Interim Evaluation ¹

Introduction and approach

The Mission was mounted to assess the performance and impact of the Project (HPM) in the context of consideration of a second phase of a new project for Viet Nam. The evaluation followed the guidelines and criteria of the IFAD Methodological Framework for Project Evaluation, with its emphasis on: rural poverty impact at household and community level; the Performance of the Project; and Performance of Partners.

Following preparatory work by OE in the provision of Approach and Issues Papers, the Mission fieldwork took place mostly in March 2004 under the aegis of the Ministry of Planning and Investment (MPI), and the Provincial and respective District Peoples Committees. Initial briefings were also held with FAO, SIDA, UNDP, UNICEF, and World Bank and the work of the Mission was greatly assisted by the active participation of Project staff and local officials. All ten districts were covered and visits made to 42 communes and 37 villages. More than 110 farmers and village people, including 42 women, were met and interviewed in one to one or in group discussions; and the IFAD evaluation matrix form was completed. Mission findings were presented in aide-memoirs at Wrap-up Meetings in Ha Giang and in Hanoi. At neither meeting was there any substantive disagreement or dissent as to the Mission findings and recommendations.

Development perspective. The economic transformation of Viet Nam since the early 1990s has been remarkable, with increasing liberalisation and private sector and foreign participation. Over the last five years, population growth and price inflation have both averaged less than 2% and debt service is now only 6% of GDP. Between 1992 and 1997, annual real growth in GDP consistently exceeded 8%; it is currently around 7%; and is planned - and forecast - to achieve 7.5% for the years 2000 to 2005. Despite this buoyancy, GDP per head remains at USD 460, so that Viet Nam remains one of the poorer countries, with poverty still widespread, especially in the densely-populated or remote rural areas, although prevalence nationally has fallen from 58% of the population in 1993, to 29% in 2002. However, poverty has marked spatial and demographic dimensions and the rate of decline of poverty incidence during the 1990s for ethnic minority groups was less than one third of that for the majority.

The agricultural sector recorded growth of over 4% per year during the 1990s, as a result of de-collectivisation and commercialisation. The change in agriculture from a predominantly peasant subsistence system to an industry that apart from providing the majority of domestic food needs is now number two in world rice and coffee exports, has been dramatic. Agriculture is estimated to employ some 70% of the workforce and to account for over 22% of GDP and about 30% of all exports.

In recent years, government policies have set out to change the development agenda and approach. Although there is no formal decentralisation policy, various decrees, laws and programmes have, since the early 1990s, had the effect of delegation of responsibility for development downstream and at the same time the emphasis on poverty alleviation has become explicit, with adoption of the Comprehensive Poverty Reduction and Growth Strategy (CPRGS) by government in 2002.

Design features and changes

Ha Giang is a large, sparsely populated province on the Chinese border with a total population of some 650 000 people or 110 000 households in 190 communes and 1 500 villages. Nearly 90% of its inhabitants belong to ethnic minority groups that have only recently started to enter mainstream society; and nearly half of its area comprises rocky, steep, mountainous terrain with harsh climatic conditions and limited natural resource potential. These characteristics prevail in the so-called Zone I, the northern border districts of Quan Ba, Yen Minh, Dong Van and Meo Vac; and to a degree in the north-western Zone II districts of Hoang Su Phi and Xin Man; the remaining southern, central and eastern Zone III districts of Bac Quang, Vi Xuyen, Bac Me and Ha Giang are more amenable, better serviced and economically more advanced.

Development in the province needs to be seen in the context of its frontier location and the dispersion and backwardness of its population. Ha Giang lies near the bottom of the national human index rankings, with an actual poverty incidence in remote communities of from 25 to over 50%. Provincial development policies and plans reflect the prime importance of infrastructure and pro-poor services. Less than 20% of the land is suitable for agriculture; about 40% is forest, with major steepness, encroachment and exploitative use problems. The prevailing total farm size is 1 to 1.5 ha, with less than 0.25 ha per family of terraced or watered land. Paddy, maize and root crops dominate the farming system -- but may not always be sufficient to meet household basic food needs -- and livestock for animal draught, as a financial reserve and a source of food and cash income are important. Customary rights of use and security of tenure for farmers are largely assured -- and are being reinforced by the ongoing allotment of land certificates.

The main elements of the rationale for HPM were the prevalence of poverty, poor health and illiteracy; the extreme difficulty of access, communication, service provision and marketing; the scarcity of arable land; the poor quality of the natural resource base and particularly the problem of potable water supply; and absence of opportunities for non-agricultural employment and income generation. Given these circumstances, the emphasis in design was on participatory local identification of priorities; local ownership of activities by implementation through existing agencies; flexibility in choice of sub-projects; and complementarity with provincial and other development operations. Hence, a strategy of direct action was adopted, to assure food security, enhance household incomes, improve access and communication, address the deficiencies of social and welfare status and inculcate and foster a local capacity for participatory planning and communal involvement in implementation.

The main changes in project design in implementation were: the amendment and curtailment of certain of the TA inputs when the project directorate had serious management problems with the inputs from the co-financier, SIDA; and, notably, the introduction of the Commune Development Fund, as a means of energising beneficiary interest and the flow of sub-project proposals.

