was delayed. In the PEM project local consultants were recruited to work alongside the external consultants for all of the sub-set of PER-related assignments; these activities appear to have been cost-effective. The decision to terminate support to KACA was done promptly and minimised waste. In the case of support to MOLG, weaknesses in project design led to delays and may have inflated costs. Support to local government reform has been by some way the largest area of expenditure on technical cooperation over the evaluation period, accounting for over half the recorded expenditure. The limited support provided to civil service reform was delivered extremely promptly and cost effectively in response to government requests, but this approach may have compromised ownership.

Impact

S14. The DFID TC projects reviewed in this study typically only addressed a limited range of capacity issues, being mostly concerned with the development of management systems, rather than with the establishment of the staffing and other capacities necessary for the sustained support of those systems. The projects reviewed therefore have the potential to have impact but this largely depends on progress with government’s commitment to reform, and on measures to address these wider aspects of organisational capacity that have not been the focus of DFID support.

Sustainability

S15. In the projects reviewed a comprehensive approach to organisational capacity development did not in fact form the focus of support. For the immediate future these systems are not sustainable in the absence of continued TC and there is a real danger of discontinuity and breakdown without either continued external support or increased government commitment – which would mainly relate to a willingness to address issues of civil service employment to attract and retain adequately skilled workers.


S16. The overall assessment is that the limited support DFID has provided has contributed to some successes in strengthening and establishing management systems relevant to economic management. However, achieving impact will require both a level of sustained political commitment to reforms that has never previously been achieved in Kenya, and measures to address fundamental issues of organisational capacity. While the overall assessment of the evaluation is that there are


61

some grounds for optimism that initiatives will be maintained and that government commitment to strengthening public finance management and civil service reform is growing and is greater than it has been in the past, the prospects remain highly uncertain.

S17. The strategic judgement about DFID’s engagement and use of TC comes down to balancing the risk that relatively modest expenditures will not yield sustainable results against the potential benefits from continuing engagement in developing systems that may facilitate future reforms even though key constraints on organisational capacity are not amenable to being addressed. The assessment of the evaluators is that in the cases reviewed, the potential benefits probably do outweigh the risks (and the same might be argued to have applied for continuing engagement with the CSR process from which DFID withdrew).

S18. Key conclusions for DFID’s future engagement are that DFID should take a long-term perspective and be prepared to continue engagement and in some cases the provision of TC even when the political context may be unfavourable, provided that there is judged to be a reasonable possibility of the situation improving in the future. In addition, DFID should work to help create a more comprehensive and robust master plan for reform, which may help to address the problems of sustainability noted in most of the projects reviewed – or at a minimum highlight the limitations on what can be achieved in terms of capacity development without wider support and reforms. DFID TC should also operate where possible within the context of a programme defined in common with other development partners as well as the GOK.


62

1. Introduction

1.1 This case study is one of four (the others being Ghana, South Africa and Zambia) in a wider study to evaluate the role of DFID Technical Cooperation (TC) in Economic Management in sub-Saharan Africa. The purpose of the wider study is:


1.2 The overall approach and conceptual framework for the country studies is discussed in the Synthesis Report (Volume 1) and in the introduction to this volume.

1.3 This study used the following main sources of information:

• Desk review of documentation on the evolution of the Kenyan economy and political system.

• Detailed review of the project files in the DFID Kenya country office.

• Interviews with key informants in the recipient organisations, the DFID Kenya advisers and discussions with development partners during a visit in June 2005 (with some subsequent interviews in February 2006).

• Presentation of the results of the study to a group of DFID Kenya country office advisers in February 2006.

1.4 In Kenya the areas of support selected for review related to four main organisations or clusters of organisations22 as defined by their role in the economic management process:

• The Ministries of Finance and of Planning and National Development (public expenditure management, financial management information system, monitoring and evaluation).

• The Office of the President and Kenya Institute of Administration (civil service reform).

• Ministry of Local Government (local government reform).

• Kenya Anti-Corruption Authority (anti-corruption policy).

22 Support through the Kenya Institute for Public Policy Research and Analysis (KIPPRA) to improving the

enabling environment for business development in Kenya was initially included in the evaluation on the grounds that it contained a small component focused on short-term initiatives to build capacity within government to help understand and respond to private sector needs. However this has been omitted from the case study report on the principal grounds that it was not possible to obtain or crosscheck sufficient information about the capacity development activities to draw firm conclusions. The information available suggested that the capacity development activities focused on government had limited impact because they were not well-integrated into the project design and were not given priority until late in project implementation. Support was also provided to the start up and initial running costs of Kenya Private Sector Alliance (KEPSA) an umbrella body for the private sector.


63

1.5 Table 1 provides information on the project expenditure for the projects reviewed over the period 1999-2004, and places this in the context of the total DFID country programme, and total TC expenditure in the more broadly defined economic and governance sectors. The most striking feature is the large gap between planned and total expenditure for those projects for which expenditure information is available. This reflects first, significant delays in implementation compared to original plans, and second the fact that much of the TC provided to Kenya has been based on a strategy of providing support to organisations and reform processes that have been identified as potentially having a catalytic impact on government performance, but with the scale of the support and its continued provision conditional on demonstrated government commitment and progress against milestones. Most of these processes have turned out to be disappointing in terms of the level of government commitment and the organisational context, with the result that disbursements have generally been well below what was originally planned. The expenditures covered by the evaluation amounted to less than £4 million over the period.

1.6 This poses some problems in relation to drawing conclusions from the case study that inform the overall evaluation objective (as well as in formulating appropriate counterfactuals to assess what alternative approaches might have been followed). The activities undertaken in Kenya were relatively dispersed in nature reflecting the attempts to provide support in areas and to organisations that appeared promising, while actual expenditures were small reflecting the assessment that there was in most cases insufficient progress to warrant the provision of additional resources. Three out of the eight projects covered were cancelled or otherwise discontinued without achieving the full planned delivery of outputs. In this context, it is considered important to make an assessment of the overall engagement strategy that DFID has followed (in relation to capacity development for economic management), as well as assessing the capacity development contribution of the relatively sparse set of activities actually undertaken. The interpretation of the potential impact of more recent activities reviewed also depends heavily on judgements about the extent to which there has been or is likely to be a strengthening in the political commitment to reform or measures to address organisational constraints that will enable progress to be built on. Such judgements are necessarily provisional in nature.

1.7 The fact that a number of the projects were discontinued early in the period covered by the evaluation also contributed to problems in finding either documentary sources about, or well-informed participants in or observers of, the activities undertaken. On the DFID side a contributory factor to this was the frequent changes in staffing and responsibility within the country team and the lack of mechanisms to maintain organisational memory.

1.8 This report is organised as follows: Section 2 reviews the political and institutional context within which DFID support has been provided and the overall strategy that has guided DFID’s approach. Section 3 assesses economic management in Kenya over the period of the evaluation. Section 4 summarises the findings of the evaluation in relation to the impact of improving organisational effectiveness of the DFID TC provided for each of the main areas of support. Section 5 summarises the conclusions in terms of the main evaluation questions (based on the detailed summaries in the Evaluation Matrix in Annex A). Section 6 discusses conclusions in relation to alternative approaches that might have been followed and implications for future DFID TC in Kenya.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

64

Ta

ble

1.

DF

ID T

C p

roje

ct

ex

pen

dit

ure

in

Ke

nya

: E

co

no

mic

an

d F

ina

nc

ial m

an

ag

em

en

t

Fis

cal

Year

(Ap

ril

- M

arc

h);

£ '0

00

s

Pro

jec

t1

99

9/0

02

000

/01

20

01

/02

20

02/0

32

00

3/0

4

Lo

ng

Te

rm

Sh

ort

Term

Co

ns

ult

an

ts

Lo

ng

Te

rm

Sh

ort

Term

Co

ns

ult

an

ts

Lo

ng

Te

rm

Sh

ort

Te

rm

Co

ns

ult

an

ts

Lo

ng

Term

Sh

ort

Te

rm

Co

ns

ult

a

nts

Lo

ng

Te

rm

Sh

ort

Term

Co

ns

ult

a

nts

To

t. E

xp

.

99

-04

Pla

nn

ed

Exp

.

CS

RF

P I

....

220

....

....

....

....

....

....

22

022

0

CS

RF

P 2

....

....

36

....

....

....

....

....

36

32

5P

ub

licS

erv

ice

Pro

v...

....

315

..6

16

92

08

734

013

6..

..65

01

50

0

Sup

po

rt t

o K

IA*

....

125

....

....

....

....

....

....

12

562

8

Sup

po

rt t

o K

AC

A..

....

....

2..

....

....

....

....

222

5

IFM

IS5

..1

16

6..

99

2..

118

27

117

....

119

59

11

05

0

PE

M..

....

....

....

....

....

..1

5..

45

60

95

0

M&

EIP

-ER

S*

....

....

....

....

....

....

....

....

10

0

PR

OLO

Gs

....

....

....

....

..8

..90

37

..1

34

51

480

500

0

KLG

RP

*..

....

....

....

....

....

....

....

..5

79

8

Urb

an P

ove

rty

Pro

j.20

....

53

....

25

8..

53

12

1..

160

4..

865

7..

PA

MN

UP

32

....

99

....

99

..3

188

....

20

..0

33

7..

KIP

PR

A*

....

....

....

....

....

....

....

....

290

0

TC

eva

luate

d57

02

41

473

36

10

752

82

02

89

25

37

380

82

01

51

7..

..T

ota

l/Y

ear

29

861

68

37

640

15

99

3990

..

% o

f T

ota

l E

co.

&

Go

v. T

C11

%1

9%

29%

24

%44%

26

%..

% o

f T

ot.

DF

ID

Co

un

try

Pro

g1

.1%

1.2

%3.1

%1.5

%6

.3%

2.3

%..

To

tal

Ec

o.

& G

ov

.

TC

2787

32

05

28

82

266

336

44

151

81

..

% o

f T

ota

l D

FID

Co

un

try

Pro

g10

%6

%1

1%

6%

14%

9%

..

To

t. D

FID

Co

un

try

Pro

g.

266

34

515

79

26

83

543

64

12

55

84

17

427

3..

The table

sum

mari

ses tota

l spend

ing

by

DF

ID o

n e

valu

ate

d T

C p

roje

cts

rela

ted t

o e

con

om

ic a

nd f

inancia

l manag

em

ent. E

xpe

nd

iture

is a

lso s

ho

wn f

or

the t

ota

l econ

om

ic a

nd g

overn

ance s

ecto

rs T

C a

nd th

e tota

l DF

ID C

ountr

y P

rogra

mm

e for

the p

eri

od 1

999-2

004

. T

he f

igure

s a

re b

ase

d o

n t

he lis

ting

of

TC

pro

jects

pro

vid

ed b

y th

e D

FID

Eva

luation D

epart

ment an

d info

rmatio

n c

olle

cte

d b

y th

e c

onsultant

duri

ng t

he f

ield

vis

it in

Ke

nya

.

* N

o info

rmatio

n o

n a

ctu

al expend

iture

by

year

availa

ble

. T

ota

l spen

d o

n K

IA 1

99

8-2

00

4 w

as £

478k. T

he p

roje

ct sta

rted in M

arc

h 1

998.


65

2. The Context for Capacity Development

The Political Context

2.1 Since independence Kenya’s political system has been characterised by a high degree of centralisation of power in the hands of the incumbent president and his close political aides and allies, and more generally in the executive. This process has also been associated with a deep-rooted culture of patronage, nepotism and corruption.23 An overall review of the political context in Kenya commissioned by DFID summarised the main features of the regime as follows (Ng’ethe et al. p. 16):

“The first two Presidents of Kenya developed and presided over a patrimonial state. In this system the structures of a modern nation state (executive, legislature, judiciary etc.) existed only in the formal sense, and power was heavily concentrated in the hands of the head of state. Power operated through a web of informal, clientelist networks based on personal ties between leaders and supporters at all levels of the political hierarchy. These informal networks permeated public institutions and subverted formal rules and decision-making. They undermined systems of public accountability and created conditions where corruption and rent-seeking behaviour could flourish.”

2.2 For almost 40 years Kenya was dominated by a single political organisation, the Kenyan African National Union (KANU), which controlled both the executive and legislative branches of government. Multi-party politics was introduced in 1992, but political parties formed on ethnic lines and narrow constituency bases, and the Kenyan political economy continued to be dominated by a strong presidency up until the December 2002 elections. These elections resulted in the first change in government since independence as voters signalled their discontent with the KANU government, the resultant poor economic and social conditions, and their desperation for change. The National Rainbow Coalition (NARC), a coalition of 16 political parties, came to power with an agenda to bring about economic growth, a reduction in poverty and an improvement in governance.

2.3 NARC began their term in office with a number of actions to change the nature of the political processes, which raised the hopes of the electorate and the international community. Examples of these were a constitutional review to reduce the powers of the executive, a suspension of corrupt judges and the passing of several public sector ethics and anti-corruption bills. Since then holding the coalition together has been a challenge for President Kibaki, and optimism about the new regime dwindled as the difficulties of reform became apparent and the true extent of political will to bring about change has been called into question. Maintaining the coalition has been regarded as a continuous process of horse-trading, which has not created a favourable environment for consistent policy making and budgetary allocation.24

Moreover, the constitutional review process on which the government expected to make rapid progress when it came to power, promising results in 100 days, has not

23 For a comparative perspective see O’Brien and Ryan (2001) and the other studies in the same volume

Devaragan, Dollar and Holmgren (2001).

24See for example the discussion of policy-making processes under the NARC government in OPM (2004).


66

led to an effective political consensus on how to move forward from the current system with its concentration of power in the Presidency and lack of effective checks on the executive.

2.4 Most recently, the issue of corruption and the NARC government’s ineffective track record in addressing it has assumed paramount political importance both domestically and in the government’s relations with donors and the international community. Following the high profile resignation in February 2005 of John Githongo as Permanent Secretary for Governance and Ethics in the Office of the President in response to the government’s failure to act on evidence of corruption involving senior ministers and officials in the current regime, the capacity of the government to demonstrate effective action has become a crucial test of the credibility of its commitment to fundamental reforms.

The institutional context – public sector administrative capacity

2.5 As power has been focused in the presidency, other public institutions, including the judiciary and the civil service, have been seriously and systematically undermined over a period of many years. The capacity of the civil service to play an effective part in designing and implementing policy, including in the field of economic management, is very weak. This loss of capacity has been brought about by a number of factors:25

• Political interference, which has often resulted in public administration being diverted to serve private aims;

• Excessive regulations and controls;

• Corruption, which is rife at all levels;

• Frequent reorganisations under the previous KANU regime, leading to confusion, lack of continuity and inefficiency;

• An authoritarian atmosphere under the previous regime, leading to an unwillingness to take decisions even at senior levels;

• The use of public employment as a principal instrument of patronage, with recruitment and promotion on grounds other than merit eroding competence, and resulting in overstaffing and high recurrent budgets;

• The difficulty of sacking civil servants protected by patronage, and of introducing a working culture based on merit and performance;

• The freedom enjoyed by senior civil servants to engage in personal business in addition to their normal responsibilities, resulting in neglect of official duties;

• Poor working conditions, contributing to high staff turnover and the inability to retain high quality and well-motivated staff.

2.6 Weak civil service capacity contributed to the erratic implementation and contradictory policy signals that rendered the government’s wide-ranging economic

25 This list draws on a number of sources including Ng’ethe et al. (2004), Office of the President (2005),

Government of Kenya (1998), and Government of Kenya (2005b).


67

reforms in the later 1980s and 1990s largely ineffective in bringing about higher growth. It also continues to contribute to poor public expenditure management, resulting in considerable discrepancies between budget allocations and actual expenditure, low completion rates for investment projects, and substantial abuse of the public procurement process.26 Weak capacity in itself has hindered attempts at reform throughout the period, as it is very difficult to introduce significant management improvements in such an environment. Despite successive reform initiatives, the GOK recognises that the public service is still faced with the challenge of poor and declining performance.27 Capacity for policy analysis in ministries is weak, and the service is unable to retain high quality and well-motivated staff. As a result, the policy making process tends to be dominated by executive directives in response to short-term political pressures leading to a wide range of confused and contradictory policy statements and a lack of institutionalisation within the bureaucracy of the policy-making process.28 Kenya has also not articulated a policy to guide its practices on the use of technical cooperation or on aid more generally, nor does there appear to be consistency in practice about how technical cooperation is obtained or managed in relation to capacity development objectives.

2.7 The environment for public administration in general, and economic management in particular, has been characterised during the period of the review and for many years before that by the chronic weakness of the civil service as a source of professional advice and expertise in the formulation and implementation of policy. Lack of capacity has been compounded by endemic corruption at all levels of the State, leading to the further erosion of competence in the public service, and waste and inefficiency in the allocation of public resources. The former government instituted a number of reforms but many have been very slow to come to fruition and little fundamental impact was achieved. The current government came to power with an explicitly reforming and anti-corruption mandate, and has introduced a number of new initiatives (including those described below in section 3) and revived others that appeared to have stalled. Whether these will in practice prove any more effective than those of its predecessor in achieving change remains uncertain.

DFID’s strategy for TC in Kenya

2.8 DFID, along with the other major donors, has been through years of trying to provide development assistance to a government whose commitment to reform has been weak and rarely sustained. This has meant that in contrast to other countries in the region, Kenya’s dependence on external financial assistance has been limited. Only 1-2 per cent of GDP has flowed to Kenya from external assistance, compared with the double-digit figures in countries such as Uganda, Tanzania and Mozambique all of which unlike Kenya have achieved conditions under which general budget support

26According to the Centre for Governance and Democracy (CGD) analysis in 1998 of the 1995-96 Controller

and Auditor General’s report, non-adherence to budgetary controls cost the economy Ksh. 107 billion during the 1995/96 financial year (around 17% of GDP). Almost half of these losses arose from the procurement system.27

See in particular, Office of the President (2005)

28 See for example the assessment of policy making in the agricultural policy area in OPM, 2004


68

can be provided. DFID also in general avoided provision of assistance through government channels because of concerns about fiduciary risk, and this led to a relatively small programme, the larger part of which was channelled directly to or through non-governmental channels. The projects relating to reforming public financial and economic reforms, several of which are studied in this review, form an exception to this wider policy of avoiding direct support for government.

2.9 DFID has like other donors faced a strategic dilemma in its cooperation with Kenya throughout the period under review. On the one hand the high and rising levels of poverty and the underlying potential of the economy point to the desirability of a substantial DFID cooperation programme. On the other hand, the poor record of commitment to poverty reduction policies and continual corruption and fiduciary concerns limit the scope for large-scale financial support to the Government. Successive DFID strategy documents, including the Country Assistance Strategy (DFID, 1998), the DFID General Briefing on Cooperation with Kenya (DFID, 2002) and the DFID Country Assistance Plan (DFID, 2004) set out the evolving approach to coping with this dilemma.

2.10 DFID’s 1998 Kenya Country Strategy Paper (CSP) recognised the challenge of establishing an effective partnership with the Government. The CSP drew attention to the centrality of governance issues highlighting widespread corruption and lack of integrity in public life; mismanagement of public resources; declining levels and quality of public services; and erosion of the rule of law; and the need for wide-ranging reforms to create a pro-poor policy framework. It established six Thematic Areas, the majority of which were oriented towards project-based support for improvement of service delivery to the poor and the promotion of civil society and political pluralism, typically operating directly with non-governmental operators. However, the CSP strategy also included as its first thematic area support for Improved Governance and Economic Policy Reform. The activities provided for in this area included support for policy reform, continued support for civil service reform, anti-corruption, development of the Public Expenditure Review and actions aimed at reducing regulatory constraints to private sector development. The emphasis in this area of the CSP lay primarily in inducing accelerated reforms, with capacity-building support envisaged as providing reinforcing support if there was progress in policy reform:

“Given clear government policies which command wide donor support, and in the context of a “high case” development assistance programme, we will offer resources to restructure public spending and improve performance, including technical assistance for capacity-building, as well as financial support to help restructure the public sector workforce”. (DFID, 1998, p. 10)

2.11 It was envisaged in contrast that technical assistance would specifically play an influencing role in the event that it was not possible to establish a “stronger development partnership” with the Kenyan government (DFID, 1998, p.9):

“Under these circumstances, Britain would continue to work to build commitment within Government to better governance and pro-poor policies, including through provision of targeted technical assistance.”


69

2.12 In April 2002, DFID issued a Briefing (DFID, 2002) to present its revised strategy in the light of the completion of the PRSP, which was seen as providing this framework of improved policies and hence as justifying increased attention to capacity development within government. This indicated a reduction in the number of thematic areas from six to four, with building support for and advocating pro-poor policy change and “developing sustainable systems, particularly in economic governance” being the two lead themes. The main feature of the revised strategy was to give emphasis to support for the development of monitoring and evaluation of the PRSP, and development of the support for local government through the launching of the project “Poverty Reduction through Optimising Local Governance Systems” (PROLOGS), which built on the former area-based projects in Nairobi and Mombasa, supporting the wider Kenya Local Government Reform Programme (KLGRP) and the introduction of the Local Authority Transfer Fund (LATF). This was highlighted as a main initiative under the theme of pro-poor policy change. In addition the 2002 Briefing committed DFID to supporting the development of the Integrated Financial Management Information System (IFMIS) in central government.

2.13 DFID’s Country Assistance Plan (CAP) was issued in April 2004, (DFID, 2004) in the wake of the change of government and the publication of the government’s Economic Reform Strategy for Wealth and Employment Creation (ERS). The CAP noted the continuing deterioration with respect to the prospects for achieving most Millennium Development Goals, and the deep-seated problem of economic and political patronage. The CAP placed emphasis on strengthening the ERS process in terms of its poverty orientation, the allocation of budgetary resources and service delivery. In relation to the use of aid instruments, the CAP envisaged support to improving public financial management as part of a process of moving towards the provision of budget support. In terms of the current evaluation the CAP comes at the end of the period under review and, accordingly, had only a limited impact on the portfolio of projects reviewed.

2.14 In practice, the activities reviewed in this evaluation can all be interpreted as directed towards building capacity to make key government functions in the area of economic management more effective, and initiatives have been taken and resources committed (such as those specified in the 2002 Briefing) in circumstances where the changes in the political and policy environment have been judged to be propitious. The extent to which these activities have progressed, delayed or ended has depended on judgements about progress within the organisations and in the wider environment. Although the 1998 Strategy did specify a role for TC as an instrument of influence (i.e. specifically to bring about policy change), it does not appear to have been used in this way in the activities reviewed here and subsequent strategy statements have moved away from a view that donors can bring about, rather than support, policy change.29

29The 2004 CAP stated (p.11) that: “DFID’s view is that international partners cannot and should not try to

drive change themselves. Instead they should support domestic drivers of change, provide technical advice and ideas to support reform, and back reforms that are genuinely owned in-country.”


70

3. Economic management in Kenya


3.1 Kenya was a strong performer in the East Africa region in the first two decades of independence, with a GDP growth rate of 7% p.a. during the 1960s and 1970s. Following the economic shocks in the early 1980s, the growth rate declined significantly to around 3% p.a. over the 1980s and 1990s. In the later 1980s and 1990s the government undertook wide-ranging economic reforms that substantially liberalised the trade, exchange, investment and financial sector regimes and privatised state-owned enterprises. These reforms were largely ineffective in bringing about higher growth due to erratic implementation, contradictory policy signals, growing corruption and macroeconomic instability.

3.2 Due to a poor record of implementing key reforms and weakening governance the relationship between the GOK and the donor community has been unstable and unpredictable. In 1997, the IMF suspended Kenya’s Enhanced Structural Adjustment Program due to the government’s failure to maintain reforms and curb corruption. A severe drought from 1999 to 2000 compounded Kenya’s problems, causing water and energy rationing and reducing agricultural output. The IMF, which resumed loans in 2000 to help Kenya through the drought, again halted lending in 2001 when the government failed to institute several anti-corruption measures.

3.3 Most recently the IMF Board approved the Poverty Reduction and Growth Framework Programme in November 2003. Donors pledged up to US$4 billion over 2004 - 2006 at the Consultative Group (CG) meeting held in November 2003, subject to rapid and substantial progress in public sector reform and improvements in the design and delivery of basic services for the poor. During January 2004, the Paris Club of Creditors agreed to reschedule US$350 million of arrears and maturities for interest and principal falling due between January 2004 and 31st December 2006.

Table 2. Key macroeconomic indicators

1999 2000 2001 2002 2003 2004

Real GDP Growth Rate (%) 1.4 -0.2 1.2 1.2 1.8 4.3

Consumer Price Index (%) 5.8 10 5.8 2 9.8 9.2

Exchange Rate (K Sh/US $) End of year 61.9 72.9 78 78.6 77.1 76.1

Total Domestic Debt/GDP (%) 28.2 26.7 30.1 27.6 27.8 26.6

Total External Debt/GDP (%) 42.6 53.8 52.9 47.7 39.4 36.4

Sources: WB (2004); GOK ( 2005a); GOK (2005b)

3.4 Growth in GDP was very low in 1999 and the years following. In 2000 the economy contracted for the first time since independence. Economic growth was only 1.2% in 2001 and 2002, but rose marginally in 2003 to 1.8%; although the new government generated considerable goodwill from the international community, the resumption of international financial assistance did little to boost growth. However, growth


71

increased further to 4.3% in 2004, and the figures for both 2003 and 2004 have met the targets set in the GOK’s Economic Recovery Strategy. The Government believes that the recovery will be consolidated in 2005. The significant reduction in inflation to only 2% in 2002 proved short lived, and the reduction in the high level of domestic debt in 2004 was nearly three percentage points less then than targeted.

Performance of Economic Management: 1999-2004

Monetary policy

3.5 A tight monetary policy over 1998-2001 and the first half of 2002 succeeded in lowering the inflation rate, the interest rates on treasury bills, and the perceived country risk premium. However the headline rate of inflation jumped to 9.8% in 2003, well outside the lower bound target of 5%, remained high in 2004 due to the impact of drought and high oil prices, and is forecast to remain at around 9% in 2005. Money supply growth (M3X) was 11.7% in 2003 compared with a target of 7%, reflecting a significant increase in credit to the private sector.

Fiscal sustainability

3.6 In recent years the fiscal framework has been unstable as a result of poor control over recurrent expenditure and donor funds not meeting expectations. Revenue performance declined, with the ratio of revenue to GDP falling from 23.3% in 1999/2000 to 20.8% in 2002/03, largely due to a reduction in import duty collection. The trend was reversed in 2003/04 with a ratio of 21.2% and an increase in revenues of 10% over the previous year. The 2005 budget relies heavily on the intensification of revenue collection efforts together with the cutting back of non-priority expenditure; the target revenue ratio is 22%, slightly less than the expected 2004/05 outturn.

3.7 The expansionary fiscal policy of recent years worsened the fiscal deficit until 2003/04, when it was reduced to – 0.39% of GDP as a result of a combination of underperformance in development expenditure and improvements in revenue collection. High domestic debt remains a major challenge for fiscal stability, a result of the government’s need to finance its deficit entirely from domestic sources. However, higher than forecast revenue and under-spending on development also led to lower than planned net domestic borrowing in 2003/04. Despite restored relations with donors, the GOK’s fiscal stance in 2005 was conservative, relying on revenue collection and expenditure control measures to close the gap rather than either an increase in domestic financing or assumptions of bilateral donor budgetary support.

External position

3.8 Kenya maintains a liberalised exchange rate policy and external trade system. The country’s trade policy has increasingly become orientated towards the pursuit of regional trade integration, primarily through its membership of COMESA and the EAC. Kenya’s external payments position improved in 2002, in a significant contrast from earlier years; exports that had previously performed badly grew substantially,


72

while the level of imports fell significantly, due to weaker demand in a slow domestic economy. In 2003, however, the increase in export volume was seriously below target (9.9% compared with 15.8%), owing to a significant appreciation in the real exchange rate and underperformance in the coffee sub-sector, and the current account deficit is projected to widen from 2% of GDP in 2003/04 to 5.6% in 2006/07 as import growth exceeds that of exports.

Budget and public financial management

Budgetary management

3.9 The level of recurrent expenditure has been on an upward trend despite the government policy objective of containing overall expenditure. The ratio of expenditure to GDP rose from 23.5% in 1999/2000 to 25.5% in 2003/4. Public resources have been primarily used for wages and debt service; as a result operations and maintenance have been under-funded and development spending has been low. In 1999/00, 52% of the recurrent budget was spent on interest payments and civil service salaries, but this was reduced to 45% in 2003/04. Since 2002/03 Kenya’s own development spending has been increasing and now stands at 1% of GDP.

Table 3. Recurrent Expenditure (Shares in Total Expenditure)

99/00 00/01 01/02 02/03 03/04 04/05

Interest Payments 16.2 13.3 13.3 13.5

Civil Service Wages 36.4 29.3 34.1 32 33

Pensions etc 2.8 2.6 3.9 3.6

Operations and Maintenance 15.6 14.4 17.1 16.6

Transfers 7.6 6.4 7.4 6.9

Defense and NSIS (National Security Intelligence Service) 6.8 7.2 9.6 8.6

Recurrent Expenditure 86.4 85.4 89.1 83.6

Sources: World Bank (2004); Government of Kenya (2005b)

3.10 According to the 2004 Public Expenditure Review, there has been a growing divergence between budgeted and actual expenditure. Budgeted and actual expenditures vary considerably across votes, and more importantly this is not factored into the later year budgeted allocations. This is true more so for development expenditure than recurrent expenditure. Over 2000/01–2002/03 less than 40 per cent of the ministerial development budget was actually spent.


73

Reform initiatives during the evaluation period

3.11 Following the 1997 Public Expenditure Review, the Government introduced a number of reforms to strengthen expenditure management:

• The adoption of a Medium Term Expenditure Framework (MTEF) approach to budget formulation in 2000/01.

• Establishment of the Budget Monitoring Department in 2000.

• Development of the Integrated Financial Management Information System (IFMIS).

• Ministerial rationalisation.

• Initiation of a comprehensive public expenditure management reform programme which includes legislative, institutional and system changes.

3.12 However, progress in implementing an efficient and effective public management expenditure system has been mixed, with delays largely due to organisational and capacity constraints (World Bank, 2004). A review of public expenditure management practices with the World Bank and other development partners in 2003 found that only three out of fifteen standard indicators were being met. Following this the GOK prepared an Enhanced Action Plan for Financial Management, to be evaluated against sixteen benchmarks. However, a World Bank progress report in April 2004 found that only one of the seven benchmarks relating to budget formulation was being met: the MTEF was still not entrenched in the budget preparation system, and there continued to be large variances between budget and outturn. Similarly, only one out of four budget execution benchmarks and two out of four reporting benchmarks were being met.

3.13 A range of public service reform measures have been implemented recently30 in an attempt to re-establish control over the wage bill, address capacity constraints and increase efficiency. For example, new regulations have been established that emphasise merit as the key determinant for promotion, and new wage-setting guidelines have been introduced. Several key ministries are being restructured, and steps have been taken to streamline and strengthen capacity in the Ministry of Finance, including consolidation of the planning and budget functions within the Ministry, the setting up of a robust debt management unit, and enhancing capacity and mechanisms for managing and monitoring macroeconomic and structural reforms.

3.14 Other measures being taken to improve public expenditure management include:

• Implementation of the Integrated Financial Management Information System (IFMIS), which is now being rolled out to line ministries.

• Strengthening of the expenditure commitment control system.

• Improving budget transparency by eliminating unclassified budget votes.

• Instituting expenditure-tracking surveys.

30 Government of Kenya (2005b), Office of the President (2005).


74

• Transforming the internal audit approach from pre-audit to risk-based systems and compliance reviews.

• Developing a cash management system.

• Strengthening decentralisation, as a means of increasing budget transparency and accountability, by improved monitoring to ensure that the Local Authority Transfer Fund and Constituency Development Fund are used efficiently for the intended development goals.

• Requiring up-to-date bank reconciliations from line ministries.

• Deepening procurement reforms, including through the establishment of a Public Procurement Oversight Authority.

3.15 Moreover, a new system of performance contracts is being implemented across the public service, in commercial and non-commercial state corporations as well as in the core civil service. Performance contracts, which are intended to improve accountability and focus resources on the attainment of key national policy priorities, are to have been signed in all corporations and civil service ministries and departments by the end of June 2005. It does not appear that any donor-funded TC has been involved in the development and rollout of this system.

Has economic management improved?

3.16 Overall during the period under review Kenya has been characterised by relatively sound macroeconomic management, which has halted and partially reversed the earlier trend of declining real economic growth, but also by a continuing poor record in terms of budgetary management with the result that budgetary allocations and expenditure have borne little relation to stated policies, a large gap between budget and expenditure out-turns has persisted, and there has been little or no progress in the improvement of service delivery with corruption remaining a major problem affecting public procurement and the budget process.

3.17 Some progress has however been made in laying the groundwork for improved budgetary performance. At the planning stage this has occurred through the development of the PRSP in 2001 and subsequently the Economic Recovery Strategy (ERS) and the development of the MTEF process since 1997. In addition, while the country continues to score poorly in terms of its public financial management benchmarking, steps are being taken that if successful may address this, including the development of an Integrated Public Financial Management Action Plan and the development of the IFMIS to provide technical support for strengthened public expenditure management. The key question (which it is not possible to answer definitively in this evaluation) is whether these measures, which provide a potential foundation for a strengthened system of economic management, will find sufficient political backing to bring about fundamental change, or whether the pattern of the last two decades in Kenya will continue and reform initiatives which may have made progress at the technical level will not be sustained.

3.18 The context for capacity development has therefore remained very difficult in general throughout the period covered by this evaluation, with little progress being made in addressing the key constraints on public sector performance, despite some promising reform initiatives.


75


4.1 This section first examines the changes in organisational capacity in the organisations that have formed the focus of the evaluation, all of which play an important or potentially important role in economic management in Kenya. It then assesses the contribution of DFID funded TC to capacity development. As has been noted in the previous sections, DFID TC in support of economic management in Kenya during the period under review has been rather diffuse, with a significant number of initiatives of varying size spread across many organisations. In order to provide some cohesion to the analysis, the projects are grouped in this section on the basis of the organisation, or cluster of organisations, which they aimed to support. This has led to focusing on four organisations or clusters:

• Organisations responsible for the management of the budgetary cycle, namely the Ministry of Finance (MF) and the Ministry of Planning and National Development (MPND), which have been the recipients of support from three of the selected projects, namely, IFMIS, the Public Expenditure Management (PEM) facility and support for the Monitoring and Evaluation Department (MED). The common strand to these in many ways disparate projects is that they all support the implementation of the planning and budget cycle.

• The Office of the President (OP) and the Kenya Institute of Administration (KIA), which relate to the support for Civil Service Reform.

• The Ministry of Local Government (MOLG).

• The Kenya Anti-Corruption Agency, KACA.

The Ministry of Finance (MF) and the Ministry of Planning and National Development (MPND)

Roles in Kenya’s economic management process

4.2 The MF and the MPND are the lead institutions in the management of national planning and the annual budgetary cycle, including budget preparation, budget execution, monitoring and evaluation and reporting. Over the years a system of planning and budgeting has become established which has in practice changed little in spite of frequent modifications in the respective roles of the two ministries. Progressive overlays of systems have witnessed the introduction of an MTEF process, the development of a PRSP (the IP-ERS), the introduction of an annual Public Expenditure Review process, and the establishment of a system for imposition of cash spending limits.

4.3 In spite of the number of reform initiatives there are major deficiencies in the system of planning and budget management, which have been highlighted by successive Public Expenditure Reviews (WB, 2003, 2004) and by the periodic assessments of the system of public expenditure management against the IMF benchmarks (GOK, 2003b, GOK 2004b). Key deficiencies are:


76

• The large gap between the approved budget and actual budget execution, which undermines the planning function.

• The pervasive problem of public sector arrears (“pending bills”).

• The lack of monitoring and evaluation and more generally of feedback from budget out-turn to the planning and budget preparation processes and most spectacularly.

• The deficiencies of the public procurement system to provide protection against corruption.

Support from DFID projects

4.4 DFID has provided support aimed at addressing the above deficiencies through three main areas of action:

1. The IFMIS project aimed at development of an integrated system, including but not limited to the development of integrated IT solutions for budget management, initiated in 2002 after a lengthy gestation and funded through a basket mechanism in cooperation with the World Bank and the Swedish International Development Agency (SIDA). Under this support programme the World Bank provided finance for hardware, SIDA financed software acquisition and DFID provided Technical Cooperation (with a budget of $0.8 million out of a total common fund budget of $4.3 million). The delivery of DFID-funded TC as a component of support under the project has largely centred on a single individual.

2. A facility established in support of Public Expenditure Management (PEM) reform of £0.5 million. This project essentially comprised a budget which could be used by DFIDK to hire technical assistance to support a range of processes, including the preparation of ministerial PERs (in 2003/04), the main PER process, a review of fiscal sustainability for the PER, expenditure-related aspects of the Civil Service Reform process and provision under the DPM of external peer review for the development of ministerial strategic plans using GOK-hired local consultants.

3. Support for the development of the national Monitoring and Evaluation system, provided to the Monitoring and Evaluation Department in the Ministry of Planning and National Development which was created in 2004 to provide a focal point for the development of M&E of the IP-ERS (Kenya’s PRSP). This was a small TC project with a total value of £0.1million. DFID recruited an external consultant, initially on a six-month contract (later extended by 2 working months over a total project period of 10 months). After some limited delay two local consultants were recruited through GOK processes with UNDP funding.

Changes in organisational capacity

4.5 As described in 3.3 above, there is evidence that significant changes have occurred during the period under review in terms of the capacity of the MOF and the MPND to provide effective management of the planning and budgetary cycle, even though the full benefits of this increased capacity have not yet been realized. The change in capacity has primarily taken the form of the creation of new processes and procedures. These have the potential to provide greatly enhanced transparency and accountability in public expenditure management. However, the human resource base for the management of these new processes (and even more so the depth of


77

political commitment to them) remains weak and this leads to questions as to whether or when the full benefits will be achieved and how sustainable the new processes will prove to be.

4.6 The key positive developments are threefold. First, and potentially of greatest significance in strengthening expenditure control, the development of the IFMIS31,which is poised for roll-out to the major spending ministries, is providing a platform for progressive introduction of controls on the commitment and execution of expenditures, for enhanced and more timely reporting on budget execution, and for the introduction of controls on the procurement function. It must be emphasised that to date none of these benefits has been fully realised, but depend on the progressive imposition of requirements for transactions (including commitments and procurement stages) to be entered into the IFMIS IT system and subjected to control processes. In the near future this should start to bite in limiting the possibility of creation of new pending bills and more generally unauthorised expenditure. Seen in this light the IFMIS is the central technical mechanism for Kenya to achieve a marked improvement in the coming years on its PEM (now PEFA) benchmark scoring. The initial progress in setting up the IFMIS, despite inauspicious circumstances during the Moi era, has paved the way for this development. This is an important technical step. However realising its benefits will depend on sustained political commitment to implementation as well as on actions to address constraints on staff capacity to use and manage the system.

4.7 The PEM facility has facilitated the PER, especially in 2004, and as such has assisted in the embedding of the PER as a routine part of the budgetary cycle. Finally, the M&E project has provided brief but effective transactional support for the new M&E department in operationalising the concepts set out in the national M&E Strategy adopted by government. This progress in the creation of an M&E capacity has been reflected in the preparation of Annual Progress Reports on the IP-ERS for 2003/4 and 2004/5, the first of which was used as the basic document for the 2005 Consultative Group Meeting.

Contribution of DFID funded TC

4.8 DFID has been the sole external financer of TC for the IFMIS project, and as such must take its share of the credit for the progress achieved. This has been delivered on the basis of a very lean package of support compared to other packages of support to the development of IFMIS projects across Africa. As an example, an equivalent process in Mozambique has to date spent $23mn, (five times the expenditure in Kenya) to achieve a comparable level of functionality with the Kenya IFMIS.

31 There have been some delays in implementation which have been attributed to opposition within some

parts of GOK to the system. Initially scheduled completion and roll-out of the system to the line ministries, originally scheduled to be completed by 2003, was rescheduled for 2004/05, but had still not been achieved as of June 2005.


78

Office of the President (Directorate of Personnel Management) (OP/DPM) and Kenya Institute of Administration (KIA)

Role in economic management

4.9 Reform of the civil service, and the public service more widely, has a potentially central role to play in improving the management of the economy in Kenya, for several reasons. In the first place, there are the fiscal and macroeconomic consequences of an over-inflated government wage bill (still estimated at 8.3% of GDP in 2004/05, compared with an average of 6% for Sub-Saharan Africa). Second, there are the burdens that a poorly motivated, poorly managed, corrupt and inefficient public services place on the population at large and on private sector development. Third, there is the need for the civil service to have the capacity to play an effective part in the formulation of economic policy, and in planning, budgeting and execution of the budget.

4.10 Whilst KIA does not in itself play a direct role in any aspect of economic management, developing capacity in this organisation was seen in the project design as part of a strategy to develop capacity across the civil service as a whole and as having the potential to contribute significantly to improvements in the quality of policy design and implementation in this as well as in other areas of public administration.

Support from DFID

4.11 DFID has a long history of support to civil service reform (CSR) in Kenya, including substantial assistance to the Civil Service Reform Programme Phase I (CSRP I), which ran from 1993 to 1998, which included support for a major Voluntary Early Retirement Scheme, assistance in the design of rationalisation, support for establishment control and personnel administration and support for development of an Integrated Personnel and Payroll Database. CSRP I ran between 1993 and 1998, and received some US$ 17m of support from DFID. It claimed the following achievements:

• A reduction of 89,349 posts through the Voluntary Early Retirement Scheme, targeted on overstaffed junior grades.

• A programme of 12 Ministry rationalisation exercises carried out (though these were regarded as very centrally driven by the Office of the President (OP), with inadequate Ministry ownership).

• Establishment control and personnel administration improved, and an Integrated Personnel and Payroll Database developed.

4.12 For the second phase, CSRP II, which was intended to run from 1998 to 2001, a new vision was articulated. GOK wished to refocus the reforms and increase the pace of implementation in order to achieve better control of the wage bill, further improve the balance between development spending and operations and maintenance spending, and promote improvements in service delivery.


79

4.13 DFID supported CSRP II through three mechanisms: initial design work for CSRP Phase II, which was undertaken in March 1998, the establishment of a budget to support implementation of Phase II, and through a support programme for the Kenya Institute of Administration (KIA), which was seen as an essential adjunct to the main thrust of the CSRP II. However, only the first of these mechanisms was fully implemented, and resulted in the very rapid preparation of a master plan for CSRP II by consultants hired by DFID.

4.14 In 1998, following the passing of the KIA Act, which established the KIA as a semi-autonomous agency, DFID embarked on a programme which was intended to be implemented in two phases, with an overall budget of £1.56 million. However, in 1999 DFID decided not to proceed with Phase II of the KIA programme, in spite of the fact that Phase I had delivered the main capacity-building outputs set out in the project document. The decision to discontinue the KIA programme was variously reported by DFID staff and documentation as resulting from non-compliance by the GOK with requirements under Phase I (required construction work had not been completed), a shortfall in the DFID budget as the overall country budget was reduced, and the perception that there was inadequate commitment to the CSRP by GOK.


4.15 During the period under review there has been little change in the organisational capacity of the Directorate of Personnel Management (DPM) within the Office of the President. However a major initiative of the GOK was the creation of a new Office for Public Sector Reform Development, in the Office of the President, headed by an officer with Permanent Secretary status. This office has been energetic and instrumental in the rapid development and rolling out of the prototype results-based management system of performance assessment for senior executives in the public service as previously described in section 3.

4.16 In the KIA there has also been some progress in capacity development, in terms of physical facilities, networks with international cooperating agencies in its technical field, and the creation of management systems designed to support the KIA to become a self-sufficient, high-quality provider of training in support of civil service reform. However, this process is by no means complete, and it is recognised that there is still a major need for upgrading of the KIA in terms of its curriculum, staff capability and capacity to deliver the scale of civil service training required – support is being sought from Canada and Sweden for this development.

4.17 The KIA is currently seen by the Office of the President as an essential institution to provide training in support of the move to establish results-based management systems in government. In the view of some government and donor representatives interviewed (but not the DFID office in Nairobi) DFID’s withdrawal from support to KIA reduced the scope for DFID to continue to play a useful role in support to CSR. It must also be recognised that this judgement is made with hindsight (after the change of regime) and conditions could equally have continued to be unfavourable if KANU had remained in power.


80


4.18 As noted above, DFID provided the initial consultancy input to the formulation of the CSRP Phase II strategy. However, this strategy never was consistently implemented, as was witnessed by the frequent changes of direction and priority in civil service reform in the following years. A similar conclusion holds for subsequent work undertaken under the CSRP II support programme in development of the Master Plan for Civil Service Reform prepared in 2002.

4.19 The new reform initiatives discussed in 3.3 above are seen within DPM as a completely new start. Whilst there is an avowed willingness to acknowledge past achievements, the present agenda is seen as owing nothing to CSRP II or other earlier reform activities. This makes it difficult to trace the extent to which the outputs of previous DFID-funded TC have had lasting impact on the development of the public service, and the costs and benefits it brings to the economy. Nevertheless, the GOK’s 2005 Medium-Term Budget Strategy asserts the need to downsize by a further 21,000 posts through Voluntary Early Retirement in the medium term, and promises that efforts to improve civil service efficiency will include pay reform and the rationalisation of Ministry structures as well as the introduction of performance based management. This indicates considerable continuity with previous agendas, but suggests that earlier projects have not solved the perennial problems of pay and of duplicated or non-productive functions in Ministries.

4.20 In the overall DFID support for CSR during the period under review the most significant results were achieved through the KIA project, even though this was discontinued on the completion of Phase I. Specific outputs produced in Phase I included the establishment of a board and management structures, preparation of strategic and financial plans, and progress in establishing a new organisational structure and strengthening management systems as well as improvements in facilities.

Ministry of Local Government (MOLG)

Role in Economic Management

4.21 Up until the mid-1990s, local government capacity had been completely eroded to the point of near collapse, although potentially of great significance for the quality of service delivery. Systematically since the 1970s power had been taken away from local government and transferred to central government; revenue collection was poor and financial management almost non-existent. As a result service delivery was very poor in Local Authorities (LAs) without exception.

4.22 The Ministry of Local Government is the central government ministry with responsibility for oversight of the Local Authority components of the decentralised structures in Kenya (the District Administrations are managed by the Office of the President).

4.23 Prior to the elections of 2002, the government emphasised the importance of local government in the Poverty Reduction Strategy Paper (PRSP); the Kenya Economic


81

Recovery Strategy for Wealth and Employment Creation, 2003-07 (ERS) recognised the importance of further decentralisation, identifying local authorities as key partners in the governance and development process. As noted above, DFID’s 2002 Strategy Briefing highlighted the central importance of strengthening local government in improving service delivery and the effective use of public expenditure.

Summary of DFID support

4.24 DFID support to local government started before its current project PROLOGS. Prior to that, DFID was funding an urban poverty alleviation project, Kenya Urban Poverty Project (KUPP). KUPP included the Partnership Approach to Meeting the Needs of the Urban Poor (PAMNUP) and the Nairobi Urban Poverty Project (NUPP), which focused on initiatives and small grants to organizations to assist in alleviating poverty in Mombasa and Nairobi respectively. This support was then amalgamated with proposed technical assistance to the Kenya Local Government Reform Program32

(KLGRP) to form a new project in 2002. The key components of the new project (PROLOGS) were as follows:

• National Component: Policy and Programme Level Technical Assistance to the Kenya Local Government Reform Programme (KLGRP).

