NGO Forum- Differing Approaches to Development Assistance in Cambodia

Cooperation and coordination

Relationships between NGOs

NGO Relationships with bilateral and multilateral agencies

NGO Relationships with government

Objectives and methodology

5. PRASAC

Origins

Design and planning

Working through government

Relationships with NGOs

6. Credit programmes

Introduction

Financial sustainability

Pricing policies

Portfolio quality

Status of fund capital

Savings service

Institutional Sustainability

Village bank sustainability

Support organisation sustainability

Sustainable apex support structure

Mutual Effects of the Programmes, Duplication, Coordination & Learning

Small/Medium Enterprise Credit

Credit (Table)

7. Domestic water programmes

Impact assessment

Working with government

Domestic Water Supply (Table)

Differing approaches

8. Management and institutional issues (table)

Project design

Mutual learning

Operational issues

Human resource issues

Monitoring and evaluation

Effectiveness of withdrawal and handover

APPENDICES

A	NGOs interviewed
B	NGOs answering questionnaire
C	Terms of reference
D	Responsibilities and contact addresses of consultants
E	Questionnaire
F	Interviews and principal contacts
G	Bibliography

Glossary

ACLEDA	Association of Cambodian Local Economic Development Agencies
AFSC	American Friends Service Committee
CCC	Cooperation Committee for Cambodia
CCRD	Credit Committee for Rural Development
CPP	Cambodian People’s Party
CRS	Catholic Relief Services
DGI DG	VIII etc Directorates General of the European Commission (very approximately equivalent to different Ministries)
EC	European Commission
ERP	EC/Cambodia Rehabilitation Programme
FUNCINPEC	Front Uni National pour un Cambodge Ind�pendent Neutre Pacifique et Coop�ratif (United National Front for an Independent Neutral Peaceful and Cooperative Cambodia)
GRET	Groupe de R�cherche et d’Echanges Technologiques
INTRAC	International NGO Research and Training Centre (Oxford)
INGO	International NGO
INGO	Local (i.e. Cambodian) NGO
NGO	Non-Governmental Organisation (here invariably development agencies often referred to as NGDOs)
PERC	Programme Europ�en de R�habilitation du Cambodge (European Rehabilitation Programme in Cambodia)
PRASAC	Programme de R�habilitation et Appui au Secteur Agricole du Cambodge (Rehabilitation and Support Programme for the Agricultural Sector of Cambodia)
RGC	Royal Government of Cambodia
TCO	Technical Cooperation Office (EC office in Phnom Penh)
WID	Women in Development (Cambodian Government)
WRD	World Relief International
WVI	World Vision International

Summary

Following discussion and debates in 1995 between NGOs and the European Commission (EC) it was mutually agreed that an independent study should be undertaken to look at comparative approaches to development in Cambodia. The study focuses on water projects and credit schemes supported by the EC and by NGOs.

The study highlights a number of policy issues in an effort to make a constructive contribution to debates over development issues in Cambodia. This summary considers some key issues facing first PRASAC and then NGOs, before outlining issues in credit programmes and domestic water schemes relevant to both the EC and NGOs.

PRASAC

PRASAC - Programme de R�habilitation et Appui au Secteur Agricole du Cambodge (Rehabilitation and Support Programme for the Agricultural Sector of Cambodia) - is a rural development programme focusing on irrigation schemes, domestic water supplies, credit schemes and small enterprise promotion. It comprises the major part of current EC development assistance to Cambodia, with a budget of approximately $44 million. PRASAC works in the six provinces adjoining Phnom Penh and was scheduled to run from January 1995 to June 1997.

PRASAC displays serious weaknesses in design and planning, primarily because the developmental sectors in which it works are not amenable to a ‘rapid impact’ approach. The sustainability of its programme is jeopardised both by the uncertainties over PRASAC’s future after June 1997, and lack of clarity surrounding the institutions PRASAC is trying to promote. These problems are particularly acute in the credit sector.

The initial emphasis on demonstrating political support to the new Cambodian government as rapidly as possible led to shortcuts in professional planning procedures and a degree of incoherence in programme design. PRASAC’s design suffered from limited baseline information, unclear assumptions, limited consultation and an unduly quantitative as well as overambitious approach. More attention should have been given to the economic status of beneficiaries and to gender issues. PRASAC’s institution building objectives are not made explicit and have led to confusion about outcomes, particularly in the credit sector. The domestic water programme is much closer to achieving its quantitative objectives, but has had to cut corners in good developmental practice to do so.

PRASAC has been unduly inward-looking, with very limited transparency, despite some practical coordination with other actors at a provincial level. Opportunities for learning between the three PRASACs and from other agencies have not been developed. PRASAC management at a national and international level has been hampered by inflexibility, and by lack of clarity over decision making procedures between the provinces, Phnom Penh, Bangkok and Brussels. Delays over procurement have added to PRASAC’s problems, though these have been mitigated by pragmatic responses at provincial level.

The solution to PRASAC’s problems of sustainability is not to cut losses and abandon the programme in 1997, but a sustained effort to redesign PRASAC and/or a long term follow-on programme on proper developmental lines. Recent management changes and the impact of the mid-term evaluation should make this more feasible, but only if this is accompanied by much greater clarity over objectives and expected outcomes, particularly at an institutional level.

NGOs

International NGO (INGO) activity in Cambodia in the sectors studied is probably quite representative of performance elsewhere in the world, though hampered by the limitations of civil society and of government capacity. Local NGO (LNGO) activities are almost equally varied: advantages in understanding society and culture are often offset by comparative inexperience. Most NGOs have quite limited capacity, though this too is very variable.

Most NGOs benefit from the comparative clarity of their objectives, often relating to poverty alleviation, gender focus and community participation, even if these objectives are often quite limited in scope. Most take some care in designing their interventions, investing time in dialogue at community level and/or in participatory rural appraisal, though only rarely in wider baseline studies. Performance indicators are sometimes unclear, and the quality of information obtained from beneficiaries often limited.

Most international NGOs attach importance to supporting local NGOs, though many do not follow this up in practice and few have well developed plans for capacity building and institutional support. Many are unclear about their future and have not developed strategies for handover and/or withdrawal, leaving the sustainability of their programmes in doubt.

Most NGOs recognise the value of coordination, cooperation, mutual learning and reasonable transparency, though some operate in comparative isolation from government and from other agencies. Some NGOs would benefit from working in a less isolated manner.

Comparative decentralisation leads to a degree of flexibility in the responses of many INGOs. Many however find difficulty in giving adequate support to fieldworkers, and experience related management problems such as lack of continuity of expatriate staff.

Assumptions about developments in Cambodian society are often not made explicit and cause uncertainties, in particular in relation to the desirability and/or appropriate methods of working with government, where approaches are particularly variable. Whilst the need for coordination and communication is very great and sometimes neglected, the desirability of working in direct support of government is less clear in current circumstances in Cambodia.

Credit schemes

Micro-credit schemes are an increasingly popular form of development intervention, and there is little doubt over the potential for such schemes in Cambodia. The national institutional framework for credit schemes in Cambodia is however in the early stages of evolution, with considerable uncertainty over the direction it will take. Any agency seriously involved in promoting credit should take careful note of this changing framework, as well as, where possible, contributing to the debates over its development. Currently there is a good deal of unrealised potential for learning between different programmes in the country.

A small number of larger or more specialised NGOs have acquired a good track record in promoting credit schemes in Cambodia and have taken a significant role in helping to develop models for the country. Many others have promoted credit schemes because they recognise their potential, because they are responsive to developmental trends, or as an entr�e to other forms of development. These agencies have been much less successful, and most should reconsider their involvement in a sector requiring specialised expertise.

PRASAC targets for credit schemes were highly overambitious even in their revised form, and will take several years to take root even if their institutional framework is clarified. The assumption that such schemes will be run effectively by local government is unlikely to be viable: non-governmental alternatives on a national scale, or possibly some variety of parastatal option, will also take time to develop. In the interim PRASAC like other agencies should ensure its initiatives fit into the wider framework - though provided its own policies can be clarified PRASAC should also take (so far neglected) initiatives to influence that framework.

Financial sustainability (as related to pricing policies, portfolio quality, and fund capital), and the feasibility or desirability of accompanying savings schemes, need to be given careful consideration by any agencies involved with credit. Institutional sustainability - at the level of the village bank, the support organisation, and the apex support structure - is no less crucial. Interaction between different programmes, and status of beneficiaries, are important considerations. The study considers these issues as they apply both to NGOs and to PRASAC.

Domestic water supply programmes

Domestic water projects are not as complex to promote as credit schemes but institutional and financial sustainability are no less central to their success, and uncertainties generally exist for both PRASAC and NGOs.

Most domestic water supply programmes lacked adequate baseline information and adequate monitoring, particularly of social aspects and of water quality. Sometimes PRASAC’s rush to meet targets has meant that social aspects have been neglected. The benefits of many interventions were liable to cooption by local social or political elites.

Different types of intervention (e.g. open wells, tube wells, village ponds) are not necessarily incompatible but could be prioritised on the basis of systematic assessment of their respective costs and benefits.

In Cambodia government-sponsored village committees such as the Village Development Committees may be considered more stable or even legitimate than alternatives but may also be subject to particular types of favouritism or political cooption. Non-governmental alternatives, whether dealing with water or credit, may not always be seen as having sufficient authority to enforce their regulations. This question needs sustained monitoring over an extended period. Dependence of local government structures is not as risky as in the credit sector, but is equally dependent on the long-term institutional viability of the structures promoted.

Recommendations

A. General recommendations for the European Commission

PRASAC’s design should be reviewed, with particular reference to its long term institutional objectives. If these can be clarified it should be extended for several years, or failing that replaced by a follow-on programme, which should take into account the considerations mentioned in the sectoral sectors below.

Future planning should pay adequate attention to baseline information, political and institutional assumptions, consultation with other development actors and qualitative aspects (including gender). General training should not be confused with institution building.

The EC should adopt an active communications policy both internally and externally, with a view to promoting mutual learning and cooperation. This also should include greater involvement in development policy debates at a national level, and more structured links with other agencies, including NGOs.

Decision making mechanisms at and between provincial, national and international levels should be changed and streamlined. A more inclusive management style should be promoted. Mutual learning between PRASACs should be improved by more structured interaction and visits.