Project description

The Project area was defined as all ten districts for the extension, animal health and income diversification activities, but only between four and nine districts for other components. In particular, the health and water supply activities were confined to the four northern and least advantaged districts. Much of the infrastructure investment was pre-targeted to commune and village level. At appraisal, targeting was proposed but no explicit criteria, nor were estimates for target group quantification provided. The target group was defined as the poor and poorest households, as per the national poverty line. This implied that the vast majority of households in the province belonged to the target group; almost entirely of ethnic groups; located in upland areas; and reliant on shifting cultivation within an increasingly diminishing land resource base; the ethnic minority population at the time was estimated at 500 000 people or 88 000 households, of whom 34 000 were considered as primary target group.

The overall goal of the Project was: to improve the living standards of the target group, that is mainly the ethnic minorities living in marginal upland areas. The specific objectives were: to raise household incomes and improve food security for the poorest households, while improving utilisation of natural resources; to support development of appropriate rural and social infrastructure; and to develop environmentally sustainable and culturally sensitive development models that can be replicated elsewhere in the northern provinces.

HMP had five components and eleven sub-components or activities, as summarised below:

• Rural Infrastructure (Base cost USD 5.26m, 32% of total):
  - Pedestrian Roads and Bridges - construction of key village paths, roads and bridges;
  - Domestic Water Supply - construction/upgrading of village water supply schemes; and
  - Irrigation - construction/upgrading of provincial (30+ha) and micro (<5ha) schemes.

• Agriculture, Livestock and Forestry Development (Base cost USD 3.50 m, 21% of total):
  - Farmer Extension - establishment of farmer-based extension systems and research support;
  - Animal Health - improvement of animal health delivery systems; and
  - Forest Protection - support for conservation models for critical watersheds/poor villages.

• Income Diversification (Base cost USD 1.02m, 6% of total):
  - Income Diversification - identification of opportunities, support/promotion, credit provision.

• Social Development (Base cost USD 4.95m, 30% of total):
  - Education - classrooms/centres, curricula/materials development; female literacy and training; and
  - Health - training/establishment of VHWs, upgrading health centres, hospitals, free drug supply.

• Project Management and Coordination (Base cost USD 1.9m, 11% of total):
  - PCUs - establishment and support of Project Coordination Units at province and districts; and
  - Monitoring and Evaluation - at province and district/department level.

Total Project costs, including contingencies, for the 6 year duration were USD 18.4m; the breakdown of financing was as follows: IFAD loan - USD 12.5m; UNDP grant - USD 2.3m; SIDA grant - USD 0.8m; government - USD 2.1m; and beneficiaries - USD 0.6m.

The implementation arrangements gave national responsibility to the Departments of Agricultural Economy and Foreign Economic Relations of MPI, which handled liaison with other Ministries. At the province level, accountability for progress and performance lay with the Provincial Peoples Committee (PPC), the Provincial Project Steering Committee (PSC); and the Provincial Project Coordination Unit (PPCU). The PPCU provided guidance, liaison and arrangement of technical support to: participating districts, for each of which there was a District Project Coordination Unit (DPCU); and to downstream agencies, principally district sectoral departments and their activity management teams. The PPCU was also responsible for liaison and dialogue with donors - IFAD, UNDP and SIDA - and associated agencies such as UNOPS, the Swedish Embassy and NGOs.

At the district level, District Peoples Committees (DPCs) were the prime movers in planning and implementation, backed up by the DPCUs. The commune is the link between villages and districts; the Commune Peoples Committee was involved in planning and oversight of Project interventions, mainly through the Commune Development Board (CDB), a special entity set up by the Project principally to manage the Commune Development Fund (CDF). Village Coordination Units (VCUs), also established by the Project, took on the role of assessing village development needs, communicating these to the CDB -- and passing on proposals and plans for sub-projects.

Summary results of implementation

Financial aspects of performance. The funds available as appraised were USD 18.4 million, equivalent, at that point, to some VND 233.9 billion. Expenditures to date are calculated to be VND 233.1 billion, that is 99.7% of the projected total in local currency terms, equivalent to some USD 16.3 million at the average USD/VND exchange rate for the Project period. In assessing these figures, it needs to be borne in mind that: the UNDP grant for TA was reduced from the original USD 2.3 million by USD 500 000 - or about VND 7 billion; the SIDA contribution amounted to only VND 4 billion, against an original target of VND 10 billion; and the Dong has depreciated against the US Dollar by over 30% during the Project period. The net result is that there remained at December 2003 outstanding amounts of VND 27 billion from IFAD funds and VND 2.2 billion from government funds to meet payments up to Project closure. Inclusion of these sums would bring total expenditure to VND 262.26 billion, or USD 18.34 million.

In comparing actual against target expenditures, there is the complicating factor of the CDF, which was not included as a cost component at appraisal but was financed from surplus funds from the low uptake of credit and social development. In the event, the main divergences by component have been: expenditures on Rural Infrastructure have been 132% of proposed, due to the heavy emphasis on infrastructure of CDF expenditures; Agricultural and Social Development and Income Diversification have reached only about two thirds of the values expected; and Project Management and Coordination costs have been over budget at 122%. In cost categories, significant divergences are: the low costs for Training and TA, at only 63% of planned, due to reduction of co-financier inputs; and very low recurrent costs, due to effective control and savings on plan, at only 45% of budget. It is clear from the financial analysis that the Project has enjoyed sound cost control and financial management.