• Nairobi Component: Reduced version of Support to the NUPP including £350,000 from a special Poverty Reduction Co-financing Fund (PRCF).

• Mombasa Component: a reduced version of support to the PAMNUP project also with a £350,000 share of the PRCF.

• Rural Component: set up as a pilot research exercise focused on Kwale and Nandi county councils each receiving £145,000 share of the PRCF.

4.25 The project budget was set at £5.8 million. All the components under PROLOGS were to be managed by a private sector “Programme Management Agent” (PMA) in order to relieve DFID of the administration of the various support activities. Following a tender process, a consulting firm was contracted in December 2002 to act as the PMA for the MOLG in managing the delivery of the agreed scope of services and accounting to DFID for the grant funds. A separate contractor who was providing Technical Assistance under the previous World Bank project, was selected to continue to provide TA to KLGRP under the DFID programme.

Changes in Organisational Capacity

4.26 MOLG, through the KLGRP and the PROLOGS project, has made significant advances in the development of systems for the oversight of local authorities. Starting from a very low base of capacity in 2001, particular progress has been

32 After several studies on the sector in the early 1990s, the World Bank provided funding for local

government reform in 1996. A secretariat, Kenya Local Government Reform Program (KLGRP), was set up to coordinate support in key areas – financial management, enabling environment for business (licensing, single business permit), information systems and stakeholder participation among others. Due to problems with a related project, World Bank funding for KLGRP stopped and DFID took over funding of TC to KLGRP in 2002.


82

recorded in the development of the processes for financial oversight of Local Authorities (LAs). In particular attention is drawn to the systems and controls developed for the release of fiscal transfers to the Local Authorities under the Local Authority Transfer Fund (LATF) and the associated promotion of Local Authority Service Delivery Plans (LASDAPs). Progress has been made by making release of LATF transfers dependent on specific management actions required of Local Authorities, notably, the presentation of a budget, the preparation of a revenue collection plan, and the completion of an acceptable LASDAP as preconditions for release of LATF funds.

4.27 The DFID OPR identified that the project has supported MOLG and LA efforts to make LATF/LASDAP work efficiently and to boost the capacity of LAs to deliver services for all the people. Little has been accomplished so far in developing pro-poor instruments including PRCF and extra poverty targeting criteria for LASDAP operation. Subsequent meetings and review of documentation during the evaluation have confirmed these findings. The pilot M&E survey undertaken by the Monitoring and Evaluation Department of MPND in April-May 2005 in 8 districts, also found that LATF projects were generally better managed than projects falling under other mechanisms at district level, such as the Constituency Development Fund.

4.28 Capacity to implement these new oversight and management systems has to a large degree depended on the staffing and capacities within the KLGRP and PROLOGS, and implementation of these functions has remained critically dependent on project staffing and financing. The main body of the MOLG has been to a degree by-passed by these developments, with the projects constituting externally funded islands of excellence in an otherwise weak ministry.


4.29 DFID TC provided under the KLGRP and PROLOGS projects has been a decisive factor in the observed increase in organisational capacities noted above partly because of particularly good counterparts who were supported by a committed Permanent Secretary. The approach/structure for KLGRP has significant ownership from government; government is contributing all funds for LATF as well as staff costs for KLGRP - DFID’s quick response to continuing support to TA at KLGRP is seen as very positive and allowed for continuity of activities. This is in direct contrast to current efforts by the Ministry to negotiate a new project with the European Union, which has been subject to long delays.

4.30 However, overall implementation has been limited by weaknesses in programme design, and the limited effectiveness of some of the TC provided. There were three main problems in the project design. First, the overall approach/structure for the project utilizing the PMA was not well integrated with government. Although ostensibly the Ministry had oversight in practice MOLG had no control over PMA/grant funds as they were to be channelled outside government funding mechanisms. Payments were effected by DFID directly after recommendations from the PMA – thus undermining government normal channels. Concern was also expressed from government that the project structure also undermined MOLG capacity, as the PMA contractor hired staff from the Ministry at a higher salary.


83

Second, the PMA was not able to combine its management and administration role with that of implementation as the project design had envisaged. Restructuring of the project under the PMA caused long delays as a result of conflict over roles between the PMA and the contractor responsible for TC provision. Third, the project design did not contain any performance indicators to measure changes in capacity in the organisation supported, nor was there a clear exit plan for sustainability after the end of support.

Kenya Anti-Corruption Agency (KACA)

4.31 The KANU government also made some ostensible efforts, under pressure from donors and the Bretton Woods institutions, to curb rampant corruption. The Kenya Anti Corruption Agency (KACA) was established in 1997 to spearhead the fight against corruption and other economic crimes. Little progress was made in developing KACA as an organisation to enable it to fulfil its legal mandate until March 1999, when the President appointed a new Director. His task was to establish KACA as a fully functioning organisation that was able to fight corruption effectively. It was in this context that DFID provided bilateral support in 2000-1 to a project aimed at assisting the Authority to develop key systems to become an efficient and effective organisation. However, the body suffered severe setbacks in December 2000 when the courts declared its prosecution powers unconstitutional, and in August 2001 when Parliament rejected a Bill that would have given it a new statutory basis.

4.32 The NARC government came to power in December 2002 with a strong public commitment to the fight against corruption, and, although disillusion has been growing about its ability to make good its pledges, has taken some positive steps. The Public Officer Ethics Act 2003 tackles civil service corruption by requiring civil servants to declare their assets, and the Anti-Corruption and Economic Crimes Act 2003 reconstituted KACA as the Kenya Anti-Corruption Commission (KACC) and established its Advisory Board. The KACC does not have any direct role in the policy-making, forecasting or planning and budgeting elements of economic management, but is intended to help ensure the probity and integrity of the financial management system (especially in relation to procurement).

4.33 DFID was briefly involved in support for the Kenya Anti-Corruption Agency. This support took the form of the initial design work for external assistance to KACA, and a contribution to (and administration of) a multi-donor basket fund to support the Director’s strategy for organisational development. However, once it became evident that KACA would have to be wound up the basket fund was terminated and the unspent balance returned to DFID for distribution pro rata to donors. While it was reasonable for DFID to attempt to support KACA at the time when it was established,33 the termination of support (albeit enforced) was timely and effectively minimised the sunk cost. The DFID contribution led to no sustained capacity development.

33 However, subsequent reviews of the experience with attempts to establish Anti-Corruption agencies in

Africa have raised questions about the appropriateness of this institutional model and have argued that the results of such initiatives have to date been disappointing. See Doig, Watt and Williams (2005).


84

Summary

4.34 DFID’s approach in the activities reviewed has generally involved providing support to strengthen systems in key organisations where there has been judged to be sufficient political commitment to justify investment, while the high risk of failure has been generally recognised in the design of the activities which provided scope for exit. The record has been mixed. There has been potentially significant progress in strengthening systems that are strategically important for economic management, but in none of the cases reviewed have fundamental issues of organisational capacity been successfully addressed. The following section summarises these findings in relation to the overall evaluation criteria.


85


Relevance

5.1 In judging the relevance of the project portfolio reviewed in this study it is important to start from recognition of the context within which most of the projects were developed – a context characterised by poor governance, a limited commitment on the part of government to addressing important areas of reform, the absence of a coherent master plan for reform and development, and a generally stand-back posture of the donor community in terms of aid flows to Kenya, especially flows channelled through the government. In this context the projects reviewed here represent support (involving a higher level of engagement with government processes) of what were seen as promising reform initiatives in circumstances that were acknowledged to be difficult. The result was a set of project initiatives in a diverse set of areas as opportunities were seen to arise: IT for central government accounting, decentralisation, civil service reform and training, private sector development and others. The projects constitute discrete activities with few direct linkages between them. While DFID’s earlier strategy documents envisaged a role for TC as an instrument for influencing policy change, in practice TC has been used throughout the period to build capacity for improving aspects of economic management as opportunities arose rather than directly to try to catalyse policy change. This role has been more explicitly recognised in DFID strategy documents from the end of the evaluation period.

5.2 The specific areas in which TC projects were developed were all relevant to the wider agenda of trying to establish the conditions, within the public sector, for the renewal of growth and improved governance. In general the objectives of the projects were linked to supporting the strengthening of organisational systems rather than full-fledged assaults on specific reforms or attempts comprehensively to address the capacity constraints faced by the organisations supported. The level of resourcing of these projects was generally modest in relation to the potential magnitude of the tasks and the organisational capacity problems faced.

5.3 The overall assessment is therefore that the approach and role used for TC was relevant and appropriate and was focused on areas of reform that did have a potentially major impact on the quality of economic management. The support was also consistent with DFID’s strategy and the approaches of other donors. Consultation processes were generally judged adequate to maintain relevance.

5.4 The limited and provisional success achieved suggests that the commitment of additional TC resources would not have led to greater impact since the environment has remained unpropitious for reform initiatives. A more difficult judgement (about which it is not possible to come to a definitive view) relates to the decisions made about the strategy for continuing engagement in particular reform areas – for instance the decision effectively to withdraw from supporting the civil service reform process. Some Kenyan observers took the view that this decision was regrettable and that sustaining involvement (even if on a limited scale) would have yielded


86

beneficial results when government focus on and commitment to civil service reform strengthened as it appears to have done subsequently. By contrast to the KIA project, DFID remained engaged on the IFMIS project even when after two years the project had made virtually no progress. It seems likely that this continuing engagement, which contrasts with the treatment of the KIA project and with DFID’s withdrawal from support to CSR more generally, was influenced by the fact that for the IFMIS, DFID was a formal and contracted co-financer with the World Bank and SIDA, and that the mechanism for reaching a common position to terminate acted as a brake on abrupt unilateral action by DFID.

Effectiveness

5.5 In general the projects reviewed in this study were effective in terms of the delivery of anticipated project outputs, with the exception of the KACA project, which was terminated when the beneficiary organisation was wound up. Indeed even with the KIA project, which was discontinued in the face of lack of committed government, the contribution was effective in the delivery of outputs. With respect to the IFMIS project the judgement on effectiveness is still pending as not all outputs of the project have been delivered.

5.6 Ownership within the organisations supported (and alignment with government policies) has been variable. For the support to the Ministry of Finance and MPND, there is evidence that ownership has increased over the period. In the case of KACA the project was terminated as the lack of government ownership became clear. In MOLG, while a level of local ownership was achieved, the projects tended to become relatively isolated islands of competence within a generally weak ministry. Harmonisation of approach with other donors was strong in all cases although the problems of integrating the World Bank support to KLGRP into a unified project structure caused difficulties. Gender issues received no prominence in any of the activities reviewed.

Efficiency

5.7 The IFMIS support appears to have been highly cost effective in general though delivery of some of the outputs was delayed in particular by opposition to some aspects of the programme within GOK. In the PEM project local consultants were recruited to work alongside the external consultants for all of the sub-set of PER-related assignments. This project also used external consultancy in a quality-control role vis-à-vis the larger operation of preparation of strategic plans by the ministries, and relied entirely on local consultants for this. These activities appear to have been cost-effective. The decision to terminate support to KACA was done promptly and minimised waste. In the case of support to MOLG weaknesses in project design led to delays and may have inflated costs. Support to local government reform has been by some way the largest area of expenditure on technical cooperation over the evaluation period, accounting for over half the recorded expenditure. The limited support provided to civil service reform was delivered extremely promptly and cost effectively in response to government requests, but this approach may have


87

compromised ownership (and hence the prospects of future success) compared to a design process with deeper GOK involvement.

Impact

5.8 The DFID TC projects reviewed in this study typically only addressed a limited range of capacity issues, being mostly concerned with the development of systems of management (IFMIS, Local Authority supervision, M&E system etc.) rather than with the establishment of the necessary staffing and other capacities necessary for the sustained support of those systems. Within these limits, the projects reviewed do have the potential to have impact, with the exception of the support to KACA. This is though largely dependent on actions being taken and continuing resources (staff and financial) being provided to allow the improved management systems to operate. These issues therefore relate fundamentally to sustainability, as discussed below.

Sustainability

5.9 An important common theme running through the projects reviewed in this study is the difficulty of achieving sustainability, resulting from the uneven progress in related areas of reform. In particular, while several of the projects have been successful in the development of various types of system (IFMIS on accounting, PROLOGS on local authority supervision, M&E in development of the approach and procedures for the national M&E system) they have not sought to create adequately skilled and trained staff for the continuation and sustained implementation of the management systems developed. While imported technical skills can relatively easily address the technical design issues required by these systems and hence deliver the outputs as these relate to management systems, the upgrading of capacity in the ministries has been, and continues to be, constrained by the problems of staffing.

5.10 In general the ministries receiving DFID TC do not have the freedom to recruit appropriately skilled staff to take on the implementation of systems developed with TC support. There are appropriately skilled staff in Kenya but pay, terms and conditions in the public service are a constraint on their recruitment. As an example, the Monitoring and Evaluation Department, which is the recipient of DFID M&E TC support has a staff entirely derived from its former staffing when it was the Poverty Eradication Commission. At the time of formation of the new department for M&E there was not a single staff member with training or background in M&E. Over the 18 months since its formation the department has attempted to give some basic orientation courses (through KIA) but has not been able to recruit the small number of experienced M&E specialists necessary for sustainability of the systems developed with TC support. The picture is very similar in the Ministry of Local Government with respect to the PROLOGS and KLGRP projects.

5.11 While the projects reviewed are generally rated well in terms of the effectiveness and efficiency of TC – including the projects judged to be more successful (such as PROLOGS and IFMIS) – have not addressed issues of organisational capacity comprehensively in the host institutions. The difficulties facing the Ministry of Local


88

Government as PROLOGS is wound up and the intended EU-funded successor project has failed to start up on time highlights the problem.

5.12 In the projects reviewed a comprehensive approach to organisational capacity development did not in fact form the focus of support. Rather the projects concentrated on the initial establishment of workable systems in key reform areas: in the IFMIS the principal project output is the functioning hardware and software system for central government accounting. While the project made provision for a substantial degree of training this was secondary to the delivery of a functioning system; likewise, in PROLOGS/KLGRP, the emphasis of the project has lain in creating workable systems for the supervision of local authorities. In the latter case the project has been successful in creating systems, but the establishment of qualified and trained staff within the MOLG has not kept pace, leading to a large role of TC in undertaking the actual management of the systems created. Thus for the immediate future these systems are not sustainable in the absence of continued TC and there is a real danger of discontinuity and breakdown without either continued external support or increased government commitment – which would mainly relate to a willingness to address issues of civil service employment to attract and retain adequately skilled workers.


89


6.1 Kenya has represented a very challenging environment for engagement by donor agencies, and has not generally provided a context within which effective capacity development has been possible. This has been recognised in the strategy that DFID has followed which has been based on the provision of limited support to promising initiatives (mainly directed at strengthening management systems) that are judged to have the potential to bring about improvements in key areas of economic management. Technical cooperation has been used as part of this process - the focus has been on supporting organisational reforms rather than either trying to bring about policy change as was originally envisaged in DFID’s strategy towards Kenya, or providing transactional support.

6.2 The overall assessment is that the limited support DFID has provided has contributed to some successes in strengthening and establishing management systems relevant to economic management. However, achieving impact will require both a level of sustained political commitment to reforms that has never previously been achieved in Kenya, combined with measures to address fundamental issues of organisational capacity. While the overall assessment of the evaluation is that there are some grounds for optimism that initiatives will be maintained and that government commitment to strengthening public finance management and civil service reform is growing and is greater than it has been in the past (as discussed in section 3), the prospects remain highly uncertain.

6.3 The support that DFID has provided has been relatively opportunist support (though support that was generally well-coordinated with other development partners) to promising reform initiatives. Under the circumstances it is hard to see that there were genuine alternatives to this approach. Refraining from any support to management system development until there was more tangible evidence that fundamental constraints on organisational capacity including the need for improvements in the wider institutional environment would have left DFID entirely disengaged from parts of government that are central to any effective improvement in economic management and would have increased the difficulty of implementing reforms in the event that conditions had improved.

6.4 The strategic judgement therefore comes down to balancing the risk that relatively modest expenditures will prove to be unsustainable against the potential benefits from continuing engagement in developing systems that may facilitate future reforms given that key constraints on organisational capacity are not amenable to being addressed. The assessment of the evaluators is that in the cases reviewed, the potential benefits probably do outweigh the risks (and the same might be argued to have applied for continuing engagement with the CSR process from which DFID withdrew).

6.5 However, there are three ways in which the approach followed might have been strengthened. First, engagement is likely to be more effective to the extent that support is designed in a jointly agreed framework within which all major constraints to capacity development in the targeted institutions are assessed and strategies to address them or to deal with problems that cannot effectively be addressed are


90

articulated. In general such an approach will only be possible where the wider framework of reform has been adequately elaborated, with a commitment on behalf of government to carry forward all the elements of the reform process. In Kenya, such an integrated reform strategy did not exist, especially during the earlier part of the period of review under the former government. The process of formulation of the IP-ERS in 2002-3 has gone a significant way towards creating such a framework. The mid-term review of the IP-ERS provides an opportunity for the country to make further progress towards the development of a coherent and complete reform strategy – to which the government will then have to demonstrate its political commitment.

6.6 Second, there is a need for greater recognition of the long-term nature of initiatives such as the IFMIS or the PROLOGS, with consequent implications for the duration of the commitment (explicit or implicit) for TC, and also to avoid pressures to adopt unrealistic exit strategies.

6.7 Third, the view of the evaluation team was that DFID Kenya has not in general made the most effective use of local consultants and that better use of local consultancy expertise could help to improve DFID’s understanding of the context, and potentially the quality of strategic dialogue with the Kenyan Government. After reviewing the project files in some detail, it is difficult to avoid the conclusion that DFIDK has lacked a coherent strategy for the use of local consultants. In particular, the recruitment process DFID has used for local consultants is open to a number of criticisms:

• The pervasive assumption of a dual market for the delivery of consultancy services, differentiated by quality and cost for local and external consultants may be questioned since many Kenyan consultants operate in a regional and international market. This assumption is built into project cost estimates. Low budget estimates for local consultants leads directly to poor quality of candidates considered and selected and the self-fulfilment of low effectiveness.

• The absence of any organised systems to facilitate search for suitably qualified local consultants available to DFIDK. In many cases only a single candidate is approached, there is no comparison of CVs, and little or no sign of real efforts to identify and recruit the best candidate for the job.

6.8 Key conclusions for DFID’s future engagement are therefore that DFID should take a long-term perspective and be prepared to continue engagement and in some cases the provision of TC even when the political context may be unfavourable, provided that there is judged to be a reasonable possibility of the situation improving in the future. In addition, DFID should work to help create a more comprehensive and robust master plan for reform, which in turn may help to address the problems of sustainability noted in most of the projects reviewed – or at a minimum highlight the limitations on what can be achieved in terms of capacity development without wider support and reforms. DFID TC should also operate where possible within the context of a programme defined in common with other development partners as well as the GOK.


91

References

Central Bureau of Statistics, Economic Survey, Statistical Abstract, (Various Issues).

Devaragan, S., D. Dollar, and T. Holmgren, (2001), Aid and Reform in Africa – Ten Case Studies, Washington DC, The World Bank.

DFID, (1998), Kenya Country Strategy Paper, Department for International Development.

DFID, (2002), British Development Cooperation with Kenya, General Briefing, Department for International Development, April.

DFID, (2004), Kenya, Country Assistance Plan, Department for International Development.

Doig, A., Watt, D., and R. Williams (2005), Measuring ‘success’ in five African Anti-Corruption Commissions - the cases of Ghana, Malawi, Tanzania, Uganda & Zambia, Utstein Anti-Corruption Resource Centre.

Government of Kenya, (1998), Draft Medium Term Strategy for Civil Service Reform, March.

Government of Kenya, (2001), Poverty Reduction Strategy Paper, Ministry of Planning and Economic Development.

Government of Kenya, (2003a), Economic Recovery Strategy for Wealth and Employment Creation, 2003-2007.

Government of Kenya, (2003b), Public Expenditure Review.

Government of Kenya, (2004a), Enhanced Financial Management Action Plan.

Government of Kenya, (2004b), Public Expenditure Review.

Government of Kenya, (2005a), Annual Progress Report on the Investment Programme for Economic Recovery and Creation of Wealth and Employment, Ministry of Planning.

Government of Kenya, (2005b), Medium Term Budgetary Strategy Paper, Ministry of Finance.

Government of Kenya, (2005c), Joint Review of the IFMIS Project, Ministry of Finance.

International Monetary Fund, (2003), Kenya, Joint Staff Assessment of the Poverty Reduction Strategy Paper, IMF Report No. 03/397.

Ng’ethe, Prof. N., K. Musambayi, and G. Williams, (2004), How can donors promote pro-poor change in Kenya? An analysis of drivers of change, Department for International Development.

O’Brien, F.S. and T.C.I. Ryan, (2001), Kenya, in Devaragan et al..


92

OPM, (2004), Agriculture in Kenya, What shapes the policy environment?, Oxford Policy Management for DFID Kenya.

Office of the President, (2005), Sensitisation/Training Manual on Performance Contracts in the Public Service, Performance Contracts Steering Committee (February 2005).

Transparency International, (2003 and 2004), The Kenya Bribery index.

Waema, T.W., (2005), Review of the Strengthening Government Finance and Accounting Systems (IFMIS), Final Report to DFIDK, August.

World Bank, (2003), Kenya, Public Expenditure Management Review.

World Bank, (2004), Public Expenditure Management, Second Assessment and Action Plan, Kenya.


93



Joyce Nyamweya Nyakeya Permanent Secretary: Public Sector Reform and Development

Office of the President

Peter Ondieki Director: Department of Monitoring and Evaluation

Ministry of Planning and National Development

Mr Anyona Principal Economist: Department of Monitoring and Evaluation

Ministry of Planning and National Development

Raphael Mwai Project Coordinator Enabling Environment Project

Sam Mwale Adviser Ministry of Planning and National Development

Roy Kelly Adviser Kenya Local Government Reform Project

Rose Okoth Project Manager Kenya Local Government Reform Project

Paul Seeds Chief Technical Adviser, IFMIS Project

Ministry of Finance

Detlef Meijer Programme Manager DFID Support to Local Government Sector

Carl Helman Economist Swedish International Development Agency, Embassy of Sweden, Nairobi

Guy Jenkinson Economic Adviser Delegation of the European Union in Kenya

John Runda Senior Economist World Bank, Nairobi

Trish Bebbington Deputy Head of Mission DFID Kenya

Tim Lamont Economic Adviser DFID Kenya

Sue Lane Senior Governance Adviser DFID Kenya

Catherine Masinde Private Sector Adviser DFID Kenya

Martin Oloo Programme Officer DFID Kenya

Robert Simuyu Adviser DFID Kenya

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

94

Su

mm

ary

of

Fin

din

gs –

Evalu

ati

on

Ma

tric

es

Re

leva

nc

e

MF

/MP

ND

O

P(D

PM

)/K

IA

MO

LG

K

AC

A

Rele

van

ce t

o

Org

an

isati

on

al

need

s &

O

pera

tio

nal

co

nte

xt

Hig

h:

IFM

IS

ad

dre

ss

ed

th

e l

on

g

term

need

fo

r a

mo

de

rn a

cc

ou

nti

ng

s

yste

m a

s t

he b

asis

fo

r en

han

ced

tr

an

sp

are

nc

y a

nd

acco

un

tab

ilit

y;

PE

M

su

pp

ort

ed

an

aly

sis

an

d f

eed

bac

k f

rom

th

e b

ud

geta

ry

ou

ttu

rn:

ME

D s

ou

gh

t to

ad

dre

ss

th

e

dis

co

nn

ect

betw

een

str

ate

gy a

nd

th

e

bu

dg

et

by p

rov

idin

g

feed

back t

o p

lan

nin

g

Su

pp

ort

un

de

r th

e C

SR

pro

jec

t w

as r

ele

van

t to

th

e n

eed

s o

f th

e D

PM

, b

ut

su

ffe

red

fro

m t

he

p

rob

lem

of

lac

k o

f c

on

sis

ten

cy

ov

er

tim

e i

n t

he

GO

K p

rio

riti

es

fo

r civ

il s

erv

ice r

efo

rm.

Hig

h r

ele

van

ce

fo

r K

IA.

PR

OL

OG

S a

nd

K

LG

RP

bo

th h

igh

ly

rele

va

nt

to t

he

n

ee

d t

o b

uil

d

ca

pa

cit

y in

MO

LG

in

su

pp

ort

of

decen

tra

lisati

on

an

d t

he n

ew

in

str

um

en

ts o

f fi

scal

decen

tra

lisati

on

(e

sp

ecia

lly t

he

L

oc

al

Go

ve

rnm

en

t T

ran

sfe

r F

un

d)

Hig

hly

re

lev

an

t to

th

e a

ck

no

wle

dg

ed

p

rob

lem

of

co

rru

pti

on

, w

hic

h

co

nsti

tute

s o

ne o

f th

e g

reate

st

ob

sta

cle

s t

o

dev

elo

pm

en

t an

d

po

ve

rty r

ed

uc

tio

n.

As

a n

ew

in

sti

tuti

on

K

AC

A u

rgen

tly

need

ed

cap

acit

y-

bu

ild

ing

su

pp

ort

Co

nsis

ten

cy

of

DF

ID’s

an

d

org

an

isati

on

’s

ob

jecti

ves

Hig

h c

on

sis

ten

cy

be

twe

en

DF

ID

ob

jec

tiv

es

an

d t

ho

se

o

f th

e c

ha

mp

ion

s o

f c

ha

ng

e w

ith

in G

OK

, n

otw

ith

sta

nd

ing

im

po

rtan

t cen

tres o

f re

sis

tan

ce t

o c

han

ge

(esp

ecia

lly w

ith

re

sp

ect

to I

FM

IS)

DF

ID f

oll

ow

ed

th

e c

ha

ng

ing

ag

en

da o

f th

e G

OK

fo

r su

pp

ort

to

CS

RP

II

On

th

e K

IA p

roje

ct

DF

ID

ob

jec

tiv

es

were

co

ns

iste

nt

wit

h

KIA

ma

na

ge

me

nt

DF

ID’s

su

pp

ort

w

as

co

ns

iste

nt

wit

h t

he G

OK

s

tate

d a

im o

f e

nh

an

cin

g l

oc

al

em

po

werm

en

t (l

att

erl

y s

tro

ng

ly

em

ph

as

ise

d in

th

e

IP-E

RS

)

Fu

lly

co

ns

iste

nt

at

the o

uts

et

of

the

pro

jec

t

Co

nsis

ten

cy

wit

h a

cti

on

s

by G

OK

an

d

oth

er

Acti

on

s a

nd

re

sis

tan

ce t

o IF

MIS

b

y i

nte

rests

ou

tsid

e

of

MF

im

pe

de

d

imp

lem

en

tati

on

DF

ID s

up

po

rt w

as

co

ns

iste

nt

wit

h t

he

ap

pro

ac

h a

do

pte

d b

y

the

oth

er

ma

jor

do

no

rs (

WB

an

d t

he E

U)

an

d t

his

was

refl

ecte

d in

a c

ollab

ora

tiv

e

Co

ns

iste

nt

wit

h t

he

cre

ati

on

of

the

LA

TF

(an

d o

ther

instr

um

en

ts o

f fi

scal

Th

e K

AC

A

co

lla

ps

ed

wh

en

it

becam

e c

lear

that

it

did

no

t c

om

ma

nd

th

e s

up

po

rt o

f th

e

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

95

MF

/MP

ND

O

P(D

PM

)/K

IA

MO

LG

K

AC

A

Develo

pm

en

t P

art

ners

su

bs

tan

tia

lly;

go

od

alig

nm

en

t w

ith

DP

s

(ba

sk

et

do

no

rs).

In

pu

ts o

n P

EM

an

d

M&

E h

av

e b

ee

n

clo

sely

harm

on

ised

w

ith

oth

er

do

no

rs t

o

the r

esp

ecti

ve a

reas

.

ap

pro

ac

h t

o t

he

fin

an

cin

g o

f sp

ecif

ic a

cti

vit

ies

decen

tra

lisati

on

s

uc

h a

s t

he

Ro

ad

M

ain

ten

an

ce L

ev

y

Fu

nd

) b

y t

he

GO

K

an

d t

he

fin

an

cin

g

for

LA

TF

pro

vid

ed

b

y t

he

EU

.

GO

K.

DF

ID s

up

po

rt

wa

s p

art

of

a m

ult

i-d

on

or

su

pp

ort

p

rog

ram

me

.

Co

nsu

ltati

on

p

rocess t

o

en

su

rere

levan

ce?

Pie

cem

eal w

ith

re

sp

ect

to I

FM

IS,

go

od

in

re

sp

ect

of

PE

M a

nd

M&

E (

in

this

cas

e

imp

lem

en

tin

g p

lan

s

of

the n

ew

ly f

orm

ed

N

ati

on

al

Ste

eri

ng

C

om

mit

tee

on

M&

E)

Th

e a

gen

da f

or

CS

R w

as d

riv

en

b

y t

he

GO

K w

hic

h o

pe

rate

d a

ste

eri

ng

co

mm

itte

e s

tru

ctu

re.

Ho

we

ve

r th

is s

ys

tem

fa

ile

d t

o

ma

inta

in a

ro

bu

st

an

d

co

nsis

ten

t ap

pro

ach

to

re

form

o

ve

r ti

me

. F

req

ue

nt

ch

an

ge

s i

n

the

GO

K p

rio

riti

es

un

de

rmin

ed

th

e e

ffecti

ven

ess o

f D

FID

su

pp

ort

Ad

eq

uate

. A

de

qu

ate

, b

oth

wit

h

the G

OK

an

d w

ith

o

ther

DP

s.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

96

Eff

ec

tive

ne

ss

MF

/MP

ND

O

P(D

PM

)/K

IA

MO

LG

K

AC

A

Were

ou

tpu

ts

gen

era

ted

? If

no

t, w

hy n

ot?

IFM

IS o

utp

uts

ha

ve b

ee

n

me

t fi

na

lly a

fte

r p

rolo

ng

ed

d

ela

ys

ari

sin

g f

rom

p

rocu

rem

en

t p

roce

ss

es a

nd

th

e n

eed

to

re

co

ncil

e IF

MIS

w

ith

na

tio

na

l IT

po

lic

y a

nd

accep

tan

ce o

f w

irele

ss

co

nn

ecti

vit

y b

etw

een

m

inis

trie

s a

nd

Tre

asu

ry.

Yes. T

he C

SP

RE

p

roje

ct

pro

du

ce

d a

n

init

ial C

SR

P I

I m

as

ter

pla

n v

ery

rap

idly

. S

pecif

ic o

utp

uts

were

g

en

era

lly d

eli

ve

red

so

lo

ng

as t

he s

up

po

rt

co

nti

nu

ed

. H

ow

ev

er,

in

eff

ect

the C

SR

p

roje

ct

acte

d a

s a

c

all

-do

wn

fa

cil

ity

, a

nd

at

a c

ert

ain

po

int

this

ceas

ed

to

be u

tili

sed

.

Yes. T

he t

wo

pro

jects

p

rov

ide

d s

tro

ng

s

up

po

rt t

o M

OL

G i

n

the c

reati

on

of

a

fram

ew

ork

fo

r im

pro

vin

g

ac

co

un

tab

ilit

y o

ve

r ti

me

, li

nk

ed

to

Lo

ca

l A

uth

ori

ty b

ud

ge

ts

an

d d

ev

elo

pm

en

t p

lan

s (

LA

SD

AP

S).

S

ign

ific

an

tly

imp

rov

ed

co

mp

lia

nc

e

by L

oc

al

Au

tho

riti

es

w

as

ac

hie

ved

du

rin

g

the lif

e o

f th

e p

roje

cts

as p

art

of

the L

AT

F

rele

ase m

ech

an

ism

.

Ve

ry p

art

iall

y:

DF

ID T

C

co

ntr

ibu

ted

to

an

a

pp

rais

al

mis

sio

n t

o

de

fin

e r

eq

uir

ed

o

ve

rall

su

pp

ort

. T

he p

roje

ct

was

dis

co

nti

nu

ed

w

hen

lack o

f G

OK

su

pp

ort

fo

r K

AC

A b

ec

am

e

ev

iden

t

Was

ow

ners

hip

m

axim

ised

?

IFM

IS o

wn

ers

hip

ha

s b

ee

n

exerc

ised

by t

he p

roje

ct

Ste

eri

ng

Co

mm

itte

e t

hat

meets

irr

eg

ula

rly a

nd

do

es

no

t c

om

ma

nd

un

ive

rsa

l s

up

po

rt w

ith

in G

OK

. P

EM

p

roces

s led

by t

he W

B b

ut

wit

h in

cre

asin

g G

OK

c

om

mit

me

nt

an

d o

wn

ers

hip

o

ve

r ti

me

. G

oo

d G

OK

o

wn

ers

hip

of

M&

E

Ow

ne

rsh

ip i

s d

iffi

cu

lt

to a

sse

ss f

or

a p

roje

ct

wh

ich

op

era

ted

lik

e a

d

raw

do

wn

fa

cil

ity.

GO

K u

tili

se

d t

he

fa

cil

ity p

eri

od

ica

lly,

bu

t to

ad

dre

ss a

s

hif

tin

g a

ge

nd

a

Wh

ile

a m

ea

su

re o

f lo

ca

l o

wn

ers

hip

wa

s

ach

iev

ed

, th

e p

roje

cts

b

ec

am

e t

ec

hn

icall

y

iso

late

d i

sla

nd

s o

f c

om

pe

ten

ce

in

an

o

therw

ise p

oo

rly

cap

acit

ate

d m

inis

try.

So

on

aft

er

init

iati

on

it

becam

e c

lear

that

GO

K

su

pp

ort

fo

r th

e

KA

CA

was

il

lus

ory

Was

ali

gn

men

t m

axim

ised

?

Th

e p

roje

cts

were

alig

ne

d

wit

h t

he o

vera

ll G

OK

pla

ns

(as

se

t o

ut

in e

.g. th

e I

P-

ER

S)

Th

ere

was a

n

un

derl

yin

g p

rob

lem

of

lack o

f cla

rity

an

d

co

ns

iste

nc

y i

n G

OK

p

oli

cie

s f

or

CS

R t

o

wh

ich

th

e d

on

ors

, in

clu

din

g D

FID

co

uld

Th

e p

roje

cts

were

fu

lly c

on

sis

ten

t w

ith

sta

ted

GO

K a

im t

o

em

po

we

r lo

ca

l c

om

mu

nit

ies

. A

lig

nm

en

t w

ith

oth

er

DP

s w

as g

en

era

lly

Su

pp

ort

fo

r an

ti-

co

rru

pti

on

wa

s

co

ns

iste

nt

wit

h

DF

ID’s

ov

era

ll

str

ate

gy o

f in

terv

en

tio

n i

n

Ken

ya

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

97

MF

/MP

ND

O

P(D

PM

)/K

IA

MO

LG

K

AC

A

alig

n t

hem

se

lves

go

od

, th

ou

gh

de

lay

s

in t

he

ap

pro

va

l o

f a

su

cce

sso

r p

roje

ct

to

be f

un

ded

by t

he E

U

cau

sed

a h

iatu

s.

Was

harm

on

isati

on

m

axim

ised

?

Th

e d

ev

elo

pm

en

t o

f IF

MIS

h

as b

een

un

dert

ak

en

by a

b

as

ke

t o

f d

on

ors

(W

B,

SID

A

an

d D

FID

) b

as

ed

on

c

om

mo

n d

ec

isio

n-m

ak

ing

in

c

oll

ab

ora

tio

n w

ith

th

e G

OK

S

tee

rin

g C

om

mit

tee

. D

uri

ng

th

e IF

MIS

pro

ject

there

ha

ve

been

dif

fere

nces o

f o

pin

ion

o

n w

heth

er

to p

roceed

(th

e

WB

was r

ead

y t

o p

ull o

ut

at

on

e p

oin

t), b

ut

the b

asket

me

ch

an

ism

ha

s m

ain

tain

ed

fi

na

l u

nit

y o

f d

ec

isio

n-

ma

kin

g.

Ha

rmo

nis

ati

on

wit

h

oth

er

DP

s w

as g

oo

d.

In p

art

icu

lar

DF

ID

lia

ise

d c

los

ely

wit

h

the W

B a

nd

th

e E

U.

Th

ere

was p

oo

r h

arm

on

isa

tio

n

betw

een

th

e t

wo

p

roje

cts

. T

his

re

su

lte

d f

rom

DF

ID’s

a

tte

mp

t to

pla

ce

on

e

pro

jec

t in

ov

era

ll

ma

na

ge

ria

l c

on

tro

l o

f tw

o s

ep

ara

te

co

ns

ult

an

t te

am

s

(co

ntr

ary

to

in

itia

l a

rra

ng

em

en

ts),

wh

ich

le

d t

o f

ric

tio

n.

Yes. D

FID

’s

inp

uts

(li

mit

ed

as t

he

y w

ere

) to

K

AC

A w

ere

h

arm

on

ise

d w

ith

o

ther

DP

s.

Were

ou

tpu

ts

gen

era

ted

in

g

en

der-

sen

sit

ive

way?

Gen

der

issu

es

gen

era

lly n

ot

co

ns

ide

red

rele

va

nt.

G

en

de

r is

su

es

do

no

t fe

atu

re s

tro

ng

ly i

n t

he

ap

pro

ach

to

th

e

pro

jects

.

No

t a

pp

lic

ab

le,

as o

utp

uts

were

n

ot

rea

lize

d.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

98

Eff

icie

nc

y

MF

/MP

ND

O

P(D

PM

)/K

IA

MO

LG

K

AC

A

Were

ou

tpu

ts

gen

era

ted

eff

icie

ntl

y i

n

term

s o

f co

st

an

dti

meli

ness?

Co

mp

ari

so

n w

ith

IF

MIS

pro

jec

ts

in o

the

r c

ou

ntr

ies

su

gg

es

ts

hig

h c

ost

eff

ecti

ven

es

s f

or

IFM

IS, b

ut

tim

eli

ne

ss

was

very

p

oo

r d

ue

to

ob

str

uc

tio

ns

to

p

rog

res

s f

rom

wit

hin

th

e G

OK

. In

pu

ts o

n P

EM

an

d M

&E

were

m

od

es

t

Pro

cu

rem

en

t fo

r th

e

CS

R p

roje

ct

typ

icall

y

rev

olv

ed

aro

un

d

co

mp

eti

ng

bid

s f

rom

tw

o l

oc

all

y b

as

ed

c

on

su

ltin

g f

irm

s.

Th

e

ne

ed

to

re

sp

on

d

rap

idly

to

GO

K

req

uests

co

nstr

ain

ed

th

e p

rocu

rem

en

t m

eth

od

.

Bo

th p

roje

cts

were

su

bje

cte

d t

o

co

mp

eti

tiv

e t

en

de

r.

Th

ere

is e

vid

en

ce

th

at

so

me c

osts

w

ere

hig

he

r th

an

th

ey

mig

ht

ha

ve

been

had

th

e p

roje

ct

been

bett

er

de

sig

ne

d.

DF

ID w

ith

dre

w in

a

tim

ely

ma

nn

er

wh

en

it

becam

e

cle

ar

that

there

w

as n

o

go

vern

men

t c

om

mit

me

nt

to

KA

CA

th

us

av

oid

ing

wa

ste

.

Wo

uld

th

ere

h

ave b

een

alt

ern

ati

ve &

m

ore

eff

icie

nt

mo

des o

f d

eli

veri

ng

sam

eo

utp

uts

?

No

. T

he

de

cis

ion

to

su

pp

ort

IF

MIS

in

vo

lved

a c

alc

ula

ted

ris

k

tha

t th

e p

roje

ct

wo

uld

fa

il,

bu

t th

at

the

be

ne

fits

fro

m e

ve

ntu

al

su

cce

ss w

ou

ld b

e larg

e.

A

larg

er

sc

ale

in

itia

l c

om

mit

me

nt

wo

uld

hav

e e

xac

erb

ate

d

pro

ble

ms

rela

tin

g t

o

pro

cu

rem

en

t.

Th

e in

itia

l d

es

ign

wo

rk

for

the

CS

RP

II

co

uld

h

av

e b

ee

n u

nd

ert

ak

en

in

a d

iffe

ren

t w

ay,

us

ing

an

ap

pro

ac

h

wit

h g

reate

r G

OK

in

vo

lve

me

nt

ins

tea

d o

f th

e v

ery

rap

id

pre

para

tio

n o

f a

maste

r p

lan

by

co

ns

ult

an

ts. T

his

m

igh

t h

av

e l

ed

to

mo

re

su

sta

ined

ow

ners

hip

o

f th

e a

pp

roach

a

do

pte

d i

n t

he

pla

n.

No

. S

uc

ces

sfu

l d

ev

elo

pm

en

t o

f fi

scal

decen

tra

lisati

on

an

d

em

po

we

rme

nt

of

local co

mm

un

itie

s

req

uir

es

str

en

gth

en

ing

th

e

MO

LG

, w

hic

h h

as

o

ve

rsig

ht.

Fu

ll

res

ult

s w

ill d

ep

en

d

on

larg

e-s

cale

su

pp

ort

to

th

e L

ocal

Au

tho

riti

es

th

em

selv

es t

o

co

mp

lem

en

t s

up

po

rt

to t

he

MO

LG

. B

ut

su

ch

su

pp

ort

can

o

nly

be d

eli

ve

red

w

ith

in a

fra

mew

ork

o

f ru

les

an

d

man

ag

em

en

t s

yste

ms c

reate

d

fro

m t

he c

en

tre, so

it

wa

s a

pp

rop

ria

te t

o

sta

rt w

ith

th

e M

OL

G.

No

. W

ith

th

e

sta

ted

GO

K

inte

nti

on

to

ma

ke

a r

eality

of

the

KA

CA

it

was t

he

ap

pro

pri

ate

ste

p

to s

up

po

rt t

his

m

ov

e, e

ve

n

tho

ug

h G

OK

’s

real

lac

k o

f c

om

mit

me

nt

su

bseq

uen

tly

be

ca

me

ma

nif

est

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

99

Imp

ac

t

MF

/MP

ND

O

P(D

PM

)/K

IA

MO

LG

K

AC

A

Did

pro

jec

ts

ach

iev

e e

xp

ecte

d

tran

sacti

on

al/

tra

ns

form

ati

on

al

imp

ac

t?

Th

e IF

MIS

pro

ject

has

gen

era

ted

a s

yste

m t

hat

is

no

w r

ea

dy f

or

roll

ou

t to

lin

e

min

istr

ies a

fter

su

cces

sfu

l p

ilo

tin

g i

n M

F/M

PN

D.

Wh

ile

th

e i

mm

ed

iate

ou

tpu

ts o

f th

e

pro

ject

are

tra

nsa

cti

on

al,

they la

y t

he b

asis

fo

r m

ajo

r tr

an

sfo

rma

tio

na

l c

ha

ng

es

in

b

ud

get

man

ag

em

en

t an

d

ac

co

un

tab

ilit

y.

PE

M h

as h

elp

ed

main

tain

an

d d

ev

elo

p t

he u

se o

f th

e

an

nu

al

PE

R a

s a

to

ol

for

imp

rov

ing

bu

dg

et

cyc

le

ma

na

ge

me

nt.

Re

su

lts

fro

m

the

PE

R a

re f

ee

din

g t

hro

ug

h

the A

nn

ua

l P

rog

ress R

ep

ort

in

th

e I

P-E

RS

in

to t

he

p

lan

nin

g a

nd

bu

dg

eta

ry

cycle

, alb

eit

no

t as s

tro

ng

ly

as

mig

ht

be

wis

he

d.

Th

e C

SR

pro

jec

t h

ad

litt

le e

ffect

in t

erm

s o

f tr

an

sfo

rma

tio

na

l im

pa

ct

larg

ely

be

cau

se

of

the

la

ck o

f co

nsis

ten

cy in

th

e a

pp

roa

ch

ov

er

tim

e

an

d t

he

lim

ite

d s

co

pe

of

the

DF

ID i

nv

olv

em

en

t.

Th

e K

IA p

roje

ct

Ph

as

e I

did

ac

hie

ve b

oth

its

tr

an

sacti

on

al an

d

tra

ns

form

ati

on

al

imp

ac

t, b

ut

the

fa

ilu

re

to p

roceed

to

Ph

as

e II

un

de

rmin

ed

th

e

su

sta

ina

bil

ity

an

d

op

tim

isa

tio

n o

f b

en

efi

ts

fro

m t

his

re

su

lt.

Yes, P

RO

LO

GS

mad

e

su

bs

tan

tia

l p

rog

res

s

in c

reati

ng

es

sen

tial

fin

an

cia

l m

an

ag

em

en

t fr

am

ew

ork

s f

or

MO

LG

to

exerc

ise

its

ro

le o

f o

vers

igh

t o

ver

LA

s

No.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

100

Su

sta

ina

bilit

y

MF

/MP

ND

O

P(D

PM

)/K

IA

MO

LG

K

AC

A

Have

tran

sfo

rm-

ati

on

al

imp

acts

been

su

sta

ined

?

Ge

ne

rall

y t

oo

ea

rly

to

ju

dg

e

de

fin

itiv

ely

. B

ut

pro

sp

ec

ts

for

full

ro

ll o

ut

of

IFM

IS a

re

go

od

.

Imp

lem

en

tati

on

an

d

ma

inte

na

nc

e o

f M

ED

p

roces

ses a

nd

syste

ms

is

lik

ely

to

co

nti

nu

e t

o b

e

co

nstr

ain

ed

by s

taff

ing

p

rob

lem

s i

n t

he

ME

D.

Fo

r th

e C

SR

pro

jec

t, g

ive

n

lim

ite

d t

ran

sfo

rma

tio

na

l im

pact

the is

su

e o

f s

us

tain

ab

ilit

y d

oe

s n

ot

ari

se.

Fo

r th

e K

IA p

roje

ct

the

tra

ns

form

ati

on

al

imp

ac

t h

as

be

en

lim

ite

d b

y f

ail

ure

to

fo

llo

w t

hro

ug

h t

o P

ha

se

II

of

the

pro

jec

t w

hic

h

wo

uld

hav

e c

on

so

lid

ate

d

the r

esu

lts

ach

iev

ed

.

Th

e m

ain

tr

an

sfo

rma

tio

na

l im

pa

ct

ha

s b

ee

n t

he

cre

ati

on

of

a

fram

ew

ork

fo

r s

up

erv

isio

n o

f b

ud

geta

ry t

ran

sfe

rs t

o

the d

ecen

tralis

ed

u

nit

s. H

ow

ev

er,

th

ese

syste

ms a

re s

till

no

t su

sta

inab

le in

th

e

ab

se

nc

e o

f s

ub

sta

nti

al

co

nti

nu

ing

ca

pa

cit

y-

bu

ild

ing

su

pp

ort

to

th

e

MO

LG

esp

ecia

lly f

or

the

Fin

an

cia

l M

an

ag

em

en

t S

ys

tem

, (a

nd

su

bseq

uen

tly t

o

the L

ocal

Au

tho

riti

es).

No

, a f

ort

iori

, as

o

utp

uts

were

no

t re

alized

.

Was t

here

a

co

here

nt

exit

str

ate

gy

pre

pare

d in

ad

van

ce?

