Agricultural Management Training in Africa (AMTA) - IOE
Agricultural Management Training in Africa (AMTA)
Interim Evaluation
Agricultural Management Training in Africa (AMTA)
Executive Summary
1. The Agricultural Management Training Programme for Africa (AMTA) was conceived by IFAD in the early 1980s and conducted in collaboration with the African Development Bank (AfDB), the World Bank's Economic Development Institute (EDI), and the Organization of African Unity/Scientific, Technical and Research Commission (OAU/STRC) as the regional sponsor and clearing house. AMTA was conceived for the following purposes:
(a) to train a cadre of agricultural sector managers;
(b) to establish a permanent agricultural management training capacity among regional and national training institutions;
(c) to devise and develop suitable curricula, training materials and methods for training agricultural sector managers;
(d) to improve, over a relatively short term, the performance of agricultural development projects; and
(e) to "increase the awareness of senior government officials of policy issues and administrative procedures that affect the implementation of agricultural development projects."
2. From 1984 to 1990, four IFAD Technical Assistance grants for a total of USD 5.3 million, with co-financing and staff support from the AfDB (USD 1.4 million) and EDI (USD 1 million), have financed four cycles of AMTA. Three sub-programmes have been completed and the fourth will be completed in mid-1992. The total costs of AMTA probably will be about USD 6.5 million, USD 2.3 million less than the original estimates. So far, mid-way through AMTA IV, a total of 411 project staff from 74 projects in 27 sub-Saharan countries have been trained, 44 management trainers from national training institutions, and about 105 agricultural officials who have attended two-day national policy seminars. A total of 49 senior officials from ministries of agriculture have attended the High Level Policy Seminars organised by IFAD in Italy.
3. Pending the conclusion of AMTA IV, this evaluation is an interim assessment of the results and the impact of AMTA since it began. It was conducted in 1991 by a team of consultants under the general direction of the Monitoring and Evaluation Division of IFAD, through missions to eight countries that participated in each of the four sub-programmes.
4. The broad aim of AMTA to introduce a management training methodology appropriate for African conditions is valid and the strategy chosen is generally suitable. The basic concepts, which focus on the policy and administrative environment of the agricultural sector and projects, the project organisation, and individual project managers and senior staff, are also valid. However, the strategy chosen emphasises project organisation, its managers and senior staff, as it should, but does not sufficiently involve senior officials1/ who are responsible for the policy and administrative apparatus of agricultural sector and project management.
5. AMTA has introduced a promising training method that combines individual and team training in the use of selected management techniques. It has been conducted effectively through residential courses, seminars, non-residential workshops and on-the-job applications. With the financial means and institutional support necessary, appropriate pedagogical guidance from training institutions, and continued training of trainers, the basic curriculum and materials developed by AMTA are adaptable to training needs of widely differing African countries. With refinements of the curriculum to fit national needs (based on a more rigorous Needs Assessment Survey (NAS) and more selective use of training materials), additional help from training institutions for on-the-job applications, the AMTA method can be an effective means of training managers and management teams. The particular feature of AMTA was its pilot nature. Consequently, the approach, curriculum and training materials of AMTA were expected to be tested and modified during the implementation of the programme. In the process, the tested methodology would become ready for wide application in the region.
6. AMTA was a massive effort to reach nearly all sub-Saharan African countries. It was adequately funded by the three sponsoring institutions (AfDB, World Bank/Economic Development Institute, and IFAD). In contrast, the commitment of regional and national training institutions, and senior agricultural officials in each country fell short of expectations. Therefore, as an externally induced, "supply driven" innovation, AMTA has not yet become an instrument of institution-building that its founders envisaged, and this for three reasons. First, too little attention has been paid to AMTA's achievements and potential by the Project Operations' staff of the three sponsoring institutions. Thus AMTA, being divorced from the operational mainstream of project design and implementation, has had a limited effect on projects. Second, even though regional and national training institutions readily adopted the AMTA curriculum and methods, there is no firm commitment on their part; and there are not enough national trainers with whom to continue AMTA (unless this activity continues to be internationally financed, preferably with grants in the short- as well as the medium-term). Third, senior officials in each country had a very limited involvement in and commitment to AMTA, compared with the scope and intensity of training provided for project management staff.
