ILO - Promoting local economic development in a war-affected country: The ILO experience in Cambodia

ILO Action Programme on Skills and Entrepreneurship Training for Countries Emerging from Armed Conflict

Promoting local economic development in a war-affected country
The ILO experience in Cambodia

Roel Hakemulder

ISBN 92-2-110515-6
First published 1997

Abstract

Click here to download the document in WordPerfect (.w61)

Training Policies and Systems Branch
Employment and Training Department
International Labour Office, Geneva, 1997

	Preface
	Acknowledgements
	Acronyms
	Executive Summary
1.	Introduction
2.	The challenge of promoting economic development at the local level
3.	Local economic development in action: Stages in the local development process
4.	Local economic development: Linking MSE development with vocational training and employment intensive works programmes
5.	Assessment of approaches
6.	Lessons learned
	Bibliography

Annexes

Annex 1	Main stages in project implementation
Annex 2	Organizational structure of ACLEDA by the end of 1995

PREFACE

This document, by Roel Hakemulder, has been prepared as an input to the ILO Action Programme on Skills and Entrepreneurship Training for Countries Emerging from Armed Conflict. This multidisciplinary and interdepartmental programme, under implementation during the 1996-97 biennium, is geared to, inter alia, making policy proposals, preparing guidelines for national and international capacity building, producing training materials, setting up a database and preparing a compendium of major training and employment-related initiatives in countries emerging from armed conflict. The programme has undertaken several country-level research activities to generate relevant data and insights. Mr. Hakemulder's study is one example. It examines a critical mechanism -- local economic development -- for economic development at the local level.

A Local Economic Development Agency (LEDA) aims at building capacity within a local or regional economy to create jobs, tackle threats being encountered and exploit opportunities emerging from rapid economic, technological and social change. In conflict-affected countries additional challenges are met at the local level including: reconstruction of conflict-affected communities, peace-building, reintegration of the conflict-affected population, rehabilitation of destroyed infrastructure, restoration of a political and administrative machinery, and perhaps most importantly, initiation and promotion of productive economic activity.

The ILO has played a key role in promoting the LEDA approach in countries emerging from armed conflict, notably in Central America and Cambodia, but also in Mozambique, Angola, Somalia, Mali, Croatia and Bosnia. The emphasis is on capacity building at the local level to enable the local population to articulate their interests, organize themselves, plan, execute and monitor activities that will lead to an improvement of their lives. Capacity-building facilitates the cooperation of private and public actors and aims at stimulating business investments and growth of local enterprises.

The LEDA approach has proven to be highly effective in meeting local needs in conflict affected countries for immediate employment creation, reconstruction and reconciliation.

The study was supervised by Haje Schutte, from the ILO's Enterprise and Cooperative Development Department, who is also a member of the team working on the current Action Programme.

ACKNOWLEDGMENTS

The author gratefully acknowledges the cooperation of Mr. In Channy, President of ACLEDA and Mr. Peter Kooi, Chief Technical Adviser of the project Alleviation of Poverty through ACLEDA's financial services; as well as the comments provided by the Enterprise and Cooperative Development Department, ILO Geneva.

ACRONYMS

ACLEDA	Association of Local Economic Development Agencies
CARERE	Cambodian Resettlement and Rehabilitation Programme
CPP	Cambodia People's Party
LEDA	Local Economic Development Agency
MSE	Micro and Small Enterprise
SEISP	Small Enterprise and Informal Sector Promotion
SNC	Supreme National Council
UNTAC	United Nations Transitional Authority for Cambodia

EXECUTIVE SUMMARY

The Small Enterprise and Informal Sector Promotion project, was implemented in Cambodia between May 1992 and December 1995. It was one of three projects in the UNDP-funded, ILO-executed Employment Generation Programme, which was itself part of the international community's response to the urgent need for social and economic rehabilitation after more than 20 years of war and civil strife.

The project aimed at contributing to Cambodia's long-term socio-economic development and raising the living standards of the disadvantaged, war-affected population groups. This was to be realized by promoting local economic development through the design and implementation of a programme of financial and non-financial services for micro and small enterprises. As such a programme did not exist, it had to be set up from scratch. Local Economic Development Agencies (LEDAs), which delivered the services, were eventually established in nine provinces, and to start with, district branch offices were set up. At the beginning of 1993, the LEDAs formed a national NGO, the Association of Cambodian LEDAs (ACLEDA), which became the project's counterpart organization.

A strategy was then developed and implemented, aiming at realizing a significant direct impact on services provided to enterprises and resulting in increased income and employment, as well as a high degree of sustainability of ACLEDA. Services consisted of a package of related elements, including business training, credit, and counselling. Separate programmes were developed for micro and small enterprises.

The sustainability of ACLEDA was addressed comprehensively, based on the project's analysis of the four aspects of sustainability: organization, managerial, technical and financial. Key interventions included supporting ACLEDA in the development of a strategic plan, in which all its staff were involved, and the establishment and implementation of a schedule for transferring authorities from the project to ACLEDA. This schedule was the basis for planning other interventions, in organizational and human resources development.

During the project period, ACLEDA assisted some 10,000 clients. Income increases were realized in an average of 45 per cent in micro and 65 per cent in small businesses. Employment in small businesses increased by an average of 1.8 jobs, at a cost of US$126 per job. The cost of lending US$1 was about US$0.61, over the entire project period. The last year, 1995, saw an exponential growth in productivity and a decrease in costs, as a result of greater routine and standardization, and a reorientation of outputs and targets in terms of lending rather than an integrated package of services. This took place in the context of ACLEDA's strategic decision to work towards becoming a micro finance institution for the poor. Training and counselling did, however, remain key components of the programme, as ACLEDA judged them to be essential for the successful use of credit.

Increased productivity also came with ACLEDA setting full self-financing as a long-term goal. In terms of organizational, managerial and technical sustainability, much was achieved over the project period, and the project's largest investment was made in this area. ACLEDA had become a fully autonomous organization, which was highly respected by its clients, the Government and Cambodian NGOs, and international organizations. The project had been scaled back in size, and only played a limited advisory and technical assistance role, especially for programme development and expansion.

With regard to financial sustainability, during 1994, ACLEDA's income amounted to 18 per cent of its total costs. Projections indicated that some 50 per cent of self-financing could be realized if the then current trends continued. This was not acceptable to ACLEDA management which, supported by the project, set in motion a process of reorientation towards micro finance services as the heart of the programme, and set full self-financing as a long-term goal. Current information indicates that this goal may be achieved by the year 2000. A follow-up project is providing support to ACLEDA in the necessary additional organizational and human resources development interventions. This project will be completed by the end of 1999. By the year 2000, ACLEDA is therefore expected to have achieved full sustainability in all aspects.

The most important lesson that may be learned from the Small Enterprise and Informal Sector Promotion project stems from its choice to develop and implement a strategy which addressed issues of institutionalization and sustainability from the very inception. In spite of the urgency of realizing an immediate impact, sustainability was not seen as a secondary issue, but was placed right at the heart of the project's strategy. The programme thus never actually functioned as a project, but operated from the start as a (nascent) organization. By ACLEDA's realization of a high degree of autonomy and sustainability in a short period of time, the project has demonstrated the feasibility and appropriateness of this approach.

The most important factor in post-war situations, where considerations of a quick impact are often uppermost in the minds of donors and politicians, is that the project was still able to assist large numbers of enterprises. The project has demonstrated that the objectives of immediate impact and sustainability can well be combined, if they are considered as two mutually reinforcing aspects of project interventions.

The project's approach towards institutionalization has been comprehensive, by addressing all aspects of sustainability. Interventions were designed and implemented to address all ACLEDA's needs as an organization, including in areas which had very little to do with enterprise development. The project and ACLEDA have demonstrated that this is a relevant approach. Projects which aim at the sustainability of an enterprise development programme should include thorough organizational audits, followed by a comprehensive programme of capacity building and organizational development.

INTRODUCTION

This paper presents a case-study of the "Small Enterprise and Informal Sector Promotion" (SEISP) project, which was implemented in Cambodia between May 1992 and December 1995. It was one of three projects which made up the UNDP-funded, ILO-executed Employment Generation Programme, which was part of the international community's response to the urgent need for the country's social and economic rehabilitation, following a period of more than 20 years of war and civil strife.⁽¹⁾ The other two projects were "Labour-based Infrastructure Rehabilitation" and "Vocational Training for Employment Generation". All three had follow-up projects, which in the case of SEISP, will be completed by the end of 1999.

The project aimed at contributing to the long-term socio-economic development of Cambodia and the raising of living standards among disadvantaged, war-affected population groups. Its strategy was the promotion of local economic development by providing financial and non-financial services to small-scale, private sector economic activities. The mechanisms used for service delivery were the Local Economic Development Agencies (LEDAs), which were eventually established in nine provinces. They assisted thousands of micro and small enterprises (MSEs), which, as indicated by impact assessments, had a significant effect on incomes and employment. The NGO in which the LEDAs joined together, the Association of Cambodian LEDAs (ACLEDA), became the most important MSE development organization in the country.

An independent evaluation conducted in May 1994, concluded that the project could be "... a model for programmes with similar objectives and in comparable circumstances in other countries".⁽²⁾ In this paper, a closer look will be taken at this proposition: the elements that make up this "model", the project's objectives, and the circumstances in which it was implemented. This will make it possible to draw some conclusions with regard to the implementation of similar programmes in other countries, including those which are war-affected.

The paper starts, in Chapter 2, with a description of the political and economic context in which the project was implemented. The main constraints on local economic development and the way the environment affected project implementation will be considered. Chapter 3 provides a description of the project, in which the most important and distinctive characteristics of its approach are highlighted. In Chapter 4 the linkages with the other two projects of the Employment Generation Programme will be briefly considered. Chapter 5 will assess the project's achievements in terms of impact, cost-effectiveness and sustainability. Chapter 6, finally, provides a summary of the "lessons learned" by implementation of the project.

2. THE CHALLENGE OF PROMOTING ECONOMIC DEVELOPMENT AT THE LOCAL LEVEL

2.1 The political context

The SEISP project was implemented in a time of great political change, initiated by the signing of the Paris Peace Accords in October 1991, which brought to an end a period of more than 20 years of war and civil war. This had seen much of Cambodia's infrastructure destroyed, while the rule of the maoist Khmer Rouge, from 1975 to 1978, had cost the lives of some 1.5 to 2 million people, including most of the country's educated population.

The parties to the Accords were the State of Cambodia's Government, which had ruled most of the country since the Vietnamese invasion which drove the Khmer Rouge from power, but had never gained international recognition; and three groups which had been fighting this Government from areas along the Thai border, the Khmer Rouge, the royalist FUNCINPEC, and the Khmer People's National Liberation Front. The agreement foresaw the creation of a Supreme National Council, comprising representatives of the four Cambodian parties to the Accords, as the country's only legitimate national authority during the period of transition to a democratically elected Government. This body became the official counterpart of the SEISP and other projects. It also envisioned the establishment of the UN Transitional Authority for Cambodia (UNTAC) whose mandate was to ensure a political environment conducive to free and fair elections, and to coordinate rehabilitation and reconstruction activities, demobilization of most of the armed forces, and the return of refugees from the border camps in Thailand.