No

t fo

r IF

MIS

, w

hic

h a

ime

d

to s

tart

a p

roces

s t

hat

it

was k

no

wn

wo

uld

exte

nd

b

eyo

nd

th

e l

ife

of

the

p

roje

ct.

Ph

as

e II

cu

rren

tly

un

de

r c

on

sid

era

tio

n.

Ex

it s

tra

teg

y n

ot

ap

pli

ca

ble

fo

r th

e m

ino

r a

nd

dif

fus

e

su

pp

ort

un

der

PE

M a

nd

M

&E

.

Th

e C

SR

pro

ject,

as a

fa

cil

ity, d

id n

ot

ha

ve

an

exit

str

ate

gy.

In e

ffect

DF

ID

wit

hd

rew

wit

ho

ut

co

mp

leti

on

of

us

e o

f th

e

bu

dg

et.

Th

e K

IA p

roje

ct

de

sig

n

pro

vid

ed

fo

r exit

at

the e

nd

o

f P

ha

se

I. T

his

op

tio

n w

as

exerc

ised

to

th

e d

etr

imen

t o

f th

e o

ve

rall

pro

jec

t im

pa

ct.

No

. T

his

has b

een

a

pro

ble

m.

Th

e e

nd

of

the p

rog

ram

cre

ate

d a

h

iatu

s p

en

din

g t

he

c

om

me

nc

em

en

t o

f su

pp

ort

fro

m a

n

alt

ern

ati

ve

do

no

r so

urc

e (

the E

U).

No

. B

ut

the

ba

sk

et

fun

darr

an

gem

en

ts

ma

de

do

no

r d

isen

gag

em

en

t an

d r

eco

very

of

un

sp

en

t b

ala

nc

es

fa

irly

str

aig

htf

orw

ard

.


101

SOUTH AFRICA CASE STUDY


102

Executive Summary

S1. This report is a case study of DFID-financed technical cooperation (TC) for economic management in South Africa focusing on the period 1999-2004. It evaluates support to two organisations involved in economic management:

• The National Treasury (NT).

• The Department of Public Enterprise (DPE).

S2. The TC was provided through three programmes. “Support for Economic Reform Projects” (SERP) I and II (1998 – 2004) and ”Support for Restructuring of Public Enterprises South Africa” SRPESA. Both long and short term TC was provided. The long term TC included staff seconded from Her Majesty’s Treasury (HMT) and ODI fellows as well as TC recruited by the management agents. South Africa was included in the evaluation as representing a middle income country in which the context for support was likely to be relatively favourable, as well as being a major recipient of DFID TC.


Political context

S3. The period covered by this evaluation has seen a consolidation of South Africa’s political progress from white minority rule to the establishment of a democratic state with a strong constitutional basis for the rule of law and respect for human rights, despite the extremely high levels of income inequality and problems of social deprivation and poverty. The period has been one of stable government with the African National Congress winning large majorities in elections in 1999 and 2004. Key political developments over the last decade have included the establishment of the new constitution and the articulation and carrying through of the Growth, Employment and Redistribution Strategy (GEAR) as the overarching framework for the government’s economic policy. The South African political system appears to have been effective in encouraging an approach to the resolution of political conflict based on seeking consensus and the balancing of interests, and a strong respect for the rule of law and constitutional principles.


S4. The democratic government of South Africa inherited a government structure and public service that was oriented towards meeting the needs of a minority of the population. It has faced the dual challenge of reorienting the focus of service provision and of making the staffing of the public service more representative of the whole population, in a context of severe educational disadvantage and a lack of exposure to management experience or opportunities for professional development for the majority of the population. The South African government has faced important capacity constraints in relation both to management systems and to skill shortages in its


103

attempts to reorient and improve the effectiveness of the public service in relation to its development objectives.

S5. The South African government has a well-articulated policy towards external aid (which is generally less than 0.5% of GDP). This specifies that aid should enhance the long-term sustainability of the government’s reconstruction and development efforts and hence can be seen as having a clear capacity development focus. Specifically the policy requires that all support should be for core development priorities, that all activities should be South Africa driven, and should be fully funded and financially sustainable beyond the provision of donor financing, while the management of TC should be integrated with normal government functioning so far as possible.


S6. DFID’s approach to the provision of TC in South Africa over the period of the evaluation strongly emphasised TC responsively provided as an instrument, but without articulating a particular strategy or framework for the use of the instrument, beyond a focus on system development and skills transfer. The DFID portfolio during the early part of the evaluation period was viewed as fragmented as a result, and its management has been rationalised through the establishment of larger programmes of support in areas identified as having the strongest strategic impact of policy and implementation with respect to poverty reduction objectives.

Economic management in South Africa

S7. South Africa has over the last decade succeeded in raising rates of economic growth, controlling inflation and achieving a high level of economic stability as measured by falling long-term interest rates. Economic growth rates however remain below the levels necessary to reduce substantially poverty or unemployment, with the high level of HIV prevalence providing a further constraint on growth.

S8. Prudent fiscal policy and structural budget reforms have been the cornerstones of the macroeconomic stabilisation in South Africa. Small budget deficits and expenditure control have contributed to reducing the burden of public debt. Savings from lower debt-service payments have been channelled to increase spending in social welfare and economic infrastructure. There has been a steady lowering of inflation, interest rates and some improvement in investment.

S9. South Africa has established a robust budget and public expenditure management process, based around a medium-term expenditure framework, clearly defined roles for the different elements of the executive and legislature (enshrined in the 1999 Public Finance Management Act), and based on the principle of holding managers accountable not only for financial control but also for the production of outputs for which they are responsible.

S10. South Africa has made progress in economic management since democratisation that has been described by the IMF (2005) as “remarkable.” The overall commitments to economic management envisaged in GEAR have been institutionalised through the constitutional arrangements and the public finance system enshrined in the 1999


104

Public Finance Management Act. This has provided a sound overall framework for improving growth and service delivery, though immense challenges remain in overcoming the legacies of poverty and disadvantage that most South Africans face.


S11. The activities in support of the National Treasury have generally been successful as a result of the NT’s clear mandate and policy direction and effective management that has enabled it to articulate and manage its needs for external support. However more attention to the selection of some of the TC providers (in matching their skills to what was demanded), as well as a more explicit capacity development focus for the activities, could have increased the impact. The effectiveness of the support to the Department of Public Enterprises was initially somewhat compromised as a result of an overestimation of the South African commitment to privatisation as a policy tool, management and staffing weaknesses within DPE and a mismatch between the approach of the programme management consultants and DPE’s requirements. These issues have been addressed through changes to staffing and project restructuring and subsequently through measures by a new Minister to clarify the Department’s mandate and approach and to improve management.


Relevance

S12. In terms of goals and objectives both the Support for Economic Reform Projects (SERP) and the Support for Public Enterprise Restructuring (SRPESA) were relevant to GSA needs and ownership of the programmes was strong in the organisations supported. Both the NT and DPE have critical roles to play in South Africa’s GEAR programme. Both programmes however suffered, viewed in relation to their capacity development objectives, from insufficient attention to the organisational context within which they were implemented, with the uncertainty about DPE’s mandate (and the level of commitment across government to its stated policy approach) providing the main problem. To some extent the initial success of SERP I may have contributed to this neglect of institutional and organisational issues. However the flexibility of design, and the willingness of all parties to overcome problems encountered reduced the negative impact of these initial design weaknesses.

Effectiveness

S13. The projects have generally been highly effective (with SERP I making a major contribution to establishing a modern system for public expenditure management for the South African context) despite some difficulties under SERP II and SRPESA. Ownership has been strong and this has been reflected in both design and management. A common factor reducing both effectiveness and efficiency in the two projects was the lack of a systematic approach to capacity building in the South African public service. This could have led to a more coherently sequenced approach which


105

might have resulted in a different emphasis in early sub-projects, and the addressing of organisational and cross cutting issues rather than simply the transfer of skills to individuals.

Efficiency

S14. Overall outputs for the NT were provided on time and within budget. In support to DPE some initial projects were not continued following the review of the project’s management in 2004. In evaluating projects it is important to distinguish between the efficiency of project management, i.e. the delivery of inputs, and the efficiency of the project interventions in achieving the desired outputs and outcomes. SERP II and SRPESA were generally efficiently managed in terms of delivering the inputs requested by the Departments it was supporting but less so in prioritising the spending of funds or in systematically evaluating outputs and outcomes.

Impact

S15. SERP I had an important impact in helping the process of establishing and implementing new budget systems. The National Treasury’s strong management and staffing capacity has enabled this contribution to be transformational and to be sustained. The impact of SERP II is less clear but the initial work with DPE was considered sufficiently useful to have generated a doubling of the Know How Fund and a spin off into SRPESA. So far the impact of SRPESA can be detected in some of the achievements of DPE in influencing pricing and regulation, but its ultimate impact will in part depend on whether it can help the GSA deliver clear objectives for state owned enterprises.

Sustainability

S16. SERP could be said to be sustainable because some of its activities have continued even after donor funds have gone. The NT has set aside some of its own funds for advice from PUK on Public Private Partnerships to the PPP Unit. More generally, NT reports that there is a growing local capacity for consultancy support in specialised areas. There are also signs that the knowledge gained will largely be retained, particularly in the NT which has, for example, launched a PPP manual, a sign that knowledge gained from consultants is being retained within the Department. However, how much of this can be attributed directly to SERP is difficult to establish.

S17. Whether or not improvements in capacity attributable to SRPESA are sustainable will depend critically on attaining stability in staffing in line departments so that transferred skills are not immediately lost to the private sector. It will also depend on whether beneficiary organisations have robust systems and procedures that will preserve the institutional memory of “how things should be done”, which can otherwise be lost even with reasonable staff retention rates.


106


S18. The role of DFID TC support has been to respond to a demand from the organisations supported, first, to access knowledge about UK government reform experience which was seen as highly relevant to South Africa’s needs, and second to use the support provided as a way of sourcing expertise that circumvented particular political and administrative constraints on using normal government channels. The approach to providing the support reflected the generally responsive and demand led nature of DFID’s TC in South Africa. The effectiveness and capacity development impact of the support provided was limited by the lack of a comprehensive analysis of the organisational context, the setting and management against capacity development targets, and some problems in sourcing appropriately skilled TC providers.

S19. The main implication is therefore the need (in line with South Africa’s policy on the use of TC) for a more rigorous assessment of the organisational context for organisations receiving support. For organisations that are less strong than the NT, this is likely to suggest that attention to issues about the role and mandate of the organisation, its internal management, and capacity to attract and retain appropriate staff, will need to be addressed for TC to be used in a way that achieves a transformational impact.


107

1. Introduction

1.1 This case study is one of four (the others being Ghana, Kenya and Zambia) in a wider study to evaluate the role of DFID TC in Economic Management in sub-Saharan Africa. The purpose of the wider study is:


1.2 The overall approach and conceptual framework for the country studies is discussed in the Synthesis Report (Volume 1) and the introduction to this volume. This study used the following main sources of information:

• Examination of a wide range of DFID management files and project documents and GSA policy documents

• Discussions with:

a. Staff and advisers of the National Treasury.

b. Staff and advisers at the Department of Public Enterprise.

c. Staff at the Independent Communications Authority of South Africa (ICASA).

d. Consultants and project managers from the main projects.

e. The SRPESA independent evaluation team.

f. Staff at DFID’s South Africa country office.

1.3 Projects for inclusion in the evaluation were selected in consultation with DFID from those falling within the scope of economic management and involving TC expenditure classified to economics or governance that it was assessed as practical to evaluate within the time and on the basis of the information available. For South Africa this process led to the identification of the following focal organisations and programmes which it was agreed with DFID Evaluation Department and the DFID South Africa office should provide the focus of the evaluation:34

• The National Treasury (Support to Economic Reform projects, SERP I and II).

• The Department of Public Enterprises (SERP II and Support for the Restructuring of South African Public Enterprises (SRPESA).

1.4 Table 1 shows expenditure on the activities reviewed as a proportion of the total DFID country programme and of total TC in the wider Economic and Governance sectors. With the exception of SRPESA, these activities represented only a small fraction of both total TC and the country programme. The major TC expenditures in these sectors which were not included in the evaluation were the Integrated Provincial Support Programme (IPSP) and the Local Government Support and

34 Although the SERP II programme provided TC to the Department on Trade and Industry it was not

practical to conduct interviews with recipients during the mission to Pretoria so it was not included in the evaluation.


108

Learning Programme (Logosul) – both of which it was considered impractical to include within the evaluation because of their focus at provincial level; Support to the South African Revenue Service; the British Investment in South Africa Promotion Scheme, and support to the Public Service Commission and the Department of Public Service and Administration.

1.5 The report is structured as follows. Section 2 examines the context for capacity development in relation to the political and institutional context, and the role of aid including DFID’s strategy. Section 3 provides an overview of economic management over the period of the evaluation. Section 4 presents the findings of the evaluation in relation to the results achieved from the TC provided to the focal organisations. Section 5 summarises the findings against the main evaluation criteria. Section 6 provides an overall conclusion and discusses issues emerging from the case study.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

109

Ta

ble

1.

DF

ID T

C p

roje

ct

ex

pen

dit

ure

in

So

uth

Afr

ica

: E

co

no

mic

an

d F

ina

nc

ial M

an

ag

em

en

t

Fis

cal

Year

(Ap

ril

- M

arc

h);

£ '000s

Pro

ject

1999/0

02000/0

12001/0

22002/0

32003/0

4

Lo

ng

Term

Sh

ort

Term

Co

ns

ult

an

ts

Lo

ng

Term

Sh

ort

Term

Co

ns

ult

a

nts

Lo

ng

Term

Sh

ort

Term

Co

nsu

ltan

ts

Lo

ng

Term

Sh

ort

Term

Co

ns

ult

a

nts

Lo

ng

Term

Sh

ort

Term

Co

ns

ult

a

nts

To

t. E

xp

.

99-0

4

Pla

nn

ed

Exp

.

SE

RP

I109

..155

68

..180

6..

69

....

....

....

587

865

SE

RP

II

....

....

....

85

..113

134

..1397

69

..664

2462

2600

SR

PR

ES

A..

....

....

....

....

....

....

..3125

3125

6800

TC

evalu

ate

d109

0155

68

0180

91

0182

134

01397

69

03789

....

To

tal/

Year

264

248

273

1531

3858

6174

..

% o

f T

ota

l E

co.

&

Gov.

TC

4%

3%

4%

11%

25%

12%

..

% o

f T

ot.

DF

ID

Countr

y P

rog

0.9

%0.8

%1.1

%4.2

%9.1

%3.8

%..

To

tal

Eco

. &

Go

v.

TC

6318

7335

7451

13643

15605

50352

..

% o

f T

ota

l D

FID

Countr

y P

rog

21%

25%

30%

37%

37%

31%

..

To

t. D

FID

Co

un

try

Pro

g.

29515

29783

25119

36445

42365

163227

..

The table

sum

mari

ses tota

l spend

ing

by

DF

ID o

n T

C p

roje

cts

eva

luate

d r

ela

ted t

o e

conom

ic a

nd f

inancia

l managem

ent. E

xpen

diture

is a

lso s

ho

wn f

or

the t

ota

l econ

om

ic a

nd g

overn

ance s

ecto

rs T

C a

nd th

e tota

l DF

ID C

ountr

y P

rogra

mm

e for

the p

eri

od 1

999-2

004

. T

he f

igure

s a

re b

ase

d o

n t

he lis

ting

of

TC

pro

jects

pro

vid

ed b

y th

e D

FID

Eva

luation D

epart

ment an

d info

rmatio

n c

olle

cte

d b

y th

e c

onsultant

duri

ng t

he f

ield

vis

it in

So

uth

Afr

ica.


110


The political context

2.1 The period covered by this evaluation has seen a consolidation of South Africa’s political progress from white minority rule to the establishment of a democratic state with a strong constitutional basis for the rule of law and respect for human rights, despite the extremely high levels of income inequality and problems of social deprivation and poverty resulting from many decades of institutionalised discrimination against the majority of the population and the violence that resulted from this and the struggle against the apartheid regime.

Figure 1. Key political events 1994-2005

2.2 Figure 1 summarises key political events over last decade. The African National Congress has won substantial victories in each of the three general elections and a new constitution has been adopted as the outcome of a long process of negotiation that spanned the period of transition to democratic government.

1990 – 1994: negotiations for an interim government; introduction of the new Constitutional dispensation. Negotiations started for new Constitution

1994 April: Elections Landslide victory for African National Congress (ANC)

1994 – 1996: Government of National Unity (GNU). President Nelson Mandela. Thabo Mbeki and F. W. De Klerk (New National Party), Deputy Presidents

1994 Reconstruction and Development Programme (RDP).

1996 Growth and Redistribution Programme (GEAR). New Constitution adopted.

1996 F. W. De Klerk resigned – never replaced

1997 NNP withdraws from government to become an independent opposition party

1999 Nelson Mandela retires, replaced as President by Thabo Mbeki

1999 ANC election victory. NNP vote falls from 20% to 7% replaced by as main opposition by Democratic Alliance (DA)

2004 ANC victory in polls. NNP disbands (vote fell to 1.7%)

2005 Deputy President Zuma charged with corruption


111

2.3 The constitution establishes a unitary rather than federal system but has three spheres of government - national, provincial and local - each with its own legislatures and executives with defined responsibilities. There is a single national judiciary. The spheres of government have distinct fields of responsibility but are required to coordinate their activities and legislation on the basis of agreed procedures – responsibility for economic and the budget is in the hands of the national Government, in particular the Treasury, but the sub national spheres are integrated into the budgetary process. The Executive is derived from the legislature, much along the Whitehall model. The President is elected by each new National Assembly and has the responsibilities of a British Prime Minister in running the Executive and of the Monarch in assenting to Legislation. He may only serve two terms, unless he takes tenure during the validity of a National Assembly. The Cabinet is selected from the legislature, which has two houses, similar to those of a federal state. This formal structure for the state emerged from the process of negotiation, and is designed to maintain checks and balances and to prevent concentration of power.35 Legislative control over the Executive is enhanced by the fact that the President is elected by the Legislature from its members as well as by systems of oversight by legislative committees including the Public Accounts Committee, Budget Committee and Portfolio (sectoral) Committees.

2.4 In the immediate aftermath of the 1994 landslide victory for the ANC, the Government of National Unity introduced a new programme aimed both at redressing past imbalances and improving South Africa’s poor economic performance – the Reconstruction and Development Programme (RDP). This focused on targets for improving service delivery for the majority of the population. With the adoption of the new constitution in 1996:

• The RDP Office was abolished and its central functions transferred to the Ministry of Finance.

• A new Minister for Finance was appointed, the first from the ANC.

• The Ministry of Finance published a new macro-economic strategy, the “Growth, Employment and Redistribution Strategy” (GEAR), which was seen as providing the macroeconomic framework within which the RDP objectives could be taken forward.

2.5 The objectives of GEAR were – and still are:

• A competitive fast - growing economy which creates sufficient jobs for all work seekers.

• A redistribution of income and opportunities in favour of the poor.

• A society in which health care, education and other services are available to all.

2.6 GEAR received widespread support from the business, international and donor communities, as well as from the Bretton Woods institutions, but has been a cause

35 Nathan (2004) argues that many of the positive features of the process of constitutional negotiation – its

inclusiveness both across the political spectrum and all sectors of civil society, and a commitment to accommodation and balancing of diverse interests through processes of consensus building - have become institutionalised in the post-apartheid period.


112

for controversy among some of the ANC’s alliance partners, particularly the Congress of South African Trade Unions (COSATU). GEAR objectives were taken by COSATU to imply privatisation was a key objective, compared to the more active role for the state that was seen as implied by the RDP. A key feature of GEAR has been the way in which a commitment to strong and effective economic management has been critical to the achievement of the government’s growth and transformation objectives.

Institutional context and policies on TC

2.7 The institutional context reflects the impact of decades of apartheid in establishing a relatively effective public service, but one that was oriented to the needs of a minority and the particular structures of the apartheid state (such as the “homelands”). The majority of the population were left undereducated and poorly equipped for formal employment. The DFID 2002 CSP noted that:

2.8 “The effects of the inadequate education available to most black people before 1994 are a particular constraint on poverty elimination. The two-tier education system strongly reinforced the majority population’s sense of disempowerment. Many of the 38% of the population who are currently unemployed lack the skills required to compete in the formal employment market, or to succeed as entrepreneurs.”

2.9 Table 2 illustrates key features of this structural inequality in the labour market. The principle of black economic empowerment is essential to the social and economic development of South Africa. The Employment Equity Act compels all employers, including the government, to set targets for the number of black staff, the number of black staff in management and the number of female staff in management at each level.

Table 2. Unequal participation

Source: MTBPS (2004)

2.10 One of the specific inheritances of the post-apartheid state was a public service that was largely oriented to providing services to (and managed at a senior level by) the white minority of the population, in a context where educational disadvantages for the majority of the population posed great difficulties in bringing about rapid staffing change to make the public service more representative and effective in addressing

Dimensions White Blacks Male Female >35 <35 Urban Rural

Average Income R8277 R2091 R3553 R2244 R3254 R2624 R3738 R1380

Unemployment 7% 35% 28% 35% 18% 43% 31% 32%

Skills/Education (%

economically active

with grade 12 and

above) 86% 32% 38% 40% 31% 46% 47% 22%


113

the underlying problems. The government has faced important capacity constraints in relation both to management systems and to skill shortages in the attempt to reorient and improve the effectiveness of the public service in relation to its development objectives.

2.11 To some extent the need to support capacity development across a wide range of individual abilities and experience can be seen as the consequence of the quite rapid progress the GSA has made in pursuing its Black Economic Empowerment and gender equity targets. Policies aimed at ensuring the workforce reflects the demographic profile of the country as whole have had the effect of bringing into positions of responsibility people from historically disadvantaged groups who have suffered the consequences both of poor educational opportunities under apartheid and of a lack of exposure to management or professional development programmes and functioning public administrations. There was a consensus among those consulted for the review that capacity in line departments of government – though not in the National Treasury – is lower than might be expected in a country of South Africa’s overall level of development. Bloch, Favis and Hargovan (2000 p. ii) noted that:

“Within government, systems are chaotic, often impeding delivery… Management capacity at all levels is a significant problem and this has a key impact on performance. Racial issues articulate uncomfortably with these problems… Affirmative Action as not addressed the deeper issues concerning negative racial attitudes and perceptions.”

2.12 The GSA policy for technical cooperation and official aid is set out in the October 2003 Policy Framework and Guidelines for the Management of Official Development Assistance. This incorporated findings from the 2000 Development Cooperation Report, which reviewed and synthesised the lessons from the experience of aid to South Africa over the period 1994-1999. Bloch, Favis and Hargovan (2000) in an evaluation of donor funded capacity building support over this period suggested that there were four key elements that should form the basis of a checklist for the management of technical assistance:

• Proper planning.

• Rigorous and consultative recruitment procedures.

• Preparing the institutional context.

• Ensuring appropriate lines of accountability.

2.13 This evaluation carried out in 2000 contributed to the formulation of GSA policy which are set out in DPSA (2003). This now includes strict criteria for acceptable technical co-operation that are intended to enhance the long-term sustainability of Government’s reconstruction and development efforts:

• All ODA-supported interventions should be core South African development priorities.

• All such interventions should be South African driven.

• All such interventions should be fully funded and financially sustainable beyond the provision of donor financing (such as counterpart staff, and recurrent operations and maintenance expenditure) and


114

• The management of such interventions is integrated with normal government functioning to the fullest extent possible.

2.14 Capacity building support from donors to address the institutional problems summarised above is actively sought, with particular merit being seen in secondments of people who have had substantial experience of public administration. Grants and technical assistance including training and capacity building are treated as additional to parliamentary appropriations (and hence falling outside the Medium Term Expenditure Framework ceilings).

2.15 The overall constitutional and political environment can therefore be regarded as highly favourable for effective economic management, and incorporates a commitment to strong and effective economic management as key to achieving government objectives. South Africa has well-articulated policies for the use of technical cooperation for capacity development, and a strong understanding of its priorities for the use of such support.36 Organisational capacity (and the capacity of individuals) within the public sector is however very variable, reflecting the challenges of the need for both a substantial reorientation of the historic role of the public sector towards serving the whole community and especially the poor, and for progress with an effective empowerment agenda that will compensate for the legacy of past discrimination. This suggests that detailed knowledge and understanding of the particular organisational context will be particularly significant for ensuring effective use of technical cooperation for capacity development.

DFID’s strategy for TC in South Africa

2.16 DFID’s 1998 Country Strategy Paper (CSP) set out DFID’s strategy as to support South African efforts to:

• Overcome the constraints to higher economic growth and promote pro-poor investment for sustainable wealth creation which allows more people to contribute to and benefit from economic growth.

• Improve the effectiveness of government and its agencies in delivering essential health, education, water and sanitation services to the poor.

• Develop sustainable rural livelihoods and initiatives with direct benefit to poor people. DFID will support the building of a public service able to put government policies into practice.

2.17 The major activities in achieving these objectives were to be developed with all three spheres of government to deliver:

• Growth and development through capacity building and integrated development planning in selected rural and urban municipalities with high unemployment, service backlogs and insufficient capacity.

36 Most of the activities reviewed were designed before the formal adoption of these policies. However they

seem in general to reflect the criteria proposed, with the main issue of uncertainty being the extent to which long-term sustainability has been addressed in the design of support to SRPESA as discussed below.


115

• Improved service delivery, through devolution of functions, publication of service standards and developing alternative delivery systems including public-private partnerships.

• More efficient and effective public services, through more efficient revenue collection, training for political and public service leadership in provinces and municipalities, transformation to meet national targets for affirmative action on race, gender and disability.

• Better use of financial resources, through reallocation of expenditure and staff to areas that add social value, and performance-based budgeting, which links service delivery outputs to the public spending used to generate them.

2.18 The 1998 CSP set out this broad framework but did not include any detailed discussion of capacity issues in South Africa or their implication for the DFID programme, other than reference (p. 5) to “administrative and capacity constraints to adoption and rapid implementation of a more radically redistributive agenda.” The CSP also did not include any discussion of the instruments to be used or the specific role of technical cooperation, though its review of areas where DFID had contributed to South African achievements focused on examples relating to advice and know how.

2.19 A performance review carried out in 2001 concluded (DFID, 2002 p. 14) that:

“well-targeted technical assistance efforts are highly valued by South African partners, and can make a valuable contribution to policy development and to strengthening implementation capacity”.

2.20 The 2002 Regional Strategy Paper for Southern Africa also noted that the main focus of the DFID programme in South Africa had been on “supporting transformation in government, and that there was a case for paying more attention to the promotion of pro-poor growth.” The RSP argued that for South Africa DFID’s “strength … in developing and delivering technical cooperation activities in particular sub-sectors … led to a diverse programme with some 150 separate interventions in 2000.” The strategy stated that (p. 15):

“In the light of the overall pattern of donor activity, we believe that DFID should develop a broad-ranging engagement on poverty rather than a highly focused programme of sector-specific interventions. We will seek to prioritise those initiatives which will have the strongest strategic impact on policy and implementation, which will help promote a holistic response to poverty, and which offer scope for strong partnership with South African counterparts.”

2.21 The focus of the 2002 CSP was on contributing to outcomes in the following areas:

• Strengthening poverty analysis and strategy.

• Promoting growth, jobs and equity.

• Strengthening democracy, governance and service delivery.

• Tackling HIV/AIDS.

2.22 Policy dialogue and technical assistance were identified as the primary instruments to be used but again the CSP did not include any discussion of issues relating to how


116

TC was to be most effectively applied. The strategy however noted in relation to organisational capacity (p.13) that:

“The government has a strong commitment to addressing the inherited problems of apartheid but faces severe capacity constraints in doing so. There are many able and dynamic policy-makers in government but implementation capacity is less strong – with weak systems, and skills shortages in many parts of government.”

2.23 DFID’s overall approach to the provision of TC in South Africa over the period of the evaluation can be summarised as one that has strongly emphasised TC responsively provided as an instrument, but without articulating a particular strategy or framework for the use of the instrument, beyond a focus on system development and skills transfer. The DFID portfolio during the early part of the evaluation period was viewed as fragmented as a result, and its management has been rationalised through the establishment of larger programmes of support in areas identified as having the strongest strategic impact on policy and implementation with respect to poverty reduction objectives.


117

3. Economic management in South Africa

Overview of economic performance

3.1 From the 1960s, the South African economy deteriorated. GDP growth fell steadily from an average of 6 per cent in the 1960s to 3 per cent in the 1970s, and 2 per cent in the 1980s. There was a continued decline in the first years of the 1990s, followed by an economic rebound in 1993. Since then, overall economic growth rates have improved (though growth remains below the levels necessary substantially to reduce poverty or unemployment); public finances have been strengthened; inflation is under control; and long-term interest rates have been falling. Table 3 summarises South Africa’s recent economic performance.

Table 3. Key macroeconomic indicators

2001 2002 2003 2004 2005 (proj)

Real GDP growth 2.7 3.6 2.8 3.7 4.0

Inflation (CPIX) 6.5 10.7 4.0 4.3 5.1

Investment (% GDP) 15.3 16.1 17.2 17.7 17.7

Overall fiscal balance (% GDP)

-1.5 -1.2 -2.0 -1.7 -2.1

Source: IMF (2005)

3.2 The depth of the structural problem of inequality is illustrated however in Table 4. The Government has set the target of halving the unemployment rate by 2014. It has initiated the Skills Development Program (SDP) and the labour intensive public works program in order to enhance skills and expand job opportunities. The high prevalence of HIV has a significant effect on economic performance, reducing both the size and effectiveness of the work force. HIV prevalence remains high at 15.6% in 2003, which has also lowered life expectancy significantly from 61.9 years in 1990 to 45.7 years in 2003. Estimates suggest that HIV could reduce GDP growth by between 0.5 and 0.25 percentage points per year (IMF 2004).


118

Table 4. Unemployment, inequality and poverty indicators

1990 1995 2001 2002 2003

Unemployment rate (total labour force) .. 17 29.5 30.5 28.2Population below US$1/day .. 11.5 .. 10.5 ..Gini coefficient (Total) 0.68 0.69 0.77 .. ..

Gini coefficient (African) 0.62 0.66 0.72 .. ..

Gini coefficient (White) 0.46 0.5 0.6 .. ..Prevalence of HIV (% of population aged

15-49) .. .. .. 15.2 15.6Life expectancy at birth (years) 61.9 58 47.8 46.5 45.7

Source: WDI (2005), MDI (2003), IMF (2004) and HSRC (2004).

Macroeconomic management: 1999-2004

3.3 Prudent fiscal policy and structural budget reforms have been the cornerstones of macroeconomic stabilisation in South Africa. Small budget deficits and expenditure control have contributed to reducing the burden of public debt. Savings from lower debt-service payments have been channelled to increase spending in social welfare and economic infrastructure. The 2005 Budget assumed a higher fiscal deficit than 2003/04. This is expected to slightly increase public debt as a proportion of GDP, which has remained under 40% of GDP since 2002. In 2004 the Government signed a three-year wage agreement with public sector unions covering national and provincial government employees, allowing the Government to align salary increases (6.2% for 2004 and projected inflation (CPIX) plus 0.4% in 2005 and 2006) with targeted inflation (MTBPS 2005).

3.4 Policy priorities have combined social and economic objectives. To reduce poverty, broaden economic participation and reduce the duality of the economy, spending has increased on social grant payments, salary progression of educators and police personnel, housing and municipal services, in addition to measures reducing the regulatory burden on businesses, and investment in skills and education. Recent initiatives include provision of HIV/AIDS drugs, child support, black economic empowerment, land reform, free basic utilities for poor households and a labour intensive public works programme.

Table 5. Fiscal development indicators

Source: IMF (2004)

3.5 There has been a steady lowering of inflation and short-term interest rates, and improvement in investment. In February 2000 government introduced the inflation-

99/00 00/01 01/02 02/03 03/04 04/05

National Deficit (% of GDP) -2 -2 -1.4 -1.1 -2.4 -3.1Expenditures (% of GDP) 26.2 25.6 26 25.4 26.9 27.7Revenues (% of GDP) 24.2 23.6 24.5 24.2 24.5 24.5National Government Debt (% of GDP) 46.5 43.6 42.7 37.1 37.3 37.8Domestic Debt (% of GDP) 43.4 40.1 34.6 30.7 31.5 32.1Foreign Debt (% of GDP) 3.1 3.5 8.1 6.5 5.8 5.7


119

targeting framework as a policy instrument for keeping inflation at a level conducive to economic growth and development. The South African Reserve Bank has kept CPIX, (headline inflation adjusted for interest rates on mortgages) within the official target range of 3-6 percent since 2003, and short-term interest rates have been falling since 2002. The improvement in macroeconomic performance has led to a continued decline in long-term interest rates.

Budget and public finance management

3.6 Folscher and Cole (2004, p.109) provide a succinct summary of the budget management process inherited by the post-apartheid government:

“In the 1992/3 fiscal year, the main budget net borrowing requirement had reached 8.7% of GDP and in the 1994/95 fiscal year public debt rose to almost 47% (a level of approximately 30 per cent, ten years earlier) leaving very little fiscal room for the state to improve the equity of public services. The annual budgeting system the new government inherited provide inadequate tools with which to stabilise fiscal balances and manage the required policy shifts. It was highly fragmented, not only in terms of a de-linking of policy, budgeting and implementation, but also institutionally, increasing budgeting uncertainty, lack of clarity and the scope for budget games. It planned and controlled for inputs and cash, with limited opportunity for systematic assessment of the effectiveness and efficiency of spending, or for relating allocations directly to policy. It was not transparent, with poor underlying information systems, hidden spending and inadequate mechanisms to extract good information for use in the budget process and for accountability purposes. The budget process itself was largely incremental, offering insufficient opportunity for the new government to identify ongoing non-priority activities and create fiscal room for higher priorities. Accountability was procedural, and the system was plagues by deeply entrenched inefficiencies.”

3.7 The South African Budget process that has now been developed has both top-down and bottom-up elements. The MTEF sets the fiscal framework but allocation of resources to programmes is based not only on political priorities but also on the translation of these into detailed national plans capable of implementation. The whole process is based on the following principles:

• Comprehensiveness and integration: the budget framework co-ordinates, integrates and disciplines policy and the budget process.

• Political oversight and a focus on policy priorities: choices between priorities are political in the final instance. The system recognises this and structures the integration of political and administrative practices to ensure that funding choices align with priorities and that spending is subject to political oversight.

• Using information efficiently: the process ensures that public information is provided transparently, accurately and on time, to improve public policy and accountability.

• Changing behaviour by changing incentives: spending department have the responsibility for setting out and justifying their own spending plans at a programme level and for getting value for money from the funds within their approved ceilings. The NT has the responsibility of challenging these bids – not to reject them but to encourage spending ministries to thoroughly review and improve their activities and plans.

• Ensuring budget stability and predictability while facilitating change at the margin.


120

3.8 The MTEF and the budget processes give a degree of certainty and policy predictability through the use of rolling baselines, although macro-economic exigencies may mean that they are subject to marginal change. This means that line Ministries need to monitor and control their spending to ensure that priorities are not at risk. Equally the NT has the responsibility to challenge spending plans to keep an eye on allocative efficiency. For example, relative price movements may mean that outputs in some areas can be met more cheaply than in the past and integrated planning may mean reallocation of funds. These principles are of course ideals to be attained in any system, but they are well understood in the NT and systems are in place for their implementation.

3.9 Under the constitution, the economic management of South Africa is the responsibility of the Executive. The Legislature must pass any necessary legislature related to economic policy such as some forms of deregulation but its primary role is scrutiny of expenditure. Provincial and local government has a role in its expenditure and limited revenue raising powers. Within the Executive, the NT has the lead on Economic Management, setting parameters within which other Departments operate and implementing policy in the field of macroeconomics, planning and budgeting and financial management.

3.10 The DPE is responsible for translating broad policy objectives into specific policies for enhancing the management of state-owned shareholder enterprises, including restructuring to promote economic efficiency. All policies are cleared through the Cabinet37 where they must have the consent of sectoral Ministers in the social as well as economic clusters. Along with the National Treasury and the Department of Public Enterprises the line Ministries are responsible for implementing sector economic policies within the framework of overall economic policy.

3.11 The role of the legislature was enhanced by the 1999 Public Finance Management Act (PFMA) The Act was introduced to improve financial management and to make sure that managers are held accountable for decisions they make. The most important innovation introduced by the Act was the requirement that managers be held accountable not only for financial control, but also for the production of outputs for which that they are responsible.

3.12 In the period leading up to the budget speech in February each year, the Ministry of Finance encourages submissions on what the budget should address. This improves participation in the economic management process. Although the general public participates in open meetings called Imbizos aimed at facilitating the signalling of the priorities of the people, most do not participate in Parliamentary hearings. Participation at consultative meetings at national and provincial governments is less proactive compared to those at local government level, where the legislative requirements for public participation are more prescriptive.

37 Managed by the Cabinet Office, part of the Office of the President


121

Summing up

3.13 South Africa has made progress in economic management since 1994 that has been described by the IMF (2005) as “remarkable.” Economic growth has been faster, inflation has been lower and more predictable, public finances have been stronger and the external position has improved. The overall commitments to economic management envisaged in GEAR have been institutionalised through the constitutional arrangements and the public finance system enshrined in the 1999 Public Finance Management Act. This has provided a sound overall framework for improving growth and service delivery, though immense challenges remain in overcoming the legacies of poverty and disadvantage that most South Africans face.


122

4. Results of Technical Cooperation: organisational capacity for economic management

4.1 This section provides an overview of the roles in the economic management process of the organisations for which DFID TC support has been evaluated, a summary of the support provided, evidence on changes in organisational capacity, and an assessment of the contribution of DFID TC to these changes.

The National Treasury


4.2 The National Treasury (NT) has responsibility for preparing policy guidelines for other departments on the interaction between the public and the private sector, for example through general principles of tariffs for utilities, subsidies, public sector grants and subsidies, public private partnership and the balance between revenue raising and other economic policy issues such as trade policy. It also plays a key role in the implementation of the Growth Employment and Redistribution Strategy (GEAR).

4.3 Since 1996 The National Treasury has developed into a coherent Finance Ministry with structures and processes based on the Whitehall model. It is not a carbon copy of the British Treasury but reflects national traditions, culture and the constitution.

4.4 There is now a coherent system for managing the process of translating political objectives into a coherent expenditure plan owned by line ministries. Nevertheless, there is general recognition of the need to develop this system and back it up with support for more detailed planning and implementation of service delivery. The commonality with other mainly developed38 Commonwealth Countries is reflected in the annual budget process, which is geared towards the allocation of resources within a fiscal envelope derived from macroeconomic forecasts and close monitoring of spend by line Ministries. Decisions are taken at the central political level – the Cabinet and its various sub committees – but supported by solid ground work within the civil service. The details of the process reflect the South African Constitution under which the three spheres of government co-operate and the regional and local spheres are integrated into the budget process not only technically but also through direct involvement at the political level.

Summary of DFID support: SERP I and SERP II

4.5 DFID supported GEAR through two consecutive programmes SERP I and SERP II, which had the objective of ensuring that public resources were allocated efficiently and equitably, with a focus on planning, budgeting and financial management. The programme developed from the secondment of a senior staff member of HM

38 But also for example Uganda.


123

Treasury in 1997, funded by DFID to assist the Budget Office, which had responsibility for public expenditure planning, macro-economic policy and inter-governmental relations.39 The overall design of the support was based around the long-term secondment supported by a “Know How Fund” providing additional resources to be used in a demand-determined way.

4.6 In November 1998 it was decided to incorporate this funding into a new programme to support economic reform – SERP. The first SERP programme ran from 1997 to 2001 when the provision of £865,000 had been spent. The highly favourable evaluation of SERP I (for instance in the 2000 DFID Output to Purpose Review) led to a decision to fund SERP II, which provided a further £1,380,000 of support for the National Treasury.

4.7 SERP II was modelled on SERP I and drawn up in close consultation with the GSA. This programme had a total budget of £2.6m running from January 2001 to December 2003. The following considerations influenced the design:

• The NT still had a shortfall of capacity in skilled professional staff, for whom it was facing increasing competition from the private sector.

• Work remained to be done on the transformation of government financial management and budgetary reform.

• Improving municipal government financial management was critical to the success of the newly-demarcated local authorities.

• There was growing pressure for restructuring of key economic infrastructure utilities and agencies.

• Improved service delivery of government anti-poverty programmes had a much higher political profile under the Mbeki Presidency.

• While NT could arguably afford to pay for this itself, DFID argued that much of the relevant UK expertise (e.g. HMT secondments) was not ‘on the market’.

4.8 The programme was extended to include the Department of Public Enterprises and the Department of Trade and Industry because together with the National Treasury they formed the key economic departments within the Economic Departmental Cluster, responsible for implementing Government’s economic reform programme and central to the fight against poverty. This extension was partially in response to requests by the relevant Departments who had recognised the useful role played by SERP I in the National Treasury.

4.9 The DFID support has comprised the following elements (with management shared between the NT, DFID and a Managing Consultant):

• Three staff seconded successively from HMT. The first worked at all levels in the National Treasury helping to design and implement the current public expenditure planning system, drawing on his wide experience within the HMT. The second

39 It is understood that the origins of this secondment lay in informal discussions between the UK Chancellor

of the Exchequer and his South African counterpart at the Commonwealth Finance Ministers’ meeting and World Bank and IMF annual meetings.


124

worked within the Budget division, helping to consolidating the work of the first. The third worked in a number of roles in different sections of the Treasury.

• “Know How Funds” for both phases of the project which were used to fund a wide range of activities. These included: two ODI fellows in line positions within the NT; short term consultancies both local and international; placements for NT officers in UK institutions and support for development of institutional links; short-term HMG secondments; and short training courses and studies. The focus was on economic policy issues rather than on administrative capacity building, although this had been one of the programme objectives. Support was provided across the whole range of National Treasury work, for example, short but intensive inputs into the development and operation of the Public Private Partnership Unit provided by Partnerships UK (PUK), training in trade policy for dedicated Treasury officials, the macro-economic impact of HIV/AIDS, and funds for attendance at relevant international course and conferences.

• Noting the lack of a baseline assessment of capacity in SERP I, SERP II included a budget of £10,000 to undertake initial capacity assessments and to put in place structured programmes of capacity development in the National Treasury and the Department of Public Enterprises. This element was not however implemented, as it was not seen as a priority.

• A fund of £100,000 was allocated to facilitate the sharing of South African experience in economic reform with other governments. A contribution to the creation of the Southern African Tax Institute (SATI) was the only project funded from the Regional Learning pool.40 However some other SERP II supported activities have begun to spin off into regional initiatives, for example the regional interest shown in the PPP Unit’s work. South Africa has also organised its own regional Budget Reform Seminar with the assistance of one of the ODI fellows funded through SERP.


4.10 The South African National Treasury is now regarded as an effective organisation with a high calibre of staff, generally strong management and a clear strategic direction over the period covered by the evaluation. This has been demonstrated in its capacity to carry through the ambitious reform agenda embodied in the 1999 Public Finance Management Act including the central role of the Medium Term Expenditure Framework (introduced in 1998/9) as part of the process of establishing a coherent top-down multi-year budgeting system. National Treasury outputs have improved considerably since the mid-nineties. Advice is provided to Ministers in a way in which they can use it for coherent decision-making. The Annual Supply Estimates now presented to Parliament clearly put the request for funds in the context of outturn in recent years and present future plans not only in terms of

40 Bureaucratic disputes stalled the original idea of a SADC Tax Institute. A self-appointed group, including

a representative from Harvard University, went ahead to create an institute. Harvard contributed $50,000 to start-up costs. SATI hosted a series of courses over a two-month period in the middle of 2002. Seventy-two participants attended the courses, 24 of them South African. The foreign participants were mainly from SADC. The teaching was conducted at the University of Pretoria. SERP II financed the airfares and subsistence costs.


125

finance and its source but also in terms of measures of activity and output in terms of service delivery (Folscher and Cole, 2004).

4.11 Examples of the way in which organisational capacity has developed over the period includes the work of the PPPU in conducting several sectoral workshops and training 800 civil servants within South Africa.41 The Unit also has a recognised international profile as a source of expertise in the Region. For example, Mauritius has approached the NT for assistance; the Unit’s Head has also spent a week in Pakistan teaching a new PPP course at the National Institute for Public Administration. The NT has also played a leading role in CABRI, which has effectively helped to create a professional planning and budgeting association in southern Africa.

4.12 The increasing managerial competence of the NT was reflected in the change in SERP management processes between phases I and II, when they prioritised their needs and effectively managed the overall budget, using the managing consultant effectively to provide support services.

4.13 The current challenge is to ensure that NT officials have equally competent counterparts in line Ministries, to ensure that sectoral interests are fully reflected in the planning and budgeting process.

Contribution of DFID Technical Co-operation

4.14 SERP I made a significant contribution to improving the performance of the Budget Office through its contribution to the strengthening of the budget management system. This arose from the high calibre of the support provided and the fact that it brought access to firsthand experience of the UK’s public expenditure reforms that were highly relevant to the public finance reform agenda that the South African government was pursuing. Its success lay in providing the right support in highly propitious circumstances.

4.15 SERP II was also highly relevant to South Africa’s needs: it sought to support the Treasury’s implementation of the Government of South Africa’s economic policy documents, in particular the 1996 Growth, Employment and Redistribution Strategy (GEAR), was designed in direct response to requests from the Government, and was sufficiently flexible in implementing instruments to be considered a programme rather than simply a set of projects. The intention was to take forward the planning and budgeting system established under SERP I, by building policy-making and other administrative capacity in operating the system for integrating political and administrative practices to ensure that funding choices aligned with the priorities of government.

4.16 However, no baseline capacity assessment was made, and the programmes had neither specific capacity development targets nor a monitoring framework focusing on measuring capacity. This is likely to have reduced the capacity development

41 The contribution of DFID support in this area is described below.


126

impact. This lack of explicit attention to capacity issues may have resulted from a failure to recognise the special circumstances that had made SERP I particularly effective and the need for a rather different approach, particularly to the role of staff seconded from HMT, in following up this initial success. The project completion report saw SERP II as having “two largely incompatible objectives: to build capacity; and to finance activities at short notice that the government was unwilling (because of risk) or unable to fund out of its own resources.” The ad hoc rather than strategic approach to capacity building could be seen as inherent from the start: half of the funds for the NT had been allocated at the time of drafting the project memorandum, despite the lack of a capacity baseline. What the design failed to take into account was that whereas SERP I had provided the necessary impetus to the development of the NT, SERP II was necessary to bring it to maturity. Not simply a strategy for capacity development but a strategy for TC withdrawal was needed. The Mid Term Review (MTR) recommended that the final year of SERP II be directed at conducting a baseline capacity assessment within the National Treasury and to supporting the development of a comprehensive, integrated capacity building programme. This was not done.

4.17 Similarly the lack of an analytical monitoring system was a serious management flaw. Although funds were set aside for evaluation, the routine reports do not provide a running analysis of expenditure which would aid evaluation by showing purpose and types of project; for example whether funds were spent on training, research or specific consultancy advice. The information is presented in the form of a list of contracts rather than an analysis of activities. Such an analysis would have needed a simple classification system that ideally would have been derived from the capacity needs assessment and strategic plan.