1/ Only ministers, permanent secretaries and senior officials participate in the 2-3 day High-Level Policy Seminars held in Italy, which focus mainly on broad policy issues. These high level officials should also be sensitized to the requirements and potential of those improved management tools which are at the base of AMTA training. This could take place in the course of policy seminars organised in Italy or (as in the case of AMTA IV) at the level of the AMTA-assisted countries.
7. In training about 411 project managers and senior staff and procurement specialists so far, AMTA has reached its original objective of 408 in about the same proportions as originally planned; with respect to national trainers, attrition during training and turn-over thereafter have kept the number of trainees down to about 25 (or less) from an enrolment of 44, which is short of the 48 originally planned.
8. A high degree of consensus was expressed by former participants that AMTA had helped to improve their administrative, financial and technical management proficiency. About 85% of participants feel that AMTA achieved most objectives. With near unanimity they assigned a satisfactory to high rating to the benefits that they personally derived from AMTA, namely, skills, greater job satisfaction, team work and, for many, career development opportunities. As many participants expressed, for them (especially senior project staff) this was the first opportunity to learn management and training techniques, and most appreciated the leadership skills that come with learning to use these techniques.
9. Overall, participants felt that as a result of AMTA the management training capacity of regional and national training institutions had been improved. The favourable outlook toward the capability of national training institutions, however, must be discounted in view of the high attrition of national trainers who initially enrolled.
10. The view expressed by 87% of participants that AMTA helped improve the performance of their project is not supported by evidence as to how AMTA was used at the project level. Upon reviewing selected project completion and evaluation reports, it is evident that other factors, beyond the control of project managers, and outside the scope of AMTA as designed, influenced the outcome: for example, AMTA could not help to correct inappropriate project design; or it came too late in the project cycle or, in the case of sector and sub-sector projects, the real project managers did not participate in AMTA. In some extreme cases (e.g. Ethiopia) other forces prevented project managers and senior staff from making effective use of AMTA. Consequently, project managers and staff who used AMTA techniques were either frustrated in their efforts (by resistance to change by colleagues who did not benefit from AMTA training, their superiors, or other institutions not involved in the AMTA process), or because trained staff were reassigned, so that AMTA training had negligible effects on the outcome of their projects.
11. Most important, AMTA was not acknowledged as an instrument of change by any of the three sponsoring institutions (AfDB, IFAD and World Bank) that also finance agricultural projects, or by senior officials of the countries. This was reflected in the strong reservation expressed by most former participants about the contribution of AMTA to generating support from senior officials in overcoming policy constraints and administrative impediments, or otherwise providing positive support to improve project performance. AMTA has not been mentioned in project supervision and completion reports or reviews that have been prepared so far.
The Curriculum and Training Materials
12. The training NAS has been the fundamental instrument of curriculum design, which determined the topics to select and emphasise, in addition to the general project management topics of the core curriculum. Because of the tedious work involved in conducting a NAS for an average of three projects per country, and the tight schedules imposed on special NAS missions and curriculum design workshops over a three-month period, the process was conducted under stress. A major flaw of the NAS was that it did not capture the environment in which projects were implemented. This is because:
(a) the choice of countries, projects and participants was largely pre-determined, rather than the end-product of the NAS. As a result, a minimum number, and not necessarily the appropriate combination, of participants were selected, and it excludes projects and participants that could have benefitted from AMTA;
(b) sectoral and sub-sectoral projects were treated in the same way as area-specific or functional projects, so that senior staff in managerial and decision-making positions were excluded and usually replaced by coordinators with little managerial authority over project performance;
(c) the survey questionnaire is too long, and lacks a scale with which to assess the relative weight of skill levels and deficiencies according to which training need priorities within projects and countries can in turn be assessed. It also lacks questions on substantive topics relevant to managers and senior staff in technically specific fields; and
(d) two important elements are explicitly left out of the NAS: (i) senior officials, especially those at the level of Director or Director-General; and (ii) national training institutions and their trainers (in order to assess their commitment to AMTA).
