3.1 Agency policies and procedures
3.2 Policies and procedures of recipient
3.3 Design of projects
The growth of the aid industry in recent years has had positive and negative effects. On the plus side there is more money available, and more experience to share. On the negative side, in the frenetic rush to find and fund projects and to speed disbursement, there is not enough time for assessing what is being done, and for using experience to improve programmes.
All aid agencies are trying to help less developed countries, but the identification, location and size of projects can all be influenced by preconceived policies or procedures or spending targets.
When evaluating projects planned in the early seventies, two facts should be borne in mind -one which distorted planning, the other which upset implementation.
Between 1970 and 1980 multilateral commitments for assistance to agriculture rose from US$ 700 million to US$ 6700 million per annum. This led to an increase in "pressure to lend", and to the setting of over-ambitious lending targets. The result was that it was too easy for marginal projects to slip through appraisal.
The second factor was the oil price rise in 1973, which knocked sideways all the assumptions about local costs and the national economies of borrower countries. To quote FAO - "Given the ubiquitous impact of the oil price shock-wave, it is hardly surprising that so few projects prepared in the 1970s were on target" (FAO 1988).
Disbursement is a major problem for most agencies. Project plans and appraisals are nearly always over-optimistic about implementation, such as the time needed to import equipment, to recruit and bring in experts, to select counterpart staff for training. But there is seldom enough flexibility in the budget to get over this problem. Perhaps some money can be transferred to the following year, but then there is a surplus to be spent in a hurry at the end of the project. Transferring funds between components because one is underspent distorts the planned balance of the project. Unplanned spending in order to meet disbursement requirements is a common failing. The problems of achieving planned disbursement have led some agencies to adopt special policies. For example the UK all-party parliamentary group on overseas development reported "The Government intends to expand what it terms 'Fast disbursing programme aid' to Africa" (All Party Parliamentary Group 1985). Concerned about slow disbursement on projects in sub-Saharan Africa, IFAD set up in 1985 a programme of seven major steps to improve matters.
Disbursing aid funds wisely and effectively is more difficult than is assumed by people outside the industry. Band-Aid is an interesting example. A large sum of money was collected for immediate-assistance projects and the intention of the Trustees was to disburse the whole amount within 12 months. Two and a half years later about one third was still unallocated. Funding good projects is more complicated than they imagined.
Some of the main factors which donors should absorb into their policy for project planning can be identified:
- There must be a strong commitment to the programme by the donor agency, the host government and the implementing agency. All must be quite clear about what they want to do and how they plan to do it.- In all the subsequent discussions of project planning and design it is emphasized that participation is critical. Most projects have some element of participation by the beneficiaries. Many seek it during operation, a few in ex-post evaluation, but few projects have sufficient participation at the planning stage. Where it does occur, it is mostly limited to central governments or national departments and agencies. It is seldom carried down the line to the level of the district, the village or farmers' groups. At project preparation, project appraisal and evaluation, the local staff tend to be junior partners of the international experts.
- Another factor which comes up as important at every stage is the question of the duration of the project. Most agencies are forced by their financial procedures to think in terms of fixed periods for projects, and there is little chance of overcoming this difficulty. But long projects are clearly more effective, and even where projects have to be funded in short-term tranches, it should be possible to assure funding over a longer period. It is not sufficient to say that after the three or five year project it will be evaluated and may be extended. This leads to unease and anxiety on all sides. What is required is an unequivocal assurance that long-term funding will be provided. If there is an element of uncertainty about whether a project will work or not, this should first be resolved by a pilot project. Projects are doomed to failure if they start with the underlying thought that "We do not know whether this will work or not, but we will try it for a few years and see".