The UNHCR repatriated nearly all 380,000 refugees between 1992 and 1993. Although cantonment was started, soldiers returned to their units when the Khmer Rouge withdrew from the agreement on the eve of the elections and resumed a campaign of violence. Elections were nevertheless held as planned in May 1993 under UNTAC administration. With a turnout of more than 95 per cent, and few irregularities, they were considered a success. FUNCINPEC became the largest party, followed closely by the ex-State of Cambodia, Cambodia People's Party (CPP). The country became a parliamentary democracy and constitutional monarchy and, in September 1993, a Government was installed which consisted mainly of FUNCINPEC and CPP members, headed by a FUNCINPEC First Prime Minister (Prince Ranaridh), and a CPP Second Prime Minister (Hun Sen). Prince Norodom Sihanouk was crowned King and became Head of State.

Since then, there has been an unremitting struggle for power within the Government, between FUNCINPEC and the CPP as well as, to a lesser extent, different factions within the parties. FUNCINPEC has been unable to gain real influence over the political and administrative apparatus in the provinces, where most decisions continue to be taken by CPP members on the basis of relations of patronage. Newspapers report that corruption is widespread, irrespective of party background .⁽³⁾ The CPP has also maintained a firm grip over the judiciary. The legal system continues to be confused as well as politicized.

At the national level, FUNCINPEC is generally thought to have lost much influence as well. What seemed an attempted coup in June 1994, and the expulsion from the country of Prince Sirivudh, a prominent FUNCINPEC member, in 1995, are indications of the continued political instability.

The preoccupation of all levels of Government with political issues, and with the Supreme National Council (abolished after the elections) at the SEISP project's official counterpart, helped to create an environment in which the project could develop a programme without significant government interference. An MSE development project did not merit close scrutiny in a time of enormous political change. The early start of the project (before elections were held) strengthened this effect. Strategic and technical choices could therefore be made based on criteria of impact and sustainability rather than politics and procedures. When eventually there were attempts to link the NGO supported by the project, ACLEDA, to ministries, and it risked becoming a prize in the battle for turf between them, its stature as an independent organization had become such that these could be blocked.

The special institutional situation in which the project functioned was recognized by UNDP and ILO, which took the approach of letting the project chart its own course and judging it on results. Like project management, they considered the project document as a starting point rather than a straightjacket. A great deal of flexibility was allowed in all aspects of the management and implementation of the project, which ensured that choices appropriate to the environment could be made.

The armed conflict with the Khmer Rouge continued throughout the period of project implementation, affecting especially the north-west of the country, and constituting an enormous drain on the Government's budget. Ownership of arms is widespread, and banditry and violent crime commonplace. This has affected project implementation negatively, for instance by periodically having to freeze operations and withdraw staff due to fighting nearing the towns and, in one instance, the Khmer Rouge taking over briefly. Vehicles were stolen at gun point and staff threatened by clients unwilling to repay their loans. On balance, however, there has been a great improvement in security and stability since the signing of the Peace Accords. This has allowed trade to flourish and gave people confidence to invest in their economic activities. The economic consequences were positive, as will be seen in the next section.

2.2 The economic context

In addition to political change, the period during which the SEISP project was implemented was one of transition from a centrally planned to a market economy, significant economic growth, and a rapid increase in the labour force. The State of Cambodia Government initiated a process of economic liberalization in the late 1980s, with the restoration of private ownership and an increased reliance on market prices. This process was continued by the new Royal Government. The main objective of its 1994 National Programme to Rehabilitate and Develop Cambodia was to strengthen the basis for a market economy through fiscal and monetary reform, and the establishment of a legal framework conducive to private entrepreneurship.

The reality has often been a high degree of economic anarchy, with state assets being sold for private gain, the growth of activities of dubious or criminal nature, and arbitrary use of authority or force. In spite of Government efforts, the regulatory framework and its enforcement remain very weak.

On the positive side, the country has experienced GDP growth rates of 7 to 8 per cent over 1991-92, and about 4 to 5 per cent in 1993-94. The economy has remained heavily dependent on agriculture, which accounts for nearly half of the GDP (1993), and has shown little growth. Industry contributes only 19 per cent of GDP, but has been growing at a rate of about 10 per cent annually. The service sector is large, accounting for 35 per cent of GDP, much of it in trading and catering, and has shown strong growth. The Government's policies have brought down inflation from more than 100 per cent in 1993 to an estimated 8 to 12 per cent in 1995.

Although this appears to be relatively encouraging, a 1995 World Bank report noted that most growth has been in Phnom Penh, and its impact has been shallow. Much of the growth can be ascribed to the presence of UNTAC (in 1992), the activities of development agencies, and the increased demand in the hotel, catering and housing sectors.⁽⁴⁾ Cambodia remains among the world's ten poorest countries, with an estimated per capita income of less than US$250. Disparity in wealth has been growing enormously. The general level of education remains low, and the country has very few higher-education graduates. On the Human Development Index (1995), Cambodia ranks 153 out of 174 countries.⁽⁵⁾

In terms of employment the agricultural sector is dominant as well: more than three-quarters of the labour force, estimated at 4 million in 1993-94, is engaged in this sector. Services account for about one-fifth of employment, and industry for just 4.5 per cent. The natural growth of the labour force is rapid, at an estimated 135,000 persons (about 3.4 per cent) per year.⁽⁶⁾ In 1992 there was an additional increase of perhaps some 150,000 due to the return of the refugees. Although open unemployment is reportedly low, at 2.3 per cent, this conceals what are expected to be very high levels of underemployment, especially in rural areas where generally only one crop can be grown per year.⁽⁷⁾

The majority of the labour force, more than 60 per cent, is self-employed, either in agriculture or MSEs. About 30 per cent work without wages for a self-employed person, and just 10 per cent of the total labour force can be counted as wage employees. Most of the latter (7.2 per cent) work for the Government. Households and individuals are generally engaged in several economic activities, especially in rural areas. This is often a combination of agriculture and micro business, with the first accounting for just 3 per cent more of total income in rural areas (44 per cent compared to 41 per cent). For Cambodia as a whole, business accounts for a much greater share of household income than agriculture (45 compared to 30 per cent).⁽⁸⁾ A recent ILO study for the World Bank estimates that the large majority of households (95 per cent, or about 1.67 million) is engaged in some micro- or small-enterprise activity.⁽⁹⁾

MSE activity is thus the main source of household income, as well as a major source of employment. The combination of liberalization and much improved security established conditions in which opportunities for MSEs grew and could be created. Although no comprehensive data are available, observation and interviews with those who work with the sector indicate that, contrary to agriculture, it has grown strongly over the past five years, and is likely to have absorbed most of the growth of the labour force.⁽¹⁰⁾ The programme set up by the SEISP project could therefore assist members of its disadvantaged target groups' clients to make use of emerging opportunities. MSE promotion services were both needed and in high demand, and the environment was conducive to the success of a programme delivering such services.

2.3 Constraints on local economic development

Given the limited scope for growth in formal sector wage employment in the short and medium term, and the importance of agriculture and MSEs in terms of employment and household income, these sectors have been the focus of most efforts to promote local economic development.⁽¹¹⁾ In this section we shall consider the main characteristics of MSEs which such efforts had and have to take into account.⁽¹²⁾

The MSE sector includes virtually all business in Cambodia. Most are one-person or family-based operations, providing generally part-time employment for several members of the household. The large majority cater to local markets. Commonly, a variety of micro enterprises is found in a community, catering to demand in the community itself, or those neighbouring it. Some may produce for the provincial market, and there is some regional specialization, where an area produces certain types of goods which are marketed nationally. This is generally related to the availability of raw materials, although tradition plays a role as well. There is some export of MSE products, especially in agro and food production and processing, from provinces bordering Thailand and Viet Nam.

Linkages among MSEs are generally limited to those between producers, wholesalers and retailers. Much enterprise is linked to the agricultural sector, including fishery, while MSEs also provide inputs into agriculture (agricultural tools).

Constraints on the further growth of the sector can be considered in terms of the environment, inputs, and skills and technology.

The environment

The main constraint in the business environment and overall is the small effective size of markets, although this has improved over the project period. Levels of income are low, and purchasing power very limited. Due to the lack of a general physical and marketing infrastructure, markets are localized, especially in rural areas. Large numbers of MSEs cater for this small market, and competition is intense. Information on market demand beyond the immediate environment is generally not available.

Inflation was an important concern during 1993 and 1994, when it was estimated to be over 100 per cent. The national currency is the riel, but most business was conducted in US dollars, which are in wide circulation and can be readily exchanged on the market. Speculation and political instability led to wild fluctuations in exchange rates especially in 1993.

Security is a significant constraint, with many businesses finding their normal activities disrupted on a regular basis. This is due to the continuation of armed conflict with the Khmer Rouge, especially, but not only, in the north-west, banditry, and the ineffectiveness of the police and the legal system.

The policy and legislative environment is not supportive of MSEs. Although the Government encourages private sector development, a longer-term strategy for the sustainable development of the MSE sector does not exist, and policies favour large foreign investors. The regulatory framework is inadequate, the administration of policy and regulations is ineffective and haphazard, and generally perceived as inequitable: there is no "level playing field". The absence of effective government control over activities in the sector has probably had positive effects on its growth.⁽¹³⁾ However, the general situation of de facto lawlessness in which, for instance, contracts cannot be enforced, property rights are not clear, and the threat of violence can be used effectively, cannot but have a negative impact on growth and development.

Inputs

Finance is most often mentioned by entrepreneurs themselves, and by MSE studies, as the most important constraint on growth of the sector. Private money lenders operate widely in urban and rural areas, but they are inadequate as a source of credit for MSEs because they generally provide short-term loans only, at very high interest rates (10 to 50 per cent per month), and require collateral which is readily exchangeable for cash. The formal banking system does not provide credit to micro enterprise at all, and serves Phnom Penh and three provincial cities only although in 1993 this was just Phnom Penh. NGO programmes, most of which started in or after 1993, have attempted to meet the demand for credit, with some success, but their outreach is insufficient: by April 1995 they had only been able to lend to 2.5 per cent of households in Cambodia.⁽¹⁴⁾

The availability of raw materials is not a significant constraint on those MSEs which use locally available materials, but it is a major obstacle to the diversification of enterprise and growth in subsectors which use processed or manufactured inputs. Few material inputs are actually available locally: there is little diversity in agricultural production, and hardly any manufacturing or processing with potential forward linkages to MSEs. Due to the lack of infrastructure and security, access to materials from other areas or imported materials is limited and unreliable, although this has improved over the project period.

The same is true of the availability of equipment and tools. Most have to be imported. Although the availability of more commonly used tools and equipment is good in urban markets, innovation and diversification is limited because of the unavailability of the required tools and equipment, or information about them. Quality is generally low, and prices relatively high.