13. Although the combination of countries -- with six or seven countries in each sub-programme -- and the number of projects and participants were based on considerations of cost-effectiveness and economies of scale, plus the need to respond to decisions made by each of the partner institutions, the same curriculum in a sub-programme for 6-7 countries could hardly take all differences between countries into account. Thus, even though the pedagogy of AMTA can be standardised in principle and kept flexible in practice, standardisation cannot apply to the entire curriculum.
14. Despite these misgivings, more than 70% of former participants felt that the curriculum responds to real training needs and expressed a relatively high degree of satisfaction. However, the evaluation missions found that participant satisfaction was as much the result of the training methods used (emphasising individual and group training) as of the variety of subjects chosen and training materials used. For example, the topic of general project management was highly rated by participants but they felt less satisfied with the general treatment of substantive topics: financial management, agricultural credit, extension, and procurement. Dissatisfaction among about 26% of participants regarding special topics (credit, marketing, extension, procurement, and reporting) seems due to a combination of factors: the shortage of trainers with sufficient expertise and broad experience in these subjects, the gaps between theory and what is normally feasible in certain countries, and the difficulties in satisfying the expectations of a heterogeneous group of participants from many countries. In contrast, more satisfaction was expressed for the subject of Monitoring and Evaluation than for other specialties.
15. Training methods and materials developed for AMTA were judged as satisfactory by at least 85% of former participants, 60% deeming them as highly satisfactory. However, participants felt, in two-thirds of cases, that the training materials on specialised topics were not sufficiently focused on their responsibilities in agriculture. For this reason, among others, they preferred case studies, because they were prepared exclusively for each sub-programme and dealt more realistically with agricultural subject matters. Nevertheless, they were deemed to be in short supply, especially during AMTA I and II, and not sufficiently diversified.
16. Team-building methods and Performance Improvement Planning (PIP) have earned the participants' appreciation, in contrast with the original in-country follow-up workshops which were not always conducted systematically. The PIP, if used creatively for peer training rather than as a rigid device, has the potential of making the greatest contribution toward binding members of a management team to formulate and pursue action plans for the entire project or for any of its components. The testing of PIP started in the early phases of AMTA, but it received all the emphasis that it deserves in AMTA III and IV, thanks to the flexibility of the programme curriculum and operations. However, the teaching of PIP application was limited to the project as a whole and not to its components.
17. By far the most positive and probably lasting effects impact of AMTA sub-programmes conducted so far are on 94-98% of participants who feel that they have acquired new knowledge and skills for project management. For a majority (70%), AMTA training has been an important step in their career development -- some having been promoted as a result, and a few having left their projects to take management positions in and out of the agricultural sector. With about 330 persons trained for 20 countries, this is an average of 16 per country, or about five per project. In view of the high mobility of trained staff, plus retirements and other causes of attrition, evidence since AMTA I shows that the effects of AMTA training and team-building rapidly dissipates because of the small number of trained people in each country and in each project.
18. In order to achieve a net increase in trained agricultural project and sector managers, training by AMTA would have to expand at a geometric rate, first to create a "critical mass" of trained cadres first at project, then at sectoral level. The concept of critical mass is important for AMTA, because it also implies a number of trained cadres sufficient to compensate for the high level of project staff mobility and attrition. The number or proportion of staff to be trained annually would depend on how quickly a critical mass should be achieved. In setting this objective AMTA would have to be a part of the manpower programme of agriculture and rural development ministries, in order to avoid the situation found during the evaluation, whereby after two years, projects have lost two or more of the five persons trained by AMTA.