B. General recommendations for NGOs

In planning development interventions, greater priority should be given to initial research, for example through baseline surveys, and to feedback from beneficiaries. Performance indicators often need to be clearer.

International NGOs should give more serious and practical consideration to supporting local NGOs, even in areas where they may need to take a long term view. Capacity building should not be confused with more generalised training. Strategies for handover and eventual withdrawal should be carefully considered.

All NGOs should give adequate priority to communication and coordination with other actors in development, including bilateral and multilateral agencies as well as the government.

Greater clarity should be sought in assumptions about political and institutional developments. In particular, the logic of working directly with government should be carefully considered. Whilst coordination and cooperation with government is important and should usually be upgraded, working in direct support of government departments may often not be the most productive type of intervention for NGOs in Cambodian circumstances.

C. Recommendations on Credit

C1. For all credit operators:

Efforts should be continued to reach a clear consensus on the key indicators determining sustainability in credit schemes, with particular reference to pricing policy, portfolio quality and institutional sustainability.

Operators should make such indicators public. This is not only in the spirit of mutual learning; it should provide ‘peer pressure’ incentive to set targets and move down the path to sustainability

Some diversity in credit methodology is healthy - there is no "right way" in credit. However the practical successes and failures of different methodologies should be compared so that guidelines for best practice in the Cambodian context emerge.

Information systems and feedback mechanisms from the villages need to be developed further. This will encourage early identification of potentially damaging unexpected effects, and build up a body of understanding about credit uptake, and the development of markets, over an extended period.

C2. For the European Commission:

The EC should urgently review how to make the PRASAC operations sustainable. This is particularly important given the scale of their operations which will soon make PRASAC the largest provider of micro credit in Cambodia. This must involve more than a review of the time scale of the EC commitment to PRASAC or and/or a follow-on programme, crucial though this is. The institutional and financial arrangements for credit schemes must be seriously considered as below.

Local government departments should not in themselves be seen as an appropriate vehicle for running credit schemes in the long term. More appropriate institutions will depend on policy developments at a national level, to which the EC should contribute. Alternatives for building up appropriate institutions - whether NGOs or new, intermediate agencies - should be pursued.

Interest rates for credit schemes should be set at levels which are sustainable in terms of covering future costs. Care should also be taken to ensure that the sustainability of other credit schemes is not undermined.

Arrangements between the three PRASACs should be reviewed, to ensure a common methodology in key areas, and a coherent and co-ordinated participation in the national debates. Structured exchanges between PRASACs should be instituted to promote mutual learning.

Credit scheme methodology at village level should be reviewed to ensure the poor are not being/ will be not be excluded from benefiting (e.g. by making smaller loans available).

C3. For NGOs working in credit:

Smaller NGOs without access to expertise in credit need seriously to consider whether they can run programmes that will be sustainable. If not they should concentrate on other priority sectors, if possible passing on their work in credit to a better placed agency. An alternative would be to work very closely with an NGO specialising in credit, but this would need to be carefully set up.

Larger NGOs could consider whether they can extend their expertise and support to smaller NGOs (including smaller LNGOs), particularly if they share funders or operate in the same geographical region.

D. Recommendations on domestic water provision

These recommendations apply to both NGOs and PRASAC, though there are variations in the degree to which the activities proposed may already be taking place.

Agencies involved with domestic water provision should give greater priority to assessing the impact of their schemes. To do so will involve baseline studies of health status and domestic labour routines (including opportunities for socialising) in villages affected.

Further research should be carried out to assess different water projects, with particular reference to costs and benefits of different types of scheme. Key facts to be taken into account are cost, availability, accessibility and water quality. The stability and performance of local committees should be carefully monitored.

Access, ownership and control of domestic water points need to be better monitored and the implications assessed, particularly with regard to beneficiaries.

Consideration should be given to alternative low-cost methods of provision such as hand-augured wells, especially by NGOs.

Systematic efforts should be made to monitor water quality in all domestic water projects and disseminate the findings.

PRASAC should play a more active liaison role in relation to water supply provision at a provincial level, in particular supporting local government in the dissemination of hydrological and other relevant data.

1. Introduction

Background

The origins of this study lie in discussions and debates in 1995 between European Commission (EC) officials and representatives of NGO Forum on Cambodia. These debates occasionally had a high profile, generating coverage in the local press. At a meeting in September 1995 it was agreed by both the EC and the NGO Forum that an independent study looking at key issues raised would be the most constructive way forward.

The nature of the new EC development assistance programmes in Cambodia were central to these debates and gave rise to this study. (Other issues are briefly considered in section 3 below.) The NGO Forum contracted the International NGO Training and Research Centre (INTRAC), a research and training agency based in Oxford, England, to carry out the study, research for which was mainly undertaken in May 1996.

Despite their original endorsement, in early 1996 uncertainty remained over the extent to which the EC welcomed the study and was willing to cooperate with it. In March 1996 INTRAC and the NGO Forum were persuaded that such cooperation would be forthcoming even though the crucial question of access to relevant EC documentation remained unresolved. Ultimately the study received good cooperation from almost all EC officials and contractors, as well as access to the most relevant reports and workplans - though one or two influential people remained opposed to extending any more cooperation than was considered strictly necessary, and we are grateful to Ambassador Caioull�t for a couple of decisive interventions in favour of cooperation and transparency. We also received good cooperation from the 35 NGOs interviewed during the course of the study, as well as from government officials both in Phnom Penh and the provinces. We would like to thank all who offered such assistance.

Methodology

The focus of the study was comparative approaches to development assistance. Consequently field work was carried out in geographical regions and developmental sectors where both the EC and NGOs were active, i.e. domestic water supplies and credit schemes in rural areas. In the case of the EC this meant focusing on their rural development programme, PRASAC (Programme de R�habilitation et Appui au Secteur Agricole du Cambodge).

The study team comprised three expatriate and three Cambodian researchers. Svay Reing and Kompong Chhnang provinces were selected, though some research was also carried out in Kompong Cham. Interviews were conducted in villages as well as with representatives of NGOs, the EC (principally PRASAC staff, though also members of the Technical Cooperation Office in Phnom Penh and officials based in the EC delegation in Bangkok and visiting from Brussels) and government officials (mainly from the Ministries of Agriculture and Rural Development) at local, provincial and national levels. Particular attention was devoted to policy issues. The interviews were supplemented by documentary research and by a questionnaire sent to all NGOs who are members of the Forum and/or of the Cooperation Committee for Cambodia (CCC). Round table meetings were held in Svay Reing, Kompong Chhnang and Phnom Penh, the last of these to discuss interim findings. The terms of reference for the study are included as Appendix C, and consultants’ responsibilities as Appendix D.

Scope and limitations of the study

Consideration of ‘comparative approaches to development’ quickly leads into a very broad field. Keeping the scope of the research (and the report) under control has not been easy. We have tried to strike a balance between a mandate encouraging us to look at a very wide range of issues and the need to keep within sight of the original debates leading to the study. This balance has conditioned the format of the report.

In looking at differences in approaches to development we have also been struck by similarities in the difficulties agencies face and in their attempts to resolve them. All agencies face problems, made more difficult by Cambodia’s unusual circumstances, in promoting programmes which are truly sustainable, in the sense that they can continue to produce benefits after the agency has left. Many of the issues we raise apply to both EC and NGOs, and indeed others working in the field, including government departments.

The limitations of this study should also be acknowledged. In an effective research period of three weeks 35 NGOs were interviewed and in many cases their projects were visited. Most too provided further documentation. 31 NGOs answered the questionnaire, with varying degrees of detail. The interest raised by the study was gratifying. However it may have served in some cases to raise expectations of feedback on individual programmes which are quite impractical. The most we can attempt is a synthesis and summary, quite enough of a challenge given the quantity of information and documentation received.

Though NGOs inevitably took up more of our time than PRASAC we have tried to give NGOs and PRASAC equal weight. However our findings on PRASAC may, in themselves, be more significant. This is not only because it easier to present an analysis of some of the work of a single organisation than of large numbers of varying NGOs, but also because PRASAC is reaching a critical point in decisions over its future. If our findings can contribute constructively to those decisions they will not have been wasted.

In early July 1996 a draft of this report was circulated to members of the NGO Forum committee and study reference group in Cambodia, and was also discussed at a meeting of Forum members in London. Many useful comments were received and most have been incorporated in this final version. The draft was also sent to the European Commission in Bangkok and Phnom Penh. The Bangkok office called for more specific recommendations. This was a reasonable comment and a welcome one, indicating a readiness to respond to the findings of the report. It can be contrasted with earlier emphasis sometimes placed on the limitations of a ‘comparative study’, including a view expressed that such a study should not extend as far as recommendations.

With such encouragement we have made our recommendations somewhat more specific, though there are limits what can follow from such a study. We were not in a position to carry out a comprehensive evaluation of PRASAC, since we focused on only some of the provinces where they are working, and gave priority to specific sectors. However we acquired enough of an overview to contribute to critical and urgent questions concerning the impact of PRASAC’s work and the future of the programme. We have now seen a summary of PRASAC’s own mid-term review, whose findings largely though not entirely coincide with ours. An external perspective such as ours may have some advantages. At any rate we hope that the EC will consider our recommendations alongside those of the internal review.

We also very much hope that the findings and recommendations to NGOs, though necessarily more general in nature, will be useful and can be incorporated into the never ending and never easy debates about development priorities in Cambodia. We hope too that the issues raised in this report will be considered by relevant authorities in the Cambodian government.

2. Contrasting actors in development

The subjects of this study are NGOs in Cambodia, considered on the selective basis outlined above, and the European Commission, considered through its largest development assistance programme in Cambodia, PRASAC. The Cambodian government is integral to PRASAC, as well as a significant actor in much development work promoted by NGOs. By way of introduction, the roles of these three actors are outlined, followed by brief comments on some similarities and differences in their approach, and on some key issues in development.

NGOs in Cambodia

The role of international NGOs in Cambodia has been exceptional. From 1979 a major international initiative was undertaken to provide Cambodia with relief from the appalling ravages of the Khmer Rouge regime. International NGOs along with bilateral and multilateral agencies played their part in this initiative.