The highly inadequate utilities infrastructure is a significant problem for most enterprises in manufacturing and services.

Skills and technology

The level of technology used in MSEs is low, as a result of the relatively high prices of imported equipment, the limited range of available equipment, and the lack of long-term credit; as well as the unawareness of improved technologies. This is further aggravated by the low level of technical skills. The large majority of workers in MSEs have been trained through apprenticeships or in the family. Their skills are adequate to the level at which enterprises are operating, but there is a lack of innovation and upgrading. Technical and vocational training by projects and institutions is accessible to just a small percentage of the population, and much of it is inappropriate in terms of employment generation in MSEs. Productivity, and the quality of the goods and services produced in the sector, are therefore low.

Business and management skills are lacking, but generally adequate at the level on which most MSEs function: the businesses are so small and simple that few skills are required. However, the lack of such skills is an obstacle to growth, diversification, and the identification and realization of new opportunities.
This chapter deals mainly with the context in which the SEISP project was implemented. This was an environment of rapid political and economic change which was, on balance, conducive to the development of the MSE sector. Economic opportunities grew, but a number of serious constraints limited the extent to which they could be used, especially by disadvantaged groups. These included a lack of business and management skills and financing, both of which the SEISP project intended to address. The next chapter provides a description and analysis of how this was achieved.

3. LOCAL ECONOMIC DEVELOPMENT IN ACTION: STAGES IN THE LOCAL DEVELOPMENT PROCESS ⁽¹⁵⁾

3.1 Project formulation

Project identification and formulation was initiated in February 1992, following an invitation to the ILO from UNDP Cambodia and the Supreme National Council, to design a programme to generate employment for soldiers to be demobilized under the Paris Peace Accords. Some 140,000 soldiers from the four parties to the Accords were expected to be demobilized between September 1992 and the May 1993 elections. Their reintegration was considered to be of key importance to the success of the peace process.⁽¹⁶⁾ It was therefore understood from the beginning that project identification and formulation would take place under a great deal of time pressure, as well as being accorded a high level of priority by everyone concerned.

A brief fact-finding mission, which identified the main elements needed in an employment generation programme targeting the demobilized was followed in March by a survey of more than 1,000 soldiers across the country, excluding the Khmer Rouge, which refused access to the areas under its control. This established an employment profile of the target group. Simultaneously, information was collected on opportunities for employment generation in the short as well longer term. This focused on the relevant institutional support mechanisms, and the general state of the country's economy and infrastructure, although agriculture was excluded as not being in the ILO's area of competence. The information was gathered through discussions with officials, study of the few available reports, and brief visits to some of the main provinces. On this basis, three separate but linked documents were formulated, covering small enterprise promotion, vocational training, and employment-intensive infrastructure rehabilitation. They were grouped together in an Employment Generation Programme and submitted to the UNDP by April.

While, under a preparatory assistance phase, the small enterprise project was started up in May, a significant reformulation took place following review of the documents by UNDP New York. For reasons of equity and long-term sustainability, the target group was changed to include other disadvantaged groups affected by the war, and the growth of the labour force due to the demobilization of soldiers and the resettlement of refugees and internally displaced persons. Women, especially heads of households, became a priority target group. This was welcomed by the ILO, which had preferred a more inclusive approach from the start. In addition, the small enterprise project was scaled down significantly, from 15 to 5 provinces. While the first draft had foreseen support to small enterprises only, the reformulated document included a large programme of assistance to the self-employed in the informal sector (mostly women). It thus became the "Small Enterprise and Informal Sector Promotion" project. There was no time to collect additional information on which to base the reformulation.

The project was finally approved in October 1992, with a budget amounting to US$4,372,523, including a preparatory phase from May to October 1992, administrative support costs, and a US$613,000 revolving fund to be managed by the project. The project was to run for a period of 30 months (including six months' preparatory assistance) to October 1994. This was later extended several times, to December 1995.

Although the project formulation process therefore did include a serious (and successful) effort to establish the needs and preferences of the initial target group (the demobilized), the final document did not have such a firm grounding. There was also no participation of the target population in project design, and only limited consultation with local authorities or other groups in the provinces covered by the project. No research was done on local economies. Project design was instead based mostly on secondary information, most of which was inadequate, and the ILO's experience elsewhere in the region.

Time pressure and the programme's change in focus were, to a large extent, responsible for this relative lack of a firm basis in research and participation. In addition, the situation in Cambodia did not lend itself very well to target group participation and greater involvement of local organizations and administration in project design. None of the local authorities were, at this stage, part of a recognized government, and relations between them and UN personnel were therefore ambiguous. There was little trust between authorities and the local population, there were no mechanisms for consultation with the population, and no local organizations which could represent them. Security was inadequate, especially outside the main towns.

The strategy was therefore to "get something going" as soon as possible, while providing for a flexible design which would allow the project to develop its strategies and approaches along the way, on the basis of studies and experience. This also recognized that the project would be implemented in a rapidly changing and unpredictable situation, and that its institutional environment was fluid.

3.2 Project strategy

The project's overall strategy evolved on the basis of the project document, during the preparatory assistance phase and the first six months of project implementation that followed it. This was a gradual and informal process, in response to unfolding events, the synergy which developed between expatriate and Cambodian project staff, and increasing knowledge of the Cambodian situation.

The first cornerstone of this strategy was that an immediate impact in terms of MSEs assisted and the creation of employment and additional income had to be realized as soon as possible. Such assistance would, in the words of the project document, enable "disadvantaged population groups to make use of and create gainful economic opportunities, and make an active contribution to the economic development of Cambodia". It was expected that this would improve living conditions, contribute to reintegration and reconciliation, and therefore eventually the consolidation of peace. Moreover, it was politically important for the project's supporters, the parties in the Supreme National Council in the first place, to be seen to "do something" urgently for those most affected by the war.

The main objective to be realized was therefore, as stated in the project document, "an integrated small enterprise and informal sector programme aimed at socially disadvantaged groups", to be delivered through five Local Economic Development Agencies (LEDAs). Over a period of two years, the LEDAs were to assist 1,400 potential entrepreneurs under a small business component, and 12,000 self-employed persons (micro businesses) under an "informal sector" component. The programme had to be set up from scratch, as no basis for it existed in the country, either in the Government or the NGO sector.

Although the achievement of an immediate impact was thus a priority, the project recognized that in terms of overall needs, its initial impact would be small while, given time, an MSE programme would be able to assist large numbers of people. As funding for this project could be expected to be finite, it therefore needed to aim at the institutionalization of the programme. This was reflected in the project document, which included, as a second objective to be realized, "tested strategies for the longer-term sustainability of the network of LEDAs", to be finalized towards the project's end in a strategy document, which would be presented to donors and the Government.

However, the project felt that fostering institutionalization and sustainability from the very start would avoid the entrenchment of a "project mode of operations", with a project culture and attitudes, which would result in dependence on expatriate experts and managers, and eventually act as a barrier to institutionalization. Too many projects have tried to transform themselves into institutions, and have found it to be a long and painstaking process which more often than not ends in failure. Operating the MSE programme from the start as an institution instead of a project therefore became the second cornerstone of the project strategy.

The third was the project's analysis of the concept of sustainability, which was defined as the ability to continue implementation and development of the MSE programme after completion of the project. This recognized that an organization needs to be able to change and develop in order to be sustainable: a capacity for the static continuation of a programme would not be sufficient. It was also understood that finance is just one factor in the continuation of a programme. Four aspects of sustainability were recognized:

Organizational sustainability, which refers to the legitimacy of an organization in the eyes of itself and its environment, and its linkages with other organizations.
Managerial sustainability, which is achieved when the organization has the organizational structure and systems, as well as the managerial and administrative capacities to function and develop as an institution and realize its objectives.
Technical sustainability: an organization should have the technical capacity to implement and further develop its programme.
Financial sustainability: an organization should be able to obtain funding for the implementation and development of its programme, through self-financing and/or donors.

It was understood that, in order to achieve sustainability of an MSE development programme, all four aspects would need to be addressed. Moreover, rather than emphasizing financial sustainability, the project's approach was to give priority to interventions addressing the other three aspects, which it understood to be essential to the achievement of financial sustainability.

Other project implementation strategies flowed, largely, from these three cornerstones. An important principle was to consider the development and delivery of services to MSEs and the institutionalization of these services as two mutually reinforcing aspects of project interventions, instead of two more or less independent issues to be addressed in different phases of the project. This meant, for instance, that if a training programme had to be designed, it needed to be done in such a way that it would foster sustainability; and if a management information system had to be designed, it had to be done in such a way that there was direct linkage to the improvement of delivery of services.

A further key principle flows from this and the realization that sustainability is as much in people as in structures and finance: the principle of participation by and responsibility of the Cambodian project staff. Instead of considering the Cambodians in the project, the large majority of whom had no experience with MSEs, as staff to be trained who would then deliver services to the enterprises, it was recognized that they needed to be partners in the development and implementation of the programme from the very start. This principle built on the strong motivation of the Cambodian staff, based on their sense of hope and optimism that Cambodia was entering a period of peace, democracy and development. There was an eagerness to make up for lost time by learning from the outside world as quickly as possible, and to take responsibility if it was offered genuinely. On the basis of the experience gained in the preparatory assistance phase, it was expected to have important benefits in terms of a sense of ownership. It would contribute to sustainability, and to higher productivity and quality, as Cambodian staff were in a much better position to know or establish what was appropriate to the country and target groups than the expatriates.

The project strategy did not include, however, strong involvement of representatives of what is generally called "the community", or community organizations, in the management of the programme. The LEDAs were conceived as professional service providers rather than community-based organizations. This was based in the first place on the experience of the ILO and others that a professional organization can be an effective and efficient provider of assistance to MSEs. It was felt that an MSE programme, if it is not also a community development programme, needs strong linkages to relevant groups and individuals in the community, but does not need to be community-based in the sense of community ownership and management. The priority of those who run MSEs is generally to obtain support as quickly and as easily as possible, and this can be well achieved by a professionally managed organization with a service delivery orientation.

In the second place, this strategy was appropriate to the social and political situation in Cambodia. Communities were fractured and in a state of flux, and Government bodies highly politicized. There were no citizens' groups, and local development NGOs did not exist, apart from one in Phnom Penh. It would have been impossible to identify legitimate community representatives to take a role in the management and ownership of the programme.

In order to achieve the required impact in terms of employment and income, the MSE programme would, however, have to be need-based. As the project would start without in-depth knowledge of the local economy, and did not foresee structured community participation in the development and implementation of the programme, it was the project's strategy to gradually develop the MSE support package based on studies and experience gained by the project. This comprised studies of economic opportunities and factors limiting MSE development in each of the areas covered, as well as training needs' assessments, and a process of continuous feedback. Programme elements were to undergo a series of pilot tests, and be the subject of ongoing evaluation and redesign, in order to ensure that they would be need-based to the greatest extent possible. In effect, this was a process similar to market research for product development in the commercial sector.