Physical achievements. The key indicators of benefit of HMP are that commune-wise, the pattern of coverage is: roads and bridges - 20; water supply - 24; irrigation - 92; farmer extension - 132; animal health - 191; income diversification - 60; education - 70; health - 67; and the CDF 78. The Project has thus had a significant, multi-component presence in at least 100 of the 191 communes and therefore had substantive impact on an estimated 50 000 households, comprising 285 000 people.

Under rural infrastructure, the roads and bridges sub-component set out to facilitate transport of people, farm inputs and produce and the outreach of social services. 105 km of commune-village roads and 7 suspension bridges were built as planned, and only the training activity fell short of 100% achievement, although the database system implemented by TA is not now operational. Physical results are impressive, but: designs were deficient; villagers engaged for construction were reported to be paid lower than market wages; community dedication to o&m is doubtful; and district budgets for works are unlikely to be sufficient to cope with costs to maintain accessibility for increasing traffic.

For domestic water supply the enormous problems of provision in the karstic limestone areas of Zone I -- and the potential solutions -- were correctly identified, but the optimal designs were not followed. Thus, although the sub-component achieved 100% of planned physical activity, covering over 5 000 new rain water tanks and 728 rainwater jars, repair of 56 communal tanks and construction of 12 gravity flow systems in 24 Zone I communes, there were serious weaknesses in design, construction quality, capacity for storage and water quality assurance; water user groups do not collect fees for repairs; a water supply database set-up by TA is not operational; and technical capacity, staffing and means of transport at district level are still insufficient.

The provincial irrigation scheme improvement work was largely pre-targeted rather than participatory and claimed farming benefits may be exaggerated. However, in all, 39 provincial schemes covering some 2 840 ha and 134 micro-schemes, including 20 new, covering 2 030 ha were assisted, with an additional water supply to over 4 300 ha and thus 100% or higher achievement of actual physical works, equipment supply and training -- providing potential benefit to over 7 300 households in 90 communes. For micro-schemes, feasibility studies were non-existent or inadequate; cost benefit analyses were not done; dams were constructed at high cost; siting and design were in many cases questionable; and little attention was paid to quality of construction and supervision and water user group establishment. On visits, several concrete canals were silted or blocked by landslides and dams were not de-silted, clearly indicating neglect of proper operation and maintenance.

Under the agricultural development component, farmer extension objectives were consistent with small farmer needs and successful insofar as yields and outputs of major cereal crops were markedly increased and interventions in livestock, forage production and agro-forestry demonstrated -- or show promise of -- expanding uptake. However, design of models appears to have ignored the value of indigenous practice, the constraints of the poorer, mountainous areas and the ability of households to cope with high cost systems. As a result, numbers of farmer field school (FFS) and other formal groups fell far below target - at 25% for FFSs - but were compensated by establishment of a wide village based extension network and on or near target performance of most of the equipment supply and training inputs. In total, some 370 FFSs and IPM clubs were formed, covering over 3 200 households in direct membership, with 266 demonstrations of improved technologies and over 80 000 person days of farmer training. Although training of staff and farmers has been impressive in numbers, the evidence now is that few FFSs or groups have survived; that there is little conviction about the methods used; and that the facilities provided and impetus achieved have not been maintained.

The animal health sub-component was implemented with: beneficiary involvement in selection of commune animal health workers (CAHWs); commune-based operation of services; provision of a necessary minimum of equipment and drugs for CAHWs; and their practical training. In total, 474 CAHWs were trained and deployed in all 191 communes, basic furniture, equipment and drug supplies provided, 9 district and 2 border veterinary stations established and a revolving vaccination fund for notifiable diseases supported, registering virtually 100% achievement. Weaknesses of the component have been at district and provincial level where there is limited evidence of maintenance or continuity of service from the equipment and training supplied. Gains in livestock outputs are difficult to gauge, but availability of vaccines and antibiotics will have had significant impact on household incomes.

The forest protection intervention recognised the importance of watershed protection and soil and water conservation. Beneficiary households gained income in return for vigilance in protection. Operation of protection contracts and achievement of objectives has been fair and at a cost of VND 8.7 billion (USD 0.55m), the economics are reasonable. 20 000 hectares of critical watershed have been protected, with over 4 100 households in 109 villages receiving direct benefits; and enrichment and new planting have reached 973 ha, or several hundred percent of target. All other aspects achieved an out-turn of 100%. The major shortcoming has been the failure to address the serious problem of exploitative arable use of steep lands and consequent degradation in many upland catchments.