However the role of international NGOs acquired greater comparative significance between 1982 and 1989 when Western bilateral and UN agencies withdrew as part of a strategy to isolate the Vietnamese-backed government. Although the number of NGOs working in the country during this period was quite limited, they included several prominent agencies who were able to provide, by NGO standards, comparatively significant amounts of aid. These NGOs developed close working relationships with central government, which nevertheless controlled and circumscribed their activities very considerably. Many of these NGOs also supported an international campaign against the isolation of the country which, at the very least, helped contribute to international awareness of the political deadlock then facing Cambodians. This campaign was largely orchestrated by the NGO Forum which has since continued its programme of advocacy on issues of importance to Cambodia both nationally and internationally.

A feature of this period was that circumstances required international NGOs to abandon their normal methods of working and to function more like bilateral or multilateral agencies. That is to say that the NGOs worked very closely with government, generally in support of infrastructural rather than community-based projects and programmes. The legacy of this situation, which continued into the 1990s, is of some significance.

From 1989 to date the numbers and variety of international NGOs and of the development assistance they have contributed has increased dramatically. Most though not all such NGOs anticipate continuing to work in the country for many years to come.

Cambodian NGOs began to operate only from 1992, since when their development has been rapid if uneven. Many, especially at first, were oriented towards human rights rather than development. A minority of local NGOs, both large and small, have proved successful; many others have struggled to raise funds and overcome handicaps. Though not the primary initiators of the debate underlying this study, some local NGOs have expressed strongly held views on the issues raised.

The number and variety of NGOs working in Cambodia makes generalisations very difficult. This needs constantly to be borne in mind when discussing NGO views and policies. We have been primarily concerned with NGOs whose focus is developmental, and particularly those with considerable local and international experience.

The European Commission in Cambodia

The growth of European Commission as a multilateral development agency reflects the complex process of European integration, making it a major player in international development assistance, with substantial resources and policy-making powers.

The developmental policies of the EC, not least as enshrined in the Maastricht Treaty, have been praised for being more poverty-focused and progressive than the corresponding policies of many European Union member states. Indeed many have looked to the EC as a means of freeing development policies, and very substantial development assistance funding, from narrower national interests. At the same time critics have claimed that the gap between policy and practice in EC development assistance programmes is even wider than in most bilateral programmes. The situation in Cambodia reflects both these perspectives.

EC aid to Cambodians from 1986 to 1993 was almost entirely to refugees and subsequently to returnees. Under the EC/Cambodia Rehabilitation Programme (ERP) approximately $37 million was allocated between 1991 and 1994. These funds were channelled through 44 multilateral agencies and international NGOs.

The new approach from 1994 came quite abruptly, as the EC, along with many bilateral donors (though later than most multilateral ones), moved to programmes of direct assistance to the new government. Some assistance to international NGOs under co-financing schemes also continued. The new initiative was substantial. Under the European Rehabilitation Programme in Cambodia (PERC) around $80 million was allocated to several programmes. The largest of these ($44 million) was PRASAC, a rural development programme focusing on irrigation schemes, domestic water supplies, credit schemes and small enterprise promotion. PRASAC works in the six provinces adjoining Phnom Penh and was scheduled to run from January 1995 to June 1997.

The detailed rationale behind this policy shift is subject to differing interpretations within the Commission. There is no doubt that a primary motive was to express support for the newly elected government of Cambodia. Indeed the PERC programme was explicitly intended to demonstrate that the government could deliver effective services, especially in rural areas, and as rapidly as possible. Most decisions over development assistance programmes are based on a combination of overlapping political, humanitarian and developmental considerations. In the case of PERC, political factors can be described as dominant.

The Royal Government of Cambodia

The Government of Cambodia has made it clear that it welcomes development assistance from all categories of donor agency. As far as international NGO aid is concerned the government’s attitude can be described as ‘open door’, though this is as much from force of circumstance as from active policy, and is unlikely to continue indefinitely. The government is more actively involved in decisions relating to bilateral and multilateral aid, much of which aims to support it directly. Even here however most of the design and a good deal of the prioritisation of aid remains largely in the hands of the donors and/or of the agencies and consultants they designate.

The government has not actively concerned itself with debates such as those between the EC and NGOs, despite the publicity they have occasionally generated. Insofar as these debates concern institutional priorities of different agencies, this is perhaps hardly surprising. Insofar as they concern broader issues of development policy however, this is unfortunate: it is hoped that this study can contribute towards broadening that debate.

Important questions surround the sustainability of much of the current government infrastructure in Cambodia, even if a reasonable degree of political stability is maintained. Most government salaries do not amount to a living wage, and are wholly insufficient, in themselves, to maintain a viable still less an effective state apparatus, not least at local level. Most observers recommend radical reform, whilst also recognising the enormous political difficulties involved. The more limited reforms which may be on the agenda will not fundamentally alter the situation.

This situation has important implications for whoever wishes to work in support of Cambodian government structures. The importance of government in development policy and planning is not in question - indeed there is a widespread need for development agencies to improve their cooperation with government at a liaison and communications level. However this is a different matter to prioritising support to structures that currently show little real chance of being sustainable, or even of retaining trained staff over a long term period. Such tough questions pose particular dilemmas for PRASAC, but need to be addressed by development agencies of all kinds.

Different and similar approaches to development

Different types of development agency (NGOs, bilaterals, multilaterals, as well as governments) are often tackling similar issues. This is particularly true when, as in the case of PRASAC and NGOs, they are working in the same field, i.e. working to promote rural development in sectors which include water and credit. Their relationships with each other, though important, are secondary to the wider problems they both face in trying to promote sustainable development. Differences of scale certainly generate different challenges, requiring differing levels of managerial and technical expertise, but the similarities of the dilemmas faced at a local level are very marked.

Differences between the approaches to development of different NGOs may be at least as significant as differences between NGOs and multilateral programmes such as PRASAC. This does not rule out generalisations about NGOs, though it does make it easier to consider ‘PRASAC approaches to development’ than ‘NGO approaches to development’. A focus on problems faced by both PRASAC and NGOs can lead to a constructive debate - provided the will is there for such a debate to take place.

Both PRASAC and NGOs must work with the realities of government and civil society in Cambodia. Both PRASAC and NGOs must face up to their own limitations, which is not always easy. Both PRASAC and most NGOs must cope with unrealistic, externally generated expectations. Many aspects of PRASAC planning (by external consultants) have proved unrealistic, whilst may NGOs, particularly international ones, struggle with expectations generated elsewhere in a different context elsewhere in the world.

Villager perceptions of different agencies are a good reflection of the greater significance of agency outputs than agency categories. Expatriate-run agencies are often referred to as NGOs whether or not they are: to villagers the distinction can be almost meaningless in comparison with the differences (or similarities) between the services on offer, about which villagers may know a good deal. Despite this the involvement or otherwise of government officials with an agency is important, and may be decisive in the authority the agency is perceived to have. Such perceptions in turn may have major implications for villager reactions, e.g. to repaying loans. External characterisations of different types of agency may not in themselves be so decisive.

Impact and sustainability

Truly systematic assessment of development assistance programmes is still uncommon, though development agencies, including NGOs, are increasingly giving greater priority to monitoring and evaluation. Key factors are the impact and sustainability of such programmes. Impact can be defined in terms of measurable changes in the status of beneficiaries, and depends on the existence of baseline information on the situation prior to the intervention. Unfortunately such information is very often lacking. Sustainability can be defined as the capacity of a programme to promote long-term processes of development after the external intervention has ended. This invariably involves institutional sustainability, i.e. the capacity of stable local organisations to manage this continuous process of development. A related goal is financial sustainability - either the ability of the development process to generate funds for continuity and expansion (the paradigm for economic development), or at least the capacity to ensure that a permanent source of funding is available (for example by securing long term government funding for health or education programmes). Assessment of sustainability is often crucially dependent on assumptions made about the environment in which the project is operating - for example concerning political stability, physical security, institutional freedoms etc.

3. Issues arising from 1995

The issues forming the main subjects of debate between the EC and NGOs in 1995 do not necessarily have the same priority a year later. Nor are issues concerning EC/NGO relations, and mutual perceptions, as significant as those concerning broader development policy. Nevertheless they comprise the background to this study and are considered briefly here.

Plans and activities of PERC and PRASAC

NGO interest in the plans and activities of PERC and PRASAC has been longstanding. The issues raised comprise a major focus of this report, but it is worth noting that the perceived lack of information available and the nature of consultation were important contributing factors to NGO anxieties (justified or otherwise), as well as to the decision to carry out this study. The main NGO concerns are considered below.

Changing relationships between NGOs and the European Commission

Following the changes in EC patterns of development assistance to Cambodia, EC funding of many NGO programmes came to an end in 1995. Although less of a public issue than those mentioned above, this also may have contributed, at least for some, to the climate of relations in 1995. This question is briefly considered in Sections 4 & 5.

The Hunting affair

In 1995 Hunting Technical Services (HTS), a development consultancy agency, was contracted by the EC to manage an EC funded Food for Work programme in Cambodia. A British NGO discovered that HTS was linked through its parent company Hunting plc to Hunting Engineering, which inter alia manufactured armaments, including landmines. The NGO Forum, which has long campaigned on the landmines issue, mounted a campaign to prevent Hunting Technical Services from working in Cambodia. The EC maintained that the lack of any operational links between Hunting Technical Services and Hunting Engineering meant that the former should not be blacklisted. However pressure generated by publicity from the campaign forced Hunting Technical Services to withdraw from the contract. The details of this affair are beyond the scope of this enquiry, but the landmines question, very understandably, generates strong feelings, contributing to the climate of relations between NGOs and the EC.

NGO perceptions of PERC and PRASAC

Criticisms of PRASAC were raised by several NGOs as soon as plans began to emerge from EC consultancy missions in 1994. Some have proved well-founded and others less so. Key concerns are summarised here, along with the findings of the study.

Quality of aid

From an early stage NGOs pointed out that PRASAC’s design was inappropriate, particularly in trying to do too much too quickly. Although based at the time on limited evidence, with only outline plans and budgets eventually reaching the public domain, this concern has proved to be fully justified and remains the most significant of the NGO criticisms, explored in later sections of this report.