National and Regional Training Institutions
19. The capacity acquired by regional and national institutions to train agricultural project managers depends on how much they were involved in AMTA, how AMTA is relevant to their training mandate, how many national trainers were trained and have remained, and what means they have to include AMTA in their training programme. Both regional and national training institutions have distinct advantages that have not yet been exploited, as well as extenuating limitations that became evident with hindsight.
20. So far the effect of AMTA on national training institutions has been less than that on regional institutions. This is due to the fact that national institutions did not begin to share real responsibilities in conducting the programme until AMTA III. Until then national institutions were only nominally involved, although special Training of Trainers (TOT) workshops were conducted for national trainers at regional seminars. Consequently, except in countries where branches of regional institutions had a dual role (Burkina Faso, Senegal, Zambia), national training institutions benefitted from AMTA only through their trainers, provided the latter remained in their jobs.
21. Since only about one-half of the 25 national trainers enrolled actually completed AMTA I and II, and fewer yet were assigned responsibility for training in agricultural project management, the benefits of AMTA for national training institutions were marginal before AMTA III. On the whole, however, the weakest element is still the relatively high drop-out rate of national trainers during a sub-programme, and their occupational mobility thereafter. Thus, out of about 45 national trainers who have been partially or fully trained during AMTA I-IV, only about 25 completed their respective sub-programme. It is estimated that when AMTA IV is completed, there will be at the most one partly or fully trained national trainer in each country, which is insufficient to assure sustainability.
22. From the above it is clear that the effort to train national trainers should be continued and receive more emphasis. Since in the past there has been no systematic effort in Africa to develop the capacity of national training institutions, the World Bank and the United Nations Development Programme (UNDP) agreed to create the Agricultural Rural Development Network (ARDNET) capitalising on AMTA's methodology and approach. The main thrust of ARDNET is to increase the capacity of national training institutions in Africa in the field of agricultural management development. IFAD is planning a close linkage between AMTA and ARDNET.
23. The value of AMTA has been recognised by some national training institutions, insofar as AMTA techniques have been incorporated in their management curriculum (e.g., the Kenya Institute of Management and the Ghana Institute for Management and Public Administration (GIMPA) in Ghana). This is a positive outcome but, to date, none of the national training institutions involved in AMTA I-III has replicated AMTA on its own initiative or on that of the agricultural authorities of their country. The reality is that in most countries national training institutions and trainers are spread over several sectors, and are not necessarily in the mainstream of agriculture. Furthermore, because of their financial self-sufficiency requirements, these institutions start new training programmes and continue established ones only if these activities contribute positively to their budgets and/or attract additional budget allocations.
24. Regional training institutions (the Pan-African Institute of Development (PAID), the Centre d'Etudes Supérieures en Administration et en Gestion (CESAG), the Eastern and Southern Africa Management Institute (ESAMI)) have contributed to AMTA, with some 20-30 regional trainers who have become proficient in using new techniques, writing new training materials on management, and developing the AMTA curriculum. However, they do not especially serve the agricultural sector (even in countries where there are no effective national training institutions) but rather other sectors, especially public enterprises, banking institutions, industry, etc. Regional training institutions have not yet pursued the possibility of continuing AMTA for projects in cooperation with national institutions. The absence of ties between them -- which AMTA showed could be temporarily established -- arise from differences in clientele, status, salaries of trainers, access to international funding, and in the amount and means of payment of institutional fees (i.e., in foreign currency vs. local currency). Although both regional and national training institutions responded positively to the demand for their services by AMTA, neither has yet cultivated the new "markets" created for their respective services.
25. There are no clear trade-offs between conducting AMTA with regional rather than national training institutions. The comparative advantage of having a national institution in each country (e.g. GIMPA in Ghana) is offset by that of having a regional institution in other countries (e.g. PAID in Cameroon), while in some other countries (e.g., Burundi), regional and national institutions can (and sometimes must) complement each other. However, the issue of the perceived inequity of remuneration between national and regional trainers is a problem that has not been faced squarely by AMTA.