The income diversification component was confined to women and seen as crucial to poverty alleviation. The implementing agency was the Womens Union (WU). Despite the claimed 100% achievement of equipment supply and training inputs, performance was well below expectation. Opportunities were missed to create linkages between saving and credit, labour saving technology and education. PRAs were undertaken in 60 communes and 318 womens savings and credit groups formed. Of these, 118 groups covering 1 538 households are still active but have little substance as groups and saving activities are absent or negligible. Total borrowing to date is VND 4 606 million (USD 287 000) or 67% of target and repayment performance so far is reported to be satisfactory at 86%. However, the training of local people has not been effective; few diversified businesses have been set up, and many have been unprofitable; and the channelling of IFAD credit through the quasi-commercial bank VBARD was not a viable approach.

In the social development component, education was the largest activity, addressing the problems of adult illiteracy and primary enrolment, especially of girls. Education continues to be extremely important as the key influence on awareness and understanding, which are the fundamental pillars of ethnic group advancement. Major achievements have included: education development plans for 70 communes; construction of 146 rooms for commune schools and 7 district continuing education centres; supply of equipment and material; and bilingual and refresher training for over 3 200 teachers. School construction has been satisfactory, but quality of workmanship was poor in some cases and proper maintenance and upkeep are absent. Most equipment and materials were of good quality and useful. There is evidence already of improvement in school access, enrolment, attendance, retention and narrowing of the gender divide. Key shortcomings have been: limitations of participation; improper use of PRA techniques; lack of upgrading to make libraries functional; and failure of the trained women literacy volunteers to activate village literacy and other programmes as envisaged.

Objectives of the health sub-component were germaine to the predicament of poor. Major achievements were: networking and training of 801 village health workers (VHWs); upgrading of facilities and equipment in 4 district health centres; construction and equipment for 5 commune health centres; training and refresher courses for doctors and other staff; and free supply of drugs. Most aspects reached 100% of target. VHWs have transformed outreach and prevention in remote villages; already reduced incidence of malnutrition and diseases is recorded; and significant increases are noted in in-patient and out-patient numbers at hospitals. Remaining concerns are: sustainability of drug supply; failure of drug revolving funds; the low percentage of women VHWs; lack of linkages between traditional birth attendants and VHWs; low educational standards and skills; and low allowances for VHWs, are issues that require attention. VHWs need significant further refresher training and capacity building and simplification of their reporting formats.

Introduction of the commune development fund in 2001 enhanced community participation, allowing CDBs to become more effective and ensuring at least some gender-balanced village representation. CDFs have spread from 10 to 78 communes, all with poverty rates of over 30%. Emphasis of CDF activities has been on infrastructure: 128 inter-village roads, 7 suspension bridges and 42 culverting or small improvement schemes; 56 water supply facilities and 289 domestic tanks; 184 irrigation construction or rehabilitation projects; one hydro-electricity dam and power cabling; and provision of 55 class rooms, 142 branch schools, 2 boarding hostels and associated furniture, have been completed. Other main interests have been in income diversification, supply of farming inputs, particularly improved seeds, forestry and herbal plants and technology transfer models.

Project management has been proficient as evidenced by the virtually 100% achievement of physical and financial targets and general progress. The provincial institutional arrangements functioned reasonably well and district structures and arrangements set up by and supported by HPM have proven to be effective. Key remaining weaknesses have been that: DPCU chairpersons have changed frequently and not devoted adequate time; CDBs and VCUs are yet to become properly established; coordination problems continue to exist, notably at districts; and appraisal, supervision and convergence between components and with other projects, as well as monitoring, have been inadequate. UNDP and SIDA TA has been important; premature withdrawal of the SIDA inputs, for reasons of unsatisfactory performance and management disagreement; and financial cuts for UNDP inputs due to budgetary restrictions, have had negative impact on the Project. The monitoring and evaluation system has captured financial and physical progress, but has had limited application as a management tool, in supervision, or as an indicator of impact or gender or ethnicity balance. Use of information technology is limited and simplified standardized formats would significantly improve the quality of monitoring and reduce the tedium of the work involved.

IFAD specific concerns. In the context of poverty targeting, since 90% of the Ha Giang population is ethnic, targeting of ethnic minorities is not a meaningful concept; so HPM concentrated on targeting on the basis of poverty level. Macro-targeting of activities is assessed as adequate, but not all components were designed to target the worst off, for instance provincial irrigation scheme improvement and use of VBARD for credit inevitably entailed disproportionate benefits for better-off groups. Targeting of communes, villages and sub-projects was generally satisfactory but if literacy and labour availability were pre-conditions for participation, the poorest, in particular women headed households, could be under-represented. Although all HPM components focused on priority needs, impacts on target groups would almost certainly have been more significant if activities had been targeted to fewer communes; and carried out in a more integrated, component-linked manner.

As far as gender is concerned, HPM has had reasonable success in addressing women's practical needs for access to services and markets, absence of gender dis-aggregation in PRAs prevented serious constraints for women such as violence due to alcoholism of husbands, excessive workloads and reproductive health problems being identified and addressed. Women's strategic needs, such as managerial and entrepreneurial capabilities and involvement in decision making remain unfulfilled, despite WU efforts and their mandatory incorporation in village and CDB affairs.