- Donor planning suffers from what the FAO Investment Centre review calls 'persuasive optimism'. Forecasts of the effect of projects are frequently over-optimistic, particularly anticipated yield increases, the effectiveness of dissemination, the rate of uptake, and the financial benefits to farmers. Benefit-cost ratios and ERRs are regularly over-estimated. World Bank has an appraisal process more rigorous than most agencies, and this commonly leads to ERR being lowered at appraisal, but this is evidently not sufficient to counter persuasive optimism. Thus:
-in the 10 years to 1984 only 44 percent of projects equalled or exceeded the appraisal ERR, falling to 23 percent in 1984;- ERR at appraisal is consistently higher than the revised ERR (RERR) estimated at ex-post evaluation. Over the 3 years 1983-1985 the average appraisal estimate was 10 percent higher than the ex-post estimate.
This tendency to over-estimate benefits is rather foolish because it can result in projects being unfairly labelled as failing to meet expectations. A project with a final RERR of 15 percent has failed to meet the target if that was set at 20 percent, but is successful if the ERR was 15 percent. The level generally considered acceptable is 10-15 percent.
In many of the projects studied, the project was impaired by problems arising on the side of the recipient. Where the problem is a weak economy or an unstable government there is nothing to be done, at least not in the short-term, by a soil conservation project. But it is always important to identify possible problems and constraints at the earliest stage of project identification. Among the following difficulties some are particularly relevant to soil conservation and agricultural development, others apply to assistance programmes in any sector.
The record shows that few countries have a satisfactory policy for the long-term development of natural resources. This requires an inventory of natural resources, or nowadays the popular phrase is a "national environmental profile", from which a long-term development policy can be constructed. Then within the policy there should be first a national conservation strategy (defining long-term objectives), and second tactical programmes for each agricultural sub-sector (defining the short-term routes towards the long-term objectives). It is quite remarkable that governments appreciate that a policy for mineral exports requires detailed study of the available mineral resources, but will cheerfully set up a policy of becoming self-sufficient in food without an assessment of how this can be achieved within the available resources of land, water and manpower. The record shows that the identification and planning of projects is often impaired by the ability of the recipient government to see how a short-term project would fit into a long-term programme. Agricultural development programmes may be directed toward implementing a strategy, more often they are tactical, but in either case they need a policy to work within. Only a small number of projects have been directed towards assisting the formulation of a national policy for agricultural development.
Another issue which comes up in every review of assistance projects is that there must be a favourable political context. Harrison puts it rather well - "The political context can interfere with the best-designed ventures. Real participation may be blocked by governments opposed to any expression of popular autonomy - or by bureaucracies and ruling parties with vested interests in maintaining the power of officials and technocrats. The economics of development may be distorted by pricing policies that favour cities over rural areas. Even low-cost, self-help projects can be crippled when the modest resources or technical assistance they need are lacking" (1987).
The question of the effect of prices on agricultural production is extremely important. There are cases where artificially low prices to the producers have seriously reduced their interest in increasing production. There are also many examples where lifting prices to realistic levels and providing market outlets have contributed to the success of projects which include the adoption of new production techniques. The question of how much donors should try to influence borrowers over prices is debatable.
The other financial issue with political implications is exchange rates. Developing countries tend on the whole to overvalue the national currency and this can have undesirable effects on projects involving imports and exports.
Closely connected with political will is political stability. Rees (1988) says, "Unstable political regimes tend to create institutions and then dissolve them before they acquire sufficient experience and expertise". He might also have added that it is difficult for government departments to pursue long-term programmes effectively when there are frequent changes of ministers. He adds "It is particularly important that environmental planning has the support and sympathy of the people".
Another question which should be resolved early in the planning is the capacity of the recipient to provide the facilities and services needed by the project. There are too many cases where a mid-term review or a termination report reproachfully complains that the host failed to provide counterparts or transport, or some other agreed input. This may be the result of a reduction in the moral support of the host for the project, or unexpected budget cuts enforced by treasury, but the most common explanation is that the ability to provide these inputs was not adequately investigated at planning.