Curriculum and Training Materials
26. A remarkable amount of training materials on the basic topics of agricultural project management has been produced since AMTA I. The early problems of bulky documentation and editorial quality have been largely overcome. However, in some volumes editorial and graphic quality still have a "makeshift," unfinished appearance, despite the establishment of a training material production facility at the PAID in Cameroon. Since they are drawn from several sources there is little uniformity in style and content, in part because some of the materials were not written for instructional purposes, while others, prepared by regional trainers exclusively for AMTA, are of a better quality. A great deal of selective editing is needed to segregate materials prepared for AMTA from background reading and references.
27. The full potential of the AMTA curriculum and training materials to create an effective demand for such training has not yet been realised. With "sunk costs" in training materials (for the basic core curriculum and specialised subjects) and the high value the former participants assigned to them, mass production and distribution of the materials would be expected. However, delays at PAID in production and distribution of a definitive version of the materials has made them unnecessarily scarce, even for former participants. Until a high quality version of at least the volume on General Project Management is available for mass circulation, a key instrument will be missing to create the demand for training for which AMTA was designed.
28. AMTA I-III experience shows that improving the performance of projects depends on more than a good management training programme. As noted above (paras 10 and 11) the sponsoring institutions' project operations staff took scant notice that the projects could benefit from the training that their institutions financed. The impact on projects was also limited by:
(a) the relatively small number of staff trained in each project, discussed above (para. 17), and their mobility, which made team-building difficult; and
(b) the focus of AMTA on the project organisation, rather than on breaking down the barriers between the project and the broader policy and administrative environment by which it is governed. For many problem projects, because of inappropriate project design and the presence of exogenous factors, AMTA was implemented too late (and the outcome of these projects would not have been different without AMTA).
Influencing Policy-Makers and Senior Officials
29. So far AMTA has not raised the awareness of senior officials sufficiently to persuade them that they can positively influence certain policies and make the administrative environment more conducive to improved project implementation. Although it may still be too early to conclude that the recently instituted national Policy Level Seminars (PLSs) have little effect, the Evaluation team found that they had little effect on promoting AMTA training and making it easier to implement projects, for the following reasons:
(a) the two PLS were not attended by the same senior officials. National PLS, introduced in AMTA III, were attended by persons who have little or no influence on policy and procedures, i.e. advisors, and occasionally assistants to Directors-General and Under-Secretaries), while the international PLSs were attended by Ministerial - Permanent Secretary-level officials; and
(b) the agenda and the topics -- although appropriately chosen -- were not generally framed in the specific context of the ongoing AMTA programme. Uneven representation and poor attendance at both types of PLSs' further reduced their usefulness. The conclusions reached at both PLSs' were usually divorced from the action that needs to be taken to make that particular phase of training an effective instrument of change.
E. The Costs of AMTA and its Cost-Effectiveness
30. The total expenditures of AMTA of about USD 4.8 million will be considerably less than the original estimate of USD 8.3. million. After deducting undisbursed and cancelled amounts from the grants made by IFAD for AMTA I-III, expenditures to IFAD amounted to USD 2 965 899 (about USD 1.0 million less than the grants approved), about USD 970 000 to the AfDB and USD 925 000 to EDI. Excluding expenses for Programme Coordination and Administration (about 17%), actual training expenditures for 326 project staff and national trainers, and for conducting policy seminars for about 150 senior officials (49 at the three-day High Level Policy Seminars in Italy, and about 100 at the two-day National Policy Seminars of AMTA III) amounted to about USD 4.0 million .