The project did not, however, plan to address all needs of MSEs. Its approach was one of specialization rather than comprehensiveness. This was based on the international experience that MSE development institutions with programmes which are limited in scope and clearly defined are more successful.⁽¹⁷⁾ It also took into account the absence of an MSE promotion capacity in Cambodia: it was more realistic to plan for developing such a capacity in some key areas, rather than in all possibly relevant fields. Credit, and business training and advice, were expected to be such key areas.

An important decision for the way the programme was developed and delivered was to consider clients responsible for the success of the services they were provided with, rather than as passive recipients. It was expected that this would make the programme more effective and appropriate, and the results more sustainable. The programme would therefore have to be highly participatory, and be built around decisions and activities of the clients.

3.3 Project implementation

Preparatory assistance

Project implementation started in May 1992, with six months of preparatory assistance. This comprised the recruitment of Cambodian staff for the LEDAs in the border camps and within the country itself, and their training. It allowed the project to make substantial progress in the establishment of an MSE support programme, while the final project document had not yet been approved, and some aspects of project design, including the target group and the size of economic activities served, were still under discussion.

Three aspects of the way the preparatory assistance was carried out are worth considering. Firstly, it was managed by the ILO's Regional Office in Bangkok and implemented with staff placed there, rather than by recruiting staff for this purpose. This made possible the rapid recruitment of staff in the Thai border camps, and an immediate start-up of activities in the country itself.

The second is the relationship developed with CARE International in Cambodia. CARE had a strong presence in the country and was one of the few organizations running a micro credit programme. The ILO made use of this capacity by contracting CARE to set up a project office in Phnom Penh, and assist in the recruitment of LEDA staff. When this was successful, under subsequent contracts CARE played an major substantive and management role in the development and implementation of the training programme for LEDA staff, and the design of the informal sector component. Contracting an organization with an established capacity in the country rather than recruiting ILO staff, enabled the project to maintain a tight schedule, at a low cost.

Finally, it had been foreseen that LEDA staff would be trained through fellowships abroad. However, following their recruitment and a training needs assessment, which made clear that few had higher education or business experience, it was decided that a tailormade programme in Cambodia, would be more effectively placed within the context in which the LEDA staff would have to work. A training programme on basic principles of small enterprise promotion was developed and implemented under ILO management, by a team of trainers from CARE and the Entrepreneurship Development Institute of India, supported by Cambodian co-trainers selected from among the future LEDA staff. This proved essential to keeping the programme relevant and on an appropriate level.

The Local Economic Development Agencies

The LEDAs were established physically by teams of newly trained Cambodian staff assigned to the different provinces, and became operational in the beginning of 1993. Their organizational structure was as had been foreseen in the project document. They were headed by a Team Leader, who supervised two or three Small Enterprise Promotion Officers and a Credit Officer, in charge of implementation of the small enterprise component. When the informal sector component had been formulated, in April 1993, two or three Informal Sector Agents joined. An administrative assistant and guards made up the support staff.

UN Volunteers placed in each of the LEDAs took the positions of LEDA Management Advisors. In accordance with the project's strategy to give responsibility to Cambodian staff as soon as possible at the LEDA level, the Team Leaders were thus in charge from the start.

Advisory Committees were established for each LEDA, with representation of relevant authorities, IOs and NGOs. They were intended to be a mechanism through which the LEDAs would be able to develop relationships with relevant actors in the local community, and build support for their programme. They thus provided for a degree of community involvement. However, their role was very limited, as they had an advisory function only.

For the functioning of the credit component, Local Credit Committees and Local Informal Sector Credit Committees were established at each LEDA. In addition to LEDA staff, these included outsiders with relevant expertise, such as business persons, and NGO or bank staff, who participated as individuals, not representatives of an organization. The Committees met every fortnight to decide on loan applications.

The programme was therefore established in accordance with the project's conception of the LEDAs as local professional service providers rather than community-based organizations. The internal structure of the LEDAs proved effective in terms of the management of the programme, and the Credit Committees fulfilled their role as expected. Although there was formal linkage to the local authorities and other organizations through the Advisory Committees, this did not extend to their participation in managing and implementing the programme. Most of the Advisory Committees did not, however play a useful role, as their members had little to contribute or were not interested. This was generally not experienced as a disadvantage, as the LEDAs preferred to keep a respectful distance from local authorities, whom they expected would attempt to further their own interests through the LEDA programme, and whom they felt did not enjoy a great deal of trust among the population.

This is not to say that the LEDAs did not develop close relationships with local communities and organizations, and persons of official or unofficial authority within them. Such relationships became strong but were informal, based on contacts from person to person rather than with institutional mechanisms. This allowed the LEDAs to develop cooperation without sacrificing autonomy, and contributed to their achieving a high degree of legitimacy.

The LEDA programme

The programme developed with and implemented by the LEDAs consisted of a small business and an informal sector component. The objective of the former was to support starting or expanding small businesses, defined loosely but pragmatically as businesses which would require a loan from the LEDA of more than US$200; and which were likely to provide full-time employment to the business owner, as well as providing work for one or more other people.

The programme elements were the joint product of international project and LEDA staff, and the result of a series of pilot tests and revisions. It included:

a selection procedure which focused on the applicant's business idea, whether he or she had the required skills and resources, and whether he or she belonged to the target group. The core event was an interview. There were no written tests or assessments of entrepreneurial potential, which were considered inappropriate to the target group;
a six-day basic business skills training programme, which revolved around the development, by the trainee, of a simple business plan, and included an informal market assessment. The training programme was simple, concrete, and participative;
application for credit on the basis of the business plan. Loans were approved by the Local Credit Committee (up to US$1,000) or by a Revolving Fund Management Committee at the national level, in Phnom Penh;
credit. Loans were in US dollars, required a 20 per cent equity contribution and were fully collateralized. Collateral included the capital goods purchased with the loan. In addition, there was a requirement for guarantors. The nominal interest rate was a flat 10 per cent per year.⁽¹⁸⁾ The maturity was a maximum of two years. The maximum loan size was US$2,000.⁽¹⁹⁾
follow-up business advisory services, which made use of check-lists, as well as loan follow-up.

In addition, the LEDAs could refer clients to support services they did not provide, such as vocational training.

The informal sector component was based on research conducted by the LEDAs and the project, in collaboration with CARE, on the needs of micro enterprises, and was developed jointly by the project and LEDA staff. It consisted of a solidarity group lending programme aimed exclusively at women in micro enterprise, defined as small economic activities which would require a loan of up to US$50, and which would generally provide part-time employment to one person only.

The main steps in the programme were:

selection of a neighbourhood or village in which there was a high concentration of target group members and potential for micro enterprise;
group and individual meetings with key persons in the village and potential clients to explain the programme and ensure that participants belonged to the target group;
formation of groups of five to ten members who guaranteed repayment of each other's loans;
a one-day training programme to help the clients decide how to use their loan;
application for credit, and approval by the Local Informal Sector Credit Committee;
provision of credit. Loans were in riel, carried an interest of 2 per cent a week on a declining balance, and had a standard maturity of three months. The joint liability of group members was the only guarantee;
follow-up to provide basic business advice and address problems which might endanger repayment.

By the end of 1993, both components had gone through several revisions based on the experience gained and client feed-back, and had reached a form which was closely geared to the clients' needs and the local economic and social situation. This was indicated by the enormous and growing demand for the services, which made promotion activities mostly superfluous. The rate of implementation of business plans by small business clients increased to some 80 per cent.

The next year saw a thorough revision of the policies, procedures and administration of credit for small as well as micro business, as it was found that the basis on which it had been established did not allow sufficiently for the expansion of the programme then under way.

Also in 1994, a village bank programme was introduced in the informal sector component. This methodology, which was being widely used in Cambodia by international NGOs, aimed at joining solidarity groups into larger units, village banks, and giving them greater responsibility for loan administration. It was expected that this would allow the LEDAs to service more clients. It was found, however, that bank formation required more time and effort than had been foreseen, and that large numbers of solidarity groups were not interested in joining. As a result, the number of clients and the outstanding portfolio decreased dramatically. The assumption that the use of the village banking by international NGOs indicated that there was a demand for it had proven incorrect, and the programme was adjusted accordingly: early 1995 saw a shift back towards the solidarity group methodology.

The content of the programme, and procedures for implementation, were strongly standardized in both components. There was a significant degree of flexibility in the contents of training, which could be adapted to the specific needs of the trainee groups. Procedures for, for instance, selection and loan approval were, however, uniform throughout the programme. Standardization aimed at ensuring the quality of the programme notwithstanding the inexperience of the staff, and the realization of higher levels of productivity which would come with routine. It was also essential, of course, to achieving a high level of accountability.

The principle of giving clients a high degree of responsibility for the success of the services was realized by making them responsible for the business idea instead of offering such ideas to them, having them carry out market assessments themselves, and making them responsible for producing business plans. Training centred around individual and group activity, with even the minimum number of brief "lecture" sessions being based on interaction between trainees and the trainer's team. Loan repayments had to be made at the LEDAs where possible and, in the informal sector component, client groups were given important tasks in loan administration. These groups were formed by the clients, not by the LEDA staff. Such measures proved effective in terms of learning, the appropriateness of decisions made by the clients, and time use of LEDA staff.

Target groups

Returned refugees, internally displaced persons, demobilized soldiers, and disabled and women heads of household were identified as the war-affected, disadvantaged groups which the project needed to target. Clients were also expected to be low-income, and priority was accorded to women in general. In the small business component this was reflected in the selection procedure, which used a scoring system to ensure that clients would be in the target group. The informal sector component selected the localities in which it operated on the basis of information on the composition of the population, and potential clients were approached with the help of knowledgeable persons in the area. It targeted low-income women only.

Over time, it became less socially acceptable to offer groups of returned refugees or the demobilized such privileged access to programmes, and the conviction gained ground in the LEDAs that reintegration and local economic development were best served by a less "discriminatory" approach. It was also recognized that most population groups were in fact war-affected. Income and gender then evolved as the best general indicators of clients being in a disadvantaged position, while women heads of households continued to receive additional priority, due to their especially weak social and economic position and their childcare responsibilities.

Women's participation in the small business component was less than expected during the first months of programme implementation. A target of 50 per cent was therefore set, which forced the LEDAs to select more women. Gender awareness programmes were run for all staff. When experience showed that the success indicators for women clients were generally better than for men, the participation of women consistently exceeded this target.

Another important way in which it was attempted to increase the number of women clients and to serve them better, was by increasing the number of women working in the programme. This had not been a priority when LEDA staff were first recruited since, at that point, demobilized soldiers were expected to be the main target group. A policy to change this was put in place in 1993, and by 1995 more than 50 per cent of the staff were women.

The Association of Cambodian Local Economic Development Agencies

The logical consequence of the strategy to foster institutionalization and sustainability from the start was the early creation of a counterpart organization distinct from the project. Management felt that the establishment of this organization needed to be addressed as soon as the programme had become operational, before the LEDAs had settled into a project mode of operation and which would have been difficult to overcome.