Geographical focus of PRASAC

Some NGOs considered that PRASAC should not have focused only on provinces adjacent to Phnom Penh, believing that this was avoiding the more difficult options in working in less accessible regions. There is some point to such arguments; however both logistical and security concerns were perfectly legitimate, particularly in Cambodian circumstances. The rural areas where PRASAC works contain much of the rural population of the country who are not greatly advantaged by their proximity to the capital. It is difficult to establish the extent to which security was an issue, but the EC can be credited with some foresight. A related NGO concern was the switch of EC support from the north and west to the south and east of the country, with some disruption, including to NGO funding, and which would have benefited from being phased over a longer period.

Consultation

NGOs complained that consultation with them was limited, and that PRASAC planning did not draw on NGO experience. Although some consultation did take place, its nature was sometimes unsatisfactory and some good advice was ignored, notably some explicit concerns re credit scheme design. More generally, ongoing consultation seems never to have been articulated as a policy for PRASAC, though a good deal has been put into practice by PRASAC management at a provincial level. At the national level of the Technical Cooperation Office (TCO) consultation has been extremely limited.

PRASAC’s relations with NGOs

PRASAC’s intended relations with NGOs were felt by many NGOs to be unclear - a legitimate concern which has continued to cause confusion. This question is considered further in the sections below on NGOs and on PRASAC.

Duplication

NGOs were concerned that PRASAC would overwhelm NGO projects or cause duplication or overlap. With a few exceptions these concerns have proved unfounded. In the provinces studied PRASAC managers have been sensitive to the issue and most incidents of duplication have been resolved on the ground. Since only a minority of communes and villages are reached by any agency this has only rarely proved difficult. However there is some danger that with the PRASAC programme continuing to expand this issue could reappear, particularly if the security situation deteriorates further and restricts areas where it is practical to work. NGO overlap, especially in relation to international NGO support to some local NGOs, may currently be as significant a problem (as may overlap between bilateral and multilateral agencies).

Competition for Cambodian staff

Some NGOs were concerned that they would lose staff to PRASAC - including in some cases government staff to whom NGOs were paying supplements. In fact this has not been a major problem, partly because PRASAC based their local salary structures on research carried out amongst NGOs, though there have been a few exceptions. In some cases - notably for example Oxfam UKI - staff movement from NGO to PRASAC coincided with changes in NGO policy away from infrastructural to more community-based projects. (In provinces not covered by the study this problem may be somewhat more significant.)

Payment of supplementary salaries

Although this criticism generally appeared later, PRASAC policy in paying salary supplements to large numbers of government employees and effectively ‘taking over’ a large part of provincial departments has been criticised by many NGOs. This is a complex issue, facing NGOs as well as the EC, and relates to assumptions about the long-term role of government and its relations to external donors that few agencies make explicit. Moreover many NGOs adopt a similar policy, if on a more limited scale, and there are few alternatives if the basic premise of working actively with government is accepted. This question is considered further in the sections below on NGOs and on PRASAC.

Political versus developmental objectives of PERC and PRASAC

The European Commission has been quite explicit about the political objectives of PERC, of which PRASAC is the major component. PERC was designed to demonstrate support for the new Cambodian government. In acknowledging this the EC naturally fed into longstanding debates over the legitimacy of tying development assistance to political objectives. To characterise such debates very crudely, NGOs and others have often argued against such linkage, principally on humanitarian grounds, whilst governments have continued to give priority to political aspects. Of course at some level separation is impossible, but the debate is substantive. Multilateral agencies have often been more sympathetic than bilaterals to efforts to allow humanitarian rather than purely political considerations to be given greater weight in decisions over development assistance.

There are two aspects to this debate. One relates to the general morality of aid allocations, and the weight which should be given to specifically humanitarian concerns. The other relates to the impact of political factors on the quality of the aid. The second aspect is particularly relevant in Cambodia. Most people would accept that supporting the new Cambodian government was in itself a reasonable political objective. However even if linking aid to such objectives is accepted, the question remains of the most appropriate and effective means for the EC to demonstrate such support, and of the impact of such political factors on the programme’s parallel developmental objectives. If the development objectives fail, the political objectives are also forfeited.

European Commission perceptions of NGOs

The European Commission has a long and profound experience of working with NGOs around the world. Its policy relating to NGOs is continuing to develop, not least with regard to facilitating support to local NGOs. Along with most multilateral donors the EC has increasingly come to consider NGOs to be key players in successful development strategies.

Such policies are however inadequately reflected in developments in Cambodia since 1994. Practical relations inevitably reflect the attitudes and perceptions of those involved. It should also be stressed that these vary considerably - indeed it may be more difficult to generalise about EC views on NGOs than about NGO views on the EC. Here we list (in italics) some of the more critical views we encountered from EC officials and employees, along with comments based on our findings. As with NGO comments on the EC, such critical comments may be of more interest, but should not obscure the positive aspects, such as a good deal of practical cooperation, particularly at provincial level, which has run parallel with such debates.

NGOs are unsympathetic to the requirements of programmes of the scale and impact of PERC/PRASAC.

Certainly some NGOs do not give adequate recognition to the desirability of maximising impact through working on a large scale, nor sometimes show much understanding of the expertise needed for a programme such as PRASAC. However most criticisms have related to specifics of the programme rather than to its general nature. NGO/bilateral relations to be successful need greater understanding from both sides. The way forward would seem to be improved dialogue based on active communications policies.

Many NGOs are not sufficiently committed to a policy of cooperation and coordination with government, nor (consequently) with multi/bilateral agencies

The situation is very variable, but most NGOs surveyed recognise the importance of government in development, and are open-minded about cooperation with multilateral and bilateral agencies. Some NGOs do operate in comparative isolation, and should be encouraged to recognise this is seldom viable in development terms.

NGO criticisms of the EC were mainly motivated by loss of funding to NGOs following the ending of the ERP and new policies developed from 1994.

Whilst this factor may have played a part with some NGOs, it was clearly not the main motivating factor behind NGO criticisms of the EC. Some EC officials tend to see NGOs as acting almost entirely out of organisational self-interest, ignoring the much wider range of concerns motivating almost all of them.

NGO performance under the ERP was in any case variable.

No doubt this was true, though it does not seem to have been a significant factor in EC planning. No doubt donor criticisms of those they fund is a sensitive area, but could be better addressed better if a more systematic and open dialogue could be established.

All these issues are explored further in Sections 4 and 5.

The following table outlines in very general terms some of differences between the approaches to development of NGOs and PRASAC. Such generalisations can be no more than indicative - they are explored more fully in the sections which follow.

NGO approaches

Small scale

Limited impact

Variable attitudes and practice towards cooperation with government

Developmental approach

Long term approach

Usually, long term commitment

Community development approach

Limited managerial and technical skills

Variable overheads, often low

Less sophisticated technology, generally replicable.

Substantial reliance on local purchase

Generally flexible

Transparency generally reasonable

Limited communications strategies

Limited accountability

Sustainability often uncertain

EC approaches (PRASAC)

Large scale

Substantial short-term impact

Working through government

‘Pre-development’ approach

‘Rapid impact’ approach

Uncertain long-term commitment

Variable approach

Substantial managerial and technical skills, occasionally inappropriate

High overheads

More sophisticated technology, some of uncertain replicability.

Substantial reliance on imported equipment; procurement delays.

Inflexible at some levels

Limited transparency

No clear communications strategy

Limited accountability

Sustainability very uncertain

4. NGOs in Cambodia

NGOs are diverse. They vary considerably in size, capacity, resources, policies and ideology. Nevertheless it is possible to make some generalisations. The study has focused on those NGOs who fall into the overlapping categories of those we were able to interview and those who answered the questionnaire. There is a bias towards those working in credit and water in the selected provinces, as well as towards those taking an active interest in the study. There is also, inevitably, a bias towards larger, development-oriented NGOs, as well as to some extent towards international NGOs - only one Cambodian NGO responded to the questionnaire, though several were interviewed. However there is reason to assume that our sample comprises a reasonable reflection of developmental NGOs working in the country, though is less representative of smaller NGOs and of those with non-developmental agendas, for example more exclusively religious ones.

NGOs and development

The contribution of NGOs to development is often difficult to assess. Evaluations of NGO projects and programmes, where they exist, are often limited in scope and may not be in the public domain. NGOs have often been accused of making uncritical assumptions about the value of their programmes, based on perceptions of how they conform to agreed criteria (poverty alleviation, participation, gender focus etc) rather than more rigorous analysis of their impact. Nevertheless in recent years some of the most wide-ranging evaluations of NGO programmes around the world indicate considerable success in achieving objectives, though these objectives are often narrowly defined. The ability of many NGOs to work at or close to community level remains one of their great assets.

Many aspects of NGO work in Cambodia reflect universal problems, whilst others are specific to the country’s unique circumstances. Amongst the strengths of NGO approaches are their commitment to poverty alleviation, to promoting participation, addressing gender issues and (less widely) environmental concerns. Generally this is combined with an ability to work at a local level. Nevertheless there is often a gap between theory and practice - the very poorest are seldom reached, the social dynamics of participation and gender relations may not be properly understood, etc. Despite improvements in recent years, and great variations between agencies, planning, appraisal, evaluation and monitoring are often weak. Few international NGOs in Cambodia, for example, have real plans concerning the length of their involvement or strategies for withdrawal, and although this in part reflects uncertainties over future developments in the country, it also indicates limitations in strategy and planning.

Cooperation and coordination

Circumstances in Cambodia can make it difficult to replicate experience gained elsewhere. Most international NGOs (INGOs) recognise the need to promote sustainable Cambodian institutions if their interventions are to have long-term impact, though others are content to pursue a purely ‘operational’ role. Many talk of the need to do themselves out of a job, but few have a focused strategy for this. In any case the most significant issue is not so much the withdrawal of the international NGO as the capacity and sustainability of the Cambodian institutions they have promoted, and from which the INGO can move on. The primary orientation of some is to build up local NGOs - in a few cases with the specific intention of promoting an organisation which will take over the work of the international NGO. Most recognise the capacity limitations of the new Cambodian NGO sector, though this may be reflected more in their intentions than in their rhetoric - only 4 out of 31 NGOs answering the questionnaire indicated that they planned to leave the country before the end of the century.