There was little scope for proactive participation or community involvement by the poor in identifying needs and shaping Project design. Because PRAs were mostly carried out by each component separately and often were simple line department surveys to identify farmer requirements from that department and activities selected in a prescriptive way, they were little used for targeting or monitoring. Improved use of PRAs in the context of CDF sub-project selection and planning showed their value, which is now recognised by province and project management as a means of strengthening decentralisation and democratisation processes. Although the province has been supporting decentralisation, it appears that participation of the local population, and particularly women, in decision-making has been limited. Local social self-help mechanisms and informal organisations have not been capitalised on; devolution of financial control has yet to materialise; and existing commune institutions could have been used better to sharpen gender and poverty focus and enhance the sustainability of Project activities and impacts.
Rural Poverty Impact

Overview. As far as practicable, HPM focused on the poorest communes; the target group poor - and women - are believed to have accounted for between 35 and 70% of the participants for most activities. For income diversification and credit women were prime beneficiaries, but they were under-represented in the CAHW and VHW cadres. Direct beneficiary numbers derive from M&E records and Project reports and indicate that between 3 000 and 17 000 households participated in the main individual Project activities, giving a gross total of benefit incidence of 136 000 households. Accounting for overlap and duplication, it is estimated that at least 50 000 households have received substantive benefits from one or more HPM interventions.

Province wide, incidence of poverty, is recorded as having fallen from a head count of 34% in 1997 to 16% at the end of 2003, using a poverty line equating to a household income of some VND 5.5 million or USD 345 per year. HPM has been one of the main factors in this improvement, alongside a raft of poverty-oriented policies and provincial programmes. Economic statistics put provincial GDP per head in 2002 at VND 2.14 million, equivalent then to about USD 165, compared to VND 1.07 million - USD 91 - in 1997, an 80% increase in real terms; agriculture, forestry and fisheries still made up 48% of GDP in 2002. In confirmation, the recent Ha Giang Participatory Poverty Assessment, reported significant improvements in the predicament of the poor in the last five years, but that village poverty head counts were higher than estimated by commune authorities and that considerable further scope for poverty alleviation existed.

Impact on Physical and Financial Assets. The impact has been positive and substantial -- from agricultural productivity and surplus produce for sale; from forestry protection revenue and rights of use; to a lesser extent from income diversification; from the facilitation of market access and potable water provision. The quantum of benefits and wider farm and household impact have been estimated for two simple farm and household models:

• for a small farm with 0.5 hectares of single season paddy, the effect of Project technology adoption would lift the annual yield from 2.5 t/ha to 3.8 t/ha, that is from 1.25 t to 1.90 t, or from bare basic food subsistence level to a surplus of 0.6 t/year, worth some VND 1.2 million (USD 76) at current prices and possibly adding between 25 and 50% to family cash income; and

• for a similar farm where irrigation improvement allows a second crop of paddy to replace a rain-fed spring crop of maize on 0.5 hectares, the effect would be an output of 1.25 t of paddy worth VND 2.5 million to replace an output of maize of 0.7 t worth VND 1.6 million, that is a benefit of VND 900 000 gross - and probably VND 600 000 (USD 40) net of additional direct costs - in just one season.

Similar or greater incremental benefits than these would have been attained for the majority of the Project interventions in crop and livestock enterprises; and those from forest protection, with only minimal exploitation of firewood and non-timber forest products, are estimated to equate to VND 486000 (USD 31) per household per year. These gains are manifested in: rights of use of additional and/or better irrigated land; possession or assured use of tree plantations; ownership of livestock; better housing; reduced indebtedness; ability to pay school boarding fees and buy shoes and clothes; and ownership of furniture, utensils, farm tools, bicycles and radios. Some 30% of first hand beneficiaries under the productive components would be enjoying more than one of these tangible livelihood gains. Project contribution to increases physical and financial assets is substantial, 3, on a scale of 1 to 4.

Impact on human assets. The Project has had some positive effects on human assets in terms of health, nutrition, education and work burden alleviation, as well as building confidence and self-reliance, but overall it is a mixed picture of achievement. The agricultural and productive interventions have contributed to adequacy of food provision and availability of cash income, primary determinants of nutrition and accessibility of health and education services. Literacy initiatives have had much less effect and the impact of drinking water supply has been chequered. Probably the major change for the better in human assets has been the outcome of providing the necessary information and support to raise the level of awareness, knowledge and skills, although there is still a long way to go to make this fully effective. Evidence of change in human assets is seen in: the proficient uptake, practice and dissemination of activities; the level of participation and interest in new developments; the success of the CAHW and VHW systems; the ability of men and women farmers to present themselves and their views, increasingly in a social and political as well as a technical context. The contribution of HPM to the increase in human assets in terms of direct health benefits is substantial, 3; in education, modest to substantial, 2-3; and that of water supply effects is modest, 2; for the impact on women's workload, modest, 2; for the imparting of skills, knowledge and self-confidence, as described, and the indirect effects on improved human assets from increased food sufficiency and incomes, the contribution is substantial to modest, 3-2.

Impact on social capital and empowerment. The Project has had only a limited impact on social capital formation and people's empowerment; there remains a large gap between what is possible and desirable from a development viewpoint in terms of participation and community self-reliance and involvement and what has been attained up till now. The impact that has been achieved has emanated partly from the demand driven and group approach adopted in the later and peak years of implementation; and has resulted partly from the uplift of financial and human assets noted above. Strong efforts have been made in farmer, womens and users organisations and in attempts to ensure gender balance and to address womens needs, but these have met with only marginal success and their impact is therefore ranked as modest, 2.