4.18 The impact of the main components of the support may be summarised as follows based on assessments from the NT and project documentation:

• The first seconded HMT staff member played an important role in helping to introduce a budgeting system based on the concepts of the MTEF as an instrument for funding service provision. The other secondments had rather less impact, reflecting a lack of clarity about the terms of reference for their roles and a difficulty in matching the skills of the candidates available from HMT (particularly those more focused on economics than the policy-making process) to the particular requirements of the NT.

• Two ODI fellows were financed from the Know How Fund and were employed in line positions. These were regarded as highly successful by NT staff and as contributing significantly to the work of the Treasury. Their terms of reference focused on the performance of operational roles and did not primarily have a capacity development function (reflecting the intentions of the Fellowship Scheme and the relatively inexperience of ODI Fellows), though there was some opportunity for skills transfer. As with other aspects of the DFID support to the NT, the value of the scheme lay in the ability to bypass normal recruitment procedures to access skilled and highly motivated staff from an international pool. In this case the rigour of the selection process was felt by NT staff to be a reason for the success, while the ODI Fellows considered that the management environment in the NT enabled them to contribute effectively.


127

• The Know-How Funds had made a real contribution to the work of the National Treasury, providing access to directly relevant consultancies and expertise in a way that would not have been practical using normal domestic budget sources. Specifically from the perspective of the recipient it provided flexible, rapid, additional funding that could be used to recruit high calibre and expensive international expertise that would have been difficult (and politically contentious) to contract using domestic sources. The support to the Public Private Partnership Unit (PPPU) using PUK was a good example of a successful use of highly specialised TC (used strategically), which probably could not have been sourced without the Know How Fund.42

The Department of Public Enterprises

The role of the Department of Public Enterprises

4.19 The Department of Public Enterprises was created in 1999 to accelerate the restructuring of State-Owned Enterprises (SOEs). It was the successor to the Office of Public Enterprises, which had been set up on a temporary basis. DPE's role is to accelerate the restructuring of South African SOEs in an integrated and coherent manner, to promote economic growth and socio-economic development. The primary aim of the Restructuring Programme is to redress the imbalances accentuated by the legacy of apartheid – not just racial, but also gender.

4.20 There have been two approaches to the restructuring of state assets:

• Outright privatisation e.g. Aventura resorts and Telkom, and

• Deregulation of markets within which some of the SOEs operate, accompanied by a requirement to pay tax and dividends to the State e.g. ESKOM, Airports Company of South Africa and the Central Energy Fund.

4.21 The DPE is responsible for only six state-owned enterprises i.e. Transnet, Eskom, Denel, Alexkor, Aventura and Safcol.

4.22 Public enterprises are of major importance in the South African economy, dominating the energy, transport and telecommunications sectors. Enterprises of this kind proliferated under the apartheid regime as a means of protecting strategic areas of the economy in the face of sanctions. The SOEs reporting to the DPE have combined assets of £15.3bn and turnover of £7.3bn, and employ 136,000 people, about 1.2% of formal sector employment. Restructuring these enterprises, intended to involve both whole or partial privatisation of state assets and a transformation of

42 The advice was “transactional” to the extent that it was highly specialised and limited to one off highly

specialised transactions, but had spin-off benefits domestically and in the region and to that extent has been transformational in that the PPPU now runs its own training courses not only in South Africa but abroad. The PPPU developed sufficient capacity in terms of the ability to handle new issues when they come along, knowing where to look and how to manage hired expertise. Thus despite the high cost the support to the PPPU was not only relevant and effective but also efficient and had both immediate and sustainable impact.


128

ownership and management arrangements, was identified by the Government of South Africa (GSA) as a key factor in promoting growth.

4.23 The infrastructure utilities were still vertically integrated near-monopolies, offering customers, and particularly the poor, little choice. Moreover, utilities designed for 6 million people now had to serve 42 million. Restructuring was intended to reshape and regulate these enterprises so that poor and low-income citizens could benefit from competition through lower prices and new investment. Nevertheless, the restructuring policy remained controversial particularly as organised labour was opposed to even partial privatisation of basic services.

4.24 In August 2000 DPE produced a “Policy Framework: An Accelerated Agenda Towards the Restructuring of State Owned Enterprises”. This stated that the GSA’s objectives were to:

• Promote economic growth and development and improve service delivery through raising efficiency and opening up sectors to competition.

• Attract foreign direct and portfolio investment.

• Promote black economic empowerment and gender equity, and uphold the government’s development (especially social and environmental) obligations.

• Raise resources for priority social and poverty reducing programmes through the sale of state assets while preserving GSA control of essential infrastructure sectors.

• Improve corporate governance and ensure improved ethics and probity.

4.25 In support of this policy, DPE was particularly keen to access UK expertise and DFID’s increasing focus on growth issues (as subsequently reflected in the 2002 CSP) prompted DFID’s willingness to extend support. DPE hoped to use the SERP II Know-How Fund to access sector-specific HMG expertise to assist in overseeing one or more of the SOE restructuring programmes.

DFID support under SERP II and SRPESA

4.26 DFID supported DPE through both SERP II and the project for Support for Restructuring Public Enterprises South Africa (SRPESA). This reflected the emphasis in the 2002 CSP on “a more effective process for restructuring public enterprises”. Both the expansion of the DPE element of SERP II and SRPESA sought in part to build on DFID's earlier support for the privatisation of commercial forests in South Africa, a long-term programme which had mainly focused on ensuring that social and environmental as well as commercial aspects were taken into account; and had led to privatisation arrangements which preserved the community land rights over the forests of local communities, and aimed to provide some important social benefits. One of the intentions of SRPESA was to apply a similar approach across a wider spectrum state owned enterprises.43

43 As with SERP I it is understood that this approach reflected in a part an initiative coming out of informal

discussions at between UK and South African ministers in this case between the DFID Secretary of State and the DPE Minister in 2002.


129

4.27 SERP II included a Know How Fund to provide both short and long term inputs to assist DPE They had sought secondments analogous to those from the HMT provided to The National Treasury but DFID recognised that the recent UK reforms meant that much relevant expertise from HMG had left the public sector and those who were remaining might not be prepared to go to South Africa. In these circumstances, funds were not set aside for secondments but allocation of the £400,000 KHF was to be based on agreement between DPE and DFID priority activities during 2001. In the early years of SERP II, DPE’s work focused on the government’s privatisation programme. Privatisation is relatively new in South Africa and it was difficult to find appropriate local expertise.

4.28 Funds from the Know-How fund were available to finance a TCO. A suitable person could not be found but SERP II funded an ODI Fellow at DPE. In addition SERP II funded an ad hoc programme of support to the State Owned Enterprises Restructuring (SOER) programme, for example: a Review of 1997 Protocol on Corporate governance; advice on the Restructuring of the ports administration; an Energy specialist for 15-18 months to act as a link with another project; and a Public Relations expert to help calm the political-industrial situation when ports restructuring was announced.

4.29 The SERPII KHF for DPE was almost doubled during this period but by 2002, it was felt to suffer from a lack of overall strategic direction, and was replaced with a substantial new single project. Support for the Restructuring of Public Enterprises in South Africa (SRPESA). When SRPESA began, the only major donors to DPE, the principal beneficiary and key stakeholder of SRPESA, were DFID and USAID. Close working relationships already existed between DFID and USAID, and in September 2002, under SERP II, they had signed a MoU, which involved them in a joint strategy, with USAID’s consultants, to support DPE in its ports restructuring efforts. It was expected that such initiatives would be pursued under SRPESA, and extended to other donors as far as feasible and practicable. USAID did join the Project Steering Committee but otherwise interaction appears to have been limited.

4.30 SRPESA was intended to run from August 2002 for three years; work eventually started in January 2003, with an overall Budget of £6.8m. SERP had been managed by a generalist project management consultancy, but a firm well known for its role in privatisation was appointed to run SRPESA, under a different style of management contract: under SERP, the know how fund covered only the fees paid to consultants, under SRPESA the managing consultant received a mark up on consultancy fees as well as a management fee. A specific budget was also included for independent monitoring and evaluation (£0.3 million).

4.31 A key role of the project was to provide access to international expertise that was difficult to obtain under GSA procurement rules. The KHF provided technical assistance across a wide range of SOE restructuring activities. This was to be accompanied by capacity building, which was in principle central to the project design. It also recognised the complexity of government structure of responsibility for SOEs. Whilst DPE had lead responsibility for restructuring, line departments had policy responsibility for SOEs and shared responsibility for the design and implementation of restructuring programmes. The regulators such as the National


130

Electricity Regulator (NER) and the Independent Communications Authority for South Africa (ICASA) and a range of other restructuring stakeholders were also supported. The whole process was to be demand led. The project was to combine advancing the GSA’s SOE restructuring programme and developing capacity inside the machinery of government, through the appointment of sector coordinators and teams of “rolling advisers” who would make a series of visits to South Africa and be accessible to counterparts by phone and email when not in country.

4.32 SRPESA got under way swiftly, with an enthusiastic response from stakeholders who were quick to propose sub-projects. By November 2003 the rate of spend was noted to be on budget with 26% of the budget spent in the first 25% of the planned project duration. Ninety-five proposed interventions had been accepted, supporting thirteen different beneficiaries. This rapid start to disbursement was thought to reflect the demand for policy advice and need for capacity building within regulators, and was regarded as a positive sign.

4.33 However, a concern from DPE was that the management approach was too focused on the advocacy of privatisation and insufficiently facilitatory in its approach. As a result the original in-country coordinator was replaced in January 2004 with the new coordinator (moved from another project post). At this point there was a freeze on activities to allow DFID the opportunity to take stock and consider changes in approach. From 2003 there has been a greater emphasis on the use of local consultants, especially legal advisers: 10 local consultants were engaged in the first quarter, and by November 2003 53 instances had been recorded. In April 2005 there was a further increase in the involvement of local consultants.

4.34 In 2005 a new DPE Minister was appointed and began a process of reform and restructuring of the Department as described below. DFID seized the opportunity to address its own concerns about project governance. The management structure was transformed to give a stronger more focused role to a smaller high-level Strategic Oversight Committee (replacing a Steering Committee) with DPE taking the lead. In the later phases of the project funds have been used mainly (about 70%) to fund long-term consultants who have been playing a key role in management of the Department, while retaining some short-term resources to address issues identified at ministerial or DG level.

4.35 Initially the M&E effort was concerned principally with monitoring the modalities of programme management, but as the project gained momentum the orientation shifted increasingly to evaluation of the efficiency and impact of selected SRPESA interventions. The great majority of the recommendations made in the quarterly monitoring reports have been accepted, and the consultants have helped to provide continuity in the direction of the project.


4.36 Two issues have been of most significance over the evaluation period in relation to DPE’s organisational capacity. The first relates to DPE’s role and the policy framework within which it has operated. The second relates to internal management and staffing:


131

• When the project started, GSA had expressed interest in privatisation as the main way of restructuring. In practice, there was no cross-government commitment to privatisation as the main strategy. Different stakeholders had different assumptions and motives, for example in terms of the relative priorities for efficiency, modernisation, employment, social impact and subsidy. Not only was the lack of a clearly articulated policy exacerbated by conflicts within the department but the DPE lacked the administrative capacity to steer a way through these difficulties.

• The problem of staff turnover has been especially acute in DPE, where there have been three Director Generals in two years. This reflects both the private sector employment options for professional staff and what was seen as a difficult working environment partly related to the expectation that DPE would have a limited life to implement privatisation. The movement of staff generally has made it difficult for consultants to function effectively. Even with subsequent greater stability in staffing, DPE capacity has been hampered by the calibre of some of the staff. Standards both of educational attainment and of administrative skills are very variable.

4.37 In both these areas, significant progress was made following the appointment of a new DPE minister after the 2004 election. In relation to the policy framework, it was made more explicit that the focus would be on the restructuring process rather than the privatisation of key SOEs. He steered a shift in policy away from privatisation to creating a profit-driven state-owned sector, whilst not excluding the possibility of private sector equity participation in state-owned enterprises once they had been transformed. The intention is now to retain many major enterprises in public ownership, and to focus on effective exercise of the government shareholder function, rather than to dispose of these assets. However, no clear vision of the future shape of the remaining SOEs, and of the boundaries between the state and the private sector, has yet been articulated, and the general view is that many large policy decisions remain to be taken, for example on the distribution end of the electricity sector, and the ports and rail sectors. The restructuring process is expected to take a further five years to bring it to completion.

4.38 In relation to internal management, progress has been made as a result of management changes following the appointment of the new minister. A new DG has been recruited from within the civil service, and the acting Deputy DG has experience of public enterprises in both South Africa and Zimbabwe. Moreover, DPE is no longer seen as a shell department, which would complete restructuring and then close down.

4.39 In this context TC support provided under SRPESA is now seen by DPE management as important to the functioning of the Department, at least prior to a new departmental reorganisation plan and the appointment of new staff to new posts.

4.40 In relation to the capacity of DPE to fulfil its functions over this period, some successes can be identified:

• Restructuring of electricity supply has been achieved, with the market opened to competition.


132

• Under the new regulatory regime electricity prices are 3% lower than they would otherwise have been, and further reductions are planned.

• A planned 40% price increase by Spoornet (the rail operator) was held to the rate of inflation.

• Separation of the regulator/landlord and operator functions was achieved in the ports sector, and a real price decrease achieved.

• A regulated competitive market has been established in the mobile telephone segment of the telecommunications sector, and capacity strengthened in regulation, for instance for the electricity industry.

Contribution of DFID Funded TC

4.41 SRPESA was envisaged as providing support focused on the mission and policy of DPE, taking on board the complexity of government structure of responsibility for SOEs with DPE’s lead responsibility for restructuring being complemented by line departments’ policy responsibility for SOEs and shared responsibility for the design and implementation of restructuring programmes. Given the central role of SOEs in the South African economy, there can be little doubt of the relevance of the objectives of the programme to the GSA’s strategy for improving the management of economy. However, the capacity of the DPE to translate strategy into policies that could be implemented was weak, and this in turn weakened coherence of the programmes and the relevance of some activities. The DFID funded activities were closely aligned with the GSA’s stated policy objectives. In practice though these objectives did not command consensus across government. The policy focus has had to shift (but has been clarified in the process) since the 2004 Election in line with changes in the government’s thinking on ownership of public enterprises – highlighting the importance of an approach that is well-aligned to government priorities and has the flexibility to change as priorities changed.

4.42 The SRPESA Project Memorandum made clear that expertise was to be procured from local, regional or international sources as appropriate, but DPE has had a view that greater use of local expertise should be made including for example through partnerships between local and international consultancy firms which has been reflected in subsequent changes to the project approach.44

4.43 At the time of the evaluation, the project had nearly two years to run and had only recently changed direction. At present it is difficult to assess both immediate impact and its sustainability. It will remain difficult to evaluate the success of SRPESA in helping the GSA to deliver its objectives for the state owned enterprises until those objectives are clear.

44 However it was also noted by GSA officials that in some circumstance local consultants had

disadvantages: Locals may be too closely embedded into the local culture and politics, being reluctant to rock the boat; They may lack experience in the necessary fields, not representing value for money; They may be perceived by counterparts as seeking their jobs but at much higher rates of pay; Consultancy rates may distort the local labour market, making it difficult to recruit and retain good public sector staff, with a loss of institutional memory.


133

4.44 The project had an explicit objective of developing capacity. However, results are hard to identify. In some beneficiary organisations such as ICASA the consultancy interventions have resulted in the transfer of skills, and this can be linked with relatively high levels of staff retention. Elsewhere there is less persuasive evidence of skills transfer (for example, ESKOM – the electricity SOE – believe that the capacity of the regulator, NER, to engage in meaningful debate with them has declined since last year, and there has been some loss of key staff there). In line departments such as DPE high staff turnover has hampered skills transfer, and it is possible that it is only in the most recent phase of the project that interventions have been seriously directed towards this aspect of capacity building.

4.45 The staff retention problem was recognised in the design of the project but insufficient attention seems to have been paid to systematically identifying the causes and proposing strategies to mitigate it. The capacity development objective remained in the picture, for example specialist consultants involved in the delivery of sub-projects in the electricity sector confirmed that the emphasis on capacity building was made clear to them from the outset, and that success was achieved is shown by changes of behaviour amongst the regulator’s officials. However, a transactional function rather than a capacity development role of sector specialist consultants was more evident. Thus the early stage of the programme was relatively ineffective in building capacity. This problem was recognised in the recasting of the programme during 2005 when it was agreed that the new long-term specialist advisors would work with newly recruited Chief Directors in DPE over a period of a year, to transfer skills and assist in policy development.

4.46 Across SRPESA, where capacity development has been addressed it has been seen primarily in terms of the transfer of skills to individuals, either through formal means (workshops, etc) or more informally through on-the-job support and coaching. Little attention was paid to other organisational aspects of capacity development. Whether or not improvements in capacity are sustainable will depend critically on attaining stability in staffing in line departments so that transferred skills are not immediately lost to the private sector. It will also depend on the existence in beneficiary organisations of robust systems and procedures that will preserve the institutional memory of “how things should be done”, which can otherwise be lost even with normal rates of attrition.

4.47 The project was designed to address social and environmental issues in a series of sub-projects, but GSA has been resistant to this approach as it regards the promotion and enforcement of policy in these matters as the business of public bodies other than DPE.

4.48 SRPESA might have been more successful, if as requested by GSA, it had it been able to provide secondment of UK officials who had been engaged in the UK privatisation programme. Such TC might have been able to provide the kind of advice implementing policy objectives traditionally provided by Whitehall civil servants. This was not practical. Such civil servants had moved on either to other policy fields or to the private sector. DFID tried to bridge the gap with long and short term TC, managed by a firm experienced in privatisation. As discussed above, this had patchy success in the early days and led to problems between some of the


134

consultancy staff and DPE. The greatest successes were indeed seen where TC providers had worked within a UK public sector enterprise during the privatisation process (for example in the electricity sector) and where the TC brought very specific technical advice of a kind that was informed by UK experience (for instance in the case of public/private partnerships). GSA continues to source such expertise using its own funds.


135

5. Results of the Technical Cooperation: summary of the evaluation results

5.1 This chapter summarises the evaluation findings in terms of the relevance, effectiveness, efficiency, impact and sustainability of the activities reviewed.

Relevance

5.2 In terms of goals and objectives both SERP and SRPESA were relevant to GSA needs and ownership of the programmes was strong in the organisations supported. Both the NT and DPE have critical roles to play in South Africa’s GEAR programme. As South Africa emerged fully into the global economy after decades of isolation, there were major needs for capacity building not only on the concepts and tools of economic management but also in the systems and practices of public administration.

5.3 Both programmes (but more particularly support to DPE) suffered, viewed in relation to their capacity development objectives, from insufficient attention to the organisational context within which they were implemented. Uncertainty about DPE’s mandate (and the level of commitment across government to its stated policy approach) was the main problem. To some extent the initial success of SERP I may have contributed to this neglect of institutional and organisational issues. However the flexibility of design, and the willingness of all parties to overcome problems encountered reduced the negative impact of these initial design weaknesses.

Effectiveness

5.4 The projects have generally been effective with SERP I making an important contribution to establishing a modern system for public expenditure management despite some difficulties under SERP II and SRPESA. Ownership has been strong and this has been reflected in both design and management. A common factor reducing both effectiveness and efficiency in both projects was the lack of a systematic approach to capacity building in the South African public service. This could have led to a more coherently sequenced approach which might have resulted in a different emphasis in early sub-projects, and the addressing of organisational and cross cutting issues rather than simply the transfer of skills to individuals.

5.5 In relation to SRPESA the project management consultants aimed for rapid implementation, as they were tasked to do, and achieved a high rate of disbursement very quickly, though at the cost of establishing an overarching strategic perspective and at some cost in terms of DPE ownership. This was subsequently addressed through staffing changes and project redesign. In retrospect, a project design which had a longer timeframe and a slower rate of spend, especially in the early stages, would have been more effective, especially in meeting the project’s capacity development objectives. The initial approach of “rolling” rather than long-term advisers was intended to reduce the risk of dependency on consultants as “doers”


136

rather than advisers. Whether it was more or less effective in developing counterpart capacity than a long-term resident adviser would have been is difficult to judge. It may though have provided the best option where highly specialised inputs were needed across a range of topics. Continuity was reinforced by the role of the sector co-ordinators, who although not resident made frequent visits and helped to consolidate the long-term relationship. In some cases, this relationship has continued after the SRPESA funding has ended, for example in the electricity sector, where it is worth noting that the consultant had worked in the UK electricity supply industry both before and after privatisation and had direct hands on experience of the restructuring process. In relation to the role of external rather than local consultants, it was also suggested from government staff that the former are likely to be perceived by officials as more independent.

5.6 DFID has expressed concerns about the limited number of activities specifically addressing social (including gender) and environmental dimensions within SRPESA. The GSA’s position is that such projects are not necessary because all issues such as gender equity, environmental protection and black economic empowerment (BEE) are all covered by laws and policies that are monitored and enforced by other public bodies, so it would be inappropriate and unnecessary for DPE to take the lead in these issues through dedicated sub-projects. However, the key issue is not whether such cross cutting projects are needed but how far in practice, DPE has the capacity to enforce compliance in its own activities.

Efficiency

5.7 Overall outputs for the NT were provided on time and within budget. In support to DPE some initial projects were not continued following the review of the project’s management in 2004. In evaluating projects it is important to distinguish between the efficiency of project management, i.e. the delivery of inputs, and the efficiency of the project interventions in achieving the desired outputs and outcomes. On the whole SERP II was efficiently managed in terms of delivering the inputs requested by the Departments it was supporting but less so in prioritising the spending of funds or in systematically evaluating outputs and outcomes.

5.8 Similar considerations apply to SRPESA. A more measured and strategic approach at the beginning of the project might have resulted in tighter linkages between the sub-projects and the project log frame. Closer scrutiny by the Steering Committee and more proactive ownership of the project by DPE on behalf of GSA might have resulted in some differences in the portfolio of sub-projects put forward and/ or accepted. Mainly because of the re-orientation of the project in 2004, a number of sub-projects were closed prematurely. Although not all of the inputs to these were lost, the M&E consultants estimate that something over £250,000 will have been wasted. The project got under way promptly, with rapid mobilisation of a number of sub-projects. Arguably a design that allowed for a slower start and more careful investigation of capacity issues might have been more effective in the longer term.


137

Impact

5.9 SERP I had an important impact in helping the process of establishing and implementing new budget systems. The National Treasury’s strong management and staffing capacity has enabled this contribution to be transformational and to be sustained. The impact of SERP II is less clear but the initial work with DPE was considered sufficiently useful to have generated a doubling of the Know How Fund and a spin off into SRPESA. So far the impact of SRPESA can be detected in some of the achievements of DPE in influencing pricing and regulation, but its ultimate impact will in part depend on whether it can help the GSA deliver clear objectives for state owned enterprises.

Sustainability

5.10 SERP could be said to be sustainable because some of its activities have continued even after donor funds have gone. The NT has set aside some of its own funds for advice from PUK on Public Private Partnerships to the PPP Unit. More generally, NT reports that there is a growing local capacity for consultancy support in specialised areas. Even in DTI, where work on two sub-projects ended prematurely, the Sector Wide Employment, Enterprise and Equity Programme (SWEEP) has been produced and the EU is interested in providing significant funding for its implementation. There are also signs that the knowledge gained will largely be retained, particularly in the NT which has, for example, launched a PPP manual, a sign that knowledge gained from consultants is being retained within the Department. However, how much of this can be attributed directly to SERP is difficult to establish.

5.11 Whether or not improvements in capacity attributable to SRPESA are sustainable will depend critically on attaining stability in staffing in line departments so that transferred skills are not immediately lost to the private sector. It will also depend on whether beneficiary organisations have robust systems and procedures that will preserve the institutional memory of “how things should be done”, which can otherwise be lost even with reasonable rates of staff retention.


138


6.1 In the cases reviewed, the role of DFID TC support has been to respond to a demand from the organisations supported, first, to access knowledge about UK government reform experience which was seen as highly relevant to South Africa’s needs, and second to use the support provided as a way of sourcing expertise that circumvented particular political and administrative constraints on using normal government channels. The approach to providing the support reflected the generally responsive and demand led nature of DFID’s TC in South Africa, within an area (improving economic management and boosting growth) that was seen as a priority in DFID’s country strategy. A wider judgement about the appropriateness of providing such support to a middle income country (or to providing this form of support rather than alternatives as a means to achieve poverty reduction) is beyond the scope of this evaluation.

6.2 The activities in support of the NT have generally been successful as a result of the NT’s clear mandate and policy direction and effective management that have enabled it to articulate and manage its needs for external support. However their impact could have been increased by more attention to the selection of some of the TC providers (in matching their skills to what was demanded), as well as a more explicit capacity development focus for the activities. The effectiveness of the support to DPE was initially somewhat compromised as a result of an overestimation of the South African commitment to privatisation as a policy tool, management and staffing weaknesses within DPE and a mismatch between the approach of the programme management consultants and DPE’s requirements. These issues have been addressed through changes to staffing and project restructuring and subsequently through measures by a new Minister to clarify the Department’s mandate and approach and to improve management.

6.3 The case study provided a range of types of TC – both long- and short-term, and national and international. Success has tended to be greatest where it has been:

• Long term with a clear remit for developing capacity through:

1. Focusing on the development of organisational systems and processes, and

2. Mentoring counterpart staff.

• Short term with a clear remit for interacting with the recipient and:

1. Where the output is intentionally transactional, and

3. Managed by the recipient and providing highly technical strategic inputs of a kind not necessarily required by the public sector in the long term.

6.4 The shift in focus of SRPESA from short term highly technical experts to long term advisers has been successful where the long term experts have been known to the host organisation through short term inputs and the host organisation has led the selection process. Both projects demonstrated the importance of the relationship between the host organisation and counterparts and the TC, for both long and short-term consultants, highlighting the potential role for the retention of long term TC with intermittent input, either at a distance, by phone and email, or in country.


139

6.5 Not all of the TC provided was appropriately skilled to maximise the organisational capacity impact that it could provide. Effectiveness might have been improved if the profile and experience of TC providers had been better suited to the needs of the organisations supported. To some extent this could have been achieved through more rigorous selection. However, the objective of sourcing directly relevant expertise from within the UK government (a key objective for the organisations supported) was not in fact achievable through the model of long-term TC because appropriate staff were not available or in a position to undertake secondments.

6.6 Insufficient attention was paid in the design of the support to capacity issues. This may have resulted from an overestimation of organisational capacity resulting from the initial success of SERP I and was reflected in the lack of clear targets for organisational capacity development and the failure to carry out baseline capacity assessments. This analysis should have taken account not only of the capacity of individuals, and of the GSA to recruit and retain good staff, but also of organisational systems, procedures and guidelines. Had this been done, it is possible that some of the constraints that militated against the effectiveness of earlier support to DPE in particular might have been addressed more directly.

6.7 The main implication is therefore the need (in line with South Africa’s policy on the use of TC) for a more rigorous assessment of the organisational context for organisations receiving support. For organisations that are less strong than the NT, this is likely to suggest that attention to issues about the role and mandate of the organisation, its internal management, and capacity to attract and retain appropriate staff, will need to be addressed for TC to be used in a way that achieves a transformational impact.


140

References

Bloch, G., M. Favis and J. Hargovan (2000), Development Cooperation Report: Evaluation of ODA to Capacity Building, Report for International Development Cooperation, Department of Finance.

DPSA (2003), Policy Framework and Procedural Guidelines for The Management of Official Development Assistance, Department of Public Service and Administration, 2003 Edition.

DFID (1998), South Africa Country Strategy Paper.

DFID (2004) Support to Economic Reform Project, Second Phase (SERP), II David Pedley, Economic Adviser, DFIDSA.

Folscher, A., and N. Cole (2004), South Africa: Transition to democracy offers opportunity for whole system reform, CABRI Budget Reform Seminar Country Case Studies, Pretoria.

Harvey, C., and J. Lynelle (2002), Mid-Term Review of Support to Economic Reform Project second phase (SERP II), DFID.

Human Science Research Council (HSRC), 2004, Fact Sheet: Poverty in South Africa, July 2004.

IMF (2004), Staff Report for the 2004 Article IV Consultation, August.

IMF (2005), Staff Report for the 2005 Article IV Consultation, September.

IPFA (2001), Training Framework for Government Finance Staff, www.treasury.gov.za08/08/05.

MDI (2003), Millennium Development Indicators.

MTBPS (2005), Medium Term Budget Policy Statement, 2005.

Nathan, L., (2004), Accounting for South Africa’s Successful Transition to Democracy, Crisis States, Development Research Centre, London School of Economics.

National Treasury (2001) Budget Review, Republic of South Africa.





National Treasury (2005) National Treasury Strategic Plan 2005-2008.


141

National Treasury (2005), Estimates of National Expenditure 2005, Trade and Industry.

NEDLAC (2005) Overview; http://www.nedlac.org.za/, downloaded on 05 August 2005.

Republic of South Africa (1996) The Constitution of South Africa Act no. 108 of 1996.

Republic of South Africa (1999) The Public Finance Management Act no. 1 of 1999, as amended by Act no. 29 of 1999.

Republic of South Africa (2001) Public Service Regulations as amended by Government Notice No. R. 785 of 28 June 2004 with effect from 1 July 2004.

Republic of South Africa (2002) PFMA Regulations as amended in June 2004.

Republic of South Africa (2003) Senior Management Service: Public Service Handbook.

Republic of South Africa (2004), Broad Based Black Economic Empowerment Bill. Simeka/SMC (2000), Development Cooperation Report: Report on Evaluations of Donor Assistance to the Republic of South Africa, 1994-9.

South African Reserve Bank (2004), Annual Economic Report 2004.

World Bank Development Indicators (WDI), 2005.

World Bank (1999), South Africa Country Assistance Strategy, May 1999.


142



Neil Cole Chief Director: Expenditure Planning Unit, Budget Office

National Treasury

Andrew Donaldson

Deputy Director General: Public Finance

National Treasury

William Dachs Head: Public Private Partnership Unit, Budget Office

National Treasury

Jonathan Dixon Chief Director: Financial Sector Policy National Treasury

Estelle van Eetveldt

Director: Budget Office National Treasury

Japie Jacobs Director: Financial and Administrative Services, Public Finance

National Treasury

Ilsa Karg Director: Economic Services National Treasury

Christopher Leowald

Acting Deputy Director General: Budget Office

National Treasury

Leslie Magagula Chief Director: Economic Services National Treasury

Shaheed Rajie Head: International Development Cooperation

National Treasury

Brenda Swart Deputy Director, Financial and Administrative Services

National Treasury

Michelle Schoch ODI Fellow, Expenditure Planning Unit, Budget Office

National Treasury

Peter Hlapolosa Independent Communications Authority of South Africa

Slingsby Wonga Mda

Chief Director: Business Operations Department of Public Enterprises

Hugh Dickson Adviser Department of Public Enterprises

DeAnneFriedholm

US Financed Resident Adviser, Public Finance

John Heath Heath Consultants, Ripon UK

Professor Chris Tapscott

Faculty of Economic and Management Sciences

University of the Western Cape

David Smith Independent consultant SRPESA evaluation team

Lara Kinley Adam Smith Institute

David Pedley Economic Adviser DFID South Africa

Lelanie Swart Deputy Programme Manager DFID South Africa

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

143

Su

mm

ary

of

Fin

din

gs –

Evalu

ati

on

Ma

tric

es

Re

leva

nc

e

Th

e N

ati

on

al

Tre

asu

ry

Th

e D

ep

art

men

t o

f P

ub

lic E

nte

rpri

ses

Rele

van

ce t

o

Org

an

isati

on

al

need

s &

O

pera

tio

nal

co

nte

xt

SE

RP

I, p

art

icula

rly

second

ed s

taff

, right pro

ject

at th

e r

ight

time. S

ER

P2 r

ele

vant b

ut

more

of

the s

am

e i.

e. d

id n

ot ta

ke

on b

oard

ca

pacity

built

und

er

SE

RP

I a

nd th

e d

iffe

ring

requir

em

ents

for

the f

utu

re.

SE

RP

II

an

d S

RP

ES

A p

rovid

ed larg

ely

tra

nsactiona

l support

ra

ther

than d

irectly

addre

ssin

g t

he o

rga

nis

ation

al pro

ble

ms o

f D

PE

an

d th

e a

ssocia

ted r

egula

tors

.

Capacity

was r

ecog

nis

ed

to

be lo

w a

nd a

rath

er

more

dir

ecte

d

appro

ach w

as r

eq

uir

ed th

an the

pro

visio

n o

f te

chnic

al expert

s.

Co

nsis

ten

cy

of

DF

ID’s

an

d

org

an

isati

on

’s

ob

jecti

ves

SE

RP

I a

nd

II h

ighly

consis

tent

with

DF

ID's

an

d th

e N

atio

nal

Tre

asury

’s o

bje

ctive

s.

SR

PE

SA

was in

ten

ded

to b

e c

onsis

tent

with

GS

A’s

sta

ted p

olic

y fo

r th

e f

utu

re o

f sta

te e

nte

rpri

ses b

ut

this

was a

polit

ically

charg

ed issue a

nd in a

sta

te o

f flux.

Follo

win

g a

chan

ge in M

inis

ter

the p

roje

ct to

ok a

ne

w d

irection

and is m

ore

cle

arl

y dri

ven b

y D

PE

, altho

ugh p

olic

ies h

ave

not

yet

bee

n d

evelo

pe

d f

or

all

SO

E s

ecto

rs

Co

nsis

ten

cy

wit

h a

cti

on

s

by G

SA

an

d

oth

er

Develo

pm

en

t P

art

ners

Consis

tent

with

actio

ns o

f G

SA

and o

ther

do

nors

D

espite initia

l inte

ntions, th

e la

ck o

f cle

ar

po

licy

focus in D

PE

m

eant th

at th

e G

SA

only

to

ok a

firm

lead o

n t

he p

roje

ct aft

er

the

appo

intm

ent of

a n

ew

Min

iste

r i.e

. duri

ng

SR

PE

SA

. .

DF

ID c

oord

inate

d c

losely

with

oth

er

don

ors

part

icu

larl

y U

SA

ID

to th

e e

xte

nt th

at U

SA

ID w

ere

repre

se

nte

d o

n t

he s

teering

com

mitt

ee.

Co

nsu

ltati

on

p

rocess t

o

en

su

rere

levan

ce?

Good c

onsu

ltatio

n w

ith G

SA

T

here

was g

oo

d c

onsu

ltatio

n w

ith th

e G

ove

rnm

ent to

ensure

re

leva

nce

but

this

on

ly b

ecam

e r

eally

eff

ectiv

e a

fter

DF

ID

work

ed w

ith t

he N

ew

Min

iste

r and h

is n

ew

se

nio

r sta

ff to

restr

uctu

re S

RP

ES

A a

nd its

managem

ent

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

144

Eff

ec

tive

ne

ss

Th

e N

ati

on

al

Tre

asu

ry

Th

e D

ep

art

men

t o

f P

ub

lic E

nte

rpri

ses

Were

ou

tpu

ts

gen

era

ted

? If

no

t, w

hy n

ot?

Excelle

nt

outc

om

e f

rom

SE

RP

I,

esta

blis

hin

g a

modern

sys

tem

for

public

expen

ditu

re p

lann

ing

an

d m

anagem

ent

suite

d to G

SA

nee

ds

Outc

om

e n

ot so g

ood u

nde

r S

ER

P II,

altho

ugh

indiv

idu

al

inp

uts

ge

nera

lly e

ffectiv

e.

Despite d

ifficulti

es o

ut p

uts

were

ge

nera

ted a

nd

som

e o

f th

e

work

has b

een t

aken f

orw

ard

with

out

dir

ect

DF

ID s

up

port

Was

ow

ners

hip

m

axim

ised

?

Yes. S

ER

P II

was th

e d

ire

ct re

sult

of

requ

ests

fro

m G

SA

fo

llow

ing s

uccess o

f S

ER

P I

T

he m

anagin

g c

onsultan

t eff

ectiv

ely

acte

d a

s a

secre

tari

at fo

r th

e s

teeri

ng c

om

mitt

ee

There

were

initia

l difficulti

es w

ith S

RP

ES

A w

he

n th

e p

olic

y to

ward

s r

estr

uctu

ring a

nd p

riva

tisation

was in a

sta

te o

f flux

and t

he m

anag

ing c

onsulta

nt

was s

ee

n a

s p

rom

otin

g a

n

appro

ach r

ath

er

tha

n f

acili

tatin

g. U

nder

the n

ew

Min

iste

r D

PE

has s

eiz

ed

ow

ners

hip

.

Was

ali

gn

men

t m

axim

ised

?

SE

RP

I b

ett

er

alig

ned

tha

n S

ER

P 2

A

lignm

ent cou

ld b

e s

aid

to h

ave b

een m

axim

ised g

ive

n the

fluid

ity

of

the G

SA

polic

y de

tails

in t

he e

arl

y sta

ges o

f S

RP

ES

A

Was

harm

on

isati

on

m

axim

ised

?

Yes, as f

ar

as c

ou

ld a

ssess o

n t

he b

asis

of

evid

ence m

ade

ava

ilab

le

Str

on

g e

ffort

s w

ere

made w

ith U

SA

ID,

no e

vid

ence o

f conflic

ts

with

oth

er

do

nors

.

Were

ou

tpu

ts

gen

era

ted

in

g

en

der-

sen

sit

ive

way?

Not p

art

icu

larl

y re

leva

nt

DP

E d

id n

ot see s

pecific

activ

itie

s u

nder

SR

PE

SA

as t

he

instr

um

ent fo

r addre

ssin

g g

ender

and

en

viro

nm

ent is

sues.

The D

PE

sa

w a

pro

active

sta

nce o

n s

uch m

atters

as the

responsib

ility

of

oth

er

Min

istr

ies.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

145

Eff

icie

nc

y

Th

e N

ati

on

al

Tre

asu

ry

Th

e D

ep

art

men

t o

f P

ub

lic E

nte

rpri

ses

Were

ou

tpu

ts

gen

era

ted

e

ffic

ien

tly in

te

rms o

f co

st

an

d t

ime

lin

es

s?

Yes o

utp

uts

gen

era

lly p

rovi

ded o

n t

ime a

nd w

ithin

bud

get.

Pri

ori

tisatio

n o

f use o

f fu

nd

s w

ell

managed

The e

arl

ier

difficulti

es o

ver

alig

nm

ent an

d p

ote

ntia

l id

eolo

gic

al c

onflic

t te

nded

to r

educe e

ffic

iency.

A

t th

e e

nd

of

2004 t

here

was a

mora

torium

on p

roje

cts

and m

an

y w

hic

h

had b

ee

n p

lan

ned

fell

by

the w

ays

ide.

Wo

uld

th

ere

h

av

e b

ee

n

alt

ern

ati

ve &

m

ore

eff

icie

nt

mo

des o

f d

eli

ve

rin

g s

am

e

ou

tpu

ts?

Arg

ua

bly

the N

ationa

l Tre

asury

could

have p

urc

hased t

he

short

term

inputs

with

dom

estic

funds, an

d a

n e

xchan

ge

pro

gra

mm

e d

eve

lop

ed w

ith H

MT

with

out D

FID

fundin

g (b

y ana

log

y w

ith H

MT

exchang

es w

ith d

eve

lop

ed C

om

mon

wea

lth

countr

ies a

nd

the

US

)

The is

sue o

f capacity

build

ing o

f both

th

e o

rgan

isation

(adm

inis

trative

sys

tem

s a

nd p

roced

ure

s)

and

of

train

ing o

f sta

ff c

ould

have b

een b

ett

er

addre

ssed u

nd

er

SE

RP

II.

The

role

of

TC

as m

ento

rs to n

atio

nal c

ounte

rpart

s c

ou

ld h

ave

been

more

cle

arl

y art

icu

late

d u

nder

SR

PE

SA

. T

his

im

pro

ved a

fter

the a

dvent

of

the n

ew

Min

iste

r.

1.1

.1

Imp

ac

t

Th

e N

ati

on

al

Tre

asu

ry

Th

e D

ep

art

men

t o

f P

ub

lic E

nte

rpri

ses

Did

pro

jects

achie

ve e

xpecte

d

transactio

na

l/ tr

ansfo

rmatio

na

l im

pact?

SE

RP

I h

ad

expecte

d tra

nsfo

rmatio

nal im

pact

SE

RP

II less s

o.

SE

RP

II

an

d S

RP

ES

A a

ch

ieve

d m

ain

ly tra

nsactio

nal im

pact

but th

ere

was s

om

e tra

nsfo

rmatio

na

l eff

ect in

th

e r

eg

ula

tors

.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

146

Su

sta

ina

bilit

y

Th

e N

ati

on

al

Tre

asu

ry

Th

e D

ep

art

men

t o

f P

ub

lic E

nte

rpri

ses

Have

transfo

rmatio

na

l im

pacts

been

susta

ine

d?

Tra

nsf

orm

atio

nal im

pact of

SE

RP

I h

ad

be

en t

hro

ug

h

develo

pm

ent of

sys

tem

s a

nd p

rocesses d

eve

lop

ed in t

he

earl

y sta

ges,

wh

ich a

re s

till

in p

lace a

nd

impro

ving in e

ffic

ient

app

lication

.

A f

ocus o

n tra

inin

g o

f in

div

idua

ls u

nd

er

SE

RP

II m

ean

s that

the initia

l im

pacts

have

not

alw

ays

been s

usta

ined –

for

exam

ple

a tra

de e

con

om

ist educate

d in t

he U

K u

nd

er

SE

RP

is

back in

So

uth

Afr

ica b

ut

work

ing f

or

the p

riva

te s

ecto

r.

Under

the n

ew

ly r

eorg

anis

ed D

PE

there

are

pro

sp

ects

of

susta

ine

d tra

nsfo

rmatio

nal im

pacts

.

Was there

a

cohere

nt

exit

str

ate

gy

pre

pare

d in

advance?

Not n

otic

eably

so.

Obje

ctives w

ere

larg

ely

ach

ieve

d

Under

the n

ew

Min

iste

r th

e D

PE

has d

evelo

pe

d it

s o

wn

exit

str

ate

gy.

D

FID

fundin

g h

as b

ee

n r

ep

hased t

o a

nd t

he

pro

ject tim

e e

xte

nde

d to

accom

modate

th

is.


147

ZAMBIA CASE STUDY


148

Executive Summary

S1. This report is a case study of DFID-financed technical cooperation (TC) for economic management in Zambia focusing on the period 1999-2004. It evaluates support to the following organisations:

• Ministry of Finance and National Planning - Budget and Economic Division (BED), Internal Audit Unit (IAU), Tax Policy Unit (TPU).

• Zambia Revenue Authority (ZRA).

• Department for Human Resources, Information and Planning in the Office of the President – the Payroll Management and Establishment Control Project (PMECP).


Political context

S2. Zambia has experienced many decades of economic difficulties during which there has been little progress in achieving sustained economic growth or export diversification or the efficient use of public resources. However, despite these chronic economic problems the country has avoided severe political violence. The system is characterised by a high degree of centralisation of power on the Presidency and patrimonial politics based on the sharing of rents from mining revenues and foreign aid. There have been a number of positive political developments showing that civil society can exercise a check on the executive but major issues about the future of the constitution remain unresolved and the political context has posed continuing challenges for the implementation of the public sector reform agenda.


S3. Zambia’s poor economic performance over decades and the politicisation of the bureaucracy has severely eroded capacity at institutional, organisational and individual levels. The delineation between the political and the bureaucratic is unclear and there is limited capacity to translate policy statements into administrative action. An IMF evaluation of technical assistance to Zambia over the period 1999-2003 noted there had been a lack of political will from the authorities and that organisations supported suffered from high staff turnover and the relatively the low pay of public employees. The framework for donor support to public sector reforms in Zambia has been provided by the Public Services Capacity Building Project (PSCAP), though since 2003/4 this is being taken forward in separate components addressing public expenditure management, payroll reform and decentralisation.

S4. Donors have also made a significant effort to address problems of poor coordination and lack of strategic focus and ownership in aid through the initiative to develop a Joint Assistance Strategy for Zambia and the leadership role give to the Government in that process - a process that has been given impetus by the desire of most major donors to move towards the provision of General Budget Support.


149

DFID’s Strategy for TC in Zambia

S5. DFID’s country strategies over the evaluation period have emphasised the central role of economic management and in particular public expenditure management in achieving poverty reduction in Zambia. The 1999 CSP did not present a strategy for addressing capacity issues in government nor did it examine specifically the role of technical cooperation.

S6. The 2004 CAP focused specifically on the problems arising from weak institutions and public sector systems, and identified three priority areas of action to address this:

• Improving formal accountability systems across government but also increasing civil society capacity and voice to demand better governance.

• Building more capacity within central government and engaging in long-term public sector reforms.

• Greater harmonisation in aid to reduce transactions costs and to improve aid effectiveness.

S7. The 2004 Country Assistance Plan emphasised working closely with the World Bank and other donors through support for the Public Expenditure Management and Financial Accountability (PEMFA) programme, with an emphasis on building systems to allow the channelling of funds through multi-donor budget support and set out a hierarchy of aid instruments including a specific role for standalone TC.

Economic management in Zambia

S8. In the 1990s Zambia’s average annual real GDP growth was negative. This has improved over the last five years, with growth averaging 4.5% and being relatively broad-based. Severe problems of fiscal management continued during the period covered by the evaluation. In 2003, civil service personal emoluments were increased by 3 % of GDP, threatening Zambia’s macroeconomic stability. By June 2003 Zambia external debt was 182 % of GDP. Budget overruns during this period were financed by domestic borrowing, which by 2003 reached 4.6 % of GDP compared to 2.1 % in 2002. The Government could not reach a new PRGF arrangement with the IMF, which in turn delayed the HIPC completion.

S9. Since 2003, however, the picture has improved. The September 2004 IMF mission considered that good progress had been made toward the HIPC Completion point. This largely reflected improvements in the fiscal position resulting from stronger economic growth but also some progress in improved fiscal management. In December 2004 a Memorandum of Understanding on a new PEMFA programme was signed. In April 2005, HIPC completion was achieved.

S10. The depth and sustainability of the improved economic performance must, given Zambia’s past record in failing to sustain economic reforms, remain in question. The extent to which economic management has in fact improved will be tested by periods of political challenge (like the 2006 election). As yet, the deep-rooted political and


150

organisational constraints to effective economic management highlighted in successive reviews of aid and technical assistance to Zambia have not been addressed.


S11. The results of DFID support may be summarised as follows:

• Advisory support to Budget and Economic Division played a major role in HIPC completion but was not designed (and was insufficiently resourced) to build capacity.

• Support to the Internal Audit Unit has provided a useful basis for the future strengthening of internal audit activities across the government of Zambia but this depends on significant organisational constraints being addressed.

• Support to the Tax Policy Unit has had limited results to date because of lack of clarity about the role of the Unit in the policy process, a mismatch between skills provided and what was required, and internal management issues in the Unit.

• Support to the Zambia Revenue Authority has been effective and has helped to develop capacity.

• Support for Payroll Management and Establishment Control Project has achieved some results but not yet realised the potential of the system whose introduction it has supported.


Relevance

S12. All the support provided was highly relevant in relation to the priorities of GRZ and addressed areas of importance for improving economic management in Zambia. For example the retention of the Long Term Technical Adviser in the Budget and Economic division was central to the objectives of Zambia being granted both HIPC completion and direct budget support. Similarly the assistance to the Internal Audit Unit and ZRA was directly sought by the organisations concerned and relevant to the building of their capacity in support of economic management.