Tentative discussions on this issue therefore took place with the national project staff just after the initial training programme for LEDA staff. These discussions gained urgency when it was found, in late 1992 that, under ILO regulations, the planned Revolving Fund could not be managed by project staff, as foreseen in the project document, but only by a separate legal entity. As there was no alternative to having the LEDAs handle credit matters, splitting the LEDAs off the project, the only way out of this dilemma seemed to be to separate the LEDAs from the project and establish them as a legally separate organization which would administer the Fund.

Opinion leaders among the national project staff were much in favour of this step, which they felt would give them an increased say in the running of the programme, and could eventually lead to complete autonomy. With regards to the LEDA's institutional status and framework, joining them to a government institution (such as the Ministry of Industry) was rejected. The existing administrative structures (there was no recognized government) were perceived as inefficient, ineffective and politicized. It was considered that they did not enjoy a high level of trust among the target groups or national project staff. The low salary levels of government staff would not allow the project to maintain its staff. In addition, international experience indicated that government institutions are generally less effective in reaching MSEs, especially those of the poor, due to bureaucratic procedures and attitudes.

Establishing the LEDAs as a commercial enterprise was considered unfeasible because it was unlikely that an MSE programme targeting the disadvantaged could be operated on a profit basis. The preferred option was therefore to establishing the LEDAs as one national or several local NGOs.

It was felt that local NGOs could have the advantage of eventually becoming more closely linked to the local business community and other relevant actors, and could be more responsive to specifically local needs. A national NGO, on the other hand, could have economies of scale, and had greater potential to develop its own capacity for staff and programme development. In addition, it would be in a stronger position to stay clear of interference from local political leaders. These arguments, combined with the esprit de corps which had begun to develop among the staff of the LEDAs, weighed most heavily, and a consensus was reached to join the LEDAs together in a national NGO.

National staff drafted the statutes of the organization, which came to be known as the Association of Cambodian Economic Development Agencies (ACLEDA). It was officially recognized by the Supreme National Council and UNTAC in early 1993, and a constituent General Assembly was held in February.

Over the years, the organizational framework for ACLEDA has undergone some changes, but its most important characteristics have remained constant (see Annex 3 for a diagram). ACLEDA's members are those working for it, and the organization is owned by its staff. Their General Assembly, held once a year, is its highest authority. ACLEDA has a Board elected from amongst the staff, whose main function is in policy-making. An Executive Committee, with an elected President and Executive Secretary, manages the organization, including its external relations, while a team of technical specialists is in charge of programme development and staff training, and provides external consultancy services on a fee basis. A Revolving Fund Management Committee, which includes external members, advises the Board on credit policies, and approves larger loans. There is no mechanism at the national level (such as an advisory board) to involve authorities or other organizations in the management of CLEDA.

The LEDAs, whose structure has already been described, are supervised directly by the Executive Committee. This relationship is characterized by a high degree of decentralization, with LEDA Team Leaders having the authority to take all necessary decisions on the operation of their programme within the regular budget allocation.

This framework provided for a high degree of accountability of policy-makers and executives to the staff which allowed a strong sense of ownership and responsibility to develop throughout the organization. Lines of authority were nevertheless well defined for the effective management of the programme. Decentralized management increased efficiency and responsiveness to local needs. The inclusion of a team of technical specialists gave the organization the capacity to sustain and further develop its programme.

The absence of a formal mechanism of participation by, and consultation with, national actors enabled the organization to develop without political interference, and further contributed to the sense of ownership among its staff. ACLEDA did develop strong relations with organizations at the national level, including in the Government, but since these were external to its own organizational structure, they allowed it to maintain autonomy while at the same time benefiting from cooperation.

The process of institutionalization

With the establishment of ACLEDA, the project's strategy to institutionalize the MSE programme became focused on developing ACLEDA as an independent and sustainable NGO for the promotion of MSEs. From the second half of 1993, the project and ACLEDA planned and implemented a series of interrelated interventions, none of which were foreseen in the project document, to develop the organization's capacity in each of the four areas of sustainability that had been identified.

The development of a Strategic Plan for ACLEDA, between October 1993 and March 1994, was a keystone in this programme. It aimed at generating a shared sense of mission among ACLEDA staff, as well as setting the general direction for the organization's development. A methodology was designed by the project in collaboration with the ACLEDA leadership, which allowed all staff to participate and contribute to the development of the Strategic Plan, while Team Leaders and the Executive Committee were responsible for the actual drafting. The Executive Committee managed this process, with guidance from the project; the impact in terms of organizational sustainability was enormous. Within a relatively short period of time, ACLEDA was transformed from an organization in name only into one to which its members felt a strong commitment, and which they believed had an important role to fulfil in rebuilding the country. The Strategic Plan was accepted by the General Assembly in March 1994.

A second key intervention was the agreement between the project and ACLEDA management on a schedule for the transfer of authority. This schedule identified the main decision-making authorities in the management of the programme, and set deadlines for a project authority to become a joint authority, and finally an ACLEDA authority. The schedule was signed by both parties as part of an official Memorandum of Understanding, and generally implemented as planned. By the end of 1994, all authorities had been transferred to ACLEDA, and the project contracted it to carry out the programme.

This approach allowed the transfer of authority to be an orderly and gradual process, linked to interventions aimed at building capacity. In addition, it enhanced ACLEDA management's confidence that the project was indeed committed to full autonomy for ACLEDA, and allowed it to counter sceptics within the organization. It thus made an important contribution to organizational sustainability.

Organizational sustainability was further strengthened by the development and implementation of a strategy to develop ACLEDA's key institutional linkages, including those with the Government. The interventions which addressed the need for capacity building for managerial and technical sustainability included:

the elaboration of the organizational structure and terms of reference for the main bodies and posts;
the establishment of a management information system, financial administration system, and personnel management system, and the appropriate manuals;
a continuous programme of management training at all levels in the organization: i.e. for the Board, the Executive Committee and team leaders;
a continuous programme of technical training for all LEDA staff members as well as for the team of technical specialists in charge of staff and programme development.

The participative manner in which these interventions were implemented by project and ACLEDA management resulted in the creation of an organizational culture which was both comparatively open and egalitarian, values which had not been encouraged in recent Cambodian history. The initial attitude had been that it was up to management to take decisions, and up to the rank and file to implement them whatever the consequences which, if they were bad, would be covered up. Gradually, however, a climate developed in which issues relating to the programme and organization could be discussed openly and a freer flow of information was possible. How difficult and yet how important a process this was, was demonstrated by the failure of village banking, which could have been detected earlier if there had been a greater ability to question management's decisions and provide it with the necessary information.

A consequence of ACLEDA's growing capacity was a notable decrease in the size of the expatriate team after the first year, and again in subsequent years. Its growing independence was reflected in a diminishing role of the project, which eventually took on a purely technical and advisory function.

With regard to financial sustainability, the project prepared a paper indicating the most important options as an input into the process of formulation of the Strategic Plan. This made a clear distinction between the concepts of self-financing and financial sustainability. It presented the possibility of achieving the latter by a mix of income from programme activities and possibly services to other organizations (self-financing), and continuing donor support. It was thought that such support would be likely to be forthcoming for a long period of time if the organization could achieve a high degree of legitimacy, and if the cost of delivery were sufficiently low to make continued support attractive.

The Strategic Plan accepted the main elements of this approach. Rather than experimenting with a variety of income-generating activities which would detract from the main function of providing MSE development services, income from interest earned on loans was seen as the mainstay of self-financing. However, it accepted that donor support would probably be necessary even in the long term.

The beginning of 1995 saw a further development in strategy, with the project's and ACLEDA management's growing awareness of the latest thinking on MSE programmes, particulary micro finance.⁽²⁰⁾ This convinced them that full self-financing would be possible, if there were a specialization on credit and if certain principles were applied to allow the programme to increase its scale and decrease its costs. These principles were generally an extension of those which had been applied already: a market orientation towards the clients; decentralization, standardization and automation; giving clients responsibility in the implementation of services, and using group approaches to credit.

Credit specialization was also felt to be appropriate because credit was the business development service most needed by the poor in Cambodia; and large numbers of the poor could benefit from financial services. The strategy thus became to develop ACLEDA as a fully self-financing, large micro finance institution for the poor, which was to be the main objective of the project which succeeded SEISP in 1996. Interest would have to become the main source of income, and rates would have to be set accordingly. Although training and counselling were considered essential for the success of financial services, and there was no question that they should be discontinued, they were no longer seen as independent services.

Increasing the scale of operations had been the intention from the start, as indicated in ACLEDA's Strategic Plan. The organization had already been able to realize an expansion from five to nine LEDAs and, in 1995, made a qualitative step forward when it obtained funding for the establishment of district offices from the UNDP-funded, OPS-executed Cambodian Resettlement and Rehabilitation Programme (CARERE).

3.4 Monitoring and evaluation

The monitoring of the project as such took place through the usual mechanism of project performance evaluation reports and tripartite review meetings. These established which activities the project had completed and which outputs produced. They, and the detailed workplans which the project produced, were important mechanisms for keeping the UNDP, the Government, and the ILO informed of the course of the project, and obtaining their support. In addition, an independent tripartite evaluation of the project was conducted in May 1994.

More important in the long term, however, was the management information and impact assessment system which the project established with ACLEDA. This was designed in the first place to strengthen ACLEDA's capacity to manage and develop its programme. A second consideration was its organizational sustainability: it had to be fully accountable to its donors, through up-to-date reporting on its inputs and outputs. The project and ACLEDA therefore designed and established systems which provided monthly updated information on all important performance indicators, including numbers of clients trained, gender of clients, numbers provided with credit, loan disbursement and portfolio outstanding, and repayment rates. A financial administration system was set up which allowed the organization to produce all necessary financial statements, including balance sheets and income, expenditure and cash flow statements. The accounts for 1994 and 1995 were audited by an international accounting firm.

Cost and ease of use were major considerations in the design of these systems, as they were not intended for the one-time use of the project, but for ACLEDA's continued implementation of the programme. The focus was on providing ACLEDA management with the key information it required, for the smallest number of inputs.

This chapter has offered a description and analysis of the strategies developed by the SEISP project and their implementation. The project's approach was one which combined immediate delivery of services to MSEs with a strong emphasis on sustainability. Both were seen as two closely related aspects of the same process, and sustainability was addressed in a comprehensive manner, taking account of all ACLEDA's needs as an organization. In the next chapter the relations of the project to other components of the Employment Generation Programme will be briefly considered, followed, in Chapter 6, by an assessment of the project's achievements in terms of its immediate impact, the sustainability of the programme and organization it has helped establish, and cost effectiveness.