The reader who queries why projects start in these circumstances might consider the remarks of a senior official who said, "Well, I do not expect this project will change anything, but it will bring in some foreign exchange, there will be a bit more transport around in the department during the project, and when it finishes the buildings will be useful". The probability of success is not the only issue in the minds of political leaders in the receiving country - nor indeed of fund managers.
It is not only the capacity of governments and departments which needs to be considered. Institution building at lower levels may also be important. A consultant to the World Bank writes, "The present concern by the development community arises from a realisation of the importance of institutions and their development as prerequisites to lasting development. Along with this goes the realisation that not only are formal organisations, ministries, and parastatals often ill-equipped to perform their functions, but there is need to develop lower-level institutions if development assistance is to be useful in its intended purpose" (Butcher 1988).
Several other writers, and reviews of projects, stress the importance of identifying at planning stage the abilities and capacities of local groupings at the lowest levels of government administration and informal groupings below this. In many countries projects cannot succeed unless they are working with the existing social structure. In some countries this has a tribal base, in others the system of power may be through chiefs and headmen, either elected or appointed by government. Or it may be party politics taken down to grass roots level. The important thing in planning is to identify the role and the capacity of institutions at all levels and to channel their abilities into the project. Sometimes quite informal groupings can play a major part, such as the Mwethya groups (Women's Work Groups) in Kenya or the Ekip (ephemeral work gangs) in Haiti.
The question of land tenure frequently arises in soil conservation projects. The conventional wisdom is that farmers will not invest in soil conservation works unless they can expect to reap the rewards. This applies particularly to works with a long pay-back period, like terracing and tree planting. This was discussed at the Puerto Rico Workshop and summed up as follows: "Obviously, the farmer who is likely to be evicted from the land he or she is working is not likely to be interested in investing heavily in conservation practices. On the other hand, participants did report on successful activities in different countries where various land tenure systems are in operation. Perhaps the lesson to be learned from this is that any soil and water conservation project must be fully aware of and understand the local land tenure system and then tailor its programme to fit in with that system" (Sanders 1988).
An interesting recent development in Lesotho may have far-reaching effects. Traditionally the exclusive right to grow crops on arable land applies only until the crop is harvested, after which the right to graze the residues reverts to the whole community. Recent legislation cancels this age-old system, and allows the land user to forbid access by other people's animals. If successful, this could open up the possibility of applying a number of useful practices, for instance stubble mulching and conservation tillage, planting grass on terrace risers, and planting trees and shrubs on corners and boundaries of arable land.
Every review of projects emphasizes the importance of getting the design right, and most reviews identify it as the most important single factor influencing project success. Projects with poor performance usually have poor design, for example:
- The World Bank review of projects in the ten years to 1984, found that 86 percent of projects with poor performance had design faults and in one third of these it was the single most important factor (World Bank 1985, quoted by Balacs 1985).- World Bank (1986) reports that poor content affected 62 percent of failed projects, and again it was the most important factor in one third of these.
- FAO reports an improvement in design of FAO projects. In 1981-82 design was good or satisfactory in 34 percent of 80 projects, 43 percent of 103 projects in 1983-4, and 61 percent of 147 projects in 1985-6. However this result still shows that 39 percent of projects had less than satisfactory design, which indicates room for further improvement.
Another constantly recurring theme in all the major project reviews is the old-style blueprint operation where everything is worked out in detail in the project document which then becomes enshrined as the operational rule book. This means that faults in the project continue even after they could and should be corrected. There is an almost unanimous recommendation for greater flexibility so that the project can learn from experience and change accordingly. This is usually associated with a recommendation that more responsibility should be delegated to project managers. NGOs and small projects are better able to profit from their ability to adapt as the technicians and the farmers learn from each other.
When considering the requirements for a soil conservation policy for Europe, Morgan and Rickson (1989) listed five points which with slight modification are equally relevant to project design:
- There must be a high level of political pressure for it to happen.- There must be a technology which has been demonstrated to work.
- There must be a clear financial incentive. In the case of a policy this must accrue to government, in the case of projects it must accrue to the participating farmers.