Most agencies interviewed, and answering the questionnaire, recognise the value of coordination and cooperation with other development actors - though here there may be a particular bias in our sample, with only the more communications-oriented agencies for example considering that the questionnaire to merited a response.

Most NGOs were found to have reasonably open and transparent attitudes towards their work, and certainly towards this study. Some, not surprisingly, can be said to present an optimistic interpretation of their activities in their reports - though others can be notably self-questioning. Annual reports and similar documents tend to be in the public domain, evaluations less so, though the study itself found a widespread preparedness amongst NGOs to release documentation of all kinds.

Structures for NGO coordination in Cambodia are quite well developed, both through the Cooperation Committee for Cambodia (CCC), which aims to represent as many NGOs as possible, and through the NGO Forum itself, which has a more specific advocacy-oriented mandate. Whilst the existence of two such structures, acting to some degree in parallel, has caused some comment and self-questioning, the different mandates of the organisations mean that they are able to work in a complementary manner. Both agencies are better placed to promote dialogue amongst NGOs, and to some degree between NGOs and bilateral and multilateral agencies, than between NGOs and government. Indeed given the overall significance of NGOs in Cambodia it is unfortunate that the government is unable to give such liaison greater priority.

Relationships between NGOs

International NGOs almost invariably indicate that they believe in supporting Cambodian NGOs, though only about half our sample would appear to give this much priority in practice. Many expressed concerns about the significant number of local NGOs whom they consider not to be sufficiently competent or functional. There are indications from the questionnaire answers that the expressed intentions of international NGOs towards supporting local NGOs are often not realised in practice, with many INGOs doing fairly little.

Local NGOs have a variety of relationships with international NGOs. Most though not all active Cambodian NGOs receive funds from international NGOs, and as with relations between international NGOs and bi/multilateral agencies this can exert a powerful influence on the relationship. Many Cambodian NGOs are also involved in training and capacity building relationships with international NGOs. Most appear to adopt a pragmatic attitude to both funding and training opportunities.

NGO relationships with bilateral and multilateral agencies

Most NGOs adopt a pragmatic attitude to working with bilateral and multilateral agencies, with many, probably most, international NGOs accepting funding from such sources. From our questionnaire sample only two NGOs professed a policy of complete independence, and only four refused to accept bilateral or multilateral funds as a matter of principle, though there are certainly some others who show little interest in practical collaboration. The nature of such collaboration is discussed in more detail below, but is worth noting that very few agencies have a policy of non-cooperation, even when they adopt a critical attitude to the agencies concerned. Most Cambodian NGOs also adopt a pragmatic attitude to cooperating with, and accepting funding from, bilateral and multilateral agencies, but for all but the larger local NGOs the opportunities for doing so are limited.

Relations between bi/multilateral agencies and NGOs (as well as between NGOs and governments) can be classified in various ways, though it is important to recognise that these are only parts of a spectrum of possible relationships:

no relationship

relationship of liaison and information exchange only

relationship of active coordination and cooperation

funding relationship. Multilateral agency funds NGO but does not determine policy.

subcontractual relationship. Multilateral agency funds NGO to carry out an aspect of its programme; multilateral agency (and/or its government counterparts) determines policy.

NGOs value their independence and most are reluctant to operate in a completely subordinate relationship. A good deal depends on initiatives at a local level. In general these have worked reasonably well, with PRASAC managers adopting a pragmatic approach, if with some variations. However relations have been hampered, sometimes severely, by confusion over EC intentions, and by a rigid approach to subcontractual relationships in particular.

As with most funding relationships (which also involve a contract) the nature of a contract where an NGO is more directly involved in implementing a multilateral agency programme should be tailored to local realities. These should include the capacity of the NGO and related needs for training, support or in some cases more direct supervision. Certainly there is a need for agreed goals and outputs - though these need not be exclusively determined by the bi/multilateral agency - and for coherence (rather than identity) in objectives. In many cases arrangements granting the NGO comparative autonomy - as for example under many existing EC funding and co-financing agreements - might be perfectly appropriate.

NGO relationships with government

The great majority of NGOs surveyed believe in the importance of collaborating with the government, and in many cases of supporting government directly, though some work independently as a matter of policy, and many in practice have limited active collaboration beyond a degree of information exchange. (A wider sample than ours of NGOs in Cambodia would almost certainly increase the proportion of those who have little practical liaison with government).

From a government perspective NGOs are secondary players in the development arena. Not surprisingly, government focuses its attention and often limited resources on the major bilateral and multilateral donors. Consequently the collective NGO perspective is not widely heard.

The question of paying salary supplements to government staff led to some particularly revealing questionnaire answers and discussions. About half of our international NGO sample, and the majority of those who actively believed in working with government, pay such ‘supplements’, which almost invariably amounts to many times the salary of the employee, which is often less than $20 per month. Yet this policy was often admitted almost reluctantly - exemplified by more than one answer stating that agency policy was not to pay such supplements but that practice was to do so. Those who refuse to adopt such a policy do so on the grounds that it is unsustainable, and those who do pay supplements are often afraid that it may be.

This issue provides a good illustration of the difficulties of grafting widely accepted development practices onto unusual circumstances. Coherent assumptions about the future role of government are often lacking. This may be understandable but is nonetheless important. Political and economic constraints mean that, even under the more optimistic projections of reform (involving increasing salaries and cutting employment in the entire state sector) a government apparatus which is both functional and financially sustainable will be difficult to achieve in Cambodia. This is not to deny the importance of the state sector in promoting coherent development programmes. It does however seriously call into question the priority given to supporting this sector, particularly for NGOs. Much of the training currently being undertaken may involve skills transfer but may have limited or even negative impact on institution building, since those trained are either going to need almost indefinite external support or are going to leave government in search of alternative employment. This may not be a disaster, but does leave a major question mark over many initiatives aimed at ‘supporting government’. This of course does not only apply to NGOs, but since NGOs do not usually work in support of government, and are doing so in Cambodia mainly for historical reasons and because of the current limitations of the Cambodian NGO sector, there are certainly arguments in favour of maintaining the trend away from such direct support. This would not remove the need for other types of coordination with government, which is likely to increase.

Such developments will not necessarily make NGO relations with government easier. Villagers may well grant government and agencies associated with government greater legitimacy than NGOs, including LNGOs - though their perceptions may not correlate with practice. (This question is briefly considered below in connection with PRASAC’s relations with government.)

Objectives and methodology

Management issues are summarised in Section 8 and only a few points are emphasised here. NGO management capacities are very variable, relating primarily to the size and experience of the agency. For international NGOs it is important to emphasise that their constituencies, structures of accountability and senior management are outside the country, sometimes providing constraints on the ability of local representatives to respond flexibly to local circumstances. For many lack of clarity over policy issues can lead either to greater flexibility, or alternatively to lengthy debates over the merits of local initiatives proposed. However for most international NGOs national offices are given a reasonable amount of autonomy, though may not be in a position to provide adequate support at a regional level.

5. PRASAC

Origins

An account of the precise decision-making processes leading to the establishment of PERC and PRASAC would require access to information and documentation not available to the study. Moreover differing interpretations are available from within the EC. Nevertheless the outlines are reasonably clear from interviews and from the reports of identification missions.

In 1993 the European Commission along with almost all bilateral and multilateral agencies was looking for ways to support the incoming Cambodian government. Initially the plan was to build on the existing EC experience of working through NGOs and multilateral agencies, which had been primarily with returnees in the north and west of the country. The first identification missions focused on replacing these programmes with others working directly with government, though still largely in the same (northern and western) regions.

Early in 1994 the plans were changed, apparently as a result of sudden, and to some people unexpected, high-level interventions in Brussels. The focus of the aid, it was now decided, should be not only to support but to demonstrate the credibility of the new government, and to shift away from the refugee and returnee communities who had long been receiving Western aid. This needs to be understood in the context not only of the high hopes entertained as Cambodia emerged from its period of isolation, but of the power-sharing arrangements between the two principal political parties - the CPP and FUNCINPEC - which were fundamental to the new dispensation. These arrangements included the establishment of new ministries to help FUNCINPEC gain some control over a government apparatus that, under a single-party system, had effectively been run by the CPP for fifteen years. Support for this dispensation in the core regions of Cambodia was seen as particularly appropriate.

As far as can be established, the thinking behind PRASAC was to devise a programme which would reach as many people as possible within a comparatively short period of time, not only as an end in itself but to demonstrate that the new government could deliver benefits to rural communities better than its predecessor. The desire to reach as many communities as possible, combined with other influences pushing developmental agencies towards a ‘grassroots’, ‘village-level’ or ‘community-based’ approach, seems to have been decisive in the overall design of PRASAC. This approach was described as ‘rapid impact’, and later justified as being a ‘pre-development’ programme (and so not subject to normal developmental criteria).

This approach represents a contrast to that of most aid projects designed to achieve political objectives. Characteristically such projects are likely to be highly visible and prestigious, for example involving major public works. As a result they can be subject to criticism that they were designed to impress the government rather than benefit the people, least of all the rural poor. The EC can be credited with some originality in trying to combine a ‘community-based’ and a ‘rapid impact’ approach. Unfortunately however this is impossible to achieve in the time frames envisaged, and this should have been clear from the start.

The financing agreement for PRASAC signed in September 1994 specified a 30-month programme in six provinces in the south and east of Cambodia: Kompong Cham and Kompong Chhnang (PRASAC I); Takeo and Kompong Speu (PRASAC II); and Prey Veng and Svay Rieng (PRASAC III). The project undertook to provide over 3000 wells, claiming a beneficiary rate of 40 families per well, as well as 450 village reservoirs (ponds). It also planned to establish about 1000 ‘village banks’ benefiting an estimated 60 families per bank. (Only much later was this figure revised to 300.) Irrigation schemes would cover 30,000 - 36,000 hectares, with an estimated beneficiary rate of one family per hectare. Microenterprise promotion would benefit 3,000 to 4,000 small entrepreneurs. Over 1,000 people, primarily government employees seconded to the project, would benefit from training, with institutional support listed as a specific objective. Overall beneficiaries were estimated at 150,000 families.