4. LOCAL ECONOMIC DEVELOPMENT: LINKING MSE DEVELOPMENT WITH VOCATIONAL TRAINING AND EMPLOYMENT-INTENSIVE WORKS PROGRAMMES

The Employment Generation Programme of which the SEISP project formed part, included two other components (projects), on "Vocational Training for Employment Generation" and "Labour-Based Infrastructure Rehabilitation". Although the projects were formulated simultaneously, they were not designed as a single system, but rather as three interlinked projects, which could also have been implemented independently. The strategy was to have three "free standing" but coordinated projects, to ensure that each would have the flexibility to develop in accordance with the changing needs of the environment, while at the same time making use of opportunities for collaboration and the mutual reinforcement of impact.

There were no joint objectives and outputs, and the project documents did not describe the conceptual relationships between the programme components. This was done only in 1994, in a concept paper which basically described the linkages which had developed between the projects, but still gave no joint objectives and outputs.⁽²¹⁾ From the point of view of the SEISP project, however, the two other projects addressed obstacles to local economic development which it could not tackle. The impact of the infrastructure project could be expected to be the increase in the size of markets, through improved opportunities for trade, and greater purchasing power. The vocational training project would address the need for upgrading of technical skills.

The project documents described the mechanisms for coordination, and stressed the need for integration. It was left to the three project managers, however, to work out what exactly needed to be integrated. The provinces covered were not the same for each project. Given, in addition, that the projects started at different dates, and under great time pressure, it was only towards the middle of 1993 that effective coordination began at the national level.

Each of the project managers took the function of programme coordinator on a rotational basis, and weekly meetings were held to discuss issues of common concern. Coordination at this level was most effective in external relations, i.e. vis-à-vis UNDP, the ILO, donors and the Government, as common programme interests were clear. With regard to internal coordination, as there were no common programme objectives or outputs, it was difficult to agree to joint activities where there were conflicting demands on time and resources. It is probably fair to say that, with some exceptions, each of the project managers gave priority to realizing the objectives of the project he managed, rather than allocating scarce resources to contributing to those of another. Although project workplans were exchanged, there was no joint programme planning, and little discussion of substantive issues.

At the local level, programme coordination as such never took root. The LEDAs in particular were very protective of their independence, and often resisted what they felt was the imposition of the interests of ILO projects on their activities. The different approaches taken by the three projects to the issue of institutionalization played a role in this: there was a certain lack of understanding of the implications of the establishment of ACLEDA in the other two projects. Eventually, productive relations were developed both at the national level and in most provinces, when there was a clear common interest in collaboration. This was nearly always on a project-to-project rather than on a programme basis.

With regard to the infrastructure project, the main instance was the involvement of ACLEDA in the training of small construction contractors, who were potential clients. With the vocational training project, a more long-term and structural relationship was developed. It contracted ACLEDA to provide the services of its small business component to vocational trainees in two provinces where LEDAs had not been established. In addition, ACLEDA was contracted to place vocational training liaison officers in each of the LEDAs. These officers had the task to work with vocational training project staff on employment needs assessments, selection of trainees, and informing them about ACLEDA's services. This resulted in a significant increase in the numbers of vocational trainees who were accepted in LEDA programmes, to some 30 per cent.

Although LEDA staff generally stated that they would have had sufficient numbers of viable clients without these trainees, the introduction of some (particularly agricultural) skills did contribute to their being able to reach larger numbers of rural clients. On the other hand, the collaboration on training for skills which were not conventional in the Cambodian context was on the whole not very effective. Trainees in these programmes generally did not reach the level of proficiency required of a business owner, or were eventually daunted by the risks involved in creating a market for a new product. With regard to conventional skills, the majority of ACLEDA clients actually made use of the traditional apprenticeship system which provided training which was both appropriate and cost effective. Although there was a need for upgrading of skills in MSEs, this could not be met by the brief, entry-level courses of a project whose objective was to train the unemployed.

A possibly more important effect of the projects working in concert was not the result of joint activities, but of joint geographical targeting. Where feasible, the LEDAs' targeted areas where roads or irrigation works were being rehabilitated by the infrastructure project. The short-term employment this created provided an infusion of cash (and food) into local communities and therefore resulted in an immediate increase in purchasing power. In the longer term, roads opened up new markets, and irrigation enabled farmers to grow more crops per year. This had a clearly visible effect on economic activity, and it is in this regard that the mutual reinforcement between this and the SEISP project was greatest. ACLEDA contributed to a boom in the micro and small business sector in such areas, as training and credit enabled its clients to make use of the new opportunities created by the infrastructure.

5. ASSESSMENT OF APPROACHES

5.1 Impact

The impact of the project in terms of local economic development, understood as the growth and diversification of economic activity in the areas covered by the LEDAs rather than just among their direct clients has not yet been assessed. At the time of writing, plans for such a study are being formulated. ACLEDA does, however, report regularly on the progress of its programme in terms of service delivery, and has also carried out studies of their impact on individual clients (the main information is summarized in table 1).

Table 1. Output and impact of ACLEDA's programme

	Small business component	Informal sector component
No. that received training only	1 106	Not applicable
No. that received training and credit	3 118	5 837
Per cent of women of those who received credit	71%	99%
Business success rate	97%	Not available
Loan recovery rate	96%	99%
Increase in income (average)	65%	45%
Jobs created per business (average)	1.8	Not available

In terms of outputs, between the start of the programme in March 1993 and the project's end in December 1995, 4,224 clients received small business training (under the small business component). Of these, 3,232 (76 per cent) implemented their business plan and loans were provided to 3,118; 41 per cent were start-ups, the others expansions; 37 per cent were in production, 21 in services, 16 in trade, and 26 in agriculture. The business failure rate was 3 per cent, while the loan recovery rate was 96 per cent.

Each of these indicators compares well internationally, and is higher than the targets set by the project document. The rate of business plan implementation was some 80 per cent for training programmes held in 1995.

Sixty-seven per cent of those trained and 71 per cent of those who received loans were women, more than targeted in the project document. On the other hand, returned refugees accounted for just 6 per cent, and the demobilized and disabled for 3 per cent each. Although the project document does not mention target percentages for these groups, this is lower than might have been expected. The main reason is that initially there was little interest among returned refugees, who had a year's supply of food, and many of whom felt they had not settled permanently. Later, ACLEDA considered it more appropriate to target those with low incomes in general, and stopped collecting information on the participation of returned refugees.

With regard to the demobilized, these were a much smaller group than had been foreseen, as the demobilization planned for 1992 was not actually carried out. In addition, ACLEDA's experience with them was quite negative: there were many deliberate defaults, and ACLEDA staff was threatened frequently. The rate of participation of the disabled, on the other hand, is respectable for a programme which is not specifically designed to meet their needs.

An ILO study for the Government has recently established the poverty line in Phnom Penh at US$149 per month; in other towns at US$89; and in rural areas at US$74.⁽²²⁾ ACLEDA reports an average monthly starting income of its clients of some US$130, which in Phnom Penh is below this poverty line, but not elsewhere. Moreover, the average probably masks a wide range, as well as regional differences.

A widely used indicator of the poverty of clients in lending programmes is the size of loans. This is based on the assumption that those with high incomes are generally not interested in small loans. The average ACLEDA small business loan size of US$475 compares well with other programmes internationally. Neither this nor the reported incomes are conclusive, however, and little more can be said than that ACLEDA's small business clients include the poor as well as the non-poor. ACLEDA recognizes this, and includes the "underserved", i.e. those who could benefit from assistance, credit in particular, but are not able to obtain it through other channels, in its target group. The rationale is that assistance to such clients contributes to local economic development, and will create employment and new opportunities for those with lower incomes.

Impact assessments which covered enterprises assisted in the first 18 months of the project found that the incomes of entrepreneurs who had implemented their business plan had increased by an average of 65 per cent.⁽²³⁾ This is an average increase of US$85 per month. Per enterprise 1.8 jobs were created (including, in start-ups, the entrepreneur and including family members). No information is available on the incomes of employees. As the assessment was carried out when 12 months or more had elapsed since completion of training, these gains in employment and income may be judged sustainable. The average increase in income is high, but the number of jobs created is lower than that foreseen in the project document, three per enterprise. This is understandable given the small size of the large majority of enterprises in Cambodia.

Under the informal sector component, 5,837 clients were provided with training and loans; 9,352 loans were disbursed. This is much less than foreseen in the project document, 12,000 over the first two years. This target was a guess made without any knowledge of the needs of micro enterprises or the methodology that would be used. At the time of project formulation, it was recognized that it was probably an over estimation.

The recovery rate for micro loans was 99 per cent, more than expected in the project document and on the high end of the range usual for such programmes (90 to 100 per cent).

Women made up 99 per cent of the clients. ACLEDA does not collect information on the extent to which this includes heads of households or returned refugees. The average income of clients was about US$65 per month, well below the poverty line. There is probably a wide diversity in client incomes: a study carried out in 1994 found that some 20 per cent of clients did not belong to "poorest 50 per cent of a community". However, the poor (as defined earlier in this paragraph) would probably make up 70 to 80 per cent of a community. The data therefore indicate that this component is well-targeted which is confirmed by the size of the loans, on average around US$3. ACLEDA does, however, believe that even this component does not reach the very poorest who, in its experience, are generally in such a marginal financial positions that loans are generally not used productively, and therefore cannot be repaid.

Two in-depth impact assessments were carried out, which found an average income increase of some 45 per cent among informal sector component clients, about one year after the provision of the first assistance.⁽²⁴⁾ The first study found that 67 per cent had increased their income, the second, 84 per cent. Improvements were also found in ownership of assets, school attendance, medical care, and housing. The studies concluded that the impact is significant, and is likely to be sustainable.

5.2 Sustainability

As described in Chapter 3, the project has implemented a range of interventions aimed at increasing the organizational, managerial, technical and financial sustainability of ACLEDA. Towards the middle of 1995, the project did a final assessment of progress made, which was included in the formulation of a follow-up project whose objective was to be the full sustainability of the organization. The context was the process set in motion by the ACLEDA leadership, in the beginning of 1995, to redefine its role as a large-scale financial institution for the poor.

This had significant implications for the analysis of ACLEDA's sustainability. It was found that ACLEDA had realized a high degree of organizational sustainability. It had a vision of its role in the development of Cambodia and the improvement of the position of disadvantaged target groups, which was accepted and supported by its members as well as its environment. Stakeholders and colleagues considered it to work effectively and efficiently on the realization of this vision. This was demonstrated, for instance, by ACLEDA's election, in 1995, by its colleague NGOs to be one of three NGO representatives on the Credit Committee for Rural Development under the Cambodia Development Council. It was the only national NGO on the Committee. ACLEDA had its own network of supporting relationships, with national and local authorities, donors, and other NGOs. It was perceived as an organization separate and autonomous from the project.

However, it was recognized that for full organizational sustainability, further work needed to be done to develop ACLEDA's vision of itself as a financial institution for the poor, and to build commitment to this vision among its members. It was foreseen that this would require limited support from the project, especially at the Board and Executive Committee level.

With regard to managerial sustainability, this had been achieved at the existing level of operations and in the realization of significant expansions, as demonstrated by ACLEDA's autonomous management of its programme and the establishment of additional LEDAs. The organizational structure functioned well and management information and financial, administrative and personnel systems were all in place. Management skills were appropriate.