31. There were 375 participants in AMTA I-III, but only 326 completed the training. The average total cost of USD 12 961 per participant seems high, but the figure is misleading because:
(a) the real training costs (excluding expenditures for programme administration, coordination and policy seminars) were reduced from USD 11 268 per participant in AMTA I to USD 10 522 in AMTA II, and to USD 10 340 in AMTA III as the result of tight cost controls imposed by IFAD management;
(b) the real training costs per participant averaged about USD 10 733 with a range of about USD 13 000 for project managers and USD 8 000 for senior project staff;
(b) costs per participant-week -- the most reliable measure of training costs -- amounted to roughly USD 1 901 for an aggregate of 1 967 participant-weeks to train 326 project cadres in AMTA I-III. This included all expenditures to prepare for the first residential seminar and follow-up activities such as team-building workshops. However, cost per participant-week did not decline with each new sub-programme as much as expected as the result of "sunk costs" in previous sub-programmes, mainly because additional training materials were produced for each sub-programme; and
(c) these costs are inflated by the relatively high cost of international high level PLSs in Italy. Expenditures for the three PLSs in Italy ranged from USD 2 100 to USD 3 647 per participant-day, for three days, compared with USD 114 per day for the non-residential national PLSs in AMTA III.
32. Although the costs per participant of AMTA are not fully comparable with those of other training programmes and institutions -- mainly because of different accounting rules for the treatment of sunk costs for training materials, institutional overhead and physical facilities -- USD 1 901 per participant-week of AMTA compares favourably with the highest costs of custom-designed EDI seminars (USD 1 800-USD 2 200). AMTA, however, is much more costly per participant week than the standard EDI courses (less than USD 1 000 per week). After allowing for the cost of travel to and subsistence in France, the United Kingdom or the United States, AMTA costs less than short (6-10 weeks) management training programmes organised by universities or special institutes in those countries.
33. AMTA has been less cost-effective than it could have been because it has been treated as an experimental programme with a number of enrolments lower than its potential capacity. Thus, one important feature for lowering the average cost of training a person, which could not be exploited by the AMTA programme, is to achieve very low marginal cost per participant. If this factor is ignored, average cost per participant-week automatically increases when the number of participants is restricted. This is accentuated for international High Level PLSs, for which high travel and subsistence costs and short duration make such seminars three to five times more expensive than similar events conducted by EDI in most national capitals. With more participants, however, the marginal cost of additional participants would be one-third of the average cost per participant-day for three days, or less (if conducted at a less expensive venue than Italy).
34. The Evaluation team concludes, therefore, that AMTA training can be cost-effective if ones bears in mind the relationship between venue, number of participants and duration of training activities, relative to preparation and other overhead (i.e. fixed) costs. This would argue in favour of seminars for a larger number of persons at the sub-regional (2-3 countries) and national levels, with a larger participation by national training institutions than in the past. Thus, when the optimum number of participants and duration of a training event is reached, the marginal cost per participant would be equivalent to only the variable cost per additional participant (for transportation and subsistence) as long as the extra participant does not require increasing fixed costs (i.e. trainers, equipment and facilities).
F. Monitoring of AMTA and Reporting
35. The 1986 Mid-Term Evaluation (MTE) recommended emphatically that a few basic indicators be used to monitor achievement of each objective. The Evaluation team is not aware that this has been done by AMTA's Secretariat. Similarly, the agreements entered into with OAU/STRC, EDI and AfDB, IFAD stipulate that periodic reports (quarterly or as specified by the Steering Committee) should be submitted to IFAD, regarding progress and achievements of AMTA. Based on information available in IFAD files, final reports were received for only the first two sub-programmes, and an interim (1988) report for AMTA III about half-way before its completion. Similar lapses for financial reporting and auditing were also evident. These were due in part to staff reduction at the AMTA Secretariat and to the highly compressed schedule of AMTA III inherited by the new AMTA Coordinator in 1988, after the position had been vacant for one year.