Impact on food security. Application of Project techniques has lifted crop yields and outputs significantly and typically moved poor, small farm beneficiary families from a position of seasonal food insecurity, to one of food self-sufficiency in most years. Estimated yield increases are of the order of over 40% -- and up to 100% in better areas -- for paddy; and 10 to 30% for maize and soybean. The farm and household models indicate that these yield increments would have ensured an output above the immediate family needs for basic energy foods and the ability to spend and invest to improve future farming operations and therefore help to break the vicious circle of poverty. Impact on food security is categorised as substantial, 3.

Other impacts. The impact on the environment where the Project has been active has been positive. However, in terms of the overall need for cessation of the profligate and exploitative arable use of steeply sloping lands and for upper catchment protection, the present coverage can only be regarded as a holding exercise. The Project has had limited impact in community-based natural resource management, or in wider application of soil conservation measures and agro-forestry. On the negative side, the intensive crop husbandry methods and the heavy fertiliser and agrochemical use that have been advocated pose dangers for both long term soil fertility and human intake of the foods produced; and the inevitable blasting and earth moving for roads construction may contribute to landslide and soil erosion propensity. Environmental impact is classified, for the present, as modest, but potentially negative in the aspects mentioned; however, the Project contribution to a changing and more stable situation and to an enhanced awareness of the dangers, is assessed as substantial, 3.

The impact on institutions, policies and regulatory framework has come from working through the organs of district governance, as well as with the local private sector. The Project has instilled a degree of orientation, training and experience, as seen in increased effectiveness of extension and the start of use of the PRA as a basis for local planning. The provincial authorities expressed their satisfaction with the level of success of HPM, so much so that the approach and design had already shaped the future development strategy and policies of the province; however, impact on district and commune institutions has been less marked. Respondents among Hanoi agencies accredited IFAD with having influenced the development philosophies, principles and concepts of other donors - and government itself - in the adoption of pro-poor policies and instruments in project design but were critical of the lack of a permanent or consistent IFAD presence in the country and the limitations of the IFAD system of supervision and liaison. HPM has thus played some part in the evolution of new policies and regulations and has utilised approaches and techniques that could influence policy and regulation formulation in future. Its current impact in this regard is classified as modest, 2.

Innovation, sustainability and overall impact. HPM had elements of successful innovation in its agricultural and environmental interventions, particularly with regard to extension methods, its community based approach and use of the decentralised local governance system. It was less so for the infrastructure activities, where accepted top-down practices that were already of dubious effect and sustainability were followed. The rating for innovation and potential scope for replication of Project approaches and activities is substantial, 3. Despite its technical proficiency, HPM has been less successful in inculcating conviction, responsibility and capability among line departments and local user groups and beneficiaries. These are key elements of future sustainability, which is also a function of the external conditions, including those of climatic change and national and international market conditions; prudence dictates that predictions of sustainability should be moderated to account for their influence. Accordingly, sustainability of Project activities and benefits is assessed as equally balanced between modest and substantial, 2/3. The overall impact of the Project is similarly ranked as between modest and substantial, 2-3.

Performance of the project

Relevance of goal and objectives. The goal of the Project, to improve the living standards of the target group, which is mainly composed of ethnic minorities living in marginal upland areas is a fair interpretation of development needs of the outlying and mountainous districts, where, in the early 1990s, many tribal groups were living in a state of serious deprivation and precarious livelihoods, as some still do. The relevance of the goal was - and is - high, 4. The objective of the roads and bridges sub-component was to improve pedestrian access in remote upland areas, the implied outputs including increased skills and sensitisation of local infrastructure agencies and improvement of village access for trading and social purposes. The relevance of these objectives is rated as high, 4. For domestic water supply, the objective was to improve both access to and quality of drinking water where availability was problematic; the relevance was therefore high, 4. For irrigation, there was a two-fold objective: to increase the area of irrigated paddy; and to improve irrigation management. The former aim has been realised to a reasonable degree; the latter, not yet so - and the relevance, for the larger, provisional schemes is regarded as modest, 2, but for micro-schemes, high, 4.

The specific objectives of agricultural development component were to develop and implement effective extension, leading to increased agricultural, livestock and forestry production; to improve delivery of animal health services; and to better manage critical watershed forests. These objectives were apposite to the predicament of forest communities at appraisal and continue to be germaine. Relevance of these objectives is high, 4. The objective of the income diversification component was to increase and diversify household income by identifying and credit funding a range of micro-enterprise options. The relevance was high, 4; the purpose is still appropriate, but the scope for, and means of achieving income diversification now need to be seriously re-assessed. The objectives for social development were to improve access to, and quality of primary education, adult literacy and vocational training programmes; and to improve health by strengthening services. The implied aims and expected outputs of the education and health activities are all crucial to the Ha Giang situation and relevance is assessed as high, 4. By the same token, the objectives of Project management and coordination, although stated at appraisal in a vague and narrow way -- and those attributed to the CDF -- basically of widening and democratising sub-project selection and design for commune and village level interventions, were -- and are -- strongly relevant, 4.