- It must not be an end in itself, but part of an environment protection programme (in the case of projects read "part of an agricultural development programme").
- There must be a local organization to take responsibility for implementation.
The similarity of the requirements for a multi-national policy and an individual project confirms the thought that while practices are site-specific, the broad issues and principles are universal (Sanders 1988).
Project objectives
The FAO Review of Field Programmes (1987) is critical of the loose definition of project objectives, and these evaluation studies confirm that it is equally a problem in the projects of other agencies. In the FAO report the most common criticism (in 37 percent of 131 projects) was the absence of a well-defined link between immediate and development objectives. The immediate objects were found to have been described too vaguely in the project document in a quarter of the projects. This problem tends to be more noticeable in smaller agencies which do not have sophisticated and refined appraisal techniques. For example several projects were found whose general objective was defined as something like, "To raise the standard of living of the rural population", and the immediate objectives loosely defined in terms such as, "To increase food production in the project area".
Another quotation from FAO is "The problem of vague project objectives still crops up too regularly. There is the frequent lack of linkage between the broad development objectives being addressed and the immediate objectives of a project." In fact FAO has introduced special in-house training courses to address this and other weaknesses in the design of FAO projects.
In project design a balance is required between aiming to meet national objectives and those of individual farmers, since these are usually not the same. The national objectives of governments may include things like increasing agricultural exports, or import substitution, which are of little interest to the subsistence farmer. The project design must therefore identify the farmer goals and be firmly addressed to them. Many big agricultural development projects of the '70s failed because this was not realized at the time. Recent evaluation studies and reviews of them now give us a much clearer understanding of what the farmer sees as problems, and what are his goals and objectives.
Particularly in the context of soil conservation there has been a major change in thinking. It is now known for example that there is little point in educational programmes intended to explain to the farmer the problems of soil degradation and the need for him to do something about them. It turns out that he usually understands perfectly well, but he has more urgent and pressing problems which require all his energies, so that soil conservation is forced into being a low priority. Several studies have shown that a way forward is to relieve the most pressing problem, for example by improving a road network, or improving food security by introducing improved agronomic practices. The farmers are then able and willing to turn their attention to soil conservation. Examples of this approach are to be found in Shaxson (1988), and other case studies in Moldenhauer and Hudson (1988).
Studies have now shown that the objectives of the commercial farmer to maximize yield and income are not necessarily those of the subsistence farmer. He is likely to be more interested in improving food security by decreasing the risk of failure, or improving the return on inputs of seed, fertilizer, or labour, or reducing the labour input, or improving the quality of life by reducing drudgery, particularly relevant in the case of women farmers. The project design must take account of the financial implications for individual farming families of the recommended practices and investments. Is the farmer able and willing to make the investment?
What projects must offer in order to be enthusiastically adopted by farmers can be defined. Firstly there must be clear on-site benefits offering short-term benefits with large increments. A new technology must offer an increase of 50-100 percent, because a 10 percent improvement will not persuade him to change. The improvement must also occur within the time-span of the farmer's planning, which usually means the next growing season, or at most the next twelve months. Donor agencies or recipient governments may appreciate the value of long-term benefits or downstream benefits, or a 10 percent increase in average national yield, but none of these are important to the subsistence farmer although it is he (or commonly she) who takes the decision on whether or not to implement project recommendations. The criteria for new technology to be acceptable are discussed in Section 4.
Project design must be based on limited attainable objectives. Reviews show that planners tend to be over-optimistic about:
- the benefits of new agricultural packages;- the ability of existing services to disseminate new ideas;
- the time required to make things happen, for example the selection of candidates for overseas training.
Associated with the turning away from big blueprint operations is a move towards a series of small changes rather than complete replacement of the technology, and towards projects which start slow and small so that the farmers can be involved as the project develops and grows.