In the framework of specialization on financial services and scaling up, it was considered that some rationalizations in the organizational structure and further decentralization would be required. ACLEDA also needed assistance to meet the legal requirements for micro finance institutions, which were then under formulation, and which could involve a change in ownership structure. Further computerization was necessary, and budgeting systems needed to be developed. Additional training at the policy making level was considered important.

ACLEDA's technical sustainability was found to be strong: a high degree of professionalism had been achieved in the LEDAs, and there was no involvement of the project at that level. ACLEDA's team of technical specialists had a demonstrated capacity to develop and implement training programmes, which was being used increasingly by other organizations. Again, however, it was felt that the specialization in financial services would require further project assistance, especially in the upgrading of the staff's financial skills. It was also found that the rapid expansion through branch offices would require additional technical inputs from the project.

By the end of the SEISP project, it might therefore have been possible to consider ACLEDA a sustainable organization in terms of legitimacy and relations, management and technical capacity, but for the strategic choice, made by the organization in 1995 to become a large scale micro finance institution for the poor. This choice was aimed primarily at reaching financial sustainability. ACLEDA's figures show that, in 1994, it could have been self-financing for 18 per cent, in 1995 for 27 per cent (table 2). Projections made in 1994 indicated that 50 per cent would probably have been the highest degree of self-financing without specialization and scaling-up. For financial sustainability to be achieved, the organization would then have been permanently dependent on donors, which was at that point no longer acceptable to ACLEDA management. The latest information from ACLEDA and the project, based on ACLEDA's new strategies, indicates that, by the year 2000, full self-financing of all operations can be achieved.⁽²⁵⁾ This is conditional on sufficient loan capital being available, continuation of the relatively positive macro-economic trends, and a liberal regulatory environment for micro finance institutions. An appraisal carried out by the World Bank concluded that "... financial trends are positive across the board ... and operational self-sufficiency is climbing steadily".⁽²⁶⁾

Table 2. ACLEDA's income and expenditure ⁽²⁷⁾

1993	1994	1995
Expenditure (US$)	298 304	407 295	517 275
Income earned (US$)	19 552	72 007	141 812
% self-financing	7	18	27

In summary: the SEISP project did not achieve full sustainability for ACLEDA, but it laid firm foundations in particular with regard to organizational, managerial and technical sustainability. The decision to aim for full self-financing required a follow-up project to provide further, though comparatively limited, technical assistance in all areas of capacity building. This project will be completed by the end of 1999, and it is expected that in the year 2000 ACLEDA will have achieved full sustainability in all its aspects.

5.3 Cost effectiveness

As the project had two main objectives, i.e. an immediate impact in terms of MSEs assisted, and sustainability of the programme, its cost effectiveness must be considered in those terms. It is therefore useful to separate the expenditure made on each account. Here this has been done by taking the cost of ACLEDA's operations as those made for immediate impact, and the expenditure on technical assistance as the cost of capacity building. This is rather simplistic, but not illogical, and in addition the only way possible.

It is then found that over the project period ACLEDA's cost per loan made was US$98 (table 3). The cost per US$ lent was US$0.61. This is relatively high, although many NGOs operate at a dollar expenditure for a dollar lent.⁽²⁸⁾ Moreover, it should be taken into account that training and counselling services, as well as the activities of the team of technical specialists in programme development, are included in this cost. This is estimated to account for 60 per cent of the total.

Table 3. Cost of delivery over project period (1993-95)

ACLEDA cost (US$)	1 217 874
Number of loans made	12 470
Total loan disbursement (US$)	2 010 327
Cost per loan (US$)	98
Cost per US$ lent	0.61

More important as an indicator for ACLEDA's future is the decrease in cost it has been able to realize. Table 4 indicates that in 1993 the cost of serving a client was US$202, while in 1995 this had decreased to US$79. In 1993, the cost per outstanding US$ was US$1.33, while in 1995 this was US$0.41. There are clearly significant improvements in cost-effectiveness, which bode well for ACLEDA's future.

Table 4. Cost over outstanding portfolio

	1993	1994	1995
Expenditure (US$)	298 304	407 295	517 275
Active loan clients	1 475	2 344	6 539
Outstanding loans (US$)	224 067	367 123	1 250 260
Cost per active client (US$)	202	174	79
Cost per US$ outstanding (US$)	1.33	1.11	0.41

Per job created (under the small business component), ACLEDA's cost is about US$126. With certain assumptions, the crude benefit-cost ratio based on additional income generated is 2.36 (table 5).

Table 5. Cost of impact over project period (1993-95)

1. ACLEDA cost (US$	1 217 874
2. Cost small business component (US$)^a	730 724
3. No. of jobs created, small business component^b	5 818
4. Additional income created (US$)^c	2 871 450
5. Cost per job created, small business component (US$)^d	126
6. Crude benefit/cost ratio^e	2.35
^a Taken at 60 per cent of the total. ^b 1.8 jobs for each small business component client. ^c 65 per cent increase for small business component clients; 45 per cent for informal sector component clients; workers in small businesses assumed to be at the same level as informal sector component clients; income increases assumed to be sustainable over a period of five years (but no further real growth). ^d Row 2 divided by row 3. this takes into account the small business component only. Information on employment generated by the informal sector component is not available. Anecdotal information suggests that clients work more hours after a loan has been given, and that in some cases additional jobs are created, mostly for family members. ^e Row 4 divided by row 1.

The investment made in ACLEDA as an organization until the time it is expected to be fully sustainable (i.e by the end of the present follow-up to the SEISP project, -- in December of 1999) is expected to amount to an estimated US$11 million, including US$5 million loan capital. Insufficient information is available for an adequate investment analysis. What can be said at this point is that a very rough estimate of the number of loans which ACLEDA could disburse over a period of 20 years from 1993 (perhaps a minimum period for a sustainable organization), without further expansion of its network of branches, could amount to some 1 million.⁽²⁹⁾ The investment in terms of technical and capital assistance provided would then amount to US$1.10 per loan made. If one assumes that clients take on average four loans each (repeat loans are the rule especially for micro businesses), the investment cost would be some US$4.40 per client.

This chapter has provided a summary assessment of the SEISP project's achievements. These can be judged to be significant, in terms of impact, and the sustainability of ACLEDA. Against this background, the following and final chapter will consider which lessons can be learned from the implementation of this project.

6. LESSONS LEARNED

The "lessons learned" or, more often, confirmed by this case-study, can be divided into those which concern the environment in which it was implemented, project design and implementation, and the linkages to the other projects in the Employment Generation Programme.

The environment

The project implemented a successful MSE promotion programme in an environment which was unstable politically, economically and with regard to security. There was, however, a good measure of peace in most of the country; the economy was in rapid transition towards a market economy; and there was significant economic growth. Many potential or actual MSE owners did not have the resources or skills to realize their business ideas. This meant that MSE development services were both needed and in demand. Economic opportunities were being created in spite of instability, and the programme was able to assist disadvantaged groups to make use of them. The project has demonstrated that it is possible to develop and implement a successful MSE programme in an unstable and difficult postwar situation, provided there is growth in economic opportunities for MSEs and they have the freedom to make use of them.

Other important factors were the project's relationship with the Government, the UNDP and the ILO. The absence of close government involvement, due to its political preoccupations and the fluid institutional environment, allowed project and ACLEDA management to take decisions with little political or bureaucratic interference. This effect was strengthened by the early start of the project, when political turmoil was at its height. The UNDP and the ILO were willing to consider the project's decisions and interventions on their strategic and technical merits, rather than using political and bureaucratic criteria or setting them against the letter of the project document, and gave their support on that basis. This case-study indicates that it is of key importance to the development of MSE programmes that a project be given such freedom and flexibility to develop its strategies in accordance with its growing knowledge and experience of the local situation. This is especially true in rapidly changing and uncertain postwar situations, which require frequent adaptation and course corrections.

Project design

The flexibility with which the project was implemented was also made possible by the project document. Although this did give detailed descriptions of the intended project strategy and the programme that should be delivered, it also recognized that little was known about local economies, the needs of MSEs, and the way the political and institutional context would develop. In general, but especially in situations of rapid social and economic transition, project documents should provide a high degree of flexibility for strategies and methodologies to be changed and developed during project implementation.

This should be so particularly when there has been little opportunity for participation of target groups and potential stakeholders in the project formulation phase. As in Cambodia, such participation may often not be feasible immediately after a peace agreement. A flexible project design may then allow project management to develop the programme step by step on the basis of immediate and continuous feedback from clients, which will result in an appropriate, demand-driven programme.

Project implementation

The project started with a period of preparatory assistance, when project formulation was still ongoing. This gave it a head start when the document was approved, and allowed it to initiate assistance to MSEs briefly afterwards. In situations where providing assistance to target groups is considered a matter of urgency, a preparatory assistance phase can be very effective.

Instead of leaving issues of institutionalization and sustainability to be dealt with in a report to be submitted by the end of the project (as foreseen in the project document), the project developed and implemented a strategy which addressed them from the very inception of the MSE programme. In spite of the urgency of realizing an immediate impact in terms of enterprises assisted, sustainability was not seen as a secondary issue, but was placed right at the heart of the project's strategy. The MSE programme thus never had the chance to function as a project (dependent on donors and "experts" and, by definition, of limited duration), but operated from the start as a (nascent) organization (to be run by Cambodians, and be sustainable). By ACLEDA's realization of a high degree of autonomy and sustainability in a short period of time, the project has demonstrated the feasibility and appropriateness of this approach. This may be the most important lesson to be learned from it.

Most important with regard to postwar situations, where considerations of a quick impact are often uppermost in the minds of donors and politicians, is that the project showed it is possible to assist large numbers of MSEs. Although it is inevitable that the shift of resources towards interventions for sustainability was at the expense of some immediate impact, stakeholders were satisfied with the numbers reached. The project has demonstrated that the objectives of immediate impact and sustainability can well be combined, if they are considered as two mutually reinforcing aspects of project interventions.

The project's approach towards institutionalization of the MSE programme has been comprehensive. It has addressed all aspects of sustainability, i.e. organizational, managerial, technical and financial sustainability. Interventions were designed and implemented to address all ACLEDA's needs as an organization, including those in areas which had very little to do with MSE development. The project and ACLEDA have demonstrated that this is a correct approach. Projects which aim at sustainability for an MSE programme should include thorough organizational audits, followed by a comprehensive programme of capacity building and organizational development. Achieving sustainability clearly requires more than the standard interventions most projects provide for, i.e. technical training and schemes to generate income.