The agreement does not specify which parts of the Cambodian government are to take responsibility for implementing the project, nor how they are to liaise with the external consultancy firms charged with sharing this responsibility - though it does specify the number of expatriate staff (6 FOR each PRASAC, i.e. 18 in all) to be provided. In practice provincial government authorities, principally from the Ministries of Agriculture and Rural Development, have been the most directly involved. In practice too - as must have been foreseen - practically all de facto management of PRASAC has devolved onto the expatriate consultants. At provincial level liaison between the consultants and government has been primarily via ‘mixed committees’ of the two government departments and PRASAC management, which generally appear to have worked quite well. Large numbers of government staff, generally the majority of relevant provincial departments, have received substantial salary supplements from PRASAC, and to a great extent have also come under PRASAC expatriate management. Formal counterpart systems have only been established to a very limited degree.

At a national level the EC established a small Technical Cooperation Office in Phnom Penh charged with overseeing EC development assistance, run by three (initially four) expatriate executive staff, only one of whom, in addition to the head of mission, had direct responsibility for the PRASAC programme. Liaison at national level was with a three-member committee drawn from the Agriculture and Rural Development Ministries - a system which may have had some advantages in terms of decentralisation, but amounted to a very limited structure for performing the significant tasks which had to be facilitated nationally.

Design and planning

Virtually everybody working with PRASAC, or monitoring its progress, now recognises that its planning was unrealistic. Fundamentally this results from an attempt to put a programme which, to succeed, must satisfy long term developmental criteria, into a ‘rapid implementation’ framework. Equally seriously the design had little detailed analysis of the capacities, constraints, and prospects of provincial and local government and society in Cambodia. Other significant actors whose work had the potential to complement PRASAC - chiefly local and international NGOs - received little consideration. Gender aspects did not receive the attention they deserve. The approach of the identification mission documents is strongly technocratic, despite proposing what amounts to a social and economic development programme. Discussion of assumptions, for example about possible political changes, or developments in local government, are largely absent.

In practice those charged with implementing the programme, particularly PRASAC’s provincial managers, have been involved in a process of modifying and adapting the plans to something more feasible given their constraints of personnel, equipment, and time. One perception was that it takes a year’s effort to prove that the plans and targets are fundamentally impossible, in an attempt to negotiate permission to pursue alternatives.

Most of the quantitative targets are proving to have been highly over-optimistic, especially those relating to credit and microenterprise (which have been greatly reduced). This is hardly surprising, since credit schemes require a suitable institutional framework, which needs to be set up with care, not to mention considerable consultation, coordination and training. Credit schemes are almost the opposite of something that can be set up quickly as part of a ‘rapid impact programme’. Water projects might appear to be more suitable for a ‘rapid impact’ approach, and certainly they present fewer obstacles than credit schemes. However successful village water schemes require adequate management and maintenance at local level, a fact recognised more fully by those charged with implementing the schemes than in the identification mission reports. Consequently the only way to attempt to reach targets has been to cut corners on social and local management aspects. The water programmes have been further hampered by very considerable delays in procurement of equipment, little of which had reached the provinces 17 months into the programme. The projects have managed to improvise quite successfully with locally hired equipment, and somewhat fortuitously, with equipment available from NGOs pulling out of infrastructural schemes.

The overall management of the programme has run into very considerable problems, undoubtedly exacerbated by the overambitious nature of PRASAC. Detailed analysis of such problems, involving as they do decision-making processes in Bangkok and Brussels as well as Phnom Penh, is beyond the scope of this study. However it is worth pointing out that a programme of this size and scope clearly merits more extensive support than can be provided by a very small number of officials with a wide range of other responsibilities. In particular communication and feedback, notably between the provinces and Phnom Penh, and between Phnom Penh, Bangkok and Brussels, need to be more flexible, responsive and clear. The TCO also needs to give greater capacity and priority to communicating with others involved in development in their fields (government, bilaterals, multilaterals, NGOs). In the opinion of many there is a pressing need for upgrading the levels of support for development programmes from within DGI, which in this regard, even by the hardworking standards of the Commission, is greatly overstretched. One consequence is that important decisions are made by individuals without the degree of consultation desirable.

PRASAC’s credit and water schemes are considered in more detail in sections which follow below. The rest of this section deals in summary form with other significant aspects of the PRASAC programme

Working through government

There are several problems relating to the sustainability of initiatives undertaken by PRASAC. The most immediate and serious is the absence of any commitment to continuing the programme or designing a follow-on to PRASAC after the end of the 30-month project period in June 1997. It would appear that the option of extending the programme for a further year finally came under review only following the internal evaluation of March/April 1996. Whilst this would be very welcome it would not be enough. The assumption that government structures will be able to sustain these programmes, particularly in credit and microenterprise, without continuing external financial and technical support, seems quite unjustified. Without a long-term commitment the chances of negating results that have been achieved is high. To this extent PRASAC is in a far weaker position than most NGOs, and on a far larger scale. It is very important that PRASAC is extended, and that the very problems it has faced will not become a reason for allowing it to collapse and its achievements to be lost.

Other aspects of sustainability are considered in the sections on credit and water below. Much greater clarity is required over the institutional framework PRASAC is attempting to promote and to which it is expecting to hand over, particularly though not only for credit schemes. Assumptions that the government will simply take over cannot be applied in the Cambodian context. Such a situation is not unique to Cambodia, nor should it have been difficult to predict.

PRASAC works through government: this is quite central to its conception. One consequence is that - in the planning if not the implementation stage - PRASAC took limited account of non-governmental actors working in the field, which was unfortunate as, despite their limitations, they should have a growing role to play in the development sectors addressed.

Different levels of working through government can be distinguished:

implementation of projects

training

capacity building

Although training takes place and capacity building is a stated objective, the primary focus of PRASAC’s work is project implementation - as is only to be expected given its ambitious targets of service delivery at a village level. Here the options are clearly practically possible, despite limitations in government capacity, though their sustainability is open to question.

Training of government staff is mostly ‘on-the-job’ - characterised by PRASAC III as ‘learning by doing’, which has been adopted almost as a corporate slogan. However questions need to be addressed as to what this training is actually for. Unless government reform is far more rapid than currently seems at all possible, it is inevitable that skills acquired will be used elsewhere in the private sector rather than in strengthening government capacity. Certainly government capacity building is much needed in Cambodia, but if it is to succeed at all it needs to be carefully planned and programmed, and will need to look very carefully at what structural reforms are feasible or sustainable.

The question of the sustainability of paying very significant salary supplements to government staff ‘working for PRASAC’ has been considered in the NGO section above. Undoubtedly this comprises the only practical way of ‘working through government’. However, at the very least, assumptions about how this is to be sustained should be made explicit, and the situation should be adequately monitored.

More broadly, this relates to the whole question of the sustainability of government structures raised in Section 1. Until there is significant progress towards creating financially viable government structures there is a grave danger of a great deal of what passes for capacity building going to waste, even when actual skills transfer has been significant.

Against this there is evidence that governmental rather than non-governmental structures are granted greater legitimacy by rural Cambodians. Undoubtedly this relates to the authoritarian traditions of the country’s recent history, and the novelty, at least at commune and village level, of non-governmental interventions. Village Development Committees are likely to be seen as an arm of government or even of a political party. To what extent this coincides with community based development or with the intentions of development agencies supporting them is at least open to question.

Relationships with NGOs

NGOs are not mentioned in the financing agreement and receive only limited attention in the identification missions. From an EC perspective this was shortsighted, since NGOs represented a valuable resource for helping to meet some PRASAC objectives, despite the varying response that could have been expected in terms of active cooperation. This is particularly true in view of the limitations of government capacity. Different PRASACs have given different degrees of priority to collaboration with NGOs, with PRASAC III making the most significant efforts to sub-contract parts of their programme. Cooperation and communication, though only more rarely active collaboration, were generally reasonable at a provincial level, with PRASAC managers pushing for more pragmatic policies. Cooperation was hampered to some degree at least by the controversies which arose in 1995, and more significantly by apparently contradictory messages from the TCO and from Brussels about whether such cooperation was desirable.

Possible frameworks for relations between a multilateral programme like PRASAC and NGOs are outlined in section 4 above. A somewhat rigid approach to subcontractual relationships in particular has been a significant constraint on better collaboration.

6. Credit programmes

Introduction

Credit schemes are an increasingly fashionable form of development intervention, despite the organisational difficulties involved and problems with including the poorest amongst the beneficiaries. Undoubtedly their potential impact and benefit is very great, though assessment is seldom straightforward. Though new in Cambodia, the demand for credit is clearly there and the short term impact appears significant and positive, though long term prospects are less clear. Credit in rural areas is primarily used for setting up micro- or family-unit businesses, or for tiding over difficult pre-harvest periods, and is able to compete successfully with traditional money-lenders. There is some evidence too that involvement with credit may give clients confidence to become involved with other development initiatives, for example in health education or literacy. A major bonus of credit programmes is that it is often feasible to target women, and this is certainly true in Cambodia.

The table at the end of this section summarises the study’s findings on credit programmes and compares and contrasts the practices of NGOs and of PRASAC. Key findings are:

the sustainability of programmes needs careful attention

there is much as yet unrealised potential for learning between programmes

there is much scope for improving feedback mechanisms from the beneficiaries - enabling better understanding of who is benefiting from the schemes and an early recognition of potentially serious unexpected effects.

We make some proposals below on how these issues could be addressed to ensure increased effectiveness of the ability of credit schemes to build on past experiences. Specifically we propose some ‘quick indicators’ for measuring progress towards sustainability.

The report only highlights perceived key issues relating to NGO and EC credit schemes in Cambodia. In practice a broad range of factors determine the success of any scheme, and this report makes no attempt to cover these comprehensively. Our primary focus is on micro credit (i.e. loan sizes up to $50 or $60), though small and medium enterprise credit, in which PRASAC but few NGOs are involved, is briefly discussed at the end of the section

Financial Sustainability

The study looked at the following four elements , the first three of which are crucial to the longer term financial viability of the schemes after the current period of high agency/donor involvement:

Pricing policies and the danger of ‘subsidised’ rates of interest

Portfolio quality - i.e. the status of repayments and defaults

Status of fund capital

Savings service accompanying credit scheme (if any)

Pricing Policies

To be sustainable any scheme, at least after an initial start up period, needs to offer loans at interest rates that are both broadly competitive and high enough to cover the full costs of lending. Broadly competitive rates are necessary for sustainability as otherwise clients will seek out the cheaper alternatives. Currently this is not so much of an issue because of the relatively strict geographical separation between different schemes - but see the section on ‘Mutual Effects of Programmes’ below for an indication of the kind of problems already emerging. Rates that will cover the full costs of lending will ensure (a) the fund is not eroded by defaults/payment delays or (b) by inflation, and that (c) the administration costs of the scheme and (d) interest on donor loans are covered.