Effectiveness. The goal achievement for the whole Project can be estimated from progress in arresting natural resource degradation, lifting the order of farming productivity, food security and income; and measurably improving living standards by provision of infrastructure and social services. Despite variation in component impact and much scope for further improvement, aggregate effectiveness is assessed as substantial, 3. The roads and bridges sub-component has accomplished its target for works and expanded its impact through the CDF. Effectiveness of meeting the roads objective is modest to substantial, 2-3. Domestic water supply has been subject to great difficulties; designs were wrong, construction standards poor and o&m, including ongoing funding, neglected. Only limited improvement of access and reduction of work load for collection were achieved; and training and mobilisation of WUGs inadequate. Effectiveness is assessed as modest, 2. In the irrigation sub-component, the provincial schemes improved and expanded were pre-selected and misgivings persist about the actual scale of improved crop production, but effectiveness is reckoned to be substantial, 3. For micro-irrigation, absence of proper feasibility and cost/benefit studies, inadequacy of siting, over-emphasis on high-cost concrete structures and uncertainty about crop production benefits render the effectiveness as minimal, 1.

The agricultural development component has largely achieved its objectives, transforming the predicament of adopting farmers and livestock keepers and providing a clear example of application of available technologies through creative extension. For forest protection specifically, effectiveness has come at high cost and more use of forest products could have been made. However, considering the original scenario of low crop and livestock outputs and the extended hunger gap for many families, the aggregate effectiveness is considered high, 4. This conclusion is corroborated by provincial statistics which show crop area, yield and livestock number increases of from 11 to 38% since 1997. Income diversification is seen to have had limited impact, with a non-viable credit scheme, chequered experience of labour saving technology introduction, lack of linkage and failure to achieve meaningful diversification of earning or substance in group formation; effectiveness is rated as negligible, 1. In the case of the social development component, the education sub-component has fulfilled its physical targets in facility, equipment and supplies provision with noticeable impact of training and software inputs in enrolment and retention, including those for girls, a major advance. The health sub-component has achieved similar and comparable results, particularly in VHW training and mobilisation, with patient numbers increasing, adequacy of district services enhanced and child malnutrition and disease incidence falling: recent provincial statistics show increases of 14% in primary girls enrolment and 45% in patients treated and a fall in child malnutrition of 40% since 1997. The weaknesses in both education and health - of local agency capability - are being ameliorated by participatory activities under the CDF, so that in terms of the parameters of participation, community involvement and gender balance, effectiveness is improved. Accordingly, for all three of these sub-components, effectiveness is assessed as substantial, 3. Operations under the Project management and coordination have worked reasonably well but impact is still only tentative in some applications, as in the case of M&E, support of PRAs and gender and district and commune capability. Effectiveness is assessed as high, 4 at provincial level - but only reasonable, substantial, 3 for downstream activities.

Efficiency. The Project in total has entailed deployment of substantial resources and funds, particularly at start-up. The crux of efficiency is the quantum and likely continuity of benefits from the high level of costs incurred. Re-calculation of the internal rate of return, based on highly conservative estimates of uptake and benefit, show an EIRR of 15%, compared to EIRRs of 14% to 27% for key components at appraisal; thus the efficiency of the whole Project is put at substantial, 3. For rural infrastructure, cost disparities and quality of design, construction and o&m for individual intervention in roads, bridges and water supply; and excessive costs for irrigation schemes, dictate that the rankings for efficiency be put at substantial, 3, for roads and bridges; modest, 2, for water supply; and minimal, 1, for irrigation. For agriculture, present and prospective expenditure is considerably below budget, partly due to cost containment strategies, and efficiency is substantial, 3. In income diversification the low costs are due to low uptakes of credit and enterprise development; given the doubtful profitability of investments, the emphasis on traditional pursuits and the unlikely sustainability of activities or benefits, efficiency is ranked as negligible, 1. Under social development, the education and health sub-components both achieved target performances well within the allocated funding in real terms. Education has registered meaningful gains in impact in most activities and efficiency is graded as substantial, 3. The health sub-component has been characterised by timely and effective procurement, supply and sound training and deployment of VHWs and other medical staff. Despite the questionable sustainability of the free drugs policy, efficiency is assessed as substantial, 3. The Project management and coordination component has incurred excessive costs and still has some weaknesses, but has had significant success in operating a complex and challenging project, to the extent that it has strongly influenced the design and conduct of similar development interventions by government and other donors. Efficiency is put as substantial, 3, a finding confirmed by the overall Project performance as assessed by UNOPS.