G. Conclusions and Main Recommendations
36. The Interim Evaluation team concludes that AMTA is an appropriate instrument for training agricultural project managers and senior staff, and, within limits, for creating more favourable conditions for project implementation. Consequently, it should be continued, bearing in mind that original assumptions about the sense of commitment of senior officials and the capacity of the training institutions must be changed to conform with reality. Moreover, AMTA can be conducted in more cost-effective ways than it has been during the last six years.
37. The basic framework of AMTA is sound but needs to include specific actions to cut across the barriers between:
(a) the project organisation and the agricultural sector ministries;
(b) officials of the sector ministries and the public sector policy and administrative environment (e.g., customs, procurement); and
(c) the project organisation and the small farmers who are the project beneficiaries. These changes would require:
(i) bolder action in translating the assumption that creating awareness among policy makers and senior officials about the necessity of changing the policy environment and administrative procedures, in order to obtain the firm commitment of senior officials to support AMTA in their country. The national and international PLSs are the appropriate means for this purpose, but the AMTA objectives and instrumentalities should be explicit in the seminar agendas. AMTA should recognise, however, that some of the changes in policies and procedures are in the domain of sector and structural adjustment lending, rather than in the context of project implementation. The timing and venue of PLSs should be reconsidered, and they should be attended by many of the same senior officials;
(ii) incorporating AMTA training in the design (with appropriate financial commitments) and implementation of projects financed by the three partners (AfDB, EDI and IFAD), and the cooperating institutions of IFAD (mainly UNDP/OPS). This should include both an assessment of the need for AMTA as a training activity in the management component of new and ongoing projects, and an acknowledgement by supervision missions of any perceived effects or deficiencies;
(iii) encouraging the three AMTA sponsors and lending institutions to finance investment in national training institutions, as part of their project financing, to create the needed capacity to conduct AMTA programmes in the future; and
(iv) training a sufficient number of managers and senior staff, instead of a selected few, so that after taking into account reassignment and other causes of attrition, the pool of trained staff will not be depleted and dispersed. This implies training more persons per project and country (i.e., more projects). Candidates for training should include de facto managers of sectoral and sub-sectoral projects.
Training Needs Assessment Survey (NAS)
38. The NAS is the most basic instrument for designing the content and process of each AMTA sub-programme. It should also assess the training needs of national trainers and institutions, and not limit itself to a few ongoing projects and only 4-5 participants from each project. The selection of projects and participants should be the last, rather than the earliest, decisions made in the NAS process. The NAS method should be revised, as follows:
(a) allow sufficient time for the process to be a deliberative one, rather than 2-3 weeks with tight deadlines;
(b) consider the management staff of all projects financed by the three institutions as candidates, and sub-sector managers as well when their function or service (e.g. agricultural extension) are crucial to the success of the projects;
(c) serve the needs of three "clients" of AMTA: ministry staff, project staff and national trainers, in a discriminating way (i.e., with an appropriate questionnaire for each "client");
(d) be particularly sensitive to the training needs of functional specialists who have become managers in their specialty (e.g. finance, credit, livestock, etc.), placing special emphasis on skills that enable project staff to improve communication with project beneficiaries (e.g., with respect to agricultural extension, research, and credit); and
(e) assess the qualifications of national trainers in order to match them with the requirements of AMTA, in terms of educational level, pedagogical and subject matter experience, and the career plans of the trainers. More than one trainer per training institution, and preferably some from ministries, should be included in the NAS.
Performance Improvement Planning (PIP)
39. The main complaints that the PIP is rigid, i.e., it entails too many steps, and that there is never enough time allowed to do it whether it is done in five days or ten days, are founded on insufficient training and the way PIP has been used (or misused) by project managers in practice, without the guidance of experienced trainers. Too many participants think of the PIP as a panacea for overcoming poor project design or conditions for its implementation, rather than a process with which to change management style.