The interest rates currently in force for the NGO micro schemes in Cambodia, lie in the range 3-5% per month, with varying practices as to how much of this remains with the village bank and how much returns to the NGO. (The village bank should cover defaults, inflation and village administration costs if this involves some remuneration of committee members. The NGO needs to cover NGO administration and donor interest.) In the case of PRASAC, there is as yet no single policy for interest rates, which are set at 2% per month in PRASAC I, and 5% in PRASACs II and III.

It is unlikely that schemes maintaining interest rates as low as 2-3% per month will achieve financial sustainability without reviewing their rates policy (although practices such as ensuring that repayments are made to a regional office will help keep administration and therefore rates down). It is possible that schemes in remote areas (involving high transport costs), or insecure areas (where staff working hours may be limited, and special arrangements made for holding cash) may need to change more than 5% per month to be sustainable.

Credit schemes in Cambodia are now entering a phase where they need to monitor these costs closely, in order to gain a good understanding of the extent to which costs are currently covered and to monitor progress towards full financial sustainability. Some NGOs are now beginning to do this, and the Credit Committee for Rural Development (CCRD) could play a valuable role over this next period in reviewing the progress (or otherwise) towards sustainability. This should also of course be done by the operating NGO itself, which can set appropriate performance targets. The percentage of costs covered - both inclusive and exclusive of expatriate costs - should also be monitored, to facilitate planning of the phasing out of expatriate involvement.

Portfolio Quality

The viability of any scheme is completely jeopardised by poor repayments (either defaults or deferred repayments). Again CCRD can play a key role in monitoring performance in this area using the indicators of (a) % of total loans outstanding and (b) % of loans in arrears over 30 days. Performance targets for sustainability could be to stay below 4% and 10% respectively.

The repayment records of the larger NGOs were impressively high (in the range 97%-100%). However smaller NGOs were often struggling with poor repayments, which threatened the viability of the scheme. The dynamic here may well be that smaller NGOs, while bringing in relatively less in material terms, are more dependent on their relationships at village level and less inclined to risk these relationships by being firm or insistent. More generally NGOs can be seen as lacking the authority of government related schemes such as PRASAC. Organisations finding difficulty in ensuring repayments could learn from the positive experience of "solidarity" groups and consider more clearly the basis on which special hardship cases might have their loans forgiven.

PRASAC was recording close to 100% repayment rates on their loan schemes, though in most cases these schemes are at a very early stage.

Status of Fund Capital

The issue of who owns (and therefore controls) the capital of the credit scheme is a further essential aspect of sustainability. NGO schemes receive their funds either from their supporting public or from donors - the latter could be providing the funds either on a grant or loan basis and this will be specified. On the withdrawal of the NGO, the capital is intended either to pass to functioning village banks (in which case their institutional sustainability is all important -see below), or, more commonly, to a local NGO which takes over the scheme. Funds provided on a loan basis could either be reimbursed or renegotiated to pass to the local NGO. In the case of the PRASAC schemes, it was unclear, and a source of great concern to project managers, what would happen to the loan capital at the end of the 30 month project period, currently June 97, or at the end of any extension to this period. This again highlights the urgent need to achieve clarity on who will operate these schemes in the future.

Accompanying Savings Service

Many schemes (both NGO and PRASAC) insist on prior savings by future clients to help inculcate the discipline needed for satisfactory repayment records. Many NGOs also insist that a savings element continues as a component of the village bank. Where this is the arrangement there is clearly the potential for building up a capital fund, owned by villagers at village level. There is currently a debate in Cambodia about the pros and cons of different arrangements, with one major NGO (GRET), which had strongly influenced the operating methodology of others, not advocating a tied savings approach and indeed moving away from the idea of a village bank holding loan capital in the village.

These are important debates, and what is important for the future success of the schemes is that lessons are drawn over time from the differing methodologies being used. Providing there is the potential for such mutual learning, there is a strong case for diversity of approach.

Institutional Sustainability

The following categories of institutional sustainability are considered:

Village bank sustainability

Support organisation sustainability

Apex support structure - i.e. structures at provincial and national levels

Village Bank Sustainability

There are different models in operation in Cambodia and the phrase "village bank" is used here to cover both those structures which account and hold the funds (the GRET 1 model) and those which have the more limited role of only managing the loans and their repayment (the GRET 2 model). Either way, for long term sustainability these structures will need to be well established and supported and for there to be a real sense of local ownership over the scheme at village level.

Most NGOs, and particularly the larger operators, have a good working methodology - with well functioning mutual liability groups ("solidarity groups" of around six clients acting as mutual collateral), a clear committee structure and a good programme for member training. Some NGOs allowed part of the interest repayments to be used for paying the village bank committee members for their services.

PRASAC had also adopted a sound working methodology for the establishment of village banks, and to date the doubts voiced about being able to motivate government employees for this grassroots work seem largely unfounded from the experience to date. However it is still far too early to comment with any assurance on the longer term sustainability of these village structures. Experience elsewhere indicates that it would take some five years to reach the point of village level structural sustainability: the central problem with the PRASAC schemes remains the lack the long term commitment necessary.

Support Organisation Sustainability

Credit schemes are most unlikely to survive without an effective support organisation to help maintain the discipline needed and to help solve problems as they arise.

The larger international NGOs are relatively far advanced in thinking through the issue of whom they will hand over to once they themselves withdraw. Usually this means setting up a local NGO, or handing over to an existing operator. In many cases they were tracking the degree of cost recovery of their schemes (see above), which in practice varies between 35% and 75%. Smaller NGOs however were much less clear on this point.

The future arrangement for the PRASAC credit programmes is quite unclear. They are currently staffed with regional government employees who are paid supplements of several times their government salaries. In the post-PRASAC arrangements will they remain government employees (with a large differential between them and their colleagues) or will they join a non-governmental structure (in which case their transferral or continued secondment will need to be negotiated and agreed with the government). These issues need early resolution before there can be any clarity over the sustainability of the programmes. The degree of cost recovery of the PRASAC programmes is unknown. More seriously, it is unlikely that major long term institutional donors will back a scheme essentially run by government, since experience elsewhere indicates these are seldom viable and there are few reasons to suppose Cambodia would be an exception.

A further important issue is the degree to which the credit scheme is integrated with non-financial services. Many NGOs tie their schemes to other development work especially health education or child health monitoring. In the case of PRASAC, credit is a separate component, but is seen as facilitating other PRASAC programme elements, particularly the provision of agricultural inputs. These factors alter the sustainability equation, leading some to argue that credit programmes should be run under a structure completely independent from other initiatives. This issue needs further debate and monitoring to ensure lessons can be drawn from current and future experience.

Sustainable Apex Support Structure

The debate on this important issue is still at an early phase in Cambodia, with different options being debated within the forum of the CCRD. A satisfactory resolution to this debate will facilitate a greater flow of funds into the credit sector from large institutional donors, as well as enabling existing operators to plan more clearly the arrangements for the sustainability of their own programmes. The larger NGOs are active in this debate, in which PRASAC, with its considerable expatriate credit expertise, could also play a role. This is made difficult at present by the lack of a common PRASAC policy on credit matters. It would also need a much stronger commitment from PRASAC to engage with policy issues at a national level.

Mutual Effect of the Programmes, Duplication, Coordination and Learning

There is considerable demand for micro credit, particularly during the programme start up period. The demand for micro credit came from the less well off rural population (amounts less than $50 are of less interest to better off people) but not from the very poorest (for whom potential default would have greater consequences in terms of loss of assets). Women are particularly benefiting from the schemes, with some agencies specifically targeting their credit at women (e.g. Women in Development, World Relief International).

Feedback from the longer established operators indicates that loan sizes tend to grow over time, but that after a certain point many villagers no longer want to continue with loans. Often an entrepreneurial individual in the solidarity group makes use of the loans of the other individuals, probably because they had built up sufficient working capital for their own operations, or because of market saturation in particular economic areas (for example basket weaving or cigarette retailing). It is important for support agencies to develop a better understanding of these issues, and operators were very variable in their monitoring of such aspects. Local NGOs such as ACLEDA have the great advantage of better linguistic and cultural access, and some long standing NGO operators - such as GRET (Groupe de R�cherche et d’Echanges Technologiques) - had carried out field research in a sample of villages where they had been operating.

In some areas PRASAC loans were uniformly $50, indicating that the poorer members of the community may not be being served. Requirements for clients to provide written evidence of land holding discriminated against the landless; in addition these requirements were being misinterpreted as efforts to treat the land as collateral.

As with all development programmes there is always the chance of unexpected effects following well-meaning interventions, or of unforeseen interpretations of actions. Establishing a good feedback mechanism from the intended beneficiaries themselves is all important. A dramatic example of this was provided by the recent experience of the UNICEF/WID programme which found out that borrowers were unsure of the loan cycle period, and that repayment may not have been occurring because villagers may simply have been borrowing off others for the period (often the same day) between "repayment" and taking out the next loan.

When PRASAC was first planned, many NGOs feared duplication, not least in the credit arena. In practice PRASAC project leaders went to some lengths to avoid duplication and few cases were found where credit was being delivered by two operators in the same village. (One exception may be with World Vision International in the Kompong Tralach area .)

However differentials in interest rates between different adjoining schemes could undermine the programmes of those with rates set justifiably higher for sustainability reasons. In Kompong Chhnang for example villages receiving WID credit at 3% per month were complaining, as PRASAC was providing to neighbouring villages at 2%.

There is relatively good coordination between the largest operators on issues of general policy, as shown by their common response to CCRD on the issues of a common interest rate and on the institutionalisation of rural credit. The model used by one long standing operator (GRET) has been adopted in differing degrees by many of the smaller players as well as by the PRASACs. However the process of mutual learning needs to develop further, in particular in relation to:

sustainability (quantitative targets, phasing out of expatriates, handover arrangements),

beneficiaries (what happens over time, what happens as local markets saturate)

other issues of best practice such as training methodology (particularly for Cambodian managerial staff).