It was the project's strategy to first build organizational, managerial and technical sustainability. Less priority was given to financial sustainability, and long-term dependence on donors was initially considered inevitable. The question should be asked whether more could have been achieved if full self-financing had been aimed at from the start. Based on the most recent international research findings, the answer should probably be positive.⁽³⁰⁾ It is also indicated by the fact that some aspects of the organization (for instance, the financial administration and credit procedures, parts of the management information system) would then have been set up differently from the start; and the enormous increase in productivity which followed ACLEDA's decision to become a self-financing micro finance institution, in 1995. On the other hand, ACLEDA taking this decision itself (instead of the project imposing it), and then realizing an expansion and rapid increase in the number of clients, would not have been possible in 1993 when operations started. A capacity had been created which was the basis for this development, and to that extent the project's approach may have been appropriate.
Most research done on MSE development programmes over the past ten years seems to suggest that financial and non-financial services should be dealt with by separate organizations. It has become usual to speak disparagingly of integrated programmes, especially with respect to their limited outreach and low potential for self-financing.⁽³¹⁾ The project's and ACLEDA's experience in this regard is not unambiguous. ACLEDA did find that the combination of financial and non-financial services was appropriate to the target group and objectives of the programme, as demonstrated by low business failure rates, high loan recovery rates and significant impact on incomes and employment. Its decision to aim at becoming a micro finance institution recognized that, in Cambodia, credit is the most important need, but did not decrease its commitment to maintain training and counselling, which it considers essential to the effectiveness of credit. In the absence of a banking system which can take on the role of credit provider, the management of an integrated programme by a single organization may also be an appropriate (or the only possible) strategy.

On the other hand, the experience with ACLEDA does confirm the limited outreach and potential for self-financing of an integrated programme which measures output in terms of businesses assisted with any service. It indicates that such a programme has a greater potential for outreach and self-financing if its output and success indicators are defined in relation to financial services and the income they generate: i.e. in terms of loan disbursement, portfolio outstanding, development of a client base, recovery rates, and income from interest in particular. Non-financial services will then be considered as contributing to the success of credit rather than as autonomous services, and may benefit from the greater financial rigour that comes with a focus on credit. The result may be an integrated programme which is sustainable, as ACLEDA's experience seems to indicate.

At least two elements of the project's programme for sustainability are important enough to mention here, as they are likely to be transferable to most similar projects. The first is the approach to the development of a strategic plan for ACLEDA. On the basis of the project's experience, it can be concluded that projects should not develop strategic plans for organizations, or be involved in determining their contents. This is the task of an organization, while a project may play a useful role in developing a methodology by which the strategic plan can be formulated, and guiding its use. Genuine involvement of all staff members of an organization in this process rather than just its management, is a powerful tool for building an organizational culture, a feeling of ownership, and a commitment among staff to the organization's goals, factors which are essential for an organization's sustainability.

Of perhaps equal importance was the project's development of a schedule for the transfer of authorities to ACLEDA, and the implementation of this schedule not in name only, but by the assumption by ACLEDA of actual programme management and implementation responsibilities. Where institutionalization of a programme is envisaged, projects should plan the process by which an institution takes over responsibility, and plan and implement its capacity building interventions on that basis.

ACLEDA is not a community-based organization, although it has very close links with members of the communities in which it works. Rather, it is a professional provider of services to MSEs for which it finds there is a demand, and for which its clients are willing to pay (in interest on loans). It has realized that this is in the interest of its own target groups as well as its own sustainability. In that way, it operates in a "business-like" way, although this is tempered by the organization's vision of its role as an institution serving the poor. Its structure, information systems and organizational culture ensure that it is a learning organization, which develops in response to the needs of its clients as well as its own internal requirements. This ensures that it can play an important role in Cambodia's society and economy, and that is has the potential for full sustainability. With regard to the institutional format, it may therefore be concluded that the project and ACLEDA have demonstrated that a professional development NGO which itself operates in many ways like a business and has a high learning capacity can be an appropriate and effective organizational form for the delivery of an MSE programme, including in a postwar situation.⁽³²⁾

Programme linkages

The project was implemented as part of a programme consisting of three projects. Implementation in the same geographical areas of the project on Labour-Based Infrastructure Rehabilitation has probably contributed to the creation of business opportunities. The LEDAs were able to support their clients in making use of these opportunities, and in this way these projects have reinforced each other's impact on local economic development. However, considering that there was a general climate of economic growth across the country, the need and demand for the LEDAs' services would probably have been more than they could meet even without this. It may be concluded that the linkage was useful, but not essential in this situation. This may, however, be different in situations of little or no economic growth, where roads and irrigation could have a greater independent impact.

The linkage with the Vocational Training for Employment Generation project allowed the LEDAs to support clients in the realization of business plans for which they did not have the vocational skills. However, training related to innovative businesses was often unsuccessful as the levels of skill reached were insufficient for an independent business owner or because in the end trainees did not dare take the risks involved. Training for conventional businesses was also being provided through the traditional apprenticeship system. Although useful, this linkage was therefore not essential either. This may be different in situations where there are fewer opportunities for low-cost training in the private sector.

From the point of view of MSE development, the strategy to have the SEISP project implemented as a free standing project linked to other components of a programme, was successful. In this situation the project was able to give the highest priority to the development of ACLEDA and its programme. It is unlikely that in an integrated programme, where programme priorities rank highest, the same could have been achieved. That said, one lesson that may be learned from the project's experience is that if the intention is to implement an integrated programme for local economic development, it should be designed as such, i.e. with programme objectives and outputs, and an integrated system of interventions; and it should be implemented under one manager, with joint work planning, and joint monitoring and evaluation. Three projects do not make a programme, although three projects may be able to achieve more than a programme each in their own field of competence.

BIBLIOGRAPHY

ACLEDA, 1993 (1): Statutes.
--, 1993 (2): Strategic plan.
--, various years, ACLEDA in figures and statistical progress report. <

Davies, M.; Hakemulder, R; Kooi, P: Small enterprise development in demobilization and reintegration; the ILO experience in Cambodia (ILO, 1995 (1)).

Edgcomb, E.; Cawley, J. (eds.). 1993. An institutional guide for enterprise development organizations; for the Small Enterprise Education and Promotion Network.

Evans, E: Small and medium-scale enterprises in Cambodia; characteristics and constraints (World Bank, 1996 (1)).

Ferrera, M.D.: Small enterprise and informal sector promotion, report of the evaluation mission (Royal Government of Cambodia/UNDP, 1994).

Folkard, A.: The urban informal sector in Cambodia; characteristics and constraints (SEISP/CARE, 1993).
--.; Phal Pisey; Ushijima R.: Informal sector component impact assessment report (SEISP/CARE, 1994).

Foster, S.; Webster L.: Financial services for the poor in Cambodia: and overview and analysis (World Bank, 1995 (2)).

ILO, 1994. ILO concept paper towards an employment generation programme for Cambodia.

Levitski, J. Innovations in the financing of small and micro enterprises in developing countries, (ILO, 1993).

Otero, M; Rhyne, E. (eds.): The new world of micro enterprise finance, 1994.

Ouk Rin; Alexander Sun; Hakemulder, R.: Opportunities for the economic reintegration of veterans (Royal Government of Cambodia/World Bank, 1996).

Royal Government of Cambodia/Caisse Française de Developpement, 1994: Etude de faisabilité d'un projet de credit rural et decentralise au Cambodge, by Horus, Banque et Finance.

Royal Government of Cambodia/National Institute of Statistics, 1995: Report on the socio-economic survey of Cambodia 1993/94 (all rounds).

Royal Government of Cambodia/UNDP, 1995: Alleviation of poverty through ACLEDA's financial services, project document.

Schmidt, R.H.; Zeitinger, C.P.; 1994: Critical issues in small and micro business finance, for Interdisciplinare Projekt Consult.

SEISP, 1993 (1): Financial sustainability, discussion paper for ACLEDA.
--, 1993 (2): Approaches to assisting ACLEDA to achieve sustainability.
--, 1993 (3): Project performance evaluation report.
--, 1994: Project performance evaluation report.

Ueda, T.: Opportunity and constraints in the informal sector in Cambodia (ILO, 1995 (2)).

UNDP, 1995: The Human Development Report.

UNDP/Government of the Netherlands/ILO/UNIDO, 1988: EM>Development of rural small industrial enterprise; lessons from experience.

UNDP/ILO, 1992: Small enterprise and informal sector promotion, project document.

Ushijima, R; Mey Chann Osaphear: Impact assessment report of the informal sector component (SEISP/ACLEDA, 1995).

Webster, L.; Tucker. B.: ACLEDA; an appraisal for CGAP (World Bank, 1996 (2)).

World Bank, 1995 (1): Cambodia; Rehabilitation Programme: Implementation and outlook.

Annex 1 - Main stages in project implementation

February to April 1992	Project identification and first formulation.
May 1992	Start of preparatory assistance.
April to June 1992	Recruitment of national staff.
August to October 1992	First training of national staff.
October 1992	Approval of project document.
November to December 1992	Establishment of first five LEDAs.
February 1993	Establishment of ACLEDA.
February 1993	Small business component becomes operational.
June 1993	Informal sector component becomes operational.
October 1993 to March 1994	ACLEDA develops its Strategic Plan.
November 1993	ACLEDA and the project sign a Memorandum of Understanding for the gradual transfer of authorities.

ILO Action Programme on Skills and Entrepreneurship Training for Countries Emerging from Armed Conflict

Promoting local economic development in a war-affected country The ILO experience in Cambodia

Table of Contents

Annexes

PREFACE

ACKNOWLEDGMENTS

ACRONYMS

EXECUTIVE SUMMARY

INTRODUCTION

2. THE CHALLENGE OF PROMOTING ECONOMIC DEVELOPMENT AT THE LOCAL LEVEL

2.1 The political context

2.2 The economic context

2.3 Constraints on local economic development

The environment

Inputs

Skills and technology

3. LOCAL ECONOMIC DEVELOPMENT IN ACTION: STAGES IN THE LOCAL DEVELOPMENT PROCESS (15)

3.1 Project formulation

3.2 Project strategy

3.3 Project implementation

Preparatory assistance

The Local Economic Development Agencies

The LEDA programme

Target groups

The Association of Cambodian Local Economic Development Agencies

The process of institutionalization

3.4 Monitoring and evaluation

4. LOCAL ECONOMIC DEVELOPMENT: LINKING MSE DEVELOPMENT WITH VOCATIONAL TRAINING AND EMPLOYMENT-INTENSIVE WORKS PROGRAMMES

5. ASSESSMENT OF APPROACHES

5.1 Impact

Table 1. Output and impact of ACLEDA's programme

5.2 Sustainability

Table 2. ACLEDA's income and expenditure (27)

5.3 Cost effectiveness

Table 3. Cost of delivery over project period (1993-95)

Table 4. Cost over outstanding portfolio

Table 5. Cost of impact over project period (1993-95)

6. LESSONS LEARNED

The environment

Project design

Project implementation

Programme linkages

BIBLIOGRAPHY

Annex 1 - Main stages in project implementation

Promoting local economic development in a war-affected country
The ILO experience in Cambodia

3. LOCAL ECONOMIC DEVELOPMENT IN ACTION: STAGES IN THE LOCAL DEVELOPMENT PROCESS ⁽¹⁵⁾

Table 2. ACLEDA's income and expenditure ⁽²⁷⁾