S13. The main reservation in relation to the relevance of the support is the extent to which individual projects took sufficient account of the wider context to address capacity at the organisational level and whether DFID’s approach might have had a greater focus on key constraints to improving government effectiveness, reflected in the lack of progress under PSCAP.

Effectiveness

S14. The effectiveness of the TC support provided varied. The support to BED and IAU was effective but is judged to be essentially transactional in character and not in itself developing organisational capacity. The support to TPU has been largely ineffective up to the point at which this evaluation occurred mainly because of unresolved issues


151

about the role of the Unit in the policy process but also the selection of TC provider in relation to the Unit’s needs. Support to ZRA has been generally effective. The support to PMECP has contributed to the establishment and use of the intended systems despite a number of problems with the design of the support and the quality of support provided. DFID activities were generally well harmonised with other donors.

Efficiency

S15. Some activities (notably the support for TC to assist in the HIPC completion process) have been extremely efficient. Some of the PMECP support was of questionable efficiency and it is likely that the results could have been delivered at lower cost.

Impact

S16. The impact of DFID support has been mixed. Support to BED has contributed significantly to HIPC completion but in itself the activity did not build capacity within government. Support to TPU and IAU has laid the groundwork for future progress if key organisational and institutional issues are addressed. PMECP has had a significant but mixed impact, in that the full potential of the system established has yet to be realised. Support to ZRA has improved the overall performance of the organisation.

Sustainability

S17. Sustainability depends on the extend to which government capacity develops to take advantage of the work that has been done – the DFID activities reviewed here have not themselves developed this capacity. A key issue is the extent to which issues about organisation roles and management are addressed (for instance TPU, IAU, PMECP), and the generally problem of the retention and motivation of staff within the public service. ZRA has a clear role and political support and pay structures that enable staff retention and so generally does not face these problems.

Conclusion and implications

S18. DFID’s overall approach to the provision of TC can be characterised as seeking to use its comparative advantages in flexibility of financing (as with the PEM support) and established experience in some reform areas (such as the long record of support to ZRA and engagement in payroll reform) within a coordinated framework that has sought to support reforms for which there is wide agreement and signs of government commitment. This approach has led to some achievements. However, more generally, the lack of progress in addressing in particular, the key incentive and management constraints on civil service performance remains a binding constraint on capacity development within government. DFID’s activities have not in general sought to address these constraints. Despite this fundamental dependence on the wider context and coordination with reform processes that address constraints on capacity, the capacity development impact of the activities reviewed could have been strengthened through a more explicit assessment of the context.


152

S19. The larger strategic question is whether more progress could have been made in the public sector reform agenda under PSCAP, and whether there were feasible actions by DFID that could have influenced this outcome. This question cannot be satisfactorily addressed in this evaluation since it would have required a much more comprehensive assessment of the overall donor framework and approach for economic management. However, taking a forward-looking approach, the findings suggest that DFID should encourage (through its own programme and through dialogue with government and other donors) a focus on addressing the wider institutional and organisational constraints, and that unless progress is made at this level, the scope for effective capacity development through the sorts of activities that have been reviewed in this evaluation will be limited.


153

1. Introduction

1.1 This case study is one of four (the others being Ghana, Kenya, and South Africa) in a wider study to evaluate the role of DFID Technical Cooperation (TC) for Economic Management in sub-Saharan Africa. The purpose of the wider study is:


1.2 The overall approach and conceptual framework for the country studies is discussed in the Synthesis Report (Volume 1). This study used the following main sources of information:

• A detailed review of project documentation including DFID management files.

• Reviews of other relevant studies of the context including IMF and World Bank reports and independent studies.

• Interviews with key informants in the organisations receiving DFID support.

• Interviews with staff of DFID and other development agencies, private sector organisations, and providers of TC.

1.3 The consultants also held a debriefing meeting with DFID staff at the end of the mission in June 2005.

1.4 The core period covered by the evaluation focused on 1999-2004. In some cases experience after the end of this period has been reviewed where it is relevant to the assessment of the activities selected but which does not match this time period exactly. The TC covered by this evaluation was selected as all that provided by DFID in Zambia falling within the scope of Economic Management as defined by:

• Economic policy and forecasting linked to setting fiscal limits for public spending.

• Introduction of management systems.

• Public administration itself through reform of the civil service.

• Reduction in the public sector – privatisations.

• Reductions in public control over the private sector – deregulation.


154

1.5 The following projects/programmes were identified as falling within the scope of the evaluation:45

1. Public expenditure management PEM

1.6 £500,000 fund available for support to the improvement of public expenditure management, primarily directed at the Ministry of Finance and National Planning.

2. Tax policy and Implementation - RIZES

1.7 A £8.6m organisation-strengthening project with the purpose of achieving an “efficient, effective and accountable taxation system, which supports Zambia’s development objectives on a sustainable basis.” It continued support to the Zambian Revenue Authority and was intended to contribute to the strengthening of the tax policy unit in Ministry of Finance and National Planning.

3.Payroll Management and Establishment control - PMECP

1.8 A £8,690,000 project with the purpose of establishing and maintaining “processes and systems, and develop GRZ management capacity, to support effective budgeting, monitoring and control of GRZ employee numbers and Personal Emolument (PE) expenditure, and underpin the achievement of GRZ’s objectives for ‘Right sizing the Public Service’.” It is relevant across the Executive of the Government of Zambia but the project itself falls within the remit of the Department for Human Resources, Information and Planning (HRIP) in the Public Service Management Division under the Office of the President.

1.9 Table 1 shows TC project expenditure in relation to the overall size of the DFID Zambia programme and the total spending on TC for the Economic and Governance Sectors. The projects covered accounted for 40% of TC spending in these sectors, but less than 3% of the total DFID country programme over the period 1999-2004. The major activities recording expenditure on TC in the economic and governance sectors that were not selected for inclusion were support to the Anti-Corruption Commission, and support to the development of a Venture Capital Fund.

1.10 The report is organised as follows. Section 2 reviews the political and institutional context within which DFID support has been provided and the overall strategy that has guided DFID’s approach. Section 3 examines the key features of economic

45 Technical assistance up to a cost of £800,000 provided to the Zambian Privatisation Agency (ZPA)

to assist the process of privatising the Zambian National Commercial Bank was also initially included within the evaluation but the findings have been omitted from this report because the support was not in fact directed towards capacity development for the ZPA. DFID provided highly relevant specialist technical advice to support ZPA’s recommendations to the Ministry of Finance and National Planning for the privatisation of Zambia National Commercial Bank (ZNCB), one of the triggers for HIPC completion. It had no direct effect on the capacity of the ZPA, nor was it intended to. Technically the advice was provided efficiently but it was ultimately ineffective because of changing external circumstances. However, the failed process should provide useful background information for future negotiations on the privatisation of the ZNCB.


155

management over the period covered by the evaluation. Section 4 summarises the findings of the evaluation in relation to the impact of DFID TC in improving organisational effectiveness in the main areas of support. Section 5 summarises the conclusions of the evaluation in terms of the evaluation criteria (based on summaries in the Evaluation Matrix in Annex A). Section 6 discusses issues emerging from the study.

.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

156

Ta

ble

1.

DF

ID T

C p

roje

ct

ex

pen

dit

ure

in

Za

mb

ia:

Eco

no

mic

an

d F

ina

nc

ial M

an

ag

em

en

t

Fis

cal

Year

(Ap

ril

- M

arc

h);

£ '000s

Pro

ject

1999/0

02000/0

12001/0

22002/0

32003/0

4

Lo

ng

Term

Sh

ort

Term

Co

ns

ult

an

ts

Lo

ng

Term

Sh

ort

Term

Co

ns

ult

a

nts

Lo

ng

Term

Sh

ort

Term

Co

nsu

ltan

ts

Lo

ng

Term

Sh

ort

Term

Co

ns

ult

a

nts

Lo

ng

Term

Sh

ort

Term

Co

ns

ult

a

nts

To

t. E

xp

.

99-0

4

Pla

nn

ed

Exp

.

PE

M*

....

....

....

4..

..89

....

91

..216

400

500

RIZ

ES

*225

..1173

88

..313

3..

....

....

....

..1802

8600

PM

EC

....

515

..126

47

..882

39

..704

47

..745

2610

8690

Priva

tisation Z

CB

....

....

....

2..

196

....

83

....

383

664

800

TC

evalu

ate

d225

01178

103

0439

52

01078

39

0787

47

01128

....

To

tal/

Year

1403

542

1130

826

1175

5076

..

% o

f T

ota

l E

co.

&

Gov.

TC

69%

15%

54%

36%

47%

40%

..

% o

f T

ot.

DF

ID

Countr

y P

rog

9.9

%1.0

%2.5

%2.0

%4.1

%2.7

%..

To

tal

Eco

. &

Go

v.

TC

2037

3588

2110

2325

2504

12564

..

% o

f T

ota

l D

FID

Countr

y P

rog

14%

6%

5%

6%

9%

7%

..

To

t. D

FID

Co

un

try

Pro

g.

14227

56030

44713

41446

28436

184852

..

The table

sum

mari

ses tota

l spend

ing

by

DF

ID o

n T

C p

roje

cts

eva

luate

d r

ela

ted t

o e

conom

ic a

nd f

inancia

l managem

ent. E

xpen

diture

is a

lso s

ho

wn f

or

the t

ota

l econ

om

ic a

nd g

overn

ance s

ecto

rs T

C a

nd th

e tota

l DF

ID C

ountr

y P

rogra

mm

e for

the p

eri

od 1

999-2

004

. T

he f

igure

s a

re b

ase

d o

n t

he lis

ting

of

TC

pro

jects

pro

vid

ed b

y th

e D

FID

Eva

luation D

epart

ment an

d info

rmatio

n c

olle

cte

d b

y th

e c

onsultant

duri

ng t

he f

ield

vis

it in

Zam

bia

.


157


The political context46

2.1 Zambia has experienced three decades of economic difficulties during which governments have failed to deliver sustained economic growth or to use public resources efficiently. However, Zambia has been successful over this period in avoiding a descent into political violence. Patrimonial politics are at the root of both poor economic performance and the relative success in avoiding deeper political crisis. Public resources (notably mining revenues) have been used to placate or buy the support of key groups to maintain political stability leading to waste and inefficiency judged against development goals. The social roots of this system are reinforced by a constitution which concentrates power in the Executive and more specifically the President. Foreign aid47 has often helped to maintain this system. For example, Rakner, van de Walle and Mulaisho (2001) reviewing Zambia reform experience in the 1980s and 1990s concluded that although there was a significant shift over this period to a market economy overall progress of reform was very limited particularly in relation to growth promotion and public sector reform and concluded that the provision of aid had reduced the pressure for reform at key stages.

2.2 Following an initial post-Independence period of consolidation and of the balancing of various interests, much of President Kenneth Kaunda’s rule was characterised by centralisation of power in State House and personalisation of the exercise of power, notably in the allocation of state resources. There was little separation between the Presidency, the single party UNIP, the civil service, and state-owned enterprises. From 1974, GDP/head fell into a secular decline, linked to the poor fortunes of mining, the weak performance of many state-owned enterprises, and the failure to diversify the economy. In 1991, largely as a result of mounting popular anger at the government’s inability to reverse the fall in per capita incomes, Kaunda was removed in one of Africa’s first peaceful electoral transfers of power. His successor, Frederick Chiluba, the leader of the new Movement of Multiparty Democracy formed to give a voice to the widespread popular opposition to Kaunda not only gained the Presidency but his party had a very large Parliamentary majority. Chiluba adopted reform policies with strong support from IMF and the World Bank and attracted hugely increased aid flows, which offset declining mineral revenues. However, this coalition was not bound by common economic interest – for instance it included business as well as unions. Some reforms, for instance to agriculture, were implemented, but the elements of the reform programme that most threatened patronage (notably mine privatisation) were delayed. The MMD rapidly began to fall apart. Chiluba engineered changes to the Constitution to prevent the re-election of Kenneth Kaunda, which ended UNIP’s effectiveness as an effective opposition. He also gradually replaced his “coalition” MMD cabinet with personal associates. He

46 This section is based on the DFID Drivers of Change study (OPM , 2003a).

47 Eberlei (2005) and Mutesa (2005).


158

won the 1996 elections with an even bigger majority. During Chiluba’s second term corruption intensified.

2.3 However, there were encouraging, and perhaps irreversible developments in political culture, with a greater openness towards public discussion of political issues, which reached a high point in the successful 2001 Oasis Forum campaign against Chiluba’s proposal to change the Constitution to permit a third Presidential term. His handpicked successor, Levy Patrick Mwanawasa won a narrow victory over his nearest rival but the MMD fared less well, entering Parliament as a minority. Mwanawasa then embarked on a vigorous anti-corruption campaign, prosecuting Chiluba and his closest allies.

2.4 At the end of 2002, the President’s election was under challenge in the Courts and he did not have full support of the MMD. However, there seemed to be a renewed policy commitment from the Government of Zambia to fight poverty, by stimulating the growth of the private sector, improving the efficiency of the public sector by tackling corruption and implementing a more genuine decentralisation process. The PRSP had been launched and steps taken to improve national planning capacity through a Transitional National Development Plan (2002 – 2005). The latter was to be the basis of an MTEF (Medium Term Expenditure Framework) and more systematic medium-term budgeting, offering more predictability for GRZ resource allocation.48

2.5 By the end of 2004, President Mwanawasa had survived the legal challenges to his election and was more secure with his party, which has strengthened its position in Parliament through some by-election successes. However, there remain major issues around electoral reform and wider constitutional reform.

The institutional context

2.6 Zambia’s poor economic performance over decades and the politicisation of the bureaucracy has led to severely eroded capacity at institutional, organisational and individual level, despite the volume of technical assistance and other forms of support provided. The 2003 DFID Drivers of Change study (OPM, 2003a) noted that (p.11):

“The decline in performance of the civil service has emerged over years as an increasingly significant political factor. Weak capacity means that even where there may be political support from the elite and where resources are available, it is difficult to pursue a coherent agenda (such as is set out in the PRSP) and to provide quality services to the bulk of the population. Capacity limitations are everywhere evident…”

48 OPM, (2003a), pp 8-11.


159

2.7 Rakner et al. (2001) reviewed the experience with the provision of technical assistance in the context of the wider aid relationship over the 1980s and 1990s and concluded that aid was itself part of cause of the institutional weaknesses:

“Technical assistance to the Central Bank and Ministry of Finance typically provide ammunition to reformist technocrats. However, very little of their work is distributed and more time is spent briefing donor missions than in promoting policy learning in the given country. In general we find that the aid process is not conducive to policy learning. In aid dependent countries such as Zambia, donor conditionality undermines genuine policy learning. Ministries become very passive once they learn that donors mean to set policy. Individual officials have negative incentives to disagree with the donors since this will only serve to delay the arrival of the much needed aid resources. Most ministries in Zambia appear paralysed by the austerity under which they have performed now for two decades.”

2.8 The more recent evaluation of IMF Technical Assistance to Zambia covering the period 1999-2003 concluded (IMF, 2005b, p.164) that:

“There are a series of factors in the environment in which TA was provided that seriously affected the ultimate effectiveness of TA. These included political will of the authorities, high turnover of key public officials, the relative low wages of public employees, and the lack of resources to implement recommendations.”

2.9 Bureaucratic capacity has been weak, in terms both of clear delineation between the political and the bureaucratic administration and the ability to translate policy statements into administrative action.

2.10 The World Bank’s Public Services Capacity Building Project (PSCAP) has provided the overall framework for public service reform over most of the period covered by the evaluation. PSCAP had five components:

• Right sizing and pay reform.

• Policy and public service management.

• Financial management, transparency and accountability.

• Judicial and legal reform.

• Decentralisation.

2.11 It was envisaged as a long- term (thirteen year) programme, in three phases, as detailed in Table 2. Following a review of progress, particularly on public expenditure management, and donor discussions in 2003-4, the idea of a third phase of PSCAP was abandoned although the concept of a public sector reform remains. However, the three main components: public expenditure management, payroll reform and decentralisation are being carried forward through separate initiatives including a programme to support strengthening of Public Expenditure Management and Financial Accountability (PEMFA).


160

Table 2. The Phasing of PSCAP

Phase Period Focus

1 2000/1- 2002/3 Systems Development a) Right sizing and retrenchment b) Results – based management c) Fiscal management – MTEF and IFMIS

2 2003/4- 2007/8 Roll-out of systems designed and piloted in phase 1

3 2008/9- 2012/13 Training

Sensitising public servants to citizens needs

Strengthening the roles of local authorities and communities.

2.12 Stevens and Teggemann (2004) reviewed progress with public sector reform in Zambia under PSCAP and judged that only limited successes has been achieved. Their findings are summarised in Table 3. They note the challenges for implementation of public sector reform in Zambia related to the weak power base of the MMD and the specific resistance that strong trade unions can offer to public sector reform .

2.13 The recognition49 of the poor record of aid in Zambia and its failure effectively to address capacity development has prompted a major effort by donors to improve aid management and coordination. In 2002, building on their experience of SWAps and basket funding, seven bilateral donors reached agreement on a framework for putting Harmonisation in Progress (HIP) between themselves and with the Minister of Finance and National Planning. Its aim was to enhance the effectiveness and efficiency of development cooperation with Zambia through improved donor coordination – based on division of labour, silent partnerships, joint financing of consultancy funds – and harmonised procedures such as planning, follow-up, reporting, etc. An HIP action framework was signed in March 2003. Proposed measures include the framing of a development assistance policy for Zambia, emphasis on the sector-wide approach (SWAp) model, joint financing and principles/preparation for budget support.

49For further reviews of the performance of aid in Zambia see also World Bank (2002) and Bigsten and

Kayizzi-Mugerwa (2000).


161

Table 3. Public sector reform intent and results in Zambia under PSCAP

Reform Intent Results achieved Assessment

Implement medium-term pay reform strategy including the following:

Consolidate allowances into salaries

Decompress salaries

Introduce selected accelerated salary enhancement scheme (SASE)

Some allowances eliminated, some decompression

Medium-term pay strategy adopted but knocked off course

Wage bill out of control by 2003

Civil service employment increasing – retrenchment not implemented

SASE under consideration

Some progress but problems re-emerging

Response on pay policy has been politically reactive

Restructuring

Right-sizing

Ensuring private sector participation

Four public sector agencies divested

Public service reduced

No progress on private sector participation

40 agencies but no overarching legal framework

Delays in retrenchment held up restructuring

Strengthen policy formulation and performance management

Establish performance improvement fund

Eight ministries developed strategic plans but little implementation

Fourteen ministries piloting work plans for staff

PIF failed and under review

Some progress but few results

Strengthen MTEF

Implement Integrated Personnel and Payroll System (IPPS)

Establish policy monitoring system

Ensure civil society consultation

Improve public information

MTEF implementation partially achieved by 2003

IFMIS tested by 2004

Work commenced on IPPS [PMEC]

Limited civil society consultation and public information campaign impact

Expenditure accountability systems weak but nascent

Administrative controls weak but under development

Civil society engagement bore little success

Conduct training in management and policy

Train legal staff

Training activities conducted, some of which not relevant

Long-term needs not addressed under PSCAP

Source: Stevens and Teggemann (2004)


162

2.14 In February 2003, the Netherlands commissioned a report on Harmonisation of Donor Practices For Aid Effectiveness in Zambia (Saasa and Claassen, 2004). It stressed the importance of GRZ ownership of the donor harmonisation process and the need for simplicity in procedures and management structures. These recommendations have been carried forward into the Memorandum of Understanding on the Joint Assistance Strategy for Zambia (JASZ) signed on 1 April 2004 by the Government of Zambia and 9 donors: Netherlands, Denmark, Norway, Finland, Sweden, Ireland, UK, the UN and the World Bank. Canada, France, Germany and Japan have since signed.

2.15 In signing the MoU, GRZ committed itself to taking “the lead through implementation of the Poverty Reduction Strategy (PRSP) and Transitional National Development Plan (TNDP), provision of solid structures for promoting and strengthening aid co-ordination and harmonisation, political decision-making, and the allocation of funds, personnel and time.” A key step in this process is the development of an Aid Policy Paper (which is being developed by the MFNP with the support of Sweden but which had not been released at the time of the evaluation). This marks the beginning on an articulated GRZ policy on foreign aid. In relation to the management of technical cooperation the draft Aid Policy for Zambia (GRZ, 2005) has proposed that:

“The Government prefer[s] full TA pooling whereby cooperating partners put resources in a common basket that is used to acquire technical assistance. TA funds under the pooling arrangement shall be included in the regular Government budgets and the State shall assume leadership in the design and implementation of the procedures for TA procurement (including the choice of source of TA), contracting and management, using its established systems.”

DFID’s strategy for TC in Zambia

2.16 DFID’s 1999 country strategy paper (DFID, 1999) had four areas of focus – governance, livelihoods, key services (health and education) and HIV/AIDS. Of the activities reviewed RIZES, PEM and PMECP were all started under this strategy, falling under the heading of Governance. DFID recognised the importance of working in concert with other donors and in particular the World Bank and the IMF on macro-economic strategy and public sector reform. The CSP did not discuss the use of aid instruments or the specific role of technical cooperation, although in relation to the improvement of macro-economic management it specified that (p.12):

“DFID will support a pro-poor budget through advocacy and technical advice on quality and composition of public expenditure and on incidence of taxation, and help with revenue collection. We will also help to build up Zambian capacity to generate and analyse the data needed to measure changes in poverty and assess the impact of policies.”

2.17 The CSP also did not discuss capacity issues in government though it noted weak policy analysis and policy-making capacity (particularly in the Ministry of Finance and Economic Development) and poor public financial management as problems. The discussion of the public service focused on the problem of overstaffing and the need for retrenchment.


163

2.18 The 2004 CAP focused specifically on the problems arising from weak institutions and public sector systems, and identified three priority areas to address this:

• Improving formal accountability systems across government but also increasing civil society capacity and voice to demand better governance

• Building more capacity within central government and engaging in long-term public sector reforms

• Greater harmonisation in aid to reduce transactions costs and to improve aid effectiveness.

2.19 The 2004 strategy foresaw budget support as the ideal funding mechanism but equally recognized that, in practice, a mix of aid instruments was likely to be appropriate over the three years of its implementation. The precise mix of aid instruments would depend not only on the different sector objectives but also on the wider Zambian context. The CAP set out a hierarchy of aid instruments from General Budget Support, through multi-donor sector investment programmes, to multi-donor pooled funding projects and finally (p.16) “co-funded or stand-alone technical assistance and direct delivery projects where they are in line with Government and other stakeholder priorities, the benefits outweigh transactions costs, and innovation benefits, or more effective delivery of outcomes are likely to result.”

2.20 The CSP also did not discuss capacity issues in government though it noted weak policy analysis and policy-making capacity (particularly in the Ministry of Finance and Economic Development) and poor public financial management as problems. The discussion of the public service focused on the problem of overstaffing and the need for retrenchment.

2.21 The strategy was developed in discussion with other donors and it was noted that:

:“In the short term, much will depend on the Government controlling the fiscal situation, achieving a new IMF PRGF mid-2004, and reaching Heavily Indebted Poor Countries (HIPC) Completion Point six months later. “ (DFID, 2004, p.16)

2.22 Thus a strengthening of Zambia’s fiscal position and financial management systems were conditions for a shift from project-based to direct budget support. The projects considered in this report are relatively small but central to the implementation of that strategy, again reflecting one of the principles in the Joint Assistance Strategy for Zambia (JASZ) – that donors should focus on areas of their comparative advantage. Over the period covered by the evaluation Zambia has indeed begun to receive budget support, as well as achieving HIPC completion.


164

3. Economic management in Zambia


3.1 Poverty in Zambia has been widespread, deep and severe, and growth performance has been weak; real per capita GDP declined at an average rate of 1.5% per annum between 1991 and 2002 (World Bank, 2004a, p. vii). In 1998 73% of the population was classified as poor (IMF, 2004). The depth of the HIV/AIDS pandemic exacerbates the situation. The prevalence of HIV/AIDS appears to be stabilising but at a high level – 16.5% among adults, varying from 8% in Northern Province to 22% in Lusaka and 20% in the Copper belt. HIV/AIDS on this scale affects all aspects of Zambian life. All public services and the development of the economy are affected. The IMF reports that HIV/AIDS could depress Zambia’s GDP growth by over 2% per year (DHG, 2004).

3.2 In the 1990s, Zambia average real GDP growth was negative (about -0.5) compared with around 4% for HIPC completion point countries. Over the last five years growth has been about 4.5 % compared with 5% in HIPC completion point countries. However problems of fiscal and monetary management led to a delay in HIPC completion and an IMF staff managed programme (SMP) was introduced in response. By the end of 2004 the economic situation had however improved sufficiently to create a favourable climate for HIPC completion that was granted in 2005. The latest economic statistics50 confirm the prospects at the time of granting HIPC completion. Economic growth has been robust, although inflation has remained high and poverty although declining, remains widespread.

3.3 Despite the broad-based economic recovery and the growth of employment in the mining sector, total formal sector employment has continued to decline, partly because of declining employment in state-owned enterprises, the short- to medium-term effect of restructuring of privatised firms and the regulatory disincentives to new employment.

3.4 Inflation has fallen sharply from the extremely high rates experienced during the 1990s, but has remained high at around 20%, well above the single-digit figures for other HIPC completion point countries.

3.5 The external position has strengthened in recent years. Copper export volumes grew by almost 70% between 2000 and 2004, and are projected to increase by another 6% in 2005. World copper prices have at the same time risen to record levels. Non-traditional exports have also grown strongly, more than doubling in U.S. dollar terms between 2000 and 2005. Although imports have also grown strongly, partly as a result of high levels of investment in the mining sector and, more recently, because of high world oil prices, the current account deficit (excluding grants) has narrowed sharply, from about 20% of GDP to 11% in 2004.

50 These figures are based on IMF (2006).


165

3.6 External grant financing, primarily in the form of project support, has fallen relative to GDP. In contrast, external budget support has been volatile but is estimated to roughly double in 2005, to about 2% of GDP.

3.7 Extensive debt relief has greatly reduced Zambia’s external debt burden. The exchange rate has appreciated strongly in real terms during the past year, but foreign exchange reserves are low. The high external value of the currency may pose a challenge for non-traditional exports, which will have to rely on improvements in productivity to enhance international competitiveness.

Macroeconomic management: 1999-2004

3.8 During the 1990s, the Government introduced a number of changes in economic policies with programs of liberalisation and deregulation, and elimination of market distortions. During the 1990s, Zambia had limited success in achieving macroeconomic stability in relation to the elimination of inflation or improved fiscal performances. Price controls were abolished, the exchange rate and interest rates were liberalised. Trade reforms simplified the tariff structure, removed quantitative restrictions, making the Zambian economy one of the most open in the sub-region. By 2005, 260 state-owned enterprises have been privatised. Zambia was granted debt relief under the enhanced HIPC framework, with the completion point originally scheduled for December 2003.

3.9 Severe problems of fiscal management continued up to 2004, but began to turn round in 2004 (the end of the period covered by the evaluation). In 2003, civil service personal emoluments were increased by 3% of GDP, threatening Zambia’s macroeconomic stability. By June 2003 Zambia external debt remained at 182 % of GDP. Budget overruns during this period were financed by domestic borrowing, which by 2003 reached 4.6 % of GDP compared to 2.1 % in 2002. The Government could not reach a new PRGF arrangement with the IMF, which in turn delayed HIPC completion.

3.10 Since 2003, however, progress has been made in improving fiscal and monetary indicators The September 2004 IMF mission considered that good progress had been made toward the HIPC Completion point: all of the triggers were being met except piloting for an Integrated Financial Management Information System (IFMIS). In December 2004, a Memorandum of Understanding on a new Public Expenditure Management and Financial Accountability programme was signed. At their meetings on 7th and 8th April 2005, the IMF and World Bank Boards agreed to HIPC completion.


166

Table 4. Progress against HIPC “Triggers”

Source: IMF and IDA (2005)

3.11 The 2005 Article IV report (IMF 2006) confirmed the relatively healthy status of the economy. GDP actually grew at 5.4% in 2004 with mining and quarrying (13.9%) and construction (22.5 %) showing the greatest growth.

Budget and public finance management

3.12 In Zambia, the formal economic management structure follows the Constitution, decisions being taken by the Head of the Executive, the President, on advice from his Cabinet. The Ministry of Finance and National Planning plays the central role in economic management in the executive. The other key organisations are the Zambian Revenue Authority, the Zambian Privatisation Authority, the Office of the President, and the Cabinet Office. The relationship between the Ministry of Finance and National Planning and other agencies of government varies: for example:

• MFNP takes the clear lead on the national planning and budgeting cycle and public financial management, proposing the fiscal framework to Cabinet, drawing up the Medium Term Expenditure Framework (Green Book) and, in consultation with line Ministries, the National Assembly and the public, translating this into the Appropriations Bill submitted to Parliament.

• MFNP designs and sets the level of taxes but once agreed by Parliament these are administered by the ZRA. ZRA participates in the revenue side of the Budget cycle.

• The size of the public sector pay bill is set by MFNP in the context of the fiscal framework for the MTEF, but responsibility for public sector manpower planning and control rests with the Public Service Management Division under the Office of the President.

• Domestic borrowing under control: 2003 5% GDP, 2004 2% GDP.

• Growth 4.6%, compared with projected 3.5%.

• Wage bill and expenditure within the Budget parameters.

• Releases to poverty reduction programmes in order.

• Reduced government domestic borrowing led to:

- Lower interest rates.

- Expansion in available credit for the private sector.

• PEMFA & overall PEM enough progress to meet the spirit of the requirement.

• ZESCO commercialisation, effectively privatisation.

• Further increase in mineral exports as new copper mines come on stream.

• Budget is in line with MTEF.

• Government expenditure will increase slightly in order to fill critical vacant posts in education and health.


167

3.13 At the start of the evaluation period, in common with other donors, DFID saw reform of PEM in Zambia as important because (as noted in the project memorandum for the PEM project):

• At the macro level, lack of fiscal discipline had resulted in successive and unsustainable public sector deficits, frustrated efforts to stabilise the economy, fuelled inflation and increased the cost of investment; the climate for long-term economic growth and poverty reduction was not promising.

• At the micro level, public expenditures rarely translated into quality public services: this had resulted in poor value for money; the link between public spending and poverty eradication was tenuous.

3.14 In 2003, a joint donor review funded by DFID (OPM, 2003b) concluded that:

“In our opinion, it is clear that the current PFM system:

• Is not capable of establishing a budget which supports deliverable and sustainable plans and policies that government wishes to implement or of delivering, controlling and measuring the financial resources required and expended on these policies, and

• Fails to deliver both an acceptable level of transparency in respect of the financial assets, liabilities, revenues and expenditure controlled by the Executive and the facilitation of its accountability for its stewardship of them.

3.15 This is compounded by weaknesses in the legislative framework and in the oversight and monitoring capacity of the legislature and its support institutions such as the Auditor General.

3.16 The DFID CAP (2004) similarly noted (p.5):

“….chronic weaknesses in public expenditure management. A system of monthly cash releases favours sectors with strong cost drivers – such as administrative and foreign exchange commitments – and allows unplanned expenditures on defence, security services and parastatal liabilities. Periodic cuts therefore often fall on variable non-wage recurrent expenditures, tending to damage service delivery in areas such as health and education. More fundamentally, the budget is not linked to the PRSP, and is incapable of providing a clear picture or basis for coordinating domestic and external resources, recurrent and capital spending and pro-poor allocations.”

3.17 OPM (2003b) went on to highlight the Government of Zambia’s decisions to address PEM reform issues and to turn the PEMFAR diagnostics into a credible action plan. By the end of 2005, Government and donors reached agreement on the project for Public Expenditure Management and Financial Accountability (PEMFA), designed to address these intrinsic weaknesses. Recruitment of the main TC was in progress during the conduct of this study. Similarly, a recent review of the Poverty Reduction and Growth Facility (PRGF) (IMF 2005) concluded that greater focus was needed on improving public expenditure management.


168

Summing up: has economic management improved?

3.18 Zambia’s macro-economic performance has improved sufficiently to allow Zambia to achieve HIPC completion. However this achievement was principally the result of improved economic growth, which enabled an improvement in the fiscal position to be realised, as well as the progress in the development of PEMFA as a framework for improving public expenditure management, rather than evidence that significant improvements in economic management capacity have been achieved. 51

3.19 The depth and sustainability of the improvements that have occurred must, given Zambia’s past record in failing to sustain promising economic reforms, remain in question until a longer record of improved performance and more sustained evidence of improved economic management can be demonstrated. How political pressures relating to the 2006 election affect economic management will be a particularly important test. The deep-rooted political and organisational constraints to effective economic management that have been highlighted in a series of studies and reviews as outlined above remain key features of the context.

51While the PEMFA appraisal noted important improvements in the budgeting process in 2005, this

assessment contrasts very markedly with the concerns about systemic weaknesses in the reviews of the public finance management system over most of the evaluation period, so caution is merited in making assessments about the depth and sustainability of the changes noted.


169


Ministry of Finance and National Planning - overview

4.1 The Ministry of Finance and Economic Planning is at the heart of economic management in Zambia and strengthening its capacity has been central to reform efforts sponsored by donors for instance under PSCAP. The allocation of the functions of economic management to sub-organisations within the Ministry is shown below, along with the nature of their outputs.

Table 5. Economic management and the structure of the Ministry of Finance and National Planning

Function/Activities

MFNP sub-organisations

Outputs

Macro–Economic Forecasting

Tracking donor funding

Budget

Investment & debt management

External technical co-operation

Fiscal discipline,

Pay policy

Tax policy

Privatisations

Liaison with line Ministries

PRSP monitoring

Planning

Budget

MTEF

Activity Based budgeting

Accounting

Auditing

Accountant General

Internal Audit

Financial oversight

Value for money etc

Payroll reform

4.2 The Ministry’s structure, and its title, has varied since the beginning of the study period. The current structure shown in Table 5 reflects the increase in importance of the planning function as a mechanism for prioritising expenditure in implementing the PRSP. It has two wings - the Budget and Economic Division and the Finance, Management and Administration Division. In principle the brigading of Budget and Planning under the same PS should help in the preparation of a budget that supports the delivery of sustainable plans.

4.3 Starting in 2001, DFID set up a fund of £500,000 for a programme of interventions designed to raise the profile of PEM and to examine those institutional issues preventing effective public expenditure management. This would cut across a range


170

of issues, including civil service salaries, job satisfaction, and corruption.52 The fund would be used to extend existing support for interventions and related studies on how best to use public expenditure to achieve policy goals. It would be short-term in outlook and opportunistic. Two direct interventions with the Ministry of Finance and National planning were supported by this fund – funding of a long-term advisor to the PS on the Budget and Economic Division and support for the revisions of the Internal Audit Manual.

4.4 The following three subsections assess the organisational performance and impact of DFID assistance in the three divisions or units of MFNP that have been the focus of DFID support.

• Budget and Economic Division.

• Internal Audit Unit.

• Tax Policy Unit.

4.5 The remaining subsections cover the other organisations that have received DFID TC:

• Zambia Revenue Authority.

• Department for Human Resources, Information and Planning, Office of the President.

4.6 DFID’s overall support for PEM is summarised in Table 6 below.

52 In practice other programmes were introduced to cover many of these specific issues, including that of

payroll reform, discussed in section 4 below.


171

Table 6. DFID support for PEM

Date Field of support Channel Cost

£’000

2001 Drafting of PRSP Ministry of Finance via World Bank pool 60.0

2002 Public Expenditure Tracking Survey Examination Council of Zambia through support to WB project

63.4

2003 Drafting Internal Audit Manual Ministry of Finance Internal Audit Manual

26.1

2003 Joint Donor Study of Direct Budget Support

53Ministry of Finance and donors 250.0

2004 Public Finance Management Study Ministry of Finance and donors 5.1

2004 Provision of Long Term TA on MTEF and Budget

Ministry of Finance 36.7

2004 PEMFA Appraisal PEMFA Procurement Guidelines

Ministry of Finance and donors 53.6

MFNP: Budget and Economic Division


4.7 The Budget and Macro Economic Division plays a central role in GRZ economic management. It is responsible for the preparation of the PRSP, the MTEF and the Budget. Planning and Budgeting are in separate departments within the Division, and co-ordination between the two seems to take place at a high, rather than at a working, level. Similarly, there are weaknesses in the linkages between the Budget and Economic Division and the Finance Management and Administration division; for example, the Budget Division do not find the current FMIS useful during the process of drawing up the annual budget.

Support from DFID

4.8 Following the delays in GRZ reaching HIPC completion, the IMF introduced a Staff Managed Programme (SMP) in 2003. DFID used the PEM funds to keep a long-standing technical adviser to MFNP in post during the difficult period of the Zambian

53 Since this study was carried out by Oxford Policy Management (OPM, 2003b) it was excluded from the

scope of this evaluation.


172

transition out of the SMP to PRGF and HIPC completion.54 The focus of the support is reflected in the advisor’s Terms of Reference, which explicitly referred to his role helping the GRZ in the process of “gaining trust from co-operating partners who wish to initiate or increase their general budget support to Zambia on the basis of agreed sectoral priorities and targets for the budget and MTEF”. He served as an advisor to the Permanent Secretary for Budget and Economic Affairs, working closely with both the Budget and Planning Divisions.


4.9 In September 2003, a study commissioned by DFID (OPM, 2003b) found that

“The current PFM system exhibits significant and pervasive weaknesses in its ability to establish a budget that supports deliverable and sustainable plans and policies and in delivering, controlling and measuring the financial resources required for and expended on these policies.”

4.10 Since the middle of 2003, the Ministry of Finance and National Planning has made great strides: HIPC completion has been achieved and important donors have undertaken to provide direct budget support. GRZ target dates were generally met sufficiently to satisfy IMF and World Bank staff and delays may be attributed to bureaucratic mishaps on both sides. On the administrative front they have achieved targeted products such as introducing the MTEF and an Activity Based Budget.

4.11 However, the use of the MTEF and measures to improve the efficiency of the budget are in their infancy in Zambia. Although economic performance had improved sufficiently for HIPC completion to be reached, efficient administrative systems have yet to be developed for producing and implementing a pro-poor multi-annual Budget based on the PRSP (IMF, 2005c, Mudenda et al,, 2005). Some sectors, notably health and education, do have costed plans linked to clear targets (as set out in multi-year sectoral strategies), which form the basis for their budget submissions to MFNP. However, there is no political and administrative system for ensuring that:

• The three year budgeted plans for individual agencies and Ministries are based on costed plans for service provision in line with the PRSP.

• The allocation of resources within the fiscal ceiling reflects priorities developed and monitored through Zambia’s Parliamentary and Cabinet - based political system.

4.12 The consultants could find no manual or detailed timetable for the construction of the MTEF and Activity Based Budget (ABB) throughout government. Considerable work remains to be done on integrating the PRSP, the planning and budgeting process and budget implementation. The first tests of the present system will lie in the 2005 budget planning process, the quality of the rolled forward MTEF, and the relationship between budget and expenditure out-turns.

54 Another donor subsequently funded him until he left Zambia in early 2005.


173


4.13 The Long Term Adviser contributed considerably to the transition from SMP to PRGF and ultimately to HIPC completion (8 April 2005) and the move towards Direct Budget Support by some key donors. Discussion with the Deputy Minister and MFNP officials showed that he was regarded as a much-missed member of staff who helped the Ministry to make progress with the donor community.

4.14 In that sense he was extremely relevant to the immediate needs of GRZ. His activities were effective in achieving the desired outcomes and short-term results. GRZ target dates were generally met sufficiently to satisfy IMF and World Bank staff and delays may be attributed to bureaucratic mishaps on both sides. Over the long term, he has made a significant contribution to the work of the Ministry including the development of the MTEF and Activity Based Budgeting systems. It is not clear how far this process advanced administrative processes and systems. The assessment of this evaluation is that the impact of this support was essentially transactional rather than transformational – in enabling GRZ to achieve specific goals rather than increasing the capacity of MFNP.

4.15 Sustainability and transformational impact will depend on the development of financial management and administrative systems that can translate PRSP priorities into the sectoral allocation of resources within the MTEF ceilings, based on targets and costs of service delivery. So far, donor programmes have focused on technical systems. Sector Advisory Groups have been set up and should be given added weight by donor harmonisation through the Joint Assistance Strategy for Zambia, but GRZ’s capacity to perform these activities will need support. The budget preparation/execution component of PEMFA aims to focus on these issues.

The Internal Audit Unit


4.16 The Internal Audit Unit is one of the departments within the Finance, Management and Administration Division of the Ministry of Finance and National Planning. It has professional responsibility for internal audit throughout the Government of Zambia, although internal auditors in Ministries, Agencies and Provinces are on the books of those bodies. The MFNP IAU sets both the structural framework and professional standards for internal audit throughout GRZ. Internal Audit units within Ministries and in the Provinces report to their line Permanent Secretary and carry out internal audits at all levels of government. Their principal role is as a management tool for the PS, evaluating both management systems and compliance with government accounting rules.

4.17 Although Internal Audit is short staffed, some Permanent Secretaries accept the importance of the work; for example the Ministry of Education requires internal audit units for education in each of the provinces. Needs are of course different for different Ministries and services. Nevertheless the service is understaffed. There


174

has been a freeze on recruitment and the inability to replace staff leaving for better jobs is a perennial problem. Some staff have CIPFA qualifications and training has been provided by the French and by New Zealand. Other support has been provided through the PSCAP programme.

Support from DFID

4.18 By 2003, the Internal Audit manual drafted in 1996 needed updating to reflect recent developments such as the Audit Act, and new financial regulations. The Internal Audit Unit approached DFID with a request for external assistance to help them prepare a new manual. DFID contacted the Association of Chartered and Certified Accountants who identified an individual to help with the production of the new GRZ manual. They subsequently reimbursed GRZ for the costs of printing the final manual.

4.19 Staff of the Internal Audit Unit prepared the new Internal Audit Manual with the help of a mentor from the Association of Charted and Certified Accountants paid by DFID.


4.20 The manual is clear, well written and fully regarded as its own by IAU. It has been rolled out to line Ministries for use by their Internal Audit Units. The World Bank’s 2003 Public Expenditure Management and Financial Accountability Review reported that the IAU has been effective in diagnosing problems in financial management procedures and the new Internal Audit Manual improves the efficiency and quality of internal audit. The capacity of the internal audit unit has been increased. The IAU is now revising its manual to reflect the latest developments in Activity Based Budgeting.

Contribution of DFID-funded TC

4.21 The technical assistance provided to the Internal Audit Unit was relevant to GRZ needs for internal Audit, and effective in providing the tool that was identified as required. Although efficiency cannot be measured in any quantitative sense by this study since no appropriate measures were built into the design of this assistance, the mode of delivery – mentoring – and the IAU’s sense of ownership suggests that the TC did represent value for money in terms of the immediate outputs both the manual and the capacity of both the MFNP Internal Audit Unit and the GRZ Internal Audit Unit.

4.22 The impact of the manual on economic management is limited by the lack of adequate resources in Internal Audit and insufficient follow-up on their recommendations. Moreover, the Permanent Secretary for Finance, Management and Administration argued that even if internal audit were adequately staffed, implementation of their reports would be seriously hampered by the quality of accounts staff who would have to implement new systems recommended by IA reports.


175

The Tax Policy Unit


4.23 In principle, Zambian tax policy is made by the Tax Policy Unit of Budget and Economic Affairs Division of the Ministry of Finance and National Planning and implemented by the Zambian Revenue Authority. In practice, after its establishment in 1993 and as its capacity developed, ZRA had become so influential that it was perceived as taking the lead on policy advice, while within MFNP tax policy analysis had fallen into abeyance. At the same time, although tax revenue was increasing as a result of improved tax administration through ZRA, there was increasing government concern about structural deficiencies in the Zambian tax system such as the narrow tax base. By 2000, the Zambian government was keen to develop its capacity to negotiate medium term revenue targets with the IMF in the context of the PRGF, and strengthening the tax policy function in the Ministry was seen as vital.

4.24 In this context, the role of the Unit is to identify policy solutions to specific revenue problems, in particular to spread the tax burden without distorting economic transactions or deflecting from social priorities. Its task is to identify the advantages and disadvantages of different revenue raising options, analysing not only the economic impact and social impact but also the prospects for effective legislation bringing them into force and viable mechanisms for their implementation. The TPU is intended to help ministers take a wider view on the economic and social impact of tax policy while ZRA focuses on providing the practical advice necessary to ensure that new taxes raise the desired revenue and provides insights into the likely response of taxpayers.

4.25 The TPU was re-established in 2000 but did not resume this policy role until 2002, when there were signs that the new Government had the political will both to combat corruption and to broaden the tax base.

Support from DFID

4.26 As part of the RIZES project launched in 2000, support for the Tax Policy Unit was slow to get off the ground, because there were difficulties in establishing precisely what was required and a lack of a clear demand from MFNP. The project aimed to strengthen the TPU as part of the policy making function within MFNP. DFID was the sole donor providing support to the TPU. By the time of the 2002 OPR, there were signs that the new Government had the political will to both combat corruption and to broaden the tax base and ZRA was actively looking for ways in which the informal sector could be brought into the tax net. As part of the 2002 RIZES OPR, the new Project Manager undertook to draft a proposal for taking the work forward with the MFNP Tax Policy Unit (TPU). The ZRA Commissioner General undertook to assist in raising awareness of TPU issues with the MFNP.

4.27 Support started in earnest in February 2003, with a consultancy to advise on the nature and internal structure of the TPU and on a research agenda and training requirements. Hardware, software and tax policy literature were provided. A further


176

two consultancies examined the institutional needs. In September 2003, a former ODI fellow from ZRA transferred to the TPU on a 9-month contract. A long-term tax policy consultant was appointed in April 2004. A fourth study was undertaken in June 2004 to review and report on the institutional and reporting arrangements of the TPU. The long-term tax policy consultant’s contract was terminated in March 2005. Further support was delayed pending a major review by MFNP of tax policy in the first half of 2006. A further review was to be carried out under PEMFA, under TORs to be developed with RIZES funding.


4.28 Over the past five years there has been little visible improvement in the capacity of the TPU to support Ministerial decision-making. That the IMF continues to engage with ZRA rather than the Ministry is symptomatic of weakness of the TPU.

4.29 The Long Term Tax Advisor identified a lack of “defined reporting procedures to ensure that the research work of the TPU had a chance of being considered by policy makers”, while the Minister regarded the advice given as too academic and remote from his needs. The June 2004 consultancy identified a fundamental problem; the TPU co-existed with the Revenue Unit of Budget Division. These two issues are related. The Revenue Unit of the Budget Division has a clear role – ensuring the tax revenue to minimise the fiscal deficit in the forthcoming Budget. The more lateral thinking required to take tax policy forward in this process, one of the functions of the TPU, has no operational place in the annual Budget process. This lack of organisational capacity is partly external to the TPU i.e. a weakness in the integration of the MTEF and the annual Budget and the lack of MFNP processes for carry forward medium term tax policymaking, but also reflects weaknesses in TPU’s management and staffing.

4.30 The plan to integrate the strengthening of the TPU into PEMFA is a step forward, but its success will critically depend upon recognition that its role is at the interface between the technical and political.

4.31 There were no readily identifiable TPU products. Both the long term and short term tax advisers contributed to the annual budget process of analysing tax proposals made by both government and private sector stakeholders. The quality of the recommendations arising from this process has improved. However, these take the form of Committee decisions, and from the information at our disposal the precise contribution of the TPU could not be identified.