The table of credit operators and the number of clients they serve recently produced by CCRD shows that of the 31 operators only 9 have substantial operations of over 1000 clients (UNICEF/WID, PRASAC, GRET, ACLEDA, World Relief International, CARE, Action Nord Sud, Concern and CRS). Most of these have credit specialists on their teams - which does help to ensure that the relatively complex working of credit schemes does get proper attention. However the past pattern of large operators is changing - and PRASAC will soon become by a long margin the largest provider of micro credit (even with its now much reduced targets). This puts a particular obligation on PRASAC to ensure that arrangements are quickly found for its future sustainability. Smaller operators, without ready access to expertise and without real prospects of achieving financial sustainability, should review whether credit programmes are the best use of their resources and skills. It may be that handing on their operations to a larger operator would be the best outcome for their programmes.

Small/Medium Enterprise (SME) Credit

Few NGOs were involved in the administration of small/medium loan schemes, as the beneficiaries here are a better off section of the population. ACLEDA is the major NGO actor providing loans up to $4000 (with only 4 to date above this amount). PRASAC also provides loan to "micro-enterprises".

Given that NGOs are not very active here, the study did not look in detail at this area. It is however a highly valid area for Cambodian development, opening up employment opportunities and national (even international markets). It is an area where the EC, as a multilateral donor, would naturally complement NGO activity. However here again a long term commitment is needed, and the sustainability of existing practices/arrangements needs examining. One example: ACLEDA is charging (after a recent policy review) 18% per year; while the PRASAC schemes on the other hand are charging 5% per month which is unlikely to be a sustainable rate, and if maintained is likely to result in poor take-up of loans.

The table on the following page summarises PRASAC and NGO approaches in credit.

MICRO / SMALL ENTERPRISE CREDIT

1. Financial sustainability

Pricing policies

interest rates need to ensure non-erosion of fund and cost recovery

Portfolio Quality

level of default

level of arrears

Status of Capital

who has final ownership

Accompanying savings service

2. Institutional Sustainability

Village banks

sense of ownership

member training

committee structure

mutual liability groups

link with non-fin services

Support Organisation

sustainable staffing

feedback systems

leadership

Apex Support Structure

3. Beneficiaries

4. Mutual effects between programmes

Duplication

Coordination

Learning

NGOs

Range 3 - 5%/m for micros (<$60) (part to village bank, part to NGO cost recovery)

18%/yr for small loans (>$60)

Repayment record for:

small NGOs very variable

larger NGOs 97%-100%

Either grant or loan, and usually clear which

Usually integrated on non-voluntary basis

Generally good methodology, often based on GRET 1 model

Sometimes credit tied to other development work - e.g. health education, child health

Mainly privately recruited staff, with large NGO cost recovery (excluding expatriates) of 35% - 75%

(others often 0%). Larger NGOs clearest about handover to/ establishment of an LNGO

Early stages still, only larger NGOs active in debate within CCRD

Usually targeted at poor (although poorest unlikely to benefit) and women

Only one NGO doing small bus'n loans on large scale

Only small degree of duplication (working in same village)

Only larger NGOs in mutual dialogue, and reasonably up to date with global experience/ donor strategies

EC (PRASAC)

PRASAC1: 2%/m for micros (sustainable?), while PRASAC 2&3: 5%/m.

All: 5%/m for small loans

Repayment record 100% to date (but too early for long term assessment)

Loan, with ownership at project end to be negotiated

Saving integrated in methodology

Good methodology, but lack of long term commitment means sustainability unlikely. Credit is separate component, but seen as facilitating. agric inputs programme element

Sustainability unlikely given poor governmental record as operator in global experience

All staffing from government with supplements, future of staff needs negotiating. Cost recovery unknown

Early stages still, needs to speak with one voice at CCRD

Poor targeted through micro loans, some doubts about beneficiaries in practice

Only initial steps taken on small loans

Genuine attempts to avoid duplication with NGOs in villages. 2% rate undermines other schemes. Seeking to build on NGO experience, but little formal coordination within PRASAC means uneven learning.

7. Domestic water programmes

Policy options surrounding domestic water programmes may be less complex than those relating to credit schemes, but this does not mean such programmes are an easy option, nor that issues of financial and institutional sustainability are any less central to their success. However diversity in such programmes is less likely to cause problems, and different approaches can more easily run in parallel. Nevertheless prioritising the most appropriate types of scheme is not easy. In particular it may depend on data that is not readily available, to this study as much as to agencies themselves.

We have focused on the domestic water supply programmes of NGOs and of PRASAC in the provinces visited. Three main types of supply are promoted, both by PRASAC and by NGOs: boreholes (tube wells), hand-dug open wells, and village ponds/reservoirs. In general boreholes are most favoured by PRASAC and open wells by NGOs. All types are supported by various types of village committee and financing arrangements. (Hand-augured wells, though largely ignored by development agencies, deserve further consideration.)

Since large scale irrigation schemes are not being implemented by NGOs in the provinces studied (though to some degree they have been in the past) they have not formed a focus for this comparative study. It is worth noting however that PRASAC planning on irrigation schemes has proved overambitious and in some cases has run into technical problems or uncertainties that were anticipated by earlier consultants who investigated these schemes. As with domestic water schemes, documentation also reveals an almost exclusive emphasis on technical and statistical analysis, with little detailed estimation of social considerations, landholding patterns etc, though we have not been able to assess the extent to which a purely technical approach has been carried over in practice.

In the narrow sense financial sustainability relates to a community’s ability to maintain and repair a water point once it has been installed, without any attempt to recover capital costs, which will normally be grants and inputs from both external and local sources. However, more broadly, capital costs are a key to sustainability and certainly to replicability: expansion of small water projects without external support will only take place if an economically competitive model is developed.

PRASAC and all NGOs are well aware of the need to establish a local committee to supervise small water projects, though most PRASAC and NGO documentation can have little information about exactly how such committees are constituted. Research reveals quite widespread, if not surprising, evidence of cooption of such committees by local political leadership, or otherwise by particular individuals or family elites, and whilst this is probably inevitable it would be reassuring if agencies found more time to try to investigate such situations and assess their impact - too often they are regarded as impenetrable if not irrelevant. In PRASAC’s case there is evidence that pressures of time and efforts to meet targets have reduced the care taken in ensuring that such committees have a genuinely representative function. In general NGOs take more time and care over consultation and attempts to ensure equity in for example the siting of a well, though in some cases they may be equally disadvantaged in terms of ability to understand local social and political dynamics. PRASAC procedures involve the establishment of a small water point fund to cover maintenance costs, but it is often unclear how this fund is to be maintained, and evidence (from Svay Reing at least) that in many cases the procedure is bypassed.

Considerable debate surrounds the most appropriate type of water point. Open wells were generally found to be the most popular with villagers, and are the most straightforward to maintain, but these may not be overriding reasons for favouring them. Water quality is very variable - here as elsewhere higher priority should be given to testing. In many places hand-dug wells cannot readily provide a year-round supply, though this also varies with location. An advantage adding to their popularity with many NGOs is that open wells provide a good opportunity for community organisation and mobilisation - water projects are often seen as an entry point for wider community development. But their comparative advantages could only be ascertained from a much more wide-ranging survey into water quality, costs and benefits than is possible in a study such as this.

Tube wells are strongly favoured by PRASAC and undoubtedly provide the most reliable basis for establishing a year-round good quality water source. The main problem is their cost - estimates range between $600 to $1000 mark - and therefore their long-term replicability. In current conditions it is inconceivable that wells at this price would be installed on a commercial basis. This is not an overriding argument against - especially since this is an area where a ‘rapid impact approach’ could have some justification - though it does put an additional premium on ensuring that maintenance and repair are properly organised. A more pressing problem is that tube wells can often provide much more water than is needed for domestic use and are likely to be coopted for irrigation schemes, leading amongst other things to sharper contests over access and control; when used for irrigation the impact on the water table and on other wells may need careful monitoring. In PRASAC III it has been recognised that proper monitoring and adequate hydrological data are needed before such schemes are expanded, and that the large diameter boreholes dug to date have been inappropriate.

Hand-augured tube wells are being dug on a purely commercial basis in parts of Cambodia but with one or two exceptions seem to be getting little attention from development agencies. There may be problems with the success rate of drilling and durability of such wells, and undoubtedly they are only appropriate in some areas, but they can be dug for a fraction of the cost of a tube well and would seem to deserve more consideration.

On the evidence available, the promotion of village ponds or reservoirs would seem to be a questionable priority, despite their popularity with villagers, who as with hand-dug reservoirs are likely to prefer the tried and tested. Use by livestock is both a benefit and an obvious hazard. The incentive to maintain filtration systems may be low, and the consequent use of contaminated water high. The safety of ponds has also been called into question, since young children can drown in them, though hard information is not available. Again a proper assessment would depend on much wider research than was possible in a study such as this: however PRASAC III at least, and many NGOs, are understandably hesitant about this option.

Different technical options are not necessarily incompatible, and have different implications in terms of year-round availability, water quality and quantity, and cost. Regional variations and differing hydrological situations are very relevant.

Impact assessment

Assessment of water projects depends on adequate baseline data, which is often absent, though some NGOs are increasingly recognising the importance of such information. The primary (if sometimes unstated) purpose of domestic water projects is to improve health, as well as to reduce the workload, particularly for women. Neither can be assessed without adequate baseline information. A programme of the size and significance of PRASAC, for example, even given the speed it intended to move, could have benefited from health surveys in a sample of the villages to be affected.

Working with government

The sustainability of the PRASAC programme is dependent on continued support from government departments, immediately raising the questions considered elsewhere (see Section 5). This problem is not as acute as for credit programmes, since in theory at least projects could continue at village level without active government support, but the speed of implementation only increases the period for which such external support is likely to be required, certainly extending well beyond the current mandate of PRASAC.

Given perceptions of government and the novelty of NGOs, NGO programmes may face different problems: that committees organised independently may lack stability or credibility in relation to other government-sponsored structures (such as village development committees, if they are separate from them). It however difficult to generalise, since many NGOs work quite closely with government.