Performance of partners

The IFAD process of formulation and design was measured and reasonable but the concept and content should have been less ambitious and complex. During implementation, the experience of the PSC and PPCU and government -- despite criticism from SIDA and UNDP -- has been that back-up has been timely and effective; performance is ranked between modest and substantial, 2/3. As cooperating institution, UNOPS, for the first four years was considered proficient but in the last two years, supervision has declined in rigour of oversight and progress monitoring; performance is assessed as only adequate, 2/3. Central Government provided dedication and continuity in Project oversight, policy guidance and donor and ministerial liaison. Provincial bodies have played a crucial role in enablement and support of project activities. Major district weaknesses have been stringency of finance and resource availability; at commune and village level there have been rather more problems of competence of staff and lack of resources; overall, government performance is satisfactory and substantial, 3. Project management, comprising the PPCU, and DPCUs, has shouldered major responsibility for driving forward local interest and participation with dedication and competence. Routine tasks of planning, budgeting and progress monitoring have been reasonably performed, with only limited weaknesses in M&E and component management. Downstream agencies became increasingly involved and confident. In aggregate, Project management performance is assessed as substantial, 3. There is a very limited presence of NGOs and the Project has therefore made little use of their expertise, but has had good liaison where existing NGO activities overlapped with HPM interventions. The Project itself has been a major player in development of CBOs and groups of various kinds, but their sustainability has been marginal. The NGO/CBO contribution to support of HPM is ranked as modest, 2. Pre-existence of the SIDA funded MRDP was a useful precedents for design of HPM and the involvement of SIDA in the planning, design and initial stages of implementation was useful and positive. However, failure to resolve practical problems of integration of the two projects and deficient communication on funding and operations led to collapse of arrangements and withdrawal of some TA inputs; performance of SIDA as a co-financier and provider of TA is rated as modest, 2. The approach and support of UNDP has been effective in both funding and technical assistance and its performance as a co-financier is put as substantial,3.

The impact and performances of the Project and partners are summarised in the tables below.

Component/ Activity	Impacts: Phys/ Fin 1	Human Assets	Social Empwt	Food Security	Envirmt	Institns/ Pol/Reg	Innovn Replen
Whole Project	3	2-/3	2	3	2	2	2-3
Project - All Impacts	2/3

 

Component/ Activity	Relevance	Effectiveness	Efficiency	Sustainability
Rural Infrastructure: - Roads and Bridges	4	2/3	3	2
- Water Supply	4	2	2	1
- Irrigation - provincial	2	3	1	2
- micro	4	1	1	1
Agricultural Development	4	4	3	2
Income Diversification	4	1	1	1
Social Development: - Education	4	3	3	2
- Health	4	3	3	2
Commune Devpt Funds	4	3	3	2
Project Mangmt/Coordn	4	3	3	1
Whole Project	4	3	3	2

 

Partner	IFAD	UNOPS	Govt	Proj Mgt	NGO/CBOs	SIDA	UNDP
Rating	2/3	2/3	3	3	2	2	3

Recommendations

Individual recommendations of key significance for design and conduct of future projects are:

• HPM has been a success in most respects, but was complex and over-ambitious in content;

• future projects should be more closely tied in to provincial and district programmes;

• further development should be restricted to the poorer, more remote communes;

• pre-determined activities should be limited to fundamental poverty alleviation areas of family food self-sufficiency and adequacy of cash income generation from established and traditional household enterprises;

• livelihood financing assistance for poor households should be based on matching grant, seed capital or simple, deferred repayment, rather than credit, arrangements;

• the CDF approach should be used as the main plank for determination of project activity across the technical and social sectors for any follow-on intervention, but with strong provisos as to the determination of feasibility, better technical oversight and real evidence of community ability and commitment for proper operation and maintenance;

• capability building, training - and the gaining of practical experience through project activity - in participation for both community and formal local organisations should be major concerns;

• assistance to government would be greatly enhanced by direct IFAD supervision of its Vietnamese projects; or by contracting supervision, with the agreement of government, to a national, or joint national/international, professional consultancy or development agency;

• if the successes of HPM can be built upon without further dissipation of initiatives and impact through delays in project preparation and approval - and the foregoing recommendations are properly incorporated in the design and in subsequent project operations - a follow-on project would be feasible and justified; and

• IFAD is to engage directly and indirectly in policy dialogue at the provincial and central level, playing a catalytic role though the Project and in close consultation and co-ordination with other international organizations. By building on evidence from the ground, IFAD could further increase and deepen its contribution to the evolution of policy in Viet Nam, especially with respect to decentralized approaches, stronger local participation and community empowerment.

---

1/ The Interim Evaluation Mission comprised: Mr Somesh Kumar, Education, Health, M&E and Management Specialist; Mr Michel van der Stricht, Rural Infrastructure Consultant; Mr Sarath Mananwatte, Economist; Ms Sarah Mader, IFAD Associate Evaluation Officer for Community and Women's Involvement and Mr Jim Semple, Team Leader and Agriculture/Institutions Specialist. The Mission was supplemented by the national counterpart team comprising selected members of the Provincial Project Coordination Unit and District cadres. Mr Ashwani Muthoo, the IFAD Senior Evaluation Officer, joined the Mission in the last week of its work in Vietnam. The National Wrap-up Meeting was chaired by the Vice Director General of the Department of Agricultural Economy, Ministry of Planning and Investment and also attended by Mr Mattia Prayer Galletti, the IFAD Country Portfolio Manager (CPM), Ms Atsuko Toda, the designate IFAD CPM and members of the formulation mission for the proposed follow-on project as observers, as well as 20 key decision makers from the two provinces and from the concerned ministries and partner agencies.