4.32 RIZES has not so far been successful in building the capacity of the Tax Policy Unit. There have been repeated attempts to identify the right organisational structure and institutional role for the organisation, but the issue remains for consideration under PEMFA. Throughout the project, the TPU was seen as conducting research on specific tax policy issues; this was reflected in the criteria used to select consultants. Both advisors were expected to promote capacity building, transfer skills to the TPU,


177

and lay a solid basis for future cooperation between the TPU and ZRA’s Executive Support Unit. The ex ODI fellow was considered to have made a useful contribution in a role that was transactional rather than transformational, with his terms of reference not including any specific mentoring or capacity development role. He brought practical skills learned from his work in the ZRA, whereas the long-term consultant’s expertise was primarily academic and analytical, without the practical experience of policy-making sought by the Ministry.

4.33 DFID support was relevant in building research capacity, but as the repeated organisational reviews indicate, it was struggling to relate research to the operational structure and processes with the Ministry. Thus it was largely ineffective and capable of having little sustainable impact on the economic management of the country.

The Zambian Revenue Authority


4.34 Since its formation in April 1994, the ZRA has gradually developed into a mature revenue collecting institution, making a significant contribution to effective fiscal policy, which is central to economic growth. There are three operational divisions each headed by a Commissioner – Customs and Excise, Direct Taxation and Value Added Tax. The collection of direct taxes, VAT and customs duties has been integrated to allow effective use of IT systems for cross compliance. ZRA has an international reputation as a highly professional revenue authority. As discussed above it has tended to usurp the role of the Ministry in the field of tax policy.

Support from DFID

4.35 Ever since its foundation in April 1993, DFID had been the sole donor supporting ZRA. By 2000 it was well on the way to becoming a sound self-sustaining revenue authority. RIZES was envisaged as the final stage in DFID support. The objective was to complete the process of helping to improve and consolidate ZRA’s own management skills and processes. On the technical side, it focused on the computerisation of Direct Tax and the creation of an Integrated Tax Administration System. In parallel, the ‘head office’ functions, both in the support divisions and directorates and the corporate management team, were provided with training, re-organisation and procedural improvements, in particular in the fields of financial management, internal audit and procurement. The RIZES project manager left Zambia in 2004. After that, ZRA received only short-term consultancies, in response to specific requests while the focus of the RIZES project has been on the TPU.


4.36 In February 2002, a Zambian was appointed Commissioner General from within the ranks of senior management, the previous acting CG staying on as Advisor for a


178

further 4 months. This completed the “Zambianisation” of the authority. The increased maturity of the organisation is reflected in the extent to which DFID support thereafter continued to be demand led by the ZRA.

4.37 All Commissioners are now Zambian and there has been a policy of internal promotions, ensuring continuity of institutional memory. Staff are paid above Zambian public service rates but are employed on relatively short-term contracts that include automatic dismissal after three investigations for fraud. If the investigation finds sufficient evidence, it proceeds to prosecution and the officer is dismissed on prosecution. There are 1500 staff stationed around the country – the Customs Service being the most dispersed, located at thirty border posts. The internal affairs division under in the Commissioner General’s division is responsible for investigation of internal corruption and fraud working with the ACC and the Zambia Police. Investigations Division was created in January 2005 and took over the investigations/ compliance units in Customs and Excise, Direct Taxes and VAT division. It has a dual mandate to deal with tax evasion and to enhance tax compliance generally, contributing to the public trust, fairness and integrity of the assessment system. The Division employs three prosecutors to pursue established offences through the Zambian legal system.

4.38 There is evidence (including from ongoing cases) that corruption in ZRA has not been eliminated but external financial and procurement audits have found no evidence of grand corruption within ZRA, although they have identified weaknesses in management systems.

4.39 Management of the Authority has improved throughout the period of the project. Working with the IMF, the Authority has now established a set of performance measures to assist in monitoring performance on internal management, operational systems and service delivery through quarterly reporting to the Board on progress against agreed action plans. By 2000, GRZ agreements with the IMF had set revenue collection benchmarks under the PRGF. Despite the worsening in Zambia’s macro-economic situation, IMF cash targets had been achieved. The 2005 IMF staff report (IMF, 2005c) reported Zambia’s tax system will undergo comprehensive review in 2006 focusing on tax policy issues (such as the low yield on VAT) but also reviewing the effectiveness of the tax administration.


179

Figure 1. ZRA revenue collection as a percentage of GDP

4.40 Figure 1 tracks Zambian Tax revenues against target. Revenues have always been above target, rather less so in 2002 -2004 than in the early days. Since 2000, Government revenues as a percentage of GDP have been consistently higher than the average for HIPC completion point countries. Although it has consistently met the overall revenue/GDP targets as set out in the IMF’s PRGF, the revenue/cost ratios were declining for much of the period. It was recognised by the IMF though that this in part reflected macroeconomic developments outside ZRA’s control.

4.41 ZRA has made a significant contribution to economic management in Zambia in terms of both consistency and predictability of revenue collection. The perceived reduction in corruption has both macro and micro-economic effects in that it has helped both the predictability and transparency of tax levels, and has improved the enabling environment for investment. In addition important changes have been made to the tax system including the introduction of VAT and the simplification of and reduction in tariff levels.

Contribution of DFID-funded TC

4.42 The activities under the RIZES project were directly relevant to the needs of ZRA – TC was demand determined by the ZRA although occasionally on the basis of incomplete information. ZRA senior staff were generally satisfied with the support provided, particularly the use of British Customs officers on VAT management and


180

saw future value in a similar secondment from Inland Revenue. On the whole it waseffective in building ZRA capacity and had a significant impact of the output of the organisation, having had a positive impact on the management and effectiveness of the ZRA which is now generally meeting the cost:revenue ratio set out in its annual corporate plan. The impact of RIZES and its predecessor seems likely to be sustainable. ZRA continues to seek improvements in its performance, and the project provided an up-graded financial management information system. Internal audit continues to monitor implementation of audit recommendations and liaison between internal audit and expected external auditors are expected to result in a strengthened audit plan.

Department for Human Resources, Information and Planning (HRIP), Public Service Management Division (PSMD), the Office of the President


4.43 The HRIP is responsible for ensuring the efficiency and effectiveness of the Zambian public service. It has overall responsibility for payroll systems and staffing levels. It plays a critical role in meeting targets for the fiscal deficit and ensuring value for money in the use of public resources. Its policy direction over the study period was to implement system-wide reforms to support improved service delivery in the public sector: specifically to establish and maintain processes and systems, and develop GRZ management capacity: to support effective budgeting, monitoring and control of GRZ employee numbers and Personal Emolument (PE) expenditure: and underpin the achievement of GRZ’s objectives for ‘Right-sizing the Public Service’. Establishment of an improved system therefore has the potential to improve economic management through more effective public expenditure control and service delivery improvement.

4.44 Development of new systems for Payroll Management and Establishment Control (PMEC) is the responsibility of the PMEC Unit working along side HRIP in the Public Service Management Division of the Office of the President. It is managed by the Project Focus Group (PFG) chaired by the PS of PSMD and including Permanent Secretaries from MDD, MFNP and MOE, the main users.

Support from DFID

4.45 PMECP represents the second recent attempt by DFID to support the strengthening of payroll management and establishment control in Zambia. The first attempt from 1997 was focused on supporting retrenchment as part of a push to reduce the wage bill in order to release resources for recurrent expenditures and to enhance pay. It was planned to cut the number of civil servants from 139,000 to 80,000 by the end of 1999 by a recruitment freeze, retrenchments, removal of ghost workers, and ministerial restructuring. An effective payroll management and payroll control system was also regarded as central (Mataka, 2002). This approach was highly politically sensitive and political opposition to retrenchment contributed to the difficulties of the DFID supported TC but there were other serious problems.


181

4.46 A January 1999 DFID Review highlighted several weaknesses: - incomplete and unclear design, weak project management, poor government-consultant-donor working relations, and “inadequate political sponsorship” of the project. GRZ and DFID agreed to close the project prematurely in 1998 and to jointly commission the design of the PMECP, which was introduced in 2000 within the framework of PSCAP. It aimed to improve the technical systems for data collection and analysis on staffing and pay, thus providing a better basis for assessing the economic consequences of changes in pay and establishment policies. Support was provided for the development of systems and revision of policies, manuals, orders and regulations and to train and sensitise direct users of the system.

4.47 A feasibility study was carried out prior to the design of the project, but it did not articulate the rationale for the advisory support proposed, which appears to have reflected a compromise between DFID and GRZ about the amount of long-term TC to be provided.55 The most precise capacity development statements in the Project Memorandum (Annex 3) focus on IT systems methodologies: PMECP “will build this capacity through on the job training of the Task Team by external adviser counterparts trained in these methodologies. Strengthening in-house capacity to design, project manage and operate complex systems change programmes will be a key feature of the project.”

4.48 DFID provided some thirteen TC years to PMECP between 2000 and 2004, about the same as Zambian staff time. Inputs were both short and long term. About 10% of TC was intended to be short term, but in practice, there were very frequent changes in the occupants of “long term” advisory posts. General skills development and training, particularly of end users, were delayed. Only 40% of planned specialised skills training activities were implemented.


4.49 There have been four main changes in the PMEC team’s outputs since 2000:

• An improvement in the quality (accuracy, detail, timeliness) and quantity of payroll and establishment data produced by GRZ.

• Apparent improved accountability for payroll management and establishment control – and for data quality, although full compliance has not been achieved by all end-users.

• New types of analyses of relevance for economic management can now be made on the improved data (earlier such analyses were either not possible or required more effort).

• Several potential and new products and services of the payroll and establishment control system are available from the technical systems but are not yet fully implemented.

55 DFID Zambia staff note that substantial discussion and consideration of organisational issues took place

during project design. However this was not articulated in the project memorandum.


182

4.50 The evidence on changes in payroll management and establishment between 2000 and 2004 suggest that the systems introduced are being used. PMECP has helped to provide valuable detailed information to policymakers, and has helped in controlling corruption. Examples cited included cases where discretionary promotions and transfers had been controlled. On the other hand, there have been cases of resistance among end users. For example the Ministry of Home Affairs (MOHA) recommended to the Head of State that PMECP be suspended. It argued that the system created confusion among security officers. PSMD meetings with MFNP, CCSD and MOHA helped to sort these problems in 2004. However, the Chair of the PFG resolved the issue and gained recognition from State House for facing down this sort of criticism.

4.51 Improved outputs have been identified by the PSMD and Budget Office in the MFNP. They felt that they had a better technical basis for the 2005 annual pay negotiations. When and if more of the functionalities of the PMEC system are utilised, the technical basis for better economic management will improve.

4.52 However, the PMEC does not yet cover the entire public service (GRZ, World Bank et al, 2005, pp 164-5).56 Some compliance problems remain, and the full potential of the new system is not yet exploited. Moreover PMEC remains in a separate unit, not yet fully integrated into HRIP. Sustainability of the PMEC system depends more on GRZ decisions, resources and commitment than on major additional donor funding. As the review by the Cabinet Office (2004, 6) noted “the current culture… has little regard for processes and controls…changing this culture is very crucial to the sustainability and stability of the system.”


4.53 DFID support to PMECP focused on one of the five PSCAP components: “Right-Sizing and pay reform of public service” and directly responded to the objective for this component, namely to put “an effective payroll and establishment control system in place.” However, the lack of explicit goal and purpose statements for the TC tended to weaken relevance in implementation.

4.54 Capacity has been developed and a system has been introduced to generate data on which to base improved economic management, but attribution is difficult to assess. Several factors have been favourable.

4.55 First, the PMEC unit’s capacity was arguably relatively high in 2000. Second, despite early staff reservations the project was pushed forward by Zambian commitment to PMEC despite frequent changes in consultant personnel. Third, the nature of donor pressure to improve payroll management and establishment control has changed from rhetoric about a ‘a bloated civil service’ and a focus on

56 The PMEC database now includes some 87,000 staff. Yet to be included are some 25,000 employees in

the Ministry of Health (due to re-structuring problems with the Health Boards) and some 6-7,000 employees without an approved post. As is usually also the case elsewhere, there are no immediate plans to include the armed forces staff in the database (although a separate base using the PMEC-system is being considered).


183

retrenchments to a more realistic view of the size and functions of the Zambian public sector and to a stronger emphasis on the tools (such as PMEC) with which its size and composition can be better controlled. Fourth, the new computerised technology – and the combination of hitherto separate payroll management and establishment databases – is a powerful tool. Last, but not least, the positive changes in organisational outcomes must be attributed to the commitment and skills among Zambian team members and users.

4.56 Only part of the TC was efficient/ cost-effective. Some of these inputs were transactional and appear expensive. Much of the assumed efficiency gains of providing long-term advisers were lost as a result of turnover among the advisory staff. In the absence of a documented assessment of the need for TC support and justification of the model presented, it is difficult to assess whether Zambian or regionally based consultants, some of whom are much cheaper than the overseas advisers, could have played a greater role and so improved cost-effectiveness

4.57 While the capacity of the PMEC unit has increased, the impact across the public sector as a whole is less certain. The systems do not yet cover the entire public service and some compliance problems remain in the provision of information to the central system. Thus their full potential is not yet exploited. Recent IMF Letters of Intent and the Interim Performance Assessment Framework (PAF) for Zambia’s Poverty Reduction Budget Support (PRBS) do require that the PMEC-system be used and coverage expanded, and this may help to widen the coverage of the system. The savings envisaged from staff retrenchments 57 have not materialised, among other factors because funds for the implementation of retrenchments have been inadequate. Actual savings are therefore smaller than envisaged, but their magnitude cannot be documented due to inadequate information and piecemeal monitoring of improvements achieved by the system.

4.58 The sustained use of the systems depends to a large extent on factors outside the control of the project and DFID i.e.: institutionalising the project staff and procedures, adequate funding of operations, political and administrative commitments to enforce the use of the system. The full extent of commitment by political leaders will be shown by the actual use in the coming years of the PMEC system and the extent to which the work of the PMEC unit is effectively institutionalised into government structures.

Summary

4.59 The DFID support reviewed has made significant contributions towards the progress that Zambia has made in achieving HIPC completion and the conditions for receiving general budget support. Support to the Internal Audit Unit and to the development of payroll systems have provided potentially important steps in the strengthening of public management systems within the wider framework of public sector reform. However, except in ZRA, wider constraints to public sector performance have not

57 Up to £6m annually according to the Project Memorandum.


184

been addressed so that there are doubts about the sustainability of, and capacity fully to implement, the improvements that have been made in systems or the progress to which transactional support (such as that provided to BED) has contributed. DFID’s strategic approach to using TC to enable progress to be made towards budget support has been successful in enabling a move up the hierarchy of aid instruments discussed in section 2 above, but this progress has not been associated with a clear improvement in organisational capacity.


185


5.1 This chapter summarises the evaluation findings in terms of relevance, effectiveness, efficiency, impact and sustainability of output, discussing under each heading the influence of various aspects of the TC as it is carried through from conception to outcome. The evaluation matrix in Annex A provides summaries for each area of support.

Relevance

5.2 All the support provided was highly relevant in relation to the priorities of GRZ and addressed areas of importance for improving economic management in Zambia. For example the retention of the Long Term Technical Adviser in the Budget and Economic division was central to the objectives of Zambia being granted both HIPC completion and direct budget support. Similarly the assistance to the Internal Audit Unit and ZRA was directly sought by the organisations concerned and relevant to the building of their capacity in support of economic management.

5.3 The reservation in relation to the relevance of the support related to the limited extent to which it addressed key issues for organisational capacity development, in particular the lack of a capacity assessment in the design of the support to PMECP. In the case of support to MFNP, DFID recognised that they had been unable to analyse the main weaknesses in Zambia’s PEM system, or to develop a comprehensive programme of reform. Thus, DFID sought to ensure relevance by introducing a programme to respond to ad hoc requests for support, relying on the World Bank lead through PSCAP projects to provide the strategic framework for the PEM programme which could be brought on stream at short notice; this ensured that any immediate but unprogrammed needs could be met quickly. However, the lack of progress in the wider public service reform agenda under PSCAP (which DFID’s activities did not directly address) does suggest that a more relevant programme would have been more explicitly focused on the core constraints to capacity development.

5.4 RIZES was intended to develop an integrated system of revenue institutions, consolidating earlier work with ZRA on tax collection, while rebuilding a tax policy unit within the MFNP. This was an important, highly relevant, objective in the context of Zambia’s fiscal policy. In the case of ZRA it built on previous success and provided highly relevant advice, although in some instances the Authority was disappointed that the advice did not give them new insights. In the case of TPU the failure to resolve issues about the organisation’s role in the policy making process as well as internal organisational issues limited the relevance of the TC support provided.


186

Effectiveness

5.5 The effectiveness of the TC support provided varied. The support to BED and IAU was effective but is judged to be essentially transactional in character and not in itself developing organisational capacity. The support to TPU has been largely ineffective because of the unresolved issues about TPU’s role in the policy process, internal management issues, and a mismatch between the skills of the long-term TC provided and the Unit’s needs. However, it is possible that these uncertainties and management issues are being resolved so that there is scope for further progress. Support to ZRA has been largely effective, reflecting the fact that ZRA (as a result of earlier support and GRZ action that addressed key organisational issues) has reached a level of organisational effectiveness where it can effectively manage and use support to meet its needs. The support to PMECP has contributed to the establishment and use of the intended systems despite a number of problems with the design of the support and the quality of support provided. DFID activities were generally well harmonised with other donors.

Efficiency

5.6 Efficiency varied in terms of the cost of achieving the outputs of the support:

• The contract extension for the MFNP long term TC cost £36,700. This is a small price to pay for the contribution involved in assisting Zambia to achieve HIPC completion.

• Similarly the £26,100 spent on the Internal Audit Manual is cost-effective.

• In contrast, the expenditure on building the TPU was largely nugatory.

• In the case of PMECP, the TC support as a whole was not efficient. Estimates from interviews with the client suggest that perhaps no more than half of the adviser inputs (measured in time) were seen as useful. Some of these inputs were transactional and although these were appreciated, they must be regarded as expensive. Most of the assumed efficiency gain of providing long-term advisors on PMECP was undermined by the turnover among the advisors.

Impact

5.7 The impact of the DFID financed TC in Zambia is mixed in terms of the immediate recipients and uncertain in terms of wider economic management:

• The support for the long-term adviser in the Budget and Economic Division of the Finance Ministry had immediate impact – achievement of HIPC completion and the production of an MTEF and Activity Based Budget, but longer-term impact will depend on factors influencing the organisational capacity of the Ministry that the DFID support did not address.

• The Internal Audit Manual has improved the quality of internal audits but there is a question as to the extent they can be implemented with current civil service staffing of accounts teams.


187

• RIZES may have failed to build a tax policy unit but the last change management adviser made a useful contribution to the way ahead in terms of organisational structure and roles. It may be that groundwork has been laid for further progress, if issues of staff recruitment and more importantly retention can be addressed.

• PMECP’s impact is mixed. There have been improvements in GRZ’s capacity to make more effective budgeting, monitoring and control of GRZ employee numbers and PE expenditures. However, the PMEC does not yet cover the entire public service, some information compliance problems remain, and the full potential of the new system is not yet exploited.

Sustainability

5.8 The sustainability of the outcomes of the TC examined depends upon the extent to which fundamental constraints on the effective performance of the Zambian civil service are addressed:

• Any benefits of the long-term advisor’s work on Activity Based Budget and the MTEF could be lost if GRZ does not develop sufficient capacity to take it forward. The sustainability of the MTEF and ABB will depend on the development of financial management and administrative systems.

• The improvements in Internal Audit procedures may atrophy if the manual is not only unused but also is seen to be ineffective because the reports are not implemented.

• The impact of support for building the TPU will depend of how the recommendations of the final institutional review are handled; the short run failure of the project may be a precursor for longer-term success.

• There is no reason to doubt the sustainability of ZRA as an effective tax collection agency on the basis of its capacity.

• The payroll and establishment systems are likely to be sustainable only if difficulties in institutionalising the PMEC unit can be overcome and there is political commitment to using the system to achieve a fundamental strengthening of payroll control.


188


6.1 The record of aid in achieving capacity development in Zambia has historically been very poor and the country still provides a difficult context for aid effectiveness despite promising signs of change in both donor and government approaches to the use of aid. The capacity of key organisations has been eroded over many years and, while it appears that democratisation is deepening, the political system still presents obstacles to coherent policy making and to reforms that touch vested interests. Over the period under review, there has been an upturn in Zambia’s economic performance. However given Zambia’s past record of failed reforms and continuing political uncertainty (especially with elections due in 2006) the sustainability of these improvements remains to be established. The extent to which any substantial improvement in economic management has actually occurred remains to be demonstrated, despite the progress towards the end of the evaluation period that was associated with achieving HIPC completion and the agreement of the PEMFA framework.

6.2 A key lesson that donors have learned is that weak donor coordination has been an important factor in explaining the ineffectiveness of aid over the last two decades or more. There have been increasingly formalised attempts to improve coordination and to agree a stronger policy and aid management framework between donors and government. The Joint Assistance Strategy is an important step in providing this and progress is being made in the development of guidelines for the use of technical cooperation and for capacity development, but this process has only gained momentum towards the end of the evaluation period. The DFID activities reviewed have taken place within the context of the wider programme frameworks for reforms related to economic management that have been provided first by PSCAP and subsequently by PEMFA. The success of PEMFA will be critical for achieving sustained improvements in economic management.

6.3 DFID’s overall approach to the provision of TC can be characterised as seeking to use its comparative advantages in flexibility of financing (as with the PEM support) and established experience in some reform areas (such as the long record of support to ZRA and engagement in payroll reform) within a coordinated framework that has sought to support reforms for which there is wide agreement and signs of government commitment. The strategic approach has also been based on a clear hierarchy of aid instruments and an objective of moving to GBS, which is accepted by most of the major donors.

6.4 Overall, DFID’s support can claim achievements:

• Within the PEM area, highly cost-effective and strategically significant assistance in helping Zambia achieve HIPC completion and developing improved procedures for internal audit.

• Long-term support to ZRA that has contributed to a strong revenue performance given macroeconomic constraints.


189

• Support to payroll reform, which has, despite difficulties, moved forward the development of systems that have the potential to generate improvements in the effectiveness of public expenditure.

6.5 Activities have been least successful where (as with the TPU) there have been unresolved issues about the role and management of the organisations supported. In the case of ZRA issues about staff retention, management and incentives and the overall mandate and role of the organisation appear to have been successfully addressed. However, more generally, the lack of progress in addressing the key incentive and management constraints on civil service performance remains a binding constraint on capacity development within government.

6.6 The DFID activities reviewed have been limited in scope and have not generally sought to address these wider determinants of organisational capacity, although the piloting of a staff incentive scheme within PMECP is an exception. As shown in Table 3, reforms under PSCAP achieved only limited impact in addressing key constraints on public sector performance.

6.7 Given that substantial progress in improving economic performance and evidence of improved government commitment to reform is quite recent and comes only at the end of the period covered by the evaluation, it is too early to make judgements about whether greater impact will be achieved and current progress will be sustained. Much will depend on the extent to which government commitment to reform deepens, the moves towards coordinated donor action through the JASZ are effective and PEMFA and associated reform frameworks prove more successful than PSCAP in addressing fundamental constraints on economic management performance and capacity development.

6.8 Despite this fundamental dependence on the wider context and coordination with reform processes that may address constraints on capacity, the capacity development impact of the activities reviewed could have been strengthened (within their relatively limited scope) through a more explicit assessment of the context:

• In the case of the PEM support this might have involved more explicit attention to skills transfer in the terms of reference for the long-term adviser.

• In the case of PMECP, it might have involved a stronger focus on institutionalisation (incorporating staffing and functions into line positions in government) to move away from a project approach.

• In relation to the IAU support, additional training would have helped with skills transfer, but would not have addressed the main constraints on the adoption and use of the manual prepared which relate both to the poor staffing of accounts divisions (in relation to properly qualified public finance accountants) as well as commitment to internal audit functions.

• Greater involvement of the partner organisation in procurement and management might have increased the effectiveness and ownership of PMECP.

6.9 The larger strategic question is whether more progress could have been made in the public sector reform agenda under PSCAP, and whether there were feasible actions


190

by DFID that could have influenced this outcome. Addressing this question satisfactorily would require a wider focus for the evaluation, which would be more appropriately addressed within the framework of a country programme evaluation, which could consider the wider issues around the strategic choices made (including the overall scale and focus of the country programme) and a fuller review of the approaches and roles of donors in relation to the PSCAP agenda. It is plausible to suggest however that an approach that was more explicitly focused on addressing the core capacity constraints might have been more relevant to the context.

6.10 Taking a forward-looking view, the findings do however strongly suggest that DFID should encourage (through its own programme and through dialogue with government and other donors) a focus on addressing the wider institutional and organisational constraints, and that unless progress is made at this level, the scope for effective capacity development through the sort of activities that have been reviewed in this evaluation will be limited. Such an approach might include, for instance, supporting the strengthening of GRZ administrative processes in conjunction with the technically based aspects of PEMFA.


191

References

Bigsten, A., and S. Kayizzi-Mugerwa (2000), The Political Economy of Policy Failure in Zambia, Working Papers in Economics 23, Department of Economics, Goteborg University

Cabinet Office (2004). Payroll management and establishment control project (PMECP): request for no-cost extension, March.

Devarajan, S., D.Dollar and T.Holmgren (2001), Aid and Reform in Africa: Lessons from Ten Case Studies, World Bank, Washington DC.

DFID (1999), Zambia Country Strategy Paper, Department for International Development

DFID (2004), Zambia Country Assistance Plan, Department for International Development

DHG (2004), Final Report on the Appraisal of the Zambia PEMFA Reform Programme, Donor Harmonisation Group, August

Eberlei W, (2005), Donor politics in Zambia: Promoting poverty reduction or fuelling neopatrimonialism?, draft, March

Folscher A., (2004), The Design Process and Achievements of Zambia’s Poverty Reduction Strategy Paper, June.

GRZ (2004), Government strategy for the priority areas of the Public Service Reform Programme (PRSP) for the period 2004-2008. Draft, April

GRZ, World Bank, DFID, Norad (2005), “Public service management component, Public Service Reform Programme: Appraisal report.” March

GRZ (2005), Draft Aid Policy Paper.

IMF (2004), Zambia: Poverty Reduction Strategy Paper Progress Report, June 2004

IMF (2005a), Second Review Under the Three –Year Arrangement Under the Poverty Reduction and Growth Facility (PRGF) and Request for Waiver and Modification of Performance Criteria, and Financing Assurances Review, April 2005

IMF (2005b), Evaluation of the Technical Assistance Provided by the International Monetary Fund, Independent Evaluation Office, International Monetary Fund.

IMF (2005c), Staff Report, December 22 2005.

IMF (2006), Staff Report for the 2005 Article IV Consultation, International Monetary Fund

IMF and IDA (2000), Zambia Decision Point Document for the Enhanced Heavily Indebted poor Countries (HIPC) Initiative, International Monetary Fund and International Development Association, Washington, November


192

IMF and IDA (2005), Zambia Enhanced Initiative for Heavily Indebted Poor Countries Completion Point Document, International Monetary Fund and International Development Association.

Kiragu, K., and H. L. Mambo (eds) (2002), Public service reform comes of age in Africa, Dar es Salaam: Mkuki na Nyota Publishers

Kiragu, K., R.S. Mukandala & D. Morin (2004), Reforming pay policy: techniques, sequencing and politics” in Levy and Kpundeh (2004).

Levy. B., and S. Kpundeh (eds), 2004, Building state capacity in Africa: new approaches, emerging lessons. Washington: The World Bank

Mataka, R. H. (2002) “Zambia: capacity building for public sector performance.” In Public service reform comes of age in Africa, eds. K. Kiragu and H. L. Mambo. Dar es Salaam: Mkuki na Nyota Publishers

MFNP (2004) 2005 -2007 Medium Term Expenditure Framework and the 2005 Budget, Green Paper, Ministry of Finance and National Planning, Republic of Zambia, October

MFNP (2005) 2005 Budget Ministry of Finance and National Planning, Republic of Zambia, February

Mudenda, D., M.Ndulo, and M.Wamulume, (2005) The Budgetary Processes and Economic Governance in Zambia: A Literature Review, NEPRU Working Paper No.,104, Namibian Economic Policy Research Unit, Windhoek, December

Mutesa F., (2005), The nexus between public resources management reforms and neo-patrimonial politics, March

Mwanawina, I., (2003) Review of Previous Budget Submissions by Nonstate Actors, Economic Association of Zambia, October

OECD/DAC International Development Statistics (IDS) 2005

OPM (2003a), Drivers of pro-poor change: an overview, March.

OPM, (2003b, Poverty Oriented Budget Support Project (POBS): Draft Final Report, November 2003

OPM, (2005), Evaluation of Development Cooperation Ireland’s Country Strategy Paper for Zambia, May

PAF (2005). “Indicators for interim Performance Assessment Framework, 2004-2005. Dated April 1, 2005.

PWC, (1999). “Project management and delivery structures required to implement GRZ’s payroll management and establishment control project.” Feasibility study report, P PriceWaterhouse Coopers

Rakner, L., N. van de Walle, and D.Mulaisho (2001), Zambia, in Devarajan et al.


193

Republic of Zambia (2005), Road Map for the National Long Term Vision (2030) , Fifth National Development Plan (2005-11) and District Strategic Plans, Ministry of Finance and National Planning, June

Saasa, O., and J. Claussen (2004), Harmonisation of Donor Practices For Aid Effectiveness in Zambia, The Royal Netherlands Embassy Lusaka

Stevens, M. and S. Teggemann (2004), “Comparative experience with public sector reform in Ghana, Tanzania and Zambia” in Levy and Kpundeh (eds)

WDI (2005), World Bank Development Indicators.

World Bank (2002), Zambia Country Assistance Evaluation, Operations Evaluation Department, World Bank.

World Bank (2004), “Zambia. Public expenditure management and financial accountability review”. A World Bank Country Study. Washington: World Bank

World Bank and IMF (2003), Public Expenditure Management Country Assessment and Action Plan, Zambia, World Bank and International Monetary Fund, December 2003

World Bank and IMF (2005), Zambia: Poverty Reduction Strategy Paper, Second Annual Progress Report - Joint Staff Advisory Note, World Bank and International Monetary Fund.

World Bank, (2003), Zambia Public Expenditure Management and Financial Accountability Review, Washington DC: World Bank

World Bank, (2004a), Country Assistance Strategy for The Republic of Zambia, March

World Bank, (2004b), “Economic management and growth credit.” Programme document. Report no 29294 ZA. November 12, 2004

World Bank, (2005), Capacity building in Africa: an OED evaluation of World Bank Support, Operations Evaluation Department. Washington: The World Bank

Zambia National Commercial Bank, Annual Reports

Zambia Revenue Authority, Annual Reports 2000, 2001.

Zambia Revenue Authority, Corporate plans, 2002- 2004 and 2005- 2007


194



Mr Mutati Deputy Minister Ministry of Finance and National Planning

Benny Chundu Director: Economic and Technical Cooperation Department

Ministry of Finance and National Planning

Daniel Chisenda Head of Tax Policy Unit Ministry of Finance and National Planning

Chasiya V. Kazembe Principal Economist: Economic and Technical Cooperation Department


Dr Mbikusita W. Lewanika Permanent Secretary: Financial Management and Administration


David Ndopu Chief Economist Ministry of Finance and National Planning

Emmanuel Ngulube Deputy Director: Budget Office Ministry of Finance and National Planning

Ledson Zulu Controller of Internal Audits Ministry of Finance and National Planning

Ignatius Kashoka Permanent Secretary: Cabinet Office

Office of the President

Ackim Sakala PMEC Manager Office of the President

Berlin Msiska Commissioner General Zambia Revenue Authority

Muyangwa Muyangwa Commissioner VAT Zambia Revenue Authority

Wisdom Nhekairo Commissioner Direct Tax Zambia Revenue Authority

Lombe Ng’Andwe Commissioner Finance Zambia Revenue Authority

Andrew Chipwenda Chief Executive Zambia Privatisation Agency

Stuart Cruikshank Advisor Zambia Privatisation Agency

Patricia Palale Public Sector Management Specialist

World Bank

Mushiba Nyamanza World Bank

Ross Worthington Senior Public Sector Management Specialist

World Bank

Lise Lindback Second Secretary (Country Economist)

Royal Norwegian Embassy


195

True Schedvin Country Economist Embassy of Sweden

Charlotte Harland Independent Consultant

Peter Henriot Director Jesuit Centre for Theological Reflection

Tom Ngenda Executive Director Economist Association of Zambia

Jan Nijhoff In-Country Project Coordinator: Food Security Research Project

Michigan State University

Oliver Saasa University of Zambia

Tim Steele Technical Cooperation Officer Anti-Corruption Commission

Richard Montgomery Deputy Head of Office DFID Zambia

Alan Harding Economics Advisor DFID Zambia

Arthur Kalila Programme Officer: Economics and Private Sector Development

DFID Zambia

Chris Murgatroyd Governance Advisor DFID Zambia

Morgan Mumbwatasi Senior Programme Officer DFID Zambia

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

196

Su

mm

ary

of

Fin

din

gs –

Evalu

ati

on

Ma

tric

es

Re

leva

nc

e

Min

istr

y of

Fin

ance

an

d N

ation

al P

lann

ing

BE

DIA

UT

PU

ZR

A

HIR

P

Rele

vance

to

Org

anis

ational

needs

&

Opera

tional

conte

xt

Sup

port

to

Bud

get D

ivis

ion

cle

arl

y n

eed

ed t

o

get Z

am

bia

on

tra

ck for

HIP

C

com

ple

tion

Hig

hly

rele

vant.

M

ento

r fo

r dra

ftin

g o

f IA

U

manual

requeste

d b

y IA

U

Pote

ntially

re

leva

nt

but

TP

U

not o

pera

tio

nally

in

tegra

ted in

Bud

get

and

MT

EF

pro

cess

Hig

hly

re

leva

nt

to

ZR

A n

eeds

and incre

asin

g

capacity

Rele

vant to

support

of

PM

EC

U

nit,

but

desig

n

ove

restim

ate

d

the n

eed

for

TC

Consi

stency

of

DF

ID’s

and

org

anis

ation’s

obje

ctiv

es

TC

hig

hly

consis

tent

with

gett

ing Z

am

bia

on tra

ck f

or

HIP

C

com

ple

tion.

Consis

tent

with

D

FID

’s a

nd I

AU

obje

ctive

s

Obje

ctiv

e o

f T

PU

hig

hly

consis

tent

with G

RZ

obje

ctive

s

support

ed b

y D

FID

Hig

hly

consis

tent

with

ZR

A a

nd

DF

ID’s

obje

ctive

s

Consis

tent

with

obje

ctive

s f

or

eff

icie

nt te

chn

ical

sys

tem

for

Pa

y and

Esta

blis

hm

ents

contr

ol

Consi

stency

w

ith a

ctio

ns

by

GR

Z a

nd o

ther

Deve

lopm

ent

Part

ners

Hig

hly

consis

tent

with

oth

er

actio

ns t

ow

ard

s

HIP

C c

om

ple

tion

Manu

al r

olle

d o

ut

thro

ug

h G

RZ

.

Incre

ased

capacity

of

IA b

ut

IA c

an

not

be f

ully

used w

ith G

RZ

because o

f w

eak

capacity

in

accounts

Consis

tent

with

G

RZ

actio

ns n

ot

dir

ectly

rela

ted t

o

oth

er

don

ors

alth

oug

h

inconsis

tent

with

IM

F w

hic

h

treate

d Z

RA

as

tax p

olic

y un

it

Fully

consis

tent

Consis

tent

with

oth

er

don

ors

actio

ns a

lthou

gh

there

were

conflic

ts b

etw

een

PM

EC

Zam

bia

n

sta

ff a

nd s

om

e

TC

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

197

Min

istr

y of

Fin

ance

an

d N

ation

al P

lann

ing

BE

DIA

UT

PU

ZR

A

HIR

P

Consu

ltation

pro

cess

to

ensu

re

rele

vance

?

Yes T

C

responsiv

e t

o

MF

NP

Short

term

input.

Consultation n

ot

rele

vant.

Yes v

ery

fre

qu

ent

inte

ractio

n

betw

een M

FN

P,

DF

ID a

nd T

C

ZR

A left

to

dea

l w

ith T

C

consulta

ncy

firm

, Z

RA

appro

ached

DF

ID w

he

n it

seem

ed

appro

priate

to

rem

ove

pro

ject

manager

positio

n

Clo

se w

ork

ing

betw

een T

C a

nd

PM

EC

unit

alth

oug

h T

C d

ea

lt directly

with

DF

ID

on c

ontr

actu

al

issues a

nd

were

at tim

es s

een a

s

wo

rkin

g f

or

DF

ID

rath

er

that

PM

EC

and its

managin

g

bod

y

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

198

Eff

ec

tive

ne

ss

Min

istr

y of

Fin

ance

an

d N

ation

al P

lann

ing

ZR

A

HIR

P

BE

DIA

UT

PU

Were

outp

uts

g

enera

ted? I

f not,

why

not?

The r

eport

s

necessary

for

HIP

C c

om

ple

tion

were

dra

fted.

T

he M

TE

F a

nd

an A

ctivity

based

Bud

get

was

pro

duce

d.

Manu

al pro

duced

TP

U w

as n

ot

str

ength

ene

d.

TP

U d

id n

ot

pro

vid

e m

inis

ter

with

the k

ind o

f advic

e h

e s

ou

ght

i.e.

bri

efin

g o

n

advanta

ges a

nd

dis

ad

vanta

ges o

f pra

ctic

al optio

ns

for

new

taxes

Genera

lly

eff

ectiv

e,

part

icu

larl

y U

K

C&

E T

C

Eff

ectiv

eness m

ixed.

Technic

al s

yste

ms in

pla

ce a

nd

used b

y som

e o

pera

tion

al

team

in g

ove

rnm

ent

in M

FN

P

Was

ow

ners

hip

m

axi

mis

ed?

The G

ove

rnm

ent

cla

ims o

wn

ers

hip

of

these

docum

ents

, but it

is n

ot cle

ar

wheth

er

the

y ca

n

be g

enera

ted b

y Z

am

bia

n s

taff

and

adm

inis

trative

sys

tem

s in

the

absence o

f th

e

TC

.

Fully

ow

ne

d b

y IA

UN

o o

wners

hip

of

TP

U a

s e

merg

ing

from

TC

Yes. Z

RA

capab

le o

f ta

kin

g

what th

ey

need

ed

and w

ere

genera

lly

satis

fied

PM

CP

sta

ff f

ully

ow

n

sys

tem

, but it

is n

ot

yet fu

lly o

wned

thro

ug

hou

t G

ove

rnm

ent

Was

alig

nm

ent

maxi

mis

ed?

Lin

kages to

secto

r pla

nn

ing

weak

Exactly

wh

at

was

requir

ed

No, d

id n

ot

pro

duce w

hat

Min

iste

r w

ante

d

Larg

ely

wha

t Z

RA

wa

nte

d

Yes, pro

vid

ed w

hat

was n

ee

ded

on t

he

technic

al s

ide

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

199

Min

istr

y of

Fin

ance

an

d N

ation

al P

lann

ing

ZR

A

HIR

P

BE

DIA

UT

PU

Was

harm

onis

ation

maxi

mis

ed?

DF

ID a

ctivi

ties

were

harm

onis

ed

with

those o

f oth

er

don

ors

Yes

Yes, n

o o

ther

donor

su

pport

for

TP

U b

ut IM

F

contin

ued

to d

ea

l w

ith Z

RA

Yes, n

o o

ther

donors

Y

es, h

arm

onis

ed w

ith

PE

SC

AP

Were

outp

uts

g

enera

ted in

gender-

sensi

tive

wa

y?

Not re

levant

Not re

levant

Not re

levant

Not re

levant

Not re

levant

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

200

Eff

icie

nc

y

Min

istr

y of

Fin

ance

an

d N

ation

al P

lann

ing

ZR

A

HR

IP

BE

DIA

UT

PU

Were

outp

uts

g

enera

ted

eff

icie

ntly

in

term

s o

f cost

and t

imelin

ess

?

Report

s

pro

duce

d w

ithin

budg

et a

nd t

ime

and t

o

satis

factio

n o

f G

RZ

and d

on

ors

.

Manu

al pro

duced

on t

ime a

nd

with

in b

ud

get

No, pro

ject

aban

don

ed

No o

vers

pe

nd

alth

oug

h Z

RA

w

ould

consid

er

not a

ll in

puts

w

ere

valu

e f

or

mone

y

No. H

alf o

f T

C

consid

ere

d

ineff

ectiv

e.

Hig

h

turn

ove

r in

long

te

rm T

C m

eant

the

y w

ere

eff

ectiv

ely

short

te

rm .A

ll tr

ain

ing

was n

ot carr

ied

out

Would

there

have

been

altern

ative

&

more

eff

icie

nt

modes

of

deliv

eri

ng

sam

e

outp

uts

?

Not o

bvio

usly

. It

would

have

taken Z

am

bia

far

lon

ger

to a

chie

ve

HIP

C c

om

ple

tion

with

out th

is

support

.

No

Yes. S

tud

y vis

its

by

se

nio

r off

icia

ls

and M

inis

ters

to

oth

er

cou

ntr

ies

e.g

. to

HM

T (

UK

) could

ha

ve b

ee

n

used to

de

velo

p

more

specific

id

eas o

f w

hat

a

TP

U s

hould

do.

No, in

term

s o

f te

chnic

al

sys

tem

s, but Z

RA

w

ould

see d

irect

assis

tance f

rom

U

K C

&E

and

IR

m

ore

eff

icie

nt

than t

hro

ugh

inte

rmedia

ry

consulta

ncy

Less T

C!

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

201

Imp

ac

t

Min

istr

y of

Fin

ance

an

d N

ation

al P

lann

ing

ZR

A

HR

IP

BE

DIA

UT

PU

Did

pro

ject

s ach

ieve

exp

ect

ed

transa

ctio

nal/

transf

orm

ational

impact

?

Tra

nsactio

nal

impact achie

ved,

not cle

ar

wh

eth

er

Zam

bia

n

sys

tem

s c

an

readily

pro

duce

meanin

gfu

l M

TE

F a

nd A

BB

Achie

ved h

igh

qua

lity

revi

se

d

manual w

ith

impact on

eff

icie

nt

opera

tio

ns o

f IA

U b

ut

impact

weakened b

y w

eak c

apacity

in

accounting a

nd

possib

ly lin

e

Min

istr

y P

erm

anent

Secre

tari

es b

uy

in

No, pro

ject

aban

don

ed

Good

contr

ibu

tion t

o

ZR

A ‘s

im

pro

ving

capacity

to r

un

an e

ffic

ient ta

x

colle

ction

serv

ice

To a

larg

e e

xte

nt

yes b

ut d

ifficult

to a

ssess e

xte

nt

of

DF

ID

contr

ibu

tion

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

202

Su

sta

ina

bilit

y

Min

istr

y of

Fin

ance

an

d N

ation

al P

lann

ing

ZR

A

HR

IP

BE

DIA

UT

PU

Have

tr

ansf

orm

ational

impact

s been

sust

ain

ed?

Not know

n

Manu

al s

till

in

use b

ut

wid

er

instit

utiona

l w

eaknesses

reduce p

ote

ntial

impact

No im

media

te

impact but

mayb

e lessons

learn

ed f

or

the

fu

ture

.

No r

easo

n to

doub

t susta

ina

bili

ty

Too s

oon to s

ay

will

depe

nd o

n

instit

utiona

lisa

tion

of

PM

EC

Was

there

a

cohere

nt

exi

t st

rate

gy

pre

pare

d

in a

dva

nce

?

No. It

was

hope

d th

at th

e

lon

g-t

erm

advis

or

wo

uld

sta

y in

the

Min

istr

y w

ith

whic

h h

e h

ad

been

associa

ted

for

nearl

y tw

enty

ye

ars

.

Negotiatio

ns

ove

r fu

ture

contr

act

were

so

pro

longe

d th

at

he left

for

a

sim

ilar

post in

nearb

y A

fric

an

countr

y

Yes.

Short

inp

ut

No. P

roje

ct

aban

don

ed

Yes. Z

am

bia

n

CG

app

oin

ted a

s

schedule

d a

nd

ex-p

at acting

Com

mis

sio

ner

Genera

l le

ft

Arg

ua

bly

in

coh

ere

nt

as a

pro

ject exte

nsio

n

was r

equ

ired


203

STUDY OF POOLING ARRANGEMENTS FOR TECHNICAL COOPERATION


204

Executive Summary

S1. This study examines experience with Technical Cooperation (TC) pooling initiatives as a means of improving the effectiveness of TC (through increasing ownership, improving transparency and accountability, reducing administration costs, simplifying procedures and improving efficiency), in the context of the substantial impetus that has been provided to donor coordination and harmonisation/alignment initiatives in recent years. The study reviews pooling initiatives in Ethiopia, Mozambique and Tanzania to examine the extent to which these objectives have been achieved in particular instances and compares the findings with an earlier review on “The Pooling of Technical Assistance” by Baser and Morgan (2001).

S2. This review suggests that there has been significant progress in improving harmonisation and alignment driven in large part by moves towards providing general budget support and sector support compared to the situation reviewed in 2001. Joint strategy processes and related organisational mechanisms have been developed to strengthen coordination. Governments are increasingly articulating the objective of moving towards receiving aid through General Budget Support (GBS) and of strengthening capacity to procure and manage TC. Policies towards capacity development and the use of technical cooperation are also being developed.

S3. In terms of the TC funding arrangements used, there has been progress towards deeper pooling, but in practice different arrangements coexist in each of the countries and sectors examined. While significant progress has been made at the policy level in the approach to TC management, in each of the countries there are constraints to effective implementation, with recognition that capacity development in procurement and financial management in particular is required.

S4. Harmonisation as a means to increasing aid effectiveness has become a more dominant theme in partner donor relationships since the Baser and Morgan report. Memoranda of understanding for Joint Assistance Strategies focus primarily on achieving measurable and time-specific targets based on the indicators set out in the Paris Declaration and tend to be built around pooled funding as a path to full General Budget Support. TC is provided as part of sector support or as an adjunct to it, either under pooling arrangements or by single donors. The most recently established TC pooling initiatives draw on financing from targeted sector, sub-sector or cross-sector support and pools associated with General Budget Support. Consequently, they tend to be integrated into national planning and budgeting systems. There are few examples of pooling arrangements that are not complemented by additional bilateral support from pooled fund partners. This move from project to sector and budget support has increased the opportunity to engage in discussions and agreements on the harmonisation of donor practices in line with partner country policies, especially in terms of TC pooling.

S5. In general it appears that the move towards using pooled financing for supporting donor activities in the form of SWAps and GBS has improved alignment on partner countries’ national development strategies, institutions and procedures.

S6. Designing and implementing common arrangements is generally a great benefit to the partner country, however they can carry considerable costs in terms of time, resources and effort for development partners. It appears that most negotiations


205

revolve around the disbursement of funds and around whose procedures should be used for implementing management if the government systems are not to be used.

S7. What looks to have worked well in some of these initiatives are coordinated bilateral arrangements. Under this arrangement donors harmonise activities in terms of an effective pool, yet deal with the implementing management individually. This reduces the development partners’ costs of harmonisation.