40. Training institutions should stress that learning the PIP process is a repetitive but flexible and iterative activity, that requires guided retraining. They should stress that it can be successfully conducted when:
(a) all senior and other project staff with key responsibilities attend;
(b) the group is free from daily routines to conduct the PIP in the time that has been set aside (one day per week or per month, for a total of three to five days);
(c) there are no boundaries among participants, so that managers and supervisors do not dominate the process, but are equal co-participants; and
(d) the sessions are run by an external trainer, to maintain the integrity of the process and ensure that there is a free flow of ideas, without intimidation whatsoever. Trainers should also encourage the more creative use of the PIP from that of a means for preparing comprehensive project action plans to its use in preparing such plans for individual project components and functions: for example, the budget process; reaching and involving the target population; improving the performance of specific components (e.g., credit, marketing, research and extension, etc.) that are vital to project success.
41. Bottlenecks and delays in the publication of training materials need to be overcome so that participants and national trainers obtain the materials in time. In order to bring AMTA to the attention of a much larger audience, an abridged version of the AMTA training materials should be published and widely distributed throughout the agricultural professions in Africa (similar to those published for the Training and Visit system of extension, and on Monitoring and Evaluation).
42. The cost-effectiveness of AMTA is not firmly established, in view of the different unit costs (per participant and participant/week) among various training activities. High Level Policy Seminars are the least cost-effective, while training activities for project staff have measured reasonably well against three cost-effectiveness criteria. Improvements can be achieved as follows:
(a) for small seminars (e.g., senior officials, project managers and procurement specialists), by applying "marginal cost" in planning training activities -- especially the venue -- and achieving economies of scale by enroling a sufficient number of participants to reduce average costs per participant-week or participant-day; and
(b) by reducing the cost effect of absenteeism and drop-out of participants before the end of a sub-programme. In addition to asking for better monitoring and help from national training institutions, absenteeism can be offset by enroling 10-15% more participants, in order to keep average cost per participant-week to its budgeted level.
Monitoring Performance43. The 1986 MTE recommendations for monitoring AMTA performances by the Secretariat are still valid and are reiterated by the Evaluation team. It would appear that the three sponsoring institution, acting through the AMTA Steering Committee, would require more systematic monitoring and reporting of progress and programme financing. Financial reporting should be according to the same accounting conventions to make it easier to measure cost-effectiveness.
Institution-Building and the Sustainability of AMTA44. A key assumption that needs to be reconsidered is whether (and how) AMTA, as an externally induced innovation, can be the instrument for institution-building that was intended. In the prevailing policy environment and administrative system of the agricultural sector ministries in most African countries in the 1980s, AMTA did not benefit from the leadership and commitment that it needed to become a sustained, "demand-driven", institutional innovation for training agricultural sector and project managers. The minimum set of conditions necessary for institution-building and the sustainability of AMTA are:
(a) strong leadership with support from both outside and within the training institution;
(b) proficiency (i.e., sufficient number and experience of trainers) to "deliver" relevant skills;
(c) an effective institutional demand for their training that justifies fixed investment, generates income and, if necessary, public sector subsidies;
(d) a broad and clear mandate (doctrine) from governments or the institution authorities, which can be readily translated into programme objectives and contents; and
(e) positive linkages with ministries and other organisations that will benefit from the innovation.
45. Only the first two conditions are met in some countries, namely where there is a well established national training institution, or locally established regional training institution -- for example GIMPA in Ghana, CESAG in Senegal, the CPF in Burundi and the three PAID national centres. Since the last three conditions are not satisfied, the task of institution building is insufficiently advanced to guarantee that AMTA will continue without a new injection of external technical and financial assistance. Leadership and commitment by African regional and national institutions, and affirmative action by senior officials in agriculture and education are needed to internalise AMTA, rather than continue it as a "supply-driven" technical assistance operation. At the regional level, the OAU/STRC, AfDB, and regional training institutions such as CESAG, PAID and ESAMI should find it in their interest to take the lead, and persuade policy-makers at the highest level, especially the ministries concerned with agriculture and rural development, to support AMTA financially, to recognise its contribution to capacity building and to better sector and project management.