S8. There seems to be an increase in transparency between donors around TC related issues under the various types of pooling arrangements. This is not necessarily so between donors and government, as donor relationships are managed through formalised structures with less frequent interaction than under bilateral arrangements. Therefore if reporting and monitoring is weak, transparency is weakened.

S9. Compared to the earlier assessment by Baser and Morgan in 2001, progress appears to have been made in several areas:




• This study has not been able to examine whether the introduction of these arrangements has led to a reduction in the quantity or cost of TC, though the arrangements may have led to an increase in the effectiveness of TC as a result of greater government ownership and more transparent selection procedures.


S10. The overall conclusion of the study is that while pooling initiatives have now begun to realise their potential for improving ownership, there remain important constraints on progress. Key strategic decisions tend to remain under joint (effectively donor) control, and there is a conflict between the ability to meet urgent demands (which can be effectively addressed by informal coordination arrangements or individual bilateral donors) and the process of capacity building for effective procurement and management of assistance. The scope for progress (and the priority to be placed on moving forward in this area, compared to lower cost and simpler processes of coordination and cooperation between donors) depends on the national context but the long-run determinants of success are likely to be the strength of government commitment to exercising more effective control over TC, and the level of success in developing organisational capacity for effective procurement and financial management.


206

1. Introduction

1.1 This study was undertaken to examine whether there is evidence that pooling arrangements for technical cooperation have the potential to lead to increased ownership by the governments and hence to increased effectiveness and efficiency in TC delivery. The purpose of the study was to complement the four country case studies and specifically to examine the hypothesis proposed in the Inception Report for the Evaluation that “TC is most likely to be effective in developing sustainable capacity for economic policy making and implementation when terms of reference for the role have been drawn up and the procurement process is managed by the host government, and processes of TC identification and management are harmonised between donors” since it was felt that the range of evidence from the case study countries was insufficient to address this question.

1.2 The study sought to:

• Update and review experience with TC pooling mechanisms in those countries where initiatives are underway.

• Clarify the differing objectives and definitions underlying pooling and related coordination and harmonisation initiatives related to TC.

1.3 The approach taken was to use as a starting point the study “The Pooling of Technical Assistance: An overview based on field research experience in six African countries” by Baser and Morgan, which had reviewed pooling experience in the period up to 2001, and seek to obtain additional evidence to see the extent to which subsequent experience might lead to a modification of their conclusions. Most of the examples reviewed related to economic management but in order to widen the range of experience some examples were taken outside this area where information was available.

1.4 The study is limited in scope. It was undertaken as a desk review58 of documentation supplemented by phone interviews. It was not possible within the resources available for the exercise comprehensively to review the experiences or to consult with all the stakeholders involved (including government representatives). The findings should therefore be regarded as suggestive only. The study was undertaken as a review of experience to cast light on one of the hypotheses for the study, not as an attempt to assess and evaluate DFID’s contribution to these initiatives. Such an evaluation would have required a much more detailed investigation, as would an attempt to review the effectiveness of the TC provided through the arrangements. As a result, no attempt is made to assess the initiatives against standard evaluation criteria. Instead, the benchmark taken is the extent to which the initiatives have made progress in improving harmonisation and alignment in line with the focus of the Paris Declaration’s assessment of the requirements for improving aid effectiveness. The assumption is that progress towards realising these principles is likely to be associated with improving aid effectiveness.

58 The review of experience in Mozambique was undertaken by a team member based in that country and

working closely with the initiative reviewed.


207

1.5 This study explores several pooling initiatives in Ethiopia, Mozambique and Tanzania59 to examine whether the above objectives have been achieved in these particular instances. The key issues are whether the benefits in terms of improved ownership, effectiveness and reduced transactions costs for government can be realised and at what cost in terms of donor long term objectives and administrative costs.

1.6 The study is structured as follows. Section 2 reviews the findings of the Baser and Morgan studies and highlights key issues that emerged from them including distinguishing types of pooling arrangement. Section 3 reviews policies towards harmonisation and alignment in the countries examined and the extent to which progress towards establishing effective pooling arrangements has been made. Section 4 examines issues emerging from the case studies and Section 5 summarises conclusions in relation to progress towards the Paris Declaration principles.

59 These are three of the six countries whose experience was reviewed in the Baser and Morgan study. The

others were Botswana, Mali and Uganda.


208

2. Framework and issues for the study

The changing international landscape for harmonisation

2.1 Baser and Morgan (2001) reviewed the experience with pooling initiatives during the period when donors and governments were beginning moves towards the use of sector wide approaches. Over the past five years, there have been substantial developments in (and increased international attention to) improving donor coordination as a means to improve aid effectiveness. There has been a move from standalone projects through sector wide approaches (SWAps) and other forms of basket funding towards the current approach of donors harmonising their activities around the Government’s national development plan, and financing through medium term expenditure frameworks (MTEFs)/annual budgets either on the basis of tied sectoral funds or General Budget Support (GBS).

2.2 At the same time the Rome and Paris Declarations60 have set out a wider view of the “Alignment and Harmonisation” agenda. The Rome Declaration committed donors to promote ownership of the development process by requiring donors harmonise and align aid delivery around partner country priorities. It looked forward to a stocktaking review in early 2005. The latter took place at the DAC High Level Forum, 28 February to 2 March 2005, which adopted the Paris Declaration on Aid Effectiveness.

2.3 The Paris Declaration had a greater emphasis on Partnership, putting partner countries rather than donors in the lead in the harmonisation process – see Figure 1. This was the culmination of a trend dating back to the mid nineties of shifting aid from a supply led to a demand driven process. Some earlier harmonisation initiatives had tended to put donors in the lead on harmonisation, reducing ownership and sustainability of the development process. The driving principle of the Paris Declaration is Partner Country ownership:

“14. Partner countries commit to:

• Exercise leadership in developing and implementing their national development strategies through broad consultative processes.

• Translate these national development strategies into prioritised results-oriented operational programmes as expressed in medium-term expenditure frameworks and annual budgets

• Take the lead in coordinating aid at all levels in conjunction with other development resources in dialogue with donors and encouraging the participation of civil society and the private sector.

60 Rome Declaration on Harmonisation February 24, 2003, Paris Declaration on Aid Effectiveness, March

2005


209

“15. Donors commit to:

• Respect partner country leadership and help strengthen their capacity to exercise it.”

2.4 Related commitments in the Declaration underpin this driving principle by emphasising capacity building, particularly in implementation of the planning, budgeting, financial management and budgeting systems.

Figure 1. Partnership Commitments under the Paris Declaration on Aid Effectiveness

Findings from the Baser and Morgan study

2.5 Initiatives for the pooling of funds for technical cooperation (TC) have been seen as a way to reduce the fragmentation of externally funded development activities, to encourage country ownership, and to reduce the considerable administrative burden on developing country governments. The Baser and Morgan (2001) study was based on reviews at field level and discussions with international development agencies and aimed to encourage linked theoretical thinking on improved aid management practice to the operational level. The study involved six country case studies (Botswana, Ethiopia, Mali, Mozambique, Tanzania, and Uganda).

Ownership: Partner countries exercise effective leadership over their development policies and strategies, and co-ordinate development actions

Alignment: Donors base their overall support on partner countries’ national development strategies, institutions and procedures

• Donors align with partners’ strategies • Donors use strengthened country systems • Partner countries strengthen development capacity with support from

donors • Strengthen public financial management capacity • Strengthen national procurement systems • Untie aid: getting better value for money

Harmonisation: Donors’ actions are more harmonised, transparent and collectively effective

• Donors implement common arrangements and simplify procedures • Complementarity: more effective division of labour • Incentives for collaborative behaviour • Delivering effective aid in fragile states • Promoting a harmonised approach to environmental assessments

Managing for results: Managing resources and improving decision-making for results

Mutual accountability: Donors and partners are accountable for development results


210

2.6 The objectives of the TC pooling initiatives as identified by Baser and Morgan (2001) are:

• Increased country ownership.

• Increased transparency and accountability.

• Reduced administrative costs.

• Simplified procedures and less bureaucracy.

• Increased efficiency.

2.7 The ultimate goal of TC pooling may be seen as the integration of TC into a single management or financing framework under recipient country ownership or with third party management support in line with the model of an ideal of TC management set out in Williams, G, S Jones, V Imber and A Cox (2003) which is discussed in detail in the Synthesis Report.

2.8 Baser and Morgan defined pooling as an activity which involves three characteristics:

• There must be some collective action among donors and national participants with the goal of producing a development result.

• There must be some sort of sharing or transfer of resources amongst the participants, and

• The pooling activity must be explicitly designed to support the agreed policy objectives of the partner country.

2.9 Baser and Morgan (2001, 2002) present an analytical framework based on the assumption that pooling is primarily a matter of managing a pool of funds set aside for technical assistance, identifying three broad categories of management mechanisms:

• Full TC pooling transfers all control to the government, including procurement, management and strategic control. The TC is fully untied.

• Mixed TC pooling: In mixed TC pooling, a third party handles procurement and personnel (e.g. a donor) and management and control are carried out by the government.

• Loose TC pooling: In loose TC pooling, tied or untied TC is supplied, contracted and managed by a donor, and the control is managed in liaison with the government.

2.10 They found that in four of the countries they examined there was an increase in the use of pooling – the exceptions being Mali (where donor collaboration is limited) and Botswana where aid resources had been fully integrated into national planning processes. They found however that “loose” pooling in their terminology was the commonest arrangement.


211

2.11 They also found that:

• Serious capacity shortages have slowed progress in TA pooling and although the goal of SWAps is to increase ownership it was frequently having the opposite effect because of the pressure imposed on capacity.

• The trend to more pooling was not yet having a major impact on national ownership.

• Small, context-aware and context-sensitive initiatives, following an incremental approach to reform stood a better chance of success than more ambitious schemes.

• The use of pooling was not reducing the amount of TA, or the costs of TA, with limited moves towards reducing tying and evidence of upward pressure on the cost of local consultancy skills.

• Efficiency gains from pooling were limited with most collaborative initiatives proving to be far more “labour intensive, time-consuming and organisationally complex than their advocates had predicted.”

• Pooling has the advantage of highlighting the comparative advantages and costs of different providers and so increases transparency.

• There was some evidence of correlation between the use of TA pooling and the crafting of better sector strategies and policies, but the impact on the development of broader programme strategies was not to be overestimated.

2.12 They concluded (2001, p.12) that their review:

“[h]as shown that pooling of TA is still in its infancy, although there are some modest but promising signs in several of the countries. The move to pooling is part of a broader change in the design and management of international development cooperation, at the core of which is the emphasis on country ownership and control. In the years ahead the challenge will be to encourage more pooling that promotes such ownership and control, while at the same time maintaining quality programming and effective implementation. At the same time, pooling of TA cannot be de-linked from structural reform efforts (e.g. public sector reforms). It also can only be implemented in an environment with adequate capacity to direct the process, or where the necessary capacities can be created within a reasonably short time span, e.g. through special incentives schemes to bring capable persons into the civil service”

Budget, procurement and management issues

2.13 Both the Baser and Morgan studies and the review of TC experience by Williams et al (2003) identified current budget and procurement systems as the main barriers to reaching the goal of strengthened country ownership.

2.14 As a result, the case studies carried out here have sought to examine how the various pooling mechanisms fit into the partner country’s planning, budgeting and financial management processes and the extent to which the pooling mechanisms contribute to Government ownership, long term sustainability of the development process and transition to the ideal of donor funds being fully integrated into the national planning and budgeting process. Figure 2 shows the alternative modes of provision of funds for TC.


212

2.15 The case studies have also examined management arrangements, distinguishing between:

• Strategic management: who decides what TC should be funded and how it is to be selected, and

• Implementing management: day to day management of finance, procurement and the TC.

2.16 The issues of financial management and procurement span both strategic and implementing management, depending on the stage of the process:

• The setting of the budget for TC and its allocation to individual areas or units of TC can be regarded as strategic; management of funding in the sense of accounting and auditing can be regard as implementation.

• The choice of procurement rules, design of TOR and final selection of bidders can be regarded as strategic, but the organisation of the process would be implementation.

2.17 Strategic management of TC critically depends on decision-making mechanisms. In the simplest case, which would demonstrate full ownership, the partner Government would decide what it wanted and how it would be procured. Where a Steering Committee is in control, chairmanship and the relationship between the chair and the secretariat will be key to determining whether the control rests with the Government of the donors. In terms of implementation, the choice over systems and reporting lines is critical to ownership: are they those of the government, a donor or group of donors.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

213

Fig

ure

2.

Fu

nd

ing

of

TC

in

re

lati

on

to

na

tio

na

l p

lan

nin

g a

nd

bu

dg

eti

ng

sys

tem

So

urc

e o

f fu

nd

sM

TE

F A

llo

ca

tio

ns

Go

ve

rnm

en

t O

wn

Fu

nd

ing

Ge

ne

ral

Bu

dg

et

Su

pp

ort

Se

cto

r B

ud

ge

t S

up

po

rt

Do

no

r P

roje

cts

in

teg

rate

d i

nto

MT

EF

Do

no

r P

roje

cts

ou

tsid

e M

TE

F

Pro

vis

ion

of

in-k

ind

TA

IDO

s p

rocu

rem

en

t, f

un

d a

nd

ma

na

ge

TA

usin

g

ow

n s

yste

ms a

nd

gu

ide

line

s.

To

tal

Go

vt

an

d D

on

or

Sp

en

d

Co

ve

rs f

ina

ncia

l a

ssis

tan

ce

as a

co

ntr

ibu

tio

n t

o

the

ove

rall

bu

dg

et

with

an

y c

on

ditio

na

lity f

ocu

se

d

on

po

licy m

ea

su

res r

ela

ted

to

ove

rall

bu

dg

et

pri

ori

tie

s.

With

in t

his

ca

teg

ory

fu

nd

s m

ay b

e

no

min

ally

acco

un

ted

fo

r a

ga

inst

ce

rta

in s

ecto

rs

bu

t th

ere

is n

o f

orm

al lim

ita

tio

n o

n w

he

re f

un

ds

ma

y a

ctu

ally

be

sp

en

t.

Co

ve

rs f

ina

ncia

l a

id e

arm

ark

ed

to

a d

iscre

te

se

cto

r o

r se

cto

rs,

with

an

y c

on

ditio

na

lity r

ela

tin

g

to t

he

se

se

cto

rs.

Ad

ditio

na

l se

cto

r re

po

rtin

g m

ay

au

gm

en

t n

orm

al g

ove

rnm

en

t a

cco

un

tin

g,

alth

ou

gh

th

e m

ea

ns o

f d

isb

urs

em

en

t is

als

o b

ase

d u

po

n

go

ve

rnm

en

t p

roce

du

res.

Ext

ern

al p

roje

ct

fun

din

g is a

no

n-d

iscre

tio

na

ry

reso

urc

e in

th

e s

en

se

th

at

su

ch

fu

nd

s c

an

on

ly b

e

sp

en

t o

n p

re-i

de

ntifie

d p

roje

ct

activitie

s a

nd

ca

nn

ot

be

re

-allo

ca

ted

.

Ext

ern

al p

roje

ct

fun

din

g is a

no

n-d

iscre

tio

na

ry

reso

urc

e in

th

e s

en

se

th

at

su

ch

fu

nd

s c

an

on

ly b

e

sp

en

t o

n p

re-i

de

ntifie

d p

roje

ct

activitie

s a

nd

ca

nn

ot

be

re

-allo

ca

ted

.

Co

nso

lid

ate

d

Fu

nd

pro

g/v

ote

1p

rog

/vo

te 1

pro

g/

vo

te 2

To

tal

Se

cto

r 1

Su

pp

ort

pro

g/

vo

te x

To

tal

Se

cto

r x

Su

pp

ort

To

tal

Se

cto

r 2

Su

pp

ort


214

3. Country experience: progress on donor harmonisation and TC pooling

Introduction

3.1 The move towards the coordination and harmonisation of donor activities in line with government policies has been challenging for donors and governments alike across Africa. Table 1 shows the extent of donor harmonisation in Ethiopia, Mozambique and Tanzania in 2004 in terms of direct Budget and Sector Support (SWAps). The amount of budget and sector support as a percentage of ODA in each of the three countries is close to the average of 40% for all the countries studied.

Table 1. Budget Support and Sector Support as a Percentage of ODA in 2004

Budget Support

(US$ millions)

Sector Support

(US$ millions)

ODA

(US$ millions)

Budget and Sector

Support as a % of ODA

Ethiopia 208 145 848 42%Mozambique 225 206 1050 41%Tanzania 405 196 1559 39%

Total 838 547 3457 40%Other countries included in the 'Total' figures are Benin, Burkino Faso, Ghana, Madagascar, Mali, Niger, Senegal, and Uganda

Source: OECD (2004)

3.2 Recent discussions on the management of aid in Ethiopia, Mozambique and Tanzania have revolved around several of the five elements of the Paris Declaration - ownership, alignment, harmonisation, managing for results and mutual accountability. Section 3 draws on government policies on harmonisation in Ethiopia, Mozambique and Tanzania and highlights any discussion on the pooling of TC in the context of the broader aid harmonisation agenda. Section 3 also looks at the extent to which current practices of pooling funds for TC adhere to the government’s vision on donor harmonisation. In discussion with DFID country offices we have identified the set of TC pooling activities as listed in Table 2. The analysis is based on documents provided by those offices and discussions with DFID and other donors in country.61

61 Particular thanks are due to the DFID country offices for the opportunity for discussions and access to

information and in particular to Alemayehu Minas, Education Adviser DFID Ethiopia, Laure Beaufils, Education Adviser DFID Ethiopia, Mieke Vogels, First Secretary Education Royal Netherlands Embassy – Chair of ESDP Donor Group, Simon van den Broek, DFID Mozambique, Kalayu Gebre-Selassie, DFID PSCAP, Ethiopia, Philip Courtnadge, Development Management Adviser, UNDP, Tanzania, Annabel Gerry, Governance Adviser, DFID Tanzania and John Piper, Economic Adviser, DFID Tanzania.


215

Table 2. Selected pooling initiatives in Ethiopia, Mozambique and Tanzania

Pooling Initiative Mandate Start Date

Ethiopia

Teacher Development Programme (TDP) To improve the quality of teaching at all levels, from primary through to tertiary

Mid-2004

Education Pooled Fund (EPF) To respond quickly and flexibly to the capacity development, research and implementation needs of the Ministry of Education (MOE)

End 2004

Public Sector Capacity Building Program (PSCAP)

To improve the scale, efficiency and responsiveness of public service delivery at the federal, regional and local levels

Early 2005

Mozambique

Strengthening of Sectoral Planning and Budgeting (FoPOS)

To address the continuing weaknesses in the planning and budgeting processes

2003

Tanzania

Public Financial Management Reform Programme (PFMRP)

To improve the financial management in the government including resource mobilization, budgeting and accounting

2003

Public Expenditure Review (PER) To facilitate transparency and participation of all stakeholders in the management of government expenditure

1998

Tax Administration Project (TAP) To strengthen the Tanzania Revenue Authority to collect and administer taxes

1999

Local Government Reform Programme (LGRP)

To facilitate the smooth administration of local government reforms

1998

Government policy on harmonisation, including TC pooling

Ethiopia

3.3 In 2002, the Ethiopian Government launched its harmonisation programme, based on the implementation of the Sustainable Development and Poverty Reduction Programme (SDPRP) is currently used as the PRS. Since then, several harmonisation initiatives, including TC for capacity building have been established, including:

• Participation in the International High Level Harmonisation Forum in Rome in February 2003 as one of the pilot countries. The national High Level Forum meets quarterly. It is chaired by the Minister of Finance and Economic Development and co-chaired by the head of the Development Assistance Group (DAG).

• Comprehensive donor-Government dialogue architecture for improved coordination.


216

• A common framework toward budget support with common conditionalities and performance indicators, and alignment of assistance with SDPRP.

• Institutionalisation of the harmonisation process through the creation of a Joint Government – Development Assistance Group Task Force on Harmonisation. Leadership of the harmonisation efforts are in the hands of the Ministry of Finance and Economic Development – Harmonisation Secretariat.

• A time bound Harmonisation Action Plan for harmonisation work undertaken at the national, sectoral and project levels, where efforts are focused on donor cooperation, financial management and accountability, procurement, M&E, and environmental safeguards.

3.4 The Development Assistance Group consists of development partners and aims to provide a forum for the discussion and coordination of donor activity. It is co-chaired by the United Nations Development Programme (UNDP) and the World Bank and the secretariat is housed in UNDP. Some donor agencies are currently discussing the idea of producing a joint strategy paper in support of the SDPRP. However, progress has been slow. The Memorandum of Understanding for the joint strategy paper remains in draft form, although there have been several MOUs signed for harmonisation in different sectors e.g. education and health. The draft MOU between donors and Government on harmonisation, alignment and aid effectiveness focuses primarily on achieving measurable and time-specific targets based on the indicators set out in the Paris Declaration.

3.5 The Government has also launched a national capacity building strategy under the direction of the Ministry of Capacity Building. The strategy articulates the need for government ownership and identification, coordination and management of capacity building initiatives including the use of TC.

Mozambique

3.6 There is no single document from the government of Mozambique setting out its strategy for aligning development assistance, but rather a series of bilateral agreements. The Government’s strategy for poverty reduction, the Plano de Acção de Redução da Pobreza Absoluta (PARPA) 2001-2005, provides the basis for partnership and provides the strategic framework necessary to increase aid effectiveness through greater coherence, harmonisation, and alignment with national systems. The relationship between government and the 15 budget support donors, formalised by a MOU, has provided a highly structured process for dialogue. The MOU establishes two joint reviews annually, focused on a performance assessment framework (PAF) indicating key outputs and actions over a three-year period.

3.7 The Government is taking steps to reduce fragmentation in capacity-building efforts. The UN - in collaboration with the Africa Capacity Building Foundation - is undertaking a capacity needs assessment to serve as a baseline for future monitoring. Several external partners, including the World Bank, are taking measures to support capacity building through a public sector reform programme.


217

Tanzania

3.8 In 1995, a set of operating rules and codes of practice to manage aid relationships were set up. In 2002, the Government launched a framework for managing foreign aid resources called the Tanzania Assistance Strategy (TAS). In mid-2004, the Government of Tanzania (GOT), DFID and the WB initiated a Joint Assistance Strategy (JAS), based on the Government’s Poverty Reduction Strategy and other programmes. The third version of the JAS Concept Paper was produced in March 2005 and is being discussed by the GOT and a representative group of development partners. The Government of Tanzania’s paper ‘The Harmonisation of Public Sector Reforms’ describes the JAS as follows:

“The Joint Assistance Strategy (JAS) is planned to move TAS to a higher stage of attaining national ownership and leadership in the development process, reduce transaction costs by enhancing harmonisation and alignment to national priorities and national systems.”

3.9 The Joint Assistance Strategy concept note states the GOT’s vision for the future development of the provision of technical assistance.

“Technical assistance will be used as one means for sustainable human and institutional capacity building. Development partners provide technical assistance that is demand-driven, untied from the source of financial assistance and procured openly by the Government, among others through arrangements of pooling TC. It will not be used for project or programme execution and will make the greatest possible use of national expertise, even though international experts may be recruited for augmenting Government capacity as a temporary gap filling measure particularly in highly specialized areas where qualified national personnel is unavailable. The recruitment, deployment, management, supervision and performance assessment of technical assistants will be led by the Government of Tanzania and integrated in the regular Government administrative system.”

3.10 The Harmonisation of Public Sector Reforms paper recommends that Government should formulate a national technical assistance policy, working closely with development partners and other stakeholders. In addition, government should also consider the option of comprehensive capacity building programme based on a capacity needs assessment of the country at large and the government in particular.

The provision of TC under the arrangements reviewed

3.11 In the cases reviewed, the provision of TC is either financed from a pool of funds for targeted sector, sub-sector or cross-government support, pool of funds for costed conditionalities associated with General Budget Support, pooled funds for TC alone and/or through complementary coordinated bilateral arrangements. Table 3 reflects the different way in which TC is funded for each pooling initiative.

3.12 The most recently established TC pooling initiatives are part of GBS or a SWAp-like pool of funds. The provision for TC in the Public Financial Management Reform Programme (PFMRP) in Tanzania is financed by a pool of funds associated with General Budget Support. The Strengthening of Sector Planning and Budgeting Programme (FoPOS) in Mozambique is part of targeted sector support to the Ministry


218

of Finance and Planning. In Ethiopia, the Teacher Development Programme (TDP) finances TC from funds used to support the education sector at a Federal Level. This can be thought of as a sub-sector pool of funds. The Public Sector Capacity Building Program (PSCAP) is a cross government pool, which includes funding for TC. This pool sits in the Ministry of Capacity Building.

3.13 In Tanzania the financing of TC within the Public Expenditure Review process (PER), the Tax Administration Project (TAP) and the Local Government Reform Programme (LGRP) are considered to have originated as donor projects and are now integrated into the National Budget. The Local Government Reform Programme was originally supported by separately funded but coordinated TC. A fund has been set up and is currently in the process of being transferred over to the President’s Office – Regional Administration and Local Government (PO-RALG). The Education Pooled Fund (EPF) in Ethiopia is the only example here of a donor project which sits outside the country’s public finance planning and management system.

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

219

Ta

ble

3.

So

urc

es o

f fu

nd

ing

fo

r p

oo

lin

g i

nit

iati

ve

s

As

so

cia

ted

wit

h G

BS

Pa

rt o

f d

ire

ct

se

cto

r,

su

b-s

ec

tor

or

cro

ss

-

se

cto

r p

oo

l

Do

no

r P

roje

ct

inte

gra

ted

in

to n

ati

on

al

bu

dg

et

Se

pa

rate

po

ol

for

TC

ou

tsid

e n

ati

on

al

bu

dg

et

Co

ord

ina

ted

bil

ate

ral

arr

an

ge

me

nts

Eth

iop

ia

Te

ach

er

De

ve

lop

me

nt

Pro

gra

mm

e

Ed

uca

tio

n P

oo

led

Fu

nd

Pu

blic

Se

cto

r C

ap

acity

Bu

ildin

g P

rog

ram

Mo

za

mb

iqu

e

Str

en

gth

en

ing

of

Se

cto

r P

lan

nin

g a

nd

Bu

dg

etin

g

Ta

nza

nia

Pu

blic

Fin

an

cia

l M

an

ag

em

en

t R

efo

rm P

rog

ram

me

Pu

blic

Exp

en

ditu

re R

evie

w P

roce

ss

Ta

x A

dm

inis

tra

tio

n P

roje

ct

Lo

ca

l G

ove

rnm

en

t R

efo

rm P

rog

ram

me

So

urc

e:

case

stu

die

s

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

220

Fig

ure

3.

Fu

nd

ing

of

TC

in

re

lati

on

to

Eth

iop

ia’s

na

tio

na

l p

lan

nin

g a

nd

bu

dg

eti

ng

sys

tem

So

urc

e o

f fu

nd

s

Po

oli

ng

of

fun

ds

for

TA

Bu

dg

et

All

oc

ati

on

s

Go

ve

rnm

en

t o

wn

fu

nd

ing

Ge

ne

ral

Bu

dg

et

Su

pp

ort

Na

tio

na

l S

ub

-Se

cto

r, S

ecto

r, C

ross-

Se

cto

r B

ud

ge

t S

up

po

rt

Do

no

r P

roje

cts

in

teg

rate

din

to N

ati

on

al

Bu

dg

et

Do

no

r P

roje

cts

ou

tsid

e B

ud

ge

t

Pro

vis

ion

of

in-k

ind

TA

Exte

rnal pro

ject

fundin

g is

a n

on-d

iscr

etionary

reso

urc

e

in t

he s

ense

that

such

funds

can o

nly

be s

pent

on p

re-

identified p

roje

ct a

ctiv

itie

s and c

annot

be r

e-a

lloca

ted

IDO

s pro

cure

, fu

nd a

nd m

anage T

A u

sing o

wn s

yst

em

s

and g

uid

elin

es

To

tal

Go

vt

an

d D

on

or

Sp

en

d

Covers

fin

anci

al ass

ista

nce

as

a c

ontr

ibution t

o t

he

overa

ll budget

with a

ny c

onditio

nalit

y f

ocu

sed o

n p

olic

y

measu

res

rela

ted t

o o

vera

ll budget

priorities.

Within

this

cate

gory

funds

may b

e n

om

inally

acc

ounte

d f

or

again

st

cert

ain

sect

ors

but

there

is

no f

orm

al lim

itation o

n w

here

funds

may a

ctually

be s

pent.

Covers

fin

anci

al aid

earm

ark

ed t

o a

dis

crete

sub-s

ect

or,

sect

or

or

sect

ors

, w

ith a

ny c

onditio

nalit

y r

ela

ting t

o t

hese

sect

ors

. Additio

nal se

ctor

report

ing m

ay a

ugm

ent

norm

al

govern

ment

acc

ounting,

although t

he m

eans

of

dis

burs

em

ent

is a

lso b

ase

d u

pon g

overn

ment

pro

cedure

s.

Exte

rnal pro

ject

fundin

g is

a n

on-d

iscr

etionary

reso

urc

e

in t

he s

ense

that

such

funds

can o

nly

be s

pent

on p

re-

identified p

roje

ct a

ctiv

itie

s and c

annot

be r

e-a

lloca

ted

Co

nso

lida

ted

F

un

d

MCB

pro

g/v

ote

xT

ota

l S

ecto

r x

Su

pp

ort

MC

B

TD

P

PS

CA

P

EP

F

MO

EO

the

r M

OE

Bila

tera

l T

C

TD

P P

oo

led

F

un

d

DF

ID

PS

CA

P

TA

Fu

nd

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

221

Fig

ure

4.

Fu

nd

ing

of

TC

in

re

lati

on

to

Mo

za

mb

iqu

e’s

na

tio

na

l p

lan

nin

g a

nd

bu

dg

eti

ng

sys

tem

So

urc

e o

f fu

nd

s

Po

oli

ng

of

fun

ds

for

TA

Bu

dg

et

All

oc

ati

on

s

Go

ve

rnm

en

t o

wn

fu

nd

ing

Ge

ne

ral

Bu

dg

et

Su

pp

ort

Na

tio

na

l S

ub

-Se

cto

r, S

ecto

r, C

ross-

Se

cto

r B

ud

ge

t S

up

po

rt

Do

no

r P

roje

cts

in

teg

rate

din

to N

ati

on

al

Bu

dg

et

Do

no

r P

roje

cts

ou

tsid

e B

ud

ge

t

Pro

vis

ion

of

in-k

ind

TA

Exte

rnal pro

ject

fundin

g is

a n

on-d

iscr

etionary

reso

urc

e

in t

he s

ense

that

such

funds

can o

nly

be s

pent

on p

re-

identified p

roje

ct

act

ivitie

s and c

annot

be r

e-a

lloca

ted

IDO

s pro

cure

, fu

nd a

nd m

anage T

A u

sing o

wn s

yst

em

s

and g

uid

elin

es

To

tal

Go

vt

an

d D

on

or

Sp

en

d

Covers

fin

anci

al ass

ista

nce

as

a c

ontr

ibution t

o t

he

overa

ll budget

with a

ny c

onditio

nalit

y f

ocu

sed o

n p

olic

y

measu

res

rela

ted t

o o

vera

ll budget

priorities.

Within

this

cate

gory

funds

may b

e n

om

inally

acc

ounte

d f

or

again

st

cert

ain

sect

ors

but

there

is

no f

orm

al lim

itation o

n w

here

funds

may a

ctually

be s

pent.

Covers

fin

anci

al aid

earm

ark

ed t

o a

dis

crete

sub-s

ect

or,

sect

or

or

sect

ors

, w

ith a

ny c

onditio

nalit

y r

ela

ting t

o t

hese

sect

ors

. Additio

nal se

ctor

report

ing m

ay a

ugm

ent

norm

al

govern

ment

acco

unting, although t

he m

eans

of

dis

burs

em

ent

is a

lso b

ase

d u

pon g

overn

ment

pro

cedure

s.

Exte

rnal pro

ject

fundin

g is

a n

on-d

iscr

etionary

reso

urc

e

in t

he s

ense

that

such

funds

can o

nly

be s

pent

on p

re-

identified p

roje

ct

act

ivitie

s and c

annot

be r

e-a

lloca

ted

Co

nso

lida

ted

F

un

d

MCB

pro

g/v

ote

xT

ota

l S

ecto

r x

Su

pp

ort

MC

B

TD

P

PS

CA

P

EP

F

MO

EO

the

r M

OE

Bila

tera

l

TC

TD

P P

oo

led

F

un

d

DF

IDP

SC

AP

TA

Fu

nd

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

222

Fig

ure

5.

Fu

nd

ing

of

TC

in

re

lati

on

to

Ta

nza

nia

’s n

ati

on

al

pla

nn

ing

an

d b

ud

ge

tin

g s

ys

tem

So

urc

e o

f fu

nd

s

Po

oli

ng

of

fun

ds

for

TA

Bu

dg

et/

MT

EF

All

oc

ati

on

s

Go

ve

rnm

en

t o

wn

fu

nd

ing

Ge

ne

ral

Bu

dg

et

Su

pp

ort

Na

tio

na

l S

ecto

r B

ud

ge

t S

up

po

rt

Do

no

r P

roje

cts

in

teg

rate

din

to N

ati

on

al

Bu

dg

et

Do

no

r P

roje

cts

ou

tsid

e B

ud

ge

t

Pro

vis

ion

of

in-k

ind

TA

Exte

rnal pro

ject

fundin

g is

a n

on-d

iscre

tionary

resourc

e

in t

he s

ense

that

such f

unds

can o

nly

be s

pent

on p

re-

identified p

roje

ct

activitie

s and c

annot

be r

e-a

llocate

d

IDO

s pro

cure

, fu

nd a

nd m

anage T

A u

sin

g o

wn s

yst

em

s

and g

uid

elin

es

To

tal

Go

vt

an

d D

on

or

Sp

en

d

Covers

fin

ancia

l assi

sta

nce a

s a

contr

ibution t

o t

he

overa

ll budget

with a

ny c

onditio

nalit

y f

ocused o

n p

olicy

measu

res

rela

ted t

o o

vera

ll budget

priorities.

Within

this

cate

gory

funds

may b

e n

om

inally

accounte

d f

or

again

st

cert

ain

secto

rs b

ut

there

is n

o f

orm

al lim

itation o

n w

here

funds m

ay a

ctu

ally

be s

pent.

Covers

fin

ancia

l aid

earm

ark

ed t

o a

dis

cre

te s

ecto

r or

secto

rs,

with a

ny c

onditio

nalit

y r

ela

ting t

o t

hese s

ecto

rs.

Additio

nal se

cto

r re

port

ing m

ay a

ugm

ent

norm

al

govern

ment

accounting, although t

he m

eans o

f

dis

burs

em

ent

is a

lso b

ased u

pon g

overn

ment

pro

cedure

s.

Exte

rnal pro

ject

fundin

g is

a n

on-d

iscre

tionary

resourc

e

in t

he s

ense

that

such f

unds

can o

nly

be s

pent

on p

re-

identified p

roje

ct

activitie

s and c

annot

be r

e-a

llocate

d

Co

nso

lida

ted

F

un

dPO

-RALG

pro

g/v

ote

xT

ota

l S

ecto

r x

Su

pp

ort

LG

RP

Ba

ske

t F

un

d

PF

MR

P

PE

R

LG

RP

TA

P

MO

FP

FM

RP

Po

ol

TA

P B

aske

t F

un

d

PE

R B

aske

t F

un

d

Bila

tera

l T

C


223

3.14 Six out of these eight initiatives are supported by additional coordinated bilateral support from pooled fund partners. This is largely due to governments and donors working to the ideal of full pooling while facing practical impediments such as weak procurement, financial management and monitoring systems. As result different combinations of TC funding modalities are implemented to satisfy both the development partners and partner country.

Ethiopia

3.15 The World Bank’s 2005 Comprehensive Development Framework assessment states that the most progress toward harmonisation has been through budget support. Structures have been put into place to strengthen government-donor dialogue. The 2004 OECD Survey on Harmonisation and Alignment found that 95% of donors supported the Government of Ethiopia’s agenda and were seeking to align their support to it. However the commentary also suggested that government and donors need to move from statements of intent to action and subsequently relations between donors and the Ethiopian government have become strained as a result of issues over human rights and the suppression of opposition political activity. The Government and donors have however been working on integrating the existing long-standing sector development programmes into an overall framework, which will facilitate general budget support for SDPRP.

3.16 The Teacher Development Programme in the education sector is an example of a government-owned initiative financed by pooled donor funds and housed within the Ministry of Education. While most TC activities are funded through a direct sub-sector support pool, there are several bilateral arrangements that supplement the pool. Government and donors aim to work towards establishing an environment where government systems are strong enough to manage implementation effectively and efficiently. In the meantime, there are incentives to using donor systems for activities that need to be implemented in a flexible manner within a short period of time, as demonstrated by the Education Pooled Fund. The aim is to move the Education Pooled Fund into the Ministry of Education (MOE) once there is enough capacity and the systems are strengthened to manage the fund effectively. In 2005, the Government launched a national capacity building programme to address the weaknesses in capacity across government. The initiative is government-led, financed by a pool of donor and government funds, implemented using government systems and World Bank procurement procedures.

3.17 The Development Assistance Group has developed a framework for donors to support the SDPRP. The framework coordinates implementation support and strategy studies and provides a forum for information sharing and arranging multi-donor reviews and joint missions. Responses to the OECD survey in 2003 indicate that approximately 200 donor missions were fielded, of which approximately 20% involved more than one agency. Since 2003, the Joint Budget Support group has followed a common framework and timing to undertake missions and engage in a dialogue with the Government.


224

3.18 Several common trends in the provision of TC seem to be emerging under PSCAP in Ethiopia. Prior to the pooling of funds, donor activity was fragmented; it did not take all of government’s needs into account; and there was a high dependence on international TC. Since the implementation of PSCAP, donor and government activity has been based on a government led plan of action:

• First, the Government seeks to request and organise donor assistance based on international best practice, country experience and comparative advantage. For example, the Government is keen to learn about the UK model for civil service reform and the German experience on higher education, construction and engineering. Such an approach maintains the need for some form of bilateral engagement within a coordinated framework. The PER process in Tanzania is seen as a successful example of such a coordinated approach.

• Secondly, the transition to this complex pooled fund arrangement has resulted in an increase in the need for TC to support the Government’s existing capacity to implement the programme.

• Thirdly, while international consultants are seen as valuable in many areas of reform, the Ethiopian government has expressed a desire to use local TC where it is known to deliver good results cost effectively.

• Fourth, some donor agencies see an advantage at least in the short-term in continuing bilateral arrangements to fund TC personnel as a way to bypass perceived bottlenecks as well as to stay informed.62

Mozambique

3.19 With all approaches to development aid represented in Mozambique, it is in direct budget and sector support that coordination processes are most developed. With other aid modalities, the role that the Government plays in donor coordination varies by sector. Overall, coordination initiatives appear to be mainly donor-driven.

3.20 Considerable efforts have been made to create a single pool for supporting capacity building in PFM through support to the public financial reform unit UTRAFE (Unidade Técnica de Reforma da Administração Financeira do Estado) and the rolling-out of SISTAFE, the new state financial administration system. Other capacity building efforts continue to take place outside these mechanisms. Respondents to the OECD 2004 Survey reported that several small and frequently overlapping technical support projects persist in budget formulation, execution, reporting and review and that no overarching plan exists to build capacity in this area. In 2003, the number of donor missions to Mozambique exceeded 130 (excluding the World Bank) and only 3% were undertaken jointly between donors. Furthermore, the respondents report that it is not clear that everyone realises the full extent of the challenges in the area of aid co-ordination and, to date, no plan has been developed to address critical weaknesses in this area, despite some suggested ideas.

62 E.g. the DFID PSCAP TA fund which is in addition to DFID’s contribution to the PSCAP pooled fund.


225

3.21 However issues relating to planning and budget formulation have not been as closely integrated into the design process for the new state financial administration system (SISTAFE) as might have been wished. In addition, the design process has to some extent militated against the adoption of a coordinated approach to planning and budgeting from across all the Directorates in the Ministry of Planning and Finance (MPF). With the on-going division of the MPF into two separate ministries for Planning and Finance, there is a danger than it will become more difficult to obtain a coordinated approach to planning, budgeting, budget execution and accounting.

Tanzania

3.22 According to the OECD Survey on Harmonisation and Alignment (OECD, 2005c), Government ownership and leadership over harmonisation appears to be limited and the effectiveness of the TAS appears to vary depending on modality and donor. The harmonisation agenda also does not appear to be supported by all donors. Government capacity weaknesses have been identified in public financial management, budget planning and execution, and aid coordination. In 2003, approximately 230 donor missions were fielded, with approximately 5% of all missions being undertaken jointly. It is likely that the number of missions undertaken jointly has increased as the recent agreements around pooled fund encourage greater harmonisation between donors.

3.23 The GOT’s harmonisation of public sector reform paper states that GBS is the Government of Tanzania’s preferred aid modality. Experience has shown that processes and systems which support common basket funds and directly funded projects have generally not been aligned to government practices, thereby creating parallel implementation and management systems and structures, and thus increasing transaction costs. Therefore, development partners are encouraged to see existing common basket funds as a transitional measure to GBS. The Local Government Reform Programme Basket Fund has been implemented by a management unit, which sits outside the PO-RALG. The programme is currently being transferred to the Ministry over a three-year period.

3.24 However, the harmonisation agenda is well established in Tanzania, and the support provided to the Poverty Reduction Strategy by many donors through budget support has helped to drive it forward. All 14 budget support development partners have agreed to the “Partnership Framework Memorandum”, which outlines the principles of co-operation and harmonisation and aims to reduce transaction costs for the Government. DFID Tanzania, as a Poverty Reduction Budget Support development partner, is keen to pursue full pooling initiatives. At present, the World Bank does not have an agreed position with GOT on the pooling of TC. Under the Public Sector Reform Programme, the World Bank has now joined the basket and therefore fully subscribes to the pooling arrangements including for TC. However, some contractual arrangements between GOT and technical advisors (financed through Bank credit) are running their course since they were part of the modality prior to the pooling of funds. WB credits finance TC but the recruitment and contracting is done by GOT.


226

3.25 Tanzania clearly has a vision of moving towards full pooling which is identified as funds to freely choose and manage TC using government systems. Government has been enthusiastic about full pooling following the ECDPM study in 2001. At the time they saw full pooling as a means to addressing the problems of ineffective TC. Full pooling initiatives have been either formed to pool together previously separate projects funded by different donors or have been formed from scratch. The latter tend to be in areas where the Government has shown a high degree of commitment during the design stage. This commitment is not necessarily followed through to the implementation stage.

Summary of progress

3.26 This review suggests that there has been significant progress in improving harmonisation and alignment driven in large part by moves towards providing general budget support and sector support compared to the situation reviewed in 2001. Joint strategy processes and related organisational mechanisms have been developed to strengthen coordination. Governments are increasingly articulating the objective of moving towards receiving aid through GBS and of strengthening capacity to procure and manage TC. Policies towards capacity development and the use of technical cooperation are also being developed.

3.27 In terms of the TC funding arrangements used, there has been progress towards deeper pooling, but in practice different arrangements coexist in each of the countries and sectors examined. While significant progress has been made at the policy level in the approach to TC management, in each of the countries there are constraints to effective implementation, with recognition that capacity development in procurement and financial management in particular is required.


227

4. TC pooling in Ethiopia, Mozambique and Tanzania

Range of pooling arrangements

4.1 Table 4 highlights the pooling initiatives chosen in each country along with the key characteristics, identified in Section 1 and the Baser and Morgan categorisation of pooling arrangements into full, mixed and loose. Baser and Morgan argued that most governments emphasise the importance of controlling the purpose and use of TC as opposed to its procurement and contracting. However, we find this is not necessarily the case and as a result there are few examples of arrangements that fall cleanly into the full, mixed, or loose categories.

4.2 In all countries studied, the long-term goal for the provision of TC appears to be full pooling that empowers the government to identify, select and manage TC based on its own assessment of need and using its own systems. The design of several recent pooling arrangements suggests that although donors are willing to support the use of government systems for implementing management, as an essential part of building their capacity in these fields, they do not yet want to relinquish a major role in strategic management issues. Therefore donors and government jointly lead strategic management, and the government can undertake the procurement and financial management. Under such arrangements, it is not obvious who – government or donors – take the upper hand in the decision making process. The CABRI report (2004) on budgeting systems suggested that in Mozambique it was the donors.

4.3 There are, however, several examples of bilateral donors supporting a fully pooled arrangement by taking responsibility under a coordinated framework to individually fund, procure and manage additional TC, e.g. the TDP in Ethiopia. These arrangements may occur because the pool cannot afford the cost of certain TC i.e. international TC or government systems may be too weak to procure and manage certain TC within tight time constraints. Additional bilateral arrangements are considered temporary by some donor staff who see full pooling as the end goal, and necessary by others who are not convinced that full pooling will lead to improved outcomes and therefore are reluctant to step back. In Baser and Morgan’s terms, this dual arrangement is neither a fully pooled fund nor a loosely pooled fund but rather a hybrid of the two.

4.4 There are several types of arrangement that could be categorised as loose pooling. These include:

• A separate TC fund under the strategic direction of a joint government-donor committee and managed by a nominated donor agency. E.g. Education Pooled Fund in Ethiopia,

• A pool of funds as part of direct sector, sub-sector or cross-sector support including SWAps or conditionalities associated with GBS, under the strategic direction of a joint donor-government committee, where each donor selects TC but these TC are


228

procured and managed by the team that manages the particular programme (which can be in government). E.g. Joint Review Mission, LGRP in Tanzania.

• A pool of donors, each willing to individually fund, select, procure and manage TC to undertake an activity from a list of activities that have been drawn up by a joint government-donor committee as has been the case with the PER process in Tanzania. The PER process is seen have worked fairly well in Tanzania, and

• Delegated co-operation arrangements by which one donor (a lead donor) acts on behalf of another donor (the delegating donor). For example, in Tanzania previously under the LGRP, DFID gave funds to Danida to procure TC. DFID has also given funds to USAID for the provision of TC to support the strengthening of Parliament..

Te

ch

nic

al

Co

-op

era

tio

n f

or

Eco

no

mic

Ma

na

ge

men

t V

olu

me

2:

Co

un

try

Ca

se

Stu

die

s

229

Ta

ble

4.

Ch

ara

cte

risti

cs

of

se

lec

ted

po

oli

ng

arr

an

gem

en

ts i

n E

thio

pia

, M

oza

mb

iqu

e a

nd

Ta

nza

nia

Cou

ntr

y l

ed

IDO

le

dS

ha

red

ro

le

Co

un

try

pro

cu

rem

en

t &

co

ntr

ac

tin

g

IDO

pro

cu

rem

en

t &

co

ntr

ac

tin

g

Eth

iop

ia

Te

ach

er

De

ve

lop

me

nt

Pro

gra

mm

eX

X*

XX

Fu

ll/L

oo

se

Ed

uca

tio

n P

oo

led

Fu

nd

XU

ND

P*

XX

Lo

os

e

Pu

blic

Se

cto

r C

ap

acity

Bu

ildin

g P

rog

ram

X -

WB

Pro

cu

rem

en

t

Pro

ce

du

res

XX

*X

XF

ull

/Lo

os

e

Mo

za

mb

iqu

e

Fo

PO

SX

XX

Fu

ll

Ta

nza

nia

Pu

blic

Fin

an

cia

l M

an

ag

em

en

t R

efo

rm P

rog

ram

me

XX

*X

XF

ull

/Lo

os

e

Pu

blic

Exp

en

ditu

re R

evie

w P

roce

ss

XX

*X

XF

ull

/Lo

os

e

Ta

x A

dm

inis

tra

tio

n P

roje

ct

XX

XX

Mix

ed

Lo

ca

l G

ove

rnm

en

t R

efo

rm P

rog

ram

me

XX

*X

XF

ull

/Lo

os

e

Ba

se

r a

nd

Mo

rga

n 2

00

1

Fu

ll P

oo

lin

gX

XX

Mix

ed

Po

oli

ng

XX

XX

Lo

os

e P

oo

lin

gX

XX

X

* A

lso

in

clu

de

s c

oo

rdin

ate

d b

ilate

ral a

rra

ng

em

en

ts o

uts

ide

po

ole

d f

un

d e

.g.

DF

ID P

SC

AP

TA

Fu

nd

Typ

e o

f

Po

oli

ng

Str

ate

gic

Ma

na

ge

me

nt

Imp

lem

en

tin

g M

an

ag

em

en

t

Tie

d T

AU

nti

ed

Aid


230

Strategic management

4.5 The strength and maturity of government institutions are critical in explaining the way in which pooling is managed. Full pooling can take place if these systems are weak but the pool is likely to be supplemented by additional coordinated bilateral arrangements. As these systems improve, Government takes more ownership and responsibility. For the TDP, PFMRP, and PSCAP, the project documentation clearly places the responsibility for managing the direction of the overall programmes with the Government.

4.6 However, in practice, decisions over the need of TC, what type of TC and which TC tend to be made in joint consultations with donors and government through the various committee structures. This may be because:

• Donors still hold and assert influential positions in the committees since they were heavily involved in the design stage and initiated many of the selected pooling initiatives.

• There is an incentive for government to take joint decisions with donors especially when this form of decision making can lead to additional resources being provided over and above pooled funds for example through bilateral arrangements

• Governments value the opinions, experience and expertise of donor agencies and while this knowledge transfer is still taking place, there is a role of donor agencies in strategic management issues.

Proposal and design of the pooling arrangement

4.7 Development partners proposed several of the initiatives studied. For example, in Mozambique under the FoPOS programme, the direct impetus for the establishment of a Technical Unit for the Reform of the State Financial System (UTRAFE), which formed the institutional basis for the shift to basket or pooled funding came from an IMF proposal in 2002, which had highlighted the need for a stronger coordinating mechanism to take forward the complex reforms implicit in the establishment of the state financial administration system.

4.8 In contrast, the PFMRP in Tanzania was developed by one of the current permanent secretaries in the Ministry of Finance and is therefore strongly supported as a concept by the top management in the main beneficiary institution. Donors were also very enthusiastic about the idea and wrote the idea into numerous appraisal documents and financing agreement for support to the Ministry (especially for general budget support). In both cases, it is difficult to draw conclusions on the impact of who proposed and designed the initiative on the ownership and direction of the programme without more information.

Joint government-donor steering committee/working groups

4.9 The effectiveness of the joint government-donor steering committees or working groups is dependent on the leadership and on the incentives for the various stakeholders to participate. Committee structures where the donor partners dominate


231

in number, chair the committees or where the secretariat is a donor agency may also affect the nature of government-donor relations. Bi-annual or quarterly joint committee meetings are generally seen to be more efficient for Government, which now only has to deal with one group of donors than with individual donors. However, a united group of donors may overwhelm the government. Such a risk has been raised in the Teacher Development Programme in Ethiopia. Here, all six Pooled Fund Partners (PFPs) attend the coordinating committee meetings and only the Vice Minister and the Head of the Teacher Development Department attend from the MOE. This risk exists despite the fact that the Vice Minister has been vocal and strong in getting his points across and PFPs generally agree to his requests. The Education Pooled Fund (EPF) is an example of where there is a small committee, which consists of one government official and two donor representatives. The UNDP acts as a secretariat. It is likely that the two development partners have a considerable say on the use of the fund.

4.10 According to the CABRI Mozambique case study (CABRI, 2004, p.84):

“Donor staff are proactive and engage in very detailed monitoring activities, providing proposals, undertaking studies and engaging in policy and technical discussions. This can overwhelm local staff and result in them feeling more accountable to donors, through this and other sectoral processes, than to Parliament.”

4.11 Similarly, a 2004 study of the budget process in Mozambique (Hodges and Tibana, 2004, p.89) found that

“There can be little doubt that the donors are at a considerable advantage relative to their national counterparts in their ability to marshal the expertise needed for analysis, policy formulation, negotiation, monitoring, evaluation and other tasks – a fact amply demonstrated by the presence of more than 100 aid officials and consultants at the joint government-donor review in March-April 2004, at which they greatly outnumbered government officials.”

Harmonising donor procedures

4.12 The costs to donor agencies of setting up joint schemes have tended to be high. In Tanzania, UNDP, working outside the PFMRP pooled fund, claims that their MOF counterparts are reluctant to bring the UNDP bilateral project into the pool due to the fact that the pooled fund is not fully functioning. There are high transaction costs in negotiating and agreeing the disbursement of funds and much energy and time has been absorbed around procedural issues. UNDP are satisfied with their current arrangement where the project works well; they have good relationships with their counterparts; they meet on a regular basis and can act quickly on request of the government. Nevertheless they feel that the working group works well in coordinating activities of pooled and non-pooled fund partners.

4.13 In Ethiopia TDP, DFID is currently chairing the PFPs and DFID’s view was that harmonising donor practices has been a time consuming exercise. There has been some tension among PFPs related to the release of funds (one partner having acted unilaterally on releasing funds), and this has taken time to manage. Financial matters have also been complex. The PFPs have to agree which agency releases how much (what percentage of earmarked funds) to the holding account at what period,


232

according to each donor’s own cycle and administrative requirements (i.e. some require audits and some require reports).

4.14 There are few examples of donors and the partner country developing a new set of harmonised procedures for procurement and financial management. In most cases government systems are used and where this is not the case a particular donors procedures will be used. In the latter case, often the pooled fund has to operate within the framework of the donor with the most restrictive rules. PSCAP progress reports covering the first six months have shown that one of the biggest challenges implementing agencies have faced has been adhering to the World Bank’s procurement procedures. During the design of the program, several donors were reluctant to use the World Bank’s procurement procedures given that they are known to be bureaucratic, time consuming and complex. However given the dominance of the World Bank in terms of relative contribution to the pooled fund and their experience in implementing PSCAP-type programs, World Bank regulations are being used.

4.15 Even where there is sufficient involvement of the host government to be considered strategic, efficiency gains still appear to be mixed. Most collaborative arrangements involving full pooling have proven to be more labour-intensive, time-consuming and organisationally complex than was thought to be the case during the design stage.

Implementation management

4.16 The management practices on implementation vary between pooling arrangements and have been conditional on the drive for government ownership and the complexities of the pooling initiatives.

Using existing government systems

4.17 Government systems for procurement and financial management are currently being used in most of the selected pooling initiatives. There are few examples of where these systems are strong enough fully to manage TC. Furthermore, the long and bureaucratic procurement procedures create incentives to extend existing TC contracts. It is recognised that these systems will lead to sub-optimal decisions about the procurement and management of TC and therefore donors are working with the government to strengthen systems. The Education Pooled Fund in Ethiopia is an example of where government systems are not being used. This fund is designed to be flexible, transparent and able to respond quickly to needs as identified by the steering committee. It is not seen to be integral to the core business of the Ministry of Education unlike the Teacher Development Programme and therefore sits outside the Ministry of Education. The existence of this particular pool for TC provision raises issues about the need for TC in addition to the pooling arrangements in place and whether external management units are appropriate structures.

Capacity gaps

4.18 Coordination mechanisms are intended to put the recipient government into the driver’s seat, but often additional TC is required to address the capacity gaps.


233

PSCAP in Ethiopia is a complex programme. The complexity of the management of the pool may be beyond the capacity of the government and require external additional TC to build that capacity, but this needs to be carefully managed. There may for example be a need for advisory and change management, and support to financial management and procurement functions of the implementing agencies. In Tanzania, for several years the Ministry of Finance believed that the PFMRP could be managed as part of the ordinary structure of the Ministry. However the MOF did not have the capacity and the departmental independence to resolve the difference between departments, and between individual donor projects (mostly with a specific departmental focus) and bring them in line under a common approach. Gradually the budget support donors increased pressure on the Ministry to employ a programme coordinator, but this process (which lasted two years) meant that the initiative shifted from the Government to the donors. As a result the programme involved an integrated structure to cover the capacity building needs of all the departments of MOF, had its own dedicated management structure, and was funded through a multi-donor common basket funding mechanism.

External management units

4.19 Using a third party agent to undertake operational management can also have constraints. The recent evaluation of GBS in Tanzania (Daima Associates and ODI, 2004) argued that the adoption of the integrated PFMRP has been driven by the move towards the provision of direct budget support. The Tanzanian Government did not see the PFMRP as such an important priority as did the PRBS Development Partners, and arguably such a programme would not have been put in place in the absence of GBS. Moving the programme out of the MOF, where capacity did not exist, failed to encourage or enhance government ownership. The programme has moved back into the structure of the MOF.

4.20 The EPF in Ethiopia provides another example of the constraints of working with external management units. There were a number of teething problems with the UNDP’s management early in the process. The contracts were not well managed and processes were delayed. Although this was temporarily resolved, the person working with at UNDP has now left and there may be a management vacuum. There has been some negative feedback about the long time taken between the acceptance of the proposal by the committee and the contracting of implementers by UNDP. UNDP has also found that managing the fund is more resource intensive than initially thought, and this arrangement may not continue. In the medium-term, it is hoped that the fund will be transferred to the Ministry of Education.

Monitoring and evaluation

4.21 The information generated from weak government financial management systems is rarely adequate to allow pooled fund partners to monitor progress. Some donor staff suggest that this is another reason for having bilateral arrangements in place. In this way, donors can have greater access to information.

4.22 The harmonisation agenda has encouraged donors to conduct joint monitoring and evaluation missions. A pool of funds for TC may not necessarily do this, but it may


234

ease negotiations around forming joint monitoring and evaluation teams as team members can be selected, procured, and managed in line with the pooling arrangements. The 2001 Joint Mid-Term Review of the Local Government Reform Programme in Tanzania highlights the benefits of loose TC pooling for arranging such reviews. All team members were financed from the basket fund. Each donor contributed to the mobilisation of expertise. This proved to be a distinct advantage, making all donors more accepting of the findings of the team. Donor representatives were not included, on the grounds that this would affect its independence. The experts had confidence in that solution because they had been identified by the donor agencies directly and, therefore, felt reassured of the contractual arrangements. It also meant that all team members were given similar contractual conditions helping to create good team spirit.


235

5. Conclusions

5.1 Harmonisation as a means to increasing aid effectiveness has become a more dominant theme in partner donor relationships since the Baser and Morgan report. Memoranda of understanding for Joint Assistance Strategies focus primarily on achieving measurable and time-specific targets based on the indicators set out in the Paris Declaration and tend to be built around pooled funding as a path to full General Budget Support. TC is provided as part of sector support or as an adjunct to it, either under pooling arrangements or by single donors. The most recently established TC pooling initiatives draw on financing from targeted sector, sub-sector or cross-sector support and pools associated with general budget support. Consequently, they tend to be integrated into national planning and budgeting systems. There are few examples of pooling arrangements that are not complemented by additional bilateral support from pooled fund partners. This move from project to sector and budget support has increased the opportunity to engage in discussions and agreements on the harmonisation of donor practices in line with partner country policies, especially in terms of TC pooling.

5.2 We have examined our findings in relation to the three most relevant of the principles of the Paris Declaration on Aid Effectiveness:

• Ownership.

• Alignment.

• Harmonisation.

Ownership

5.3 Baser and Morgan argued that most governments emphasise the importance of controlling the purpose and use of TC as opposed to its procurement and contracting. This study in so far as there was evidence available, suggests that current practices appear to be resulting in the government taking control of the procurement and financial management aspects of TC management even though systems are weak. However, joint government-donor committees have heavily influenced strategic management decisions so ownership cannot be seen as complete.

Alignment

5.4 In general it appears that the move towards using pooled financing for supporting donor activities in the form of SWAps and GBS has improved alignment on partner countries’ national development strategies, institutions and procedures. This is most evident in Tanzania where DFID is very keen to use the support the PRBS process and to use government’s institutions and procedures where possible. However the GBS study on Tanzania (Daima Associates and ODI, 2004) commented:

“The development of an integrated Public Finance Management Reform Programme (PFMRP) represents something of a departure from the earlier pattern of TA use. It involves


236

an integrated structure to cover the capacity building needs of all the departments of MOF as well as the CAG’s office; it has its own dedicated management structure, a multi-donor common basket funding mechanism and a long term resident co-ordinator, who is a Tanzanian public finance specialist. The finalisation of the PFMRP action plan and management structure was a prior action within the PAF and the failure to complete this action by the due date of March 2004 led to a delay in the disbursement of the PRSC.”

5.5 It seems that the Tanzanian Government did not see the PFMRP as such an important priority as did the PRBS Development Partners, and arguably such a programme would not have been put in place in the absence of GBS. Whether it will speed up the process of PFM improvements, only time will tell. Clearly, it responds to a belief by the PRBS development partners that an accelerated rate of improvement in PFM can be achieved through a more substantial level of TA and a more formalised process of coordination. This may be true and the Tanzania Revenue Authority could be quoted as an example of the potential of such a strategy. On the other hand, it is not entirely consistent with the pattern of organisational strengthening successfully pursued in the Ministry of Finance over the past decade.

5.6 Ethiopia and Mozambique seem to be following in the same direction, however the lack of clear policy statements compared to Tanzania do not allow for the same degree of alignment. The extent to which public expenditure proposals are scrutinised and genuinely questioned continues to be limited. The Tanzania GBS evaluation refers to the relative weakness of the “budget challenge function” within the planning units of sector ministries, the Budget department of the MOF and within Parliament. The reform of planning and budgetary processes in Mozambique, which forms the core objective for the FoPOS continues to suffer from the lack of a well-articulated reform plan which has the command of key stakeholders (National Directorate of Planning and Budget, the other Directorates of the Ministries of Finance and Planning, the IMF and World Bank, the sectoral ministries). As a result the FoPOS project has, over its duration, not been able to effectively support a reform process. The switch to a pooling mechanism could not in itself resolve this underlying problem.

5.7 Coordinated bilateral TC is particularly effective for short assignments that can be commissioned pieces of work and need to be undertaken with little notice. Furthermore individual donors may not have a regular source of funding for these activities, and therefore one donor may fund a bilateral arrangement one year, and another donor may fund it the following year. Donors in the Education Pooled Fund in Ethiopia use these arrangements especially for Joint Review Missions as they are flexible and practical. Where procurement and financial management systems are weak, additional support tends to be provided to strengthen these systems. An example of such support is the World Bank funded procurement specialist in the MOF in Tanzania. Bilateral arrangements can also be a way in which the development partners monitor progress on the ground, especially if government monitoring systems are not credible. It is unclear what affect this type of additional TC has on promoting ownership. As mentioned earlier, additional bilateral arrangements are considered temporary by some donor staff that see full pooling as the end goal and necessary by others who are not convinced that full pooling will lead to improved outcomes and therefore are reluctant to step back.


237

Harmonisation

5.8 The pooling of funds creates an incentive for donors to harmonise procedures and practices. Designing and implementing common arrangements is generally a great benefit to the partner country, however they can carry considerable costs in terms of time, resources and effort for development partners. It appears that most negotiations revolve around the disbursement of funds and around whose procedures should be used for implementing management if the government systems are not to be used.

5.9 What looks to have worked well in some of these initiatives are coordinated bilateral arrangements. Under this arrangement donors harmonise activities in terms of an effective pool, yet deal with the implementing management individually. This reduces the development partners’ costs of harmonisation.

5.10 There seems to be an increase in transparency between donors around TC related issues under the various types of pooling arrangements. This is not necessarily so between donors and government, as donor relationships are managed through formalised structures with less frequent interaction than under bilateral arrangements. Therefore if reporting and monitoring is weak, transparency is weakened.

Overall assessment

5.11 The specific country context plays a significant role in the choice and the experience of the different types of pooling. The effectiveness of the arrangement has depended on the relationships between the personnel involved (donors, government and TC) and the quality and effectiveness of existing institutional arrangements to facilitate the management of aid.

5.12 Compared to the earlier assessment by Baser and Morgan in 2001, progress appears to have been made in several areas:

• The progress towards the provision of direct budget support (and the attention towards harmonisation and alignment issues in the Rome and Paris Declarations) has provided a strong impetus towards strengthened donor coordination around government led programmes.





238

• This study has not been able to examine whether the introduction of these arrangements has led to a reduction in the quantity or cost of TC, though to the extent that government ownership has increased and selection procedures have become more transparent, there are reasons to think that the arrangements may have led to an increase in the effectiveness of TC.


5.13 The overall conclusion of the study is that while pooling initiatives have now begun to realise their potential for improving ownership, there remain important constraints on progress. Key strategic decisions tend to remain under joint (effectively donor) control, and there is a conflict between the ability to meet urgent demands (which can be effectively addressed by informal coordination arrangements or individual bilateral donors) and the process of capacity building for effective procurement and management of assistance. The scope for progress (and the priority to be placed on moving forward in this area, compared to lower cost and simpler processes of coordination and cooperation between donors) depends on the national context but the long-run determinants of success are likely to be the strength of government commitment to exercising more effective control over TC, and the level of success in developing organisational capacity for effective procurement and financial management.


239

References

Baser, H., and P. Morgan (2001), The Pooling of Technical Assistance: An overview based on Field Research Experience in Six African Countries, ECDPM for the Ministry of Foreign Affairs of the Netherlands.

Baser, H., and P. Morgan (2002), Harmonising the Provision of Technical Assistance: Finding the Right Balance and Avoiding the New Religion, ECDPM.

CABRI (2004), Case Studies on Ethiopia, Mozambique and Tanzania, Collaborative African Budget Reform Seminar, December

Daima Associates Limited and ODI (2004), Joint Evaluation of General Budget Support – Tanzania 1995-2004. November

DFID (2005), Discussion Paper: A Long-Term Joint Strategy for Capacity Strengthening, Mozambique, Department for International Development, Maputoo

Hodges, T. and R. Tibana (2004) Political Economy of the Budget in Mozambique, Oxford Policy Management for DFID

Imber, V., C. Byaruhanga and F. Ronsholt (2004), Government of Rwanda Joint Review of Support for the Ministry of Finance and Economic Planning, Oxford Policy Management

OECD (2004), Survey of the Alignment of Budget Support and Balance of Payments Support with National PRSP Processes (2004), OECD-DAC Statistics.

OECD (2005a), Survey on Harmonisation and Alignment: Ethiopia

OECD (2005b) Survey on Harmonisation and Alignment: Mozambique

OECD (2005c), Survey on Harmonisation and Alignment: Tanzania

OECD (2005d), OECD/DAC joint progress toward enhanced development and effectiveness, harmonisation, alignment, and results

Pavignani, E. and V. Hauck (2002), Pooling of Technical Assistance in Mozambique: Innovative Practices and Challenges, ECPDM

UNDP (2002), Development Policy Journal: Special Issue Technical Cooperation, vol. 2 December

UNDP (2004), Issue Paper for UNDP: Official Development Assistance as Direct Budget Support

Wangwe, S. and L. Madete (2002), Pooling of Technical Assistance in the context of Sector Wide Approaches in Tanzania, ECDPM

Williams, G., S. Jones, V. Imber and A. Cox with A. Balihuta and A. Revi (2003), A Vision for the Future of Technical Assistance in the International Development System, Oxford Policy Management for DFID


240

Worku, S. (2002), Pooling of Technical Assistance in the context of Sector-Wide Approaches (SWAps): Ethiopian Case Study

World Bank (2005), Enabling Country Capacity to Achieve Results, CDF Progress Report, Washington DC.


241

Synthesis References

Baser, H., and P. Morgan (2001), The Pooling of Technical Assistance: An overview based on Field Research Experience in Six African Countries, ECDPM for the Ministry of Foreign Affairs of the Netherlands. European Centre for Development Policy Management, Maastricht http://www.ecdpm.org/Web_ECDPM/Web/Content/Navigation.nsf/index2?readform&http://www.ecdpm.org/Web_ECDPM/Web/Content/Content.nsf/80ba021853007405c1256c790053145c/42c720dbf697d1ccc1256caa0033eb71

Baser, H., and P. Morgan (2002), Harmonising the Provision of Technical Assistance: Finding the Right Balance and Avoiding the New Religion, ECDPM Discussion Paper 36, European Centre for Development Policy Management, Maastricht. http://domino.ecdpm.org/Web_ECDPM/Web/Content/Navigation.nsf/index2?readform&http://domino.ecdpm.org/Web_ECDPM/Web/Content/Content.nsf/80ba021853007405c1256c790053145c/a16f40945c97e9d2c1256c7e00337429?OpenDocument

Berg, E., (1993), Rethinking Technical Cooperation, UNDP, New York.

Bloch, G., M. Favis, and J. Hargovan (2000), Development Cooperation Report: Evaluation of ODA to Capacity Building, Department of Finance, International Development Cooperation, South Africa http://www.treasury.gov.za/documents/dcr/capacity.pdf

Browne, S., (ed) (2002), Developing Capacity through Technical Cooperation: Country Experiences, Earthscan/UNDP

Commission for Africa (2005), Our Common Interest: Report of the Commission for Africa, Commission for Africa, March http://www.commissionforafrica.org/english/report/thereport/english/11-03-05_cr_report.pdf

Devarjan, S., D. Dollar and T. Holmgren (eds) (2001), Aid and Reform in Africa: Lessons from Ten Case Studies, The World Bank: Washington DC http://www.worldbank.org/research/aid/africa/release/aid.htm

DFID (2003a), Promoting Institutional and Organisational Development: A Source Book of Tools and Techniques, Department for International Development http://www.dfid.gov.uk/pubs/files/prominstdevsourcebook.pdf

DFID (2005b), Why we need to work more effectively in fragile states, Department for International Development, January, London, UK. http://www.dfid.gov.uk/pubs/files/fragilestates-paper.pdf

Fukuda-Parr, S., C. Lopes and K. Malik (eds) (2002), Capacity for Development: New Solutions to Old Problems, Earthscan/UNDP

IDD (2006), Joint Evaluation of General Budget Support 1994-2004: Synthesis Report, International Development Department, School of Public Policy, University of Birmingham, January http://www.oecd.org/dataoecd/42/38/36685401.pdf


242

IMF (1999), Review of Fund Technical Assistance: From Technical Assistance to Technical Consultation and Cooperation, International Monetary Fund, Washington DC.

IMF (2005), Evaluation of the Technical Assistance Provided by the International Monetary Fund, Independent Evaluation Office, International Monetary Fund, January 31st,Washington DC, USA http://www.imf.org/external/np/ieo/2005/ta/eng/013105.htm

Levy, B. (2004), Governance and Economic Development in Africa: Meeting the Challenge of Capacity Building, in Levy and Kpundeh (2004)

Levy, B., and S. Kpundeh (eds) (2004), Building State Capacity in Africa – New Approaches, Emerging Lessons, WBI Development Studies, World Bank Institute, Washington DC, USA

Morgan, P., (2002), Technical Assistance: Correcting the Precedents, Development Policy Journal, Volume 2, December

Morgan, P., T. Land and H. Baser (2005), Study on Capacity, Change and Performance: Interim Report, Discussion Paper No 59A, European Centre for Development Policy Management, Maastricht, April http://www.ecdpm.org/Web_ECDPM/Web/Content/Navigation.nsf/index2?readform&http://www.ecdpm.org/Web_ECDPM/Web/Content/Content.nsf/vwDocID/F14147635159014CC1256FE4003E2026?OpenDocument

OECD (2005), The Paris Declaration on Aid Effectiveness, OECD DAC High Level Forum, February – March, Paris http://www.oecd.org/dataoecd/11/41/34428351.pdf

OECD (2006), The Challenge of Capacity Development: Working Towards Good Practice, DAC Network on Governance, February. http://www.oecd.org/dataoecd/4/36/36326495.pdf

Stevens M., and S. Teggeman (2004), Comparative Experience with Public Service Reform in Ghana, Tanzania and Zambia in Levy and Kpundeh (2004)

Teskey, Graham(2005a), Capacity Development and State Building: Issues, Evidence and Implications for DFID, Governance and Social Development Group, Policy Division, Department for International Development, October, London, UK http://www.ecdpm.org/Web_ECDPM/Web/Content/Download.nsf/0/2213EFD5C384BF61C125714F00498FD2/$FILE/Teskey_CD&state%20building-issues%20evidence&implications%20for%20%E2%80%A6.pdf

Watson, D., (2006), Monitoring and evaluation of capacity and capacity development, Discussion Paper 58b, February, European Centre for Development Policy Management, Maastricht http://www.ecdpm.org/Web_ECDPM/Web/Content/Navigation.nsf/index2?readform&http://www.ecdpm.org/Web_ECDPM/Web/Content/Content.nsf/0/134BDFCA6B69DFD3C125716B003486D7?OpenDocument

Williams, G., S. Jones, V. Imber and A. Cox (2003), A Vision for the future of technical assistance, Oxford Policy Management for the Department for International Development. Available from www.opml.co.uk


243

World Bank (2005a), Capacity Building in Africa: An OED Evaluation of World Bank Support, Operations Evaluation Department, The World Bank, April http://lnweb18.worldbank.org/oed/oeddoclib.nsf/24cc3bb1f94ae11c85256808006a0046/5676a297fe57caf685256fdd00692e32/$FILE/africa_capacity_building.pdf

World Bank (2005b), Building Effective States, Forging Engaged Societies, Report of the World Bank Task Force on Capacity Development in Africa, The World Bank, September


244

Appendix A: Terms of Reference

DEVELOPING CAPACITY?

AN EVALUATION OF DFID-FUNDED TC PERSONNEL FOR

ECONOMIC MANAGEMENT IN SUB-SAHARAN AFRICA

1. Introduction

A team of suitably qualified consultants (the evaluation team) will be contracted to carry out an evaluation of DFID-funded technical cooperation (TC) personnel in Sub-Saharan Africa over the last five to ten years. The focus will be on TC in support of public sector management, and specifically economic management in the form of economic experts employed in partner country government ministries and agencies. The evaluation will assess the contribution of this type of TC to capacity expansion in the short-term and longer-term capacity development in partner countries.

2. Background and rationale

Technical cooperation (TC) is the provision of advice and/or skills, in the form of specialist personnel, training and scholarships, grants for research and associated costs. TC can be classified (among other ways) into transactional and transformational depending on its purpose. Transactional TC aims to bridge capacity constraints in order to achieve operational goals, whereas transformational TC aims to develop partner countries’ capacity in the longer term. In practice, most TC will combine these two purposes to some extent.

Partner country capacity is crucial to the achievement of the Millennium Development Goals. It is particularly important in the context of the discussions on the scaling up of aid, as increased resources need to be accompanied by increased capacity to manage them. Given TC is a key input in bridging capacity constraints and developing capacity it follows that TC has a potentially important role to play in making progress towards the MDGs.

Moreover, over £500 million of DFID’s bilateral programme was used to fund TC in 2003/04, about half of which is accounted for by TC personnel. These amounts have remained relatively stable over the past five years. Despite the lack of a formal policy on TC in DFID (see appendix 1 for a Policy Division core brief on technical assistance), there have been some interesting developments in the past five years, such as the untying of aid (including TC) and the halving of expenditure on training and scholarships to just £23 million.

Despite this, there is little empirical evidence (particularly DFID-specific) supporting or attempting to establish the links from TC inputs to capacity outcomes and impact in a systematic way (appendix 2 provides a succinct summary of the evidence). The demand for this DFID-specific evidence is increasing both within and outside the organisation.

3. Scope and purpose

3.1 Scope


245

The evaluation will focus on DFID-funded TC personnel for public sector management, and specifically economic management, in Sub-Saharan Africa over the last five years to ten years. TC personnel is defined for DFID purposes as assistance provided to recipient countries in the form of specialists, including consultants. Personnel may be engaged long term (one year or longer) or short term (less than one year). Of particular interest will be the role of experts seconded to partner country Ministries of Finance (including Overseas Development Institute fellows). Other projects in the economic and governance sectors (e.g. public sector reform, privatisation and deregulation) that include TC personnel (particularly Technical Cooperation Officers, TCOs) based in partner country public sector bodies will also be considered for inclusion in the evaluation.

Covering this specific type of DFID TC will allow an in-depth analysis of an aid modality of growing importance and interest to DFID within the timescales suggested (the publication Statistics on International Development shows that DFID expenditure on TC personnel in the economic and governance sectors to Africa was approximately £200 million over the period from 1999 to 2004).

The evaluation will also provide lessons that are relevant more generally to: (1) TC to the economic and governance sectors, which accounted for £200 million overall in 2003/04, over a third of all TC; (2) TC to Africa, the region that receives the largest share of DFID TC (30 per cent); and (3) more importantly TC personnel, accounting for over half of all TC.

The applicability of the evaluation’s findings to other sectors, regions and types of TC is likely to be limited. Nevertheless, it is expected that it will provide a framework and methodology that can be usefully applied in future studies of TC. Moreover, some of the counterfactuals that will be considered will include questions about whether other types of TC (e.g. training or the use of local consultants) might have been better suited to deliver the desired outcomes. Finally, the feasibility of carrying out a broader survey of TC in Africa as part of this evaluation will be considered.

3.2 Purpose

The main purpose of the evaluation is to map and test the chain of results from DFID-funded TC personnel inputs to enhanced partner country capacity and better policies and service delivery. In doing so, lessons will be drawn on the best ways to provide TC personnel in different contexts in order to maximise effectiveness, efficiency and impact on partner country capacity. The key broad questions that will be answered are:

• To what extent does DFID TC aim and achieve to (1) bridge capacity constraints in order to deliver operational outputs (“getting the job done”) and (2) support partner country capacity development?

• What have the most efficient and effective ways of providing long-term TC personnel to Africa been for DFID (focusing on differences between consultancy and other personnel such as TCOs and ODI fellows, short and long-term personnel, local and expatriate consultants)?

• What is the role of recipient country ownership in the delivery of DFID TC? What difference does it make to the effectiveness, efficiency, impact and sustainability of TC?

• How does the context in which DFID TC is delivered influence the relationships mapped above?


246

The evaluation will also explore possible counterfactuals, addressing the question of whether another type of TC or aid modality (or indeed doing nothing) would have been better suited to deliver capacity outcomes than TC personnel in specific circumstances and contexts.

3.3 On capacity and capacity development

For the purposes of this evaluation different types of capacity will need to be considered. These capacities relate to the stage the policy process is at, whether that is planning, implementation or monitoring and evaluation. Capacity development will need to be explored at a number of different levels: individual, institutional and societal. Moreover, at least six key facets of capacity development should be addressed: knowledge acquisition, institution building, institutional and environment partnerships, policy environment, country commitment and autonomy and results and accountability. For a more detailed discussion of these issues see Browne (2002).

4. Key Evaluation Themes and Issues

Figure 1 provides a preliminary map of the chain of results from TC inputs to outcomes and poverty impact. It also lists many of the recommendations for the delivery of TC that have been made in the literature over the past few years, starting with those contained in the Berg (1993) report. Although the evaluation team will be expected to expand and modify this chain of results, it provides the context for dealing with the key themes below.

Figure 1. Chain of results from TC inputs to capacity outcomes

Standard

recommendations for

delivery of TC inputs

Outputs

POLICY RECOMMENDATIONS AND REFORM PROPOSALS

FOR TECHNICAL COOPERATION (1993-2004)

Improved policy process (planning, implementation and monitoring and evaluation) and service delivery leading to:

• More favourable macro environment for growth and

poverty reduction

• Improvements in broader development outcomes

Consultancy Short-term

personnel

Long-term

personnel

Training and

scholarships

Knowledge

and research TC inputs

More effective and efficient TC as demonstrated by

• More transparency in TC pricing

• Lower transaction costs

• Government capacity enhanced in the short-term

and/or developed in the longer term

• Clear roles and objectives in strategies

• TC lines up with national priorities

• TC should be integrated into national structures, e.g. gap filling

• TC should be delivered untied

• TC should be delivered on budget

• TC should be harmonised, e.g. through pooling arrangements

• Increased use of local consultants

• More short term advisers and coaching

• Increased flexibility in delivery channels

• More monitoring and evaluation should be in place

Outcomes

POVERTYImpact

Co

nte

xt

of

infl

uen

ce:

fact

ors

DF

ID h

as i

nfl

uen

ce

ov

er,

e.g.

par

tner

co

un

try o

wn

ersh

ip

Co

nte

xt

of

ap

pre

cia

tio

n:

Str

uctu

ral a

nd in

stitu

tion

al f

act

ors

be

yon

d D

FID

’s in

flue

nce


247

4.1 Relevance

Although there is DFID guidance available as well as the recommendations for delivery summarised in Figure 1, DFID does not have a formal policy on TC and therefore practice varies. The evaluation should establish what the current state of play is in terms of DFID practice and what the main changes have been over time. It should also establish the extent to which practice, guidance and the recommendations in the literature are not consistent with each other, offering explanations for observed inconsistencies.

4.2 Effectiveness and efficiency

The evaluation will assess how effective and efficient DFID-funded TC personnel have been in expanding or developing partner country capacity and improving its policy-making, public sector and economic management and service delivery. It should also make an assessment on how efficiency and effectiveness vary depending on the way in which TC is delivered, and whether DFID experience bears out the evidence that the standard recommendations summarised in Figure 1 deliver more transparent TC pricing and lower transaction and overall costs.

4.3 Impact and sustainability

At the outcome level the study should focus on the impact of enhanced and/or developed capacity on partner country public sector, economic and financial management, policy processes (planning, implementation and M&E) and service delivery. The evaluation should explore both intended and unintended outcomes. Testing the links to lower level outcomes, such as improvements to the macro environment and to broader development outcomes, falls outside the scope of this evaluation.

The final DAC evaluation criterion to be addressed is sustainability. The key question is what are the circumstances in which the impacts of DFID-funded TC personnel are sustained beyond the project’s lifetime. Where this happens capacity will have been developed; where impacts are not sustained capacity constraints will merely have been bridged in the short term.

4.4 Counterfactual

As stated above the evaluation will explore a number of possible counterfactuals, addressing the question of whether another type of TC or aid modality (or indeed doing nothing) would have delivered capacity outputs more effectively and efficiently than TC personnel. In particular:

• Would training and coaching, or increased use of local consultancies or repeated visits have better met partner countries’ capacity needs?

• Would other donors have been better placed to provide assistance aimed at addressing the identified capacity needs (what is DFID’s comparative advantage)?

4.5 Gender

The evaluation will also explore the gender dimension, particularly in terms of the level of understanding and mainstreaming of gender issues in TC projects.


248

4.6 Context

• A key finding that emerges from the evidence on TC is that the success of specific interventions is as dependent on the context in which it is delivered as it is on its design and implementation. Thus, the evaluation will address the following questions: How does the institutional context in which DFID TC is delivered influence the relationships mapped above? Focus should be placed on issues such as corruption, institutional weakness, low initial partner country capacity, political commitment (the presence of a project champion?), aid dependency and failing states.

• Is the performance of DFID-funded TC personnel (as defined by the DAC evaluation criteria) dependent on: the mix of aid instruments being employed; whether other resources accompany TC; the overall level of TC; other donors’ activities?

• How is the performance of TC personnel influenced by partner country ownership at the following levels: ideas and strategies, processes, resources and outcomes? How does ownership fit into the chain of results – e.g. is it just part of the context, is it in some instances treated as a desired outcome?

These questions should be assessed giving consideration to Boesen and Therkildsen’s (2004) classification of context. Firstly, there is the context of appreciation, which refers to “structural and institutional factors, and actions of agents beyond influence” by DFID in this case. On the other hand the context of influence refers to “structural and institutional factors, and actions of agents within some influence” of DFID. The evaluation should assess the extent to which DFID influence over some contextual elements has influenced the chain of results from inputs to outcomes and impact. It is likely that placing the analysis of ownership in this framework will be of particular relevance.

5. Methodology and Evaluation Phases

5.1 Inception

During this phase the evaluation team will develop further the chain of results to be tested, the approach and methods for the evaluation in preparation for the country case study phase. An initial stakeholder analysis will also be required. In developing the methodology the evaluation team will draw on:

• The UNDP’s work on TC and capacity development (see Fukuda-Parr et al, 2002, Browne, 2002, and Lopes and Theisohn, 2003);

• SIDA’s work on ownership (summarised in Edgren, 2003);

• DANIDA’s work on evaluating capacity development support in sector programmes (Boesen and Thekildsen, 2003);

• The review of Technical Assistance literature and issues carried out by Oxford Policy Management (OPM, 2003) for DFID.

The evaluation team should also discuss arrangements for ensuring the robustness of their findings. These arrangements are likely to include triangulation, e.g. by cross-checking the consistency of findings obtained from different stakeholder groups and by different methods.

This phase will also include a desk review of relevant and readily available project documentation, which should be analysed in the framework provided by the methodology


249

developed above. All projects and programmes in Kenya, Zambia, Ghana and possibly South Africa recording some expenditure since 1999 and having an element of public sector and economic management TC personnel since 1999 should be identified and included in the review. A selection of the most relevant TC projects in other countries should also be included if the broader survey of TC projects is deemed feasible. This review should enable the identification of specific TC interventions to be explored and key people to be interviewed during the country case study phase and cross-country survey.

The end product of this phase will be an inception report and a presentation to the DFID reference panel. The country case study phase will only start once this report has been approved by Evaluation Department.

5.2 Country case studies

This phase of the evaluation will consist of 15-day field visits to Kenya, Zambia, Ghana and possibly South Africa. The criteria for this selection of countries were: (1) the likely availability of data (proxied by the DFID’s TC expenditure); (2) including a range of diverse policy and institutional contexts; (3) avoiding overlap with other on-going and planned evaluations.

The country case studies will rely mainly on the collection of primary data and key informant interviews and any further methods approved in the inception phase. Key informants will include DFID and other donors’ staff, previous and current job-holders of DFID-funded TC posts where possible and, more importantly, staff in partner countries’ institutions that have been supported by TC personnel (ministries and government agencies). As wide a cross-section of staff that would have had dealings with TC personnel should be interviewed. It will also be important to gather views from broader civil society in partner countries to establish changes in institutions’ capacity.

For this phase at least, the evaluation team will include local consultants, who should play a key role in this phase of the evaluation. Given the tight timescale, the evaluation team should carry out at least two of these case studies simultaneously.

The preliminary findings should be shared with partners at the end of the field mission through a stakeholder workshop. Comments at the workshop will feed into the three separate final country case study reports. These should be presented in the same format and be of publishable quality.

5.3 Use of a broader cross-country survey

Given the small sample of three countries the evaluation team should consider the appropriateness and feasibility of carrying out a broader (telephone based?) survey addressing the same issues as the key informant interviews. A more succinct format would enable a wider (if more superficial) coverage of the issues, in particular dealing with other African countries and perhaps other TC inputs. The evidence gathered could be used in order to tackle some of the counterfactual questions and to carry out triangulation of the findings from the country case studies.

The survey would take place at the same time as the field phase and results would be analysed, with the findings presented in a separate report (concluded at the same time and in the same format as those for the country case studies).


250

5.4 Synthesis

The findings from the country case studies and the broader survey will be summarised in a synthesis report, which should include recommendations for the future use of DFID-funded TC personnel in Sub-Saharan Africa.

The synthesis report will include a stand-alone two-page executive summary and should be of publishable quality.

References

Berg, E. (1993), Rethinking Technical Cooperation: Reforms for Capacity Building in Africa, New York, UNDP

Boesen, N. and O. Therkildsen (2003), Capacity Development Evaluation – Step 3: Draft methodology for evaluation of Capacity Development, Copenhagen, DANIDA

http://www.um.dk/NR/rdonlyres/248E3D29-DA6E-4448-976E-B9134369CB72/0/CDEstep3_Methodology.pdf

Boesen, N. and O. Therkildsen (2004), Between Naivety and Cynicism: A Pragmatic Approach to Donor Support for Public Sector Capacity Development, Copenhagen, DANIDA

http://remote4.acdi-cida.gc.ca/Extranet/policy/cdbboard.nsf/516c6e077d8de1dd85256b45007704ea/0f10590870e4ed1985256eba0067816b/$FILE/2004-06%20Between%20Naivety%20and%20Cynicism%20Final%20July%20LE.doc

Browne, S. (2002), Developing Capacity through Technical Cooperation: Country Experiences, London, Earthscan

Edgren, G. (2003), Donorship, Ownership and Partnership, Stockholm, SIDA Evaluation 03/03, http://www.sida.se/content/1/c6/01/99/56/45636%20Studies%2003-03.pdf

Fukuda-Parr, S., C. Lopes and K. Malik (2002), Capacity for Development: New Solutions to old problems, London, Earthscan

Lopes, C. and T. Theisohn (2003), Ownership, Leadership and Transformation: Can We Do Better for Capacity Development?, London, Earthscan

Williams, G., S. Jones, V. Imber and A. Cox with A. Balihuta and A. Revi (2003), A Vision for the Future of Technical Assistance in the International Development System, Oxford Policy Management, http://www.sti.ch/pdfs/swap383.pdf


251

Appendix B: Evaluation Matrix for Country Case Studies

Evaluation Criteria

Questions Information sources

Method and responsibility

A. CONTEXT FOR CAPACITY DEVELOPMENT

Main features of the state and political system in terms of structures, institutions and agents – role of political elites, donor agencies, civil society. Key economic and political developments over 1999-2004 (and longer where necessary)

Drivers of change studies

DFID CAP, World Bank CAS and other studies

EIU reports and similar sources

Preliminary work by researchers and national consultant

Key features of economic management performance over period – inflation, fiscal management, policy development and implementation

World Bank and IMF reports

External assessments (e.g. CFAA, CPIA)


National policies and practices on technical cooperation and capacity development and relation to overall national priorities over the period

Poverty Reduction Strategy

Other government policy documents



252

Evaluation Criteria



Classification of contextual factors potentially affecting capacity and organisational effectiveness, including the extent of effective political demand for organisational capacity

Whole team

B. ORGANISATIONAL CAPACITY FOR ECONOMIC MANAGEMENT

Overview of organisational roles in economic management and selection of focal organisations for the evaluation

Mapping of organisational processes – formal and informal – and products and services of the selected organisations

Documentation on national budget process etc

Political economy studies

Interviews with key informants

Review of documents produced by organisation

Organisational development specialist using agreed framework and listing of types of outputs

Identification and assessment of role of stakeholders in the selected economic management processes

Interviews with key informants

Stakeholder analysis

National consultant using agreed instrument for stakeholder classification and analysis


253

Evaluation Criteria



Benchmarking of status of organisational processes and functions – [in relation to period of specific DFID interventions]

Identification of changes in outputs and organisational processes

Assessment of changes in capacity

Survey of stakeholders (users of outputs)

Organisational development consultant

C. CONTRIBUTIONS TO CAPACITY CHANGE [FOR EACH FOCAL ORGANISATION]

Identify changes in factors affecting capacity over the period (including budget, staffing), including inputs provided as part of planned capacity development activities

Information from the partner organisation and other development partners

Derived from analysis above under A and information from partner organisation

Organisational development consultant


254

Evaluation Criteria



Documentation and analysis of key DFID supported activities – what inputs provided and approach, process, management arrangements

DFID project documentation

Preliminary analysis by researchers supplemented by field visits

Consideration of “counterfactuals” – how might greater improvement have been achieved?

Brainstorming discussion with stakeholders

D. ASSESSING THE DFID INTERVENTIONS

Relevance What were the objectives of the DFID TC? To what extent was the provision of the TC part of a strategy for improving capacity for economic management

Did the design and approach of the DFID TC take account of the context?

Was the approach consistent with actions of other development partners and other capacity development inputs?


Partner organisation documentation

Interviews with partner organisation (using structured interview format)

Interviews with partners

Interviews with service providers


255

Evaluation Criteria



Effectiveness What outputs were provided through the DFID TC?

Did the approach have sufficient ownership from the partner organisation?

Was the approach coordinated/harmonised with other capacity development efforts?

Was the approach gender-and culture-sensitive?


Partner organisation documentation

Interviews with partner organisation (using structured interview format)

Interviews with partners

Interviews with service providers

Develop preliminary conclusions on this for validation/triangulation with views of stakeholders in the process

Efficiency Costs of provision in relation to possible alternatives?

Timeliness of provision against original schedules


Comparison with alternative approaches

Comparison with counterfactuals … were there models for more efficient delivery?

Outcomes What were the results of the DFID activity in terms of (a) the products and services of the organisation and (b) improvements in capacity associated with the outputs provided?

Review of organisation products and services and information on DFID activities

Impact Were there improvements in economic management as a result of the DFID TC?


256

Evaluation Criteria



Sustainability Was there a capacity improvement beyond the period during which the DFID input was provided?

Was a coherent exit strategy prepared in advance?

Review of organisation products and services

Consultation with stakeholders

DFID Project documentation

E. CONCLUSIONS

Overall assessment of the DFID activities evaluated – what was achieved? What could have been done better?

What hypotheses emerge in terms of potential lessons or issues for further investigations?

Prin

ted

& S

uppl

ied

by J

ohn

McC

orm

ick

& C

o. L

td.

Tel:

0141

-429

422

2 R

ef: 6

2701

DEPARTMENT FOR INTERNATIONAL DEVELOPMENT

DFID, the Department for International Development: leading the Britishgovernment’s fight against world poverty.

One in five people in the world today, over 1 billion people, live in poverty onless than one dollar a day. In an increasingly interdependent world, manyproblems – like conflict, crime, pollution and diseases such as HIV and AIDS –are caused or made worse by poverty.

DFID supports long-term programmes to help tackle the underlying causes ofpoverty. DFID also responds to emergencies, both natural and man-made.

DFID’s work forms part of a global promise to

• halve the number of people living in extreme poverty and hunger• ensure that all children receive primary education• promote sexual equality and give women a stronger voice• reduce child death rates• improve the health of mothers• combat HIV and AIDS, malaria and other diseases• make sure the environment is protected• build a global partnership for those working in development.

Together, these form the United Nations’ eight ‘Millennium DevelopmentGoals’, with a 2015 deadline. Each of these Goals has its own, measurable,targets.

DFID works in partnership with governments, civil society, the private sectorand others. It also works with multilateral institutions, including the WorldBank, United Nations agencies and the European Commission.

DFID works directly in over 150 countries worldwide, with a budget of some£4.6 billion in 2005. Its headquarters are in London and East Kilbride, nearGlasgow.

DFID1 Palace Street London SW1E 5HE

and at:

DFIDAbercrombie HouseEaglesham RoadEast KilbrideGlasgow G75 8EA

Switchboard: 020 7023 0000 Fax: 020 7023 0016Website: www.dfid.gov.ukEmail: [email protected] Enquiry Point: 0845 300 4100From overseas: + 44 1355 84 3132ISBN: 1 86192 820 3

EV667 Developing capacity? An evaluation of DFID-funded ...

Documents

Transcript of EV667 Developing capacity? An evaluation of DFID-funded ...