The element of risk is important in small-scale farming, which is a gamble against the weather, pests, disease and prices. What a project can hope to do is either to shorten the odds against the farmer by decreasing the risk, or to increase his ability to invest by credit schemes. Improving the odds is by far the more common function of development projects. It is remarkable how even in the poorest community, money can be found to invest in what is seen as a certain winner.
Project targets
In agricultural development projects, and particularly soil conservation projects, it is difficult to establish targets where progress can be quantified. It is common to use a measurable item as surrogate for the real target, but this can introduce distortions. For example increase in crop yield has been used as a surrogate for measuring effectiveness of a soil conservation programme. But the programme will almost certainly include a package of practices, and a measured increase in yield claimed as the result of soil conservation practices may in fact be the result of improved water conservation, use of improved seed or fertilizer, or indeed factors external to the project such as improved marketing or pricing.
In forestation projects the number of tree nurseries established is frequently used as the measure of effectiveness. But this is a poor surrogate. The number of nurseries established does not necessarily reflect the number of seedlings produced; the number of seedlings in the nursery is not the same as the number distributed to farmers; this is not the same as the number planted which is in turn not the same as the number which survive. The number surviving after a year will be further reduced by drought, pest, disease, and attack by livestock. So the real measure of the effectiveness of the project, which is the number of new trees growing after several years, has only a very tenuous link with the chosen yardstick of the number of tree nurseries established.
A common deficiency in project design is the absence of an annual workplan which sets out programme activities against a timescale. The plan must be realistic and attainable, not a catalogue of hopes which are unlikely to be achieved. Most experienced development workers will be aware that in some countries the record of accomplishment submitted at the required time intervals tends to mirror the targets, with little relation to what has actually been done. Monitoring of progress must be accurate and reliable to be of any use.
Institutional arrangements and local participation
Good project design must include detailed plans of how the project will be implemented. It is a common complaint of evaluation missions that, first, insufficient attention was given to this matter at project preparation and at appraisal, and second that too much is assumed by the planners, with insufficient discussion with the recipients, at all levels. The mechanics of how the project interacts with existing departments, agencies and institutions is often not worked out in sufficient detail. For example many projects have been impaired by not having worked out the duties and responsibilities of project managers vis-a-vis a counterpart. Minor issues left unresolved at planning can reduce goodwill and cooperation - examples are control of expenditure and allocation of transport.
It is difficult for a project to be successful if it does not fit into a national policy or a long-term development plan for the sector. The lack of a long-term national strategy for agriculture in some countries in Africa has hindered development projects for many years.
There must be full agreement and cooperation between donor and recipient government. The project or programme cannot work if there is conflict with the indigenous regime over major political issues. Most agencies include in the project proposal a section on issues, i.e. conditions to be satisfied before the project will proceed. However from the evaluation studies it is clear that the urge to get projects started has on many occasions softened the resolve of the donor and led them to start projects without obtaining complete agreement and support. This is shown by the World Bank study which showed that in 79 percent of unsatisfactory projects, borrower support was unsatisfactory. On the other hand a firm stand by a donor can lead to a change of heart by the recipient.
An example is when World Bank pulled out of a huge nationwide programme in the face of what it felt was the intransigent refusal of the government to increase food prices to the producer. Whether as a direct result of this action or not, farmer prices were increased some years later.
The question of central or dispersed control of soil conservation activities is quite common. Many countries have a structure similar to that in the United States, where conservation is a federal responsibility but extension is a state responsibility. Hopefully this will become less important with the growing acceptance of the concept that soil conservation (and hence the soil conservation component of development projects) should no longer be treated as a separate discipline, but considered as an integral part of extension work (Moldenhauer and Hudson 1988; Shaxson et al. 1989).
Another weakness identified in project design is insufficient consideration of the capacity of local institutions to play their part in a proposed project. As a result this ability is frequently overestimated. It is not sufficient to list what inputs the project looks for; it requires a careful study, in cooperation with the local agency, of the resources of the agency, particularly manpower, since this often reveals the need for a training component in the project. Any other constraints should also be carefully discussed. For example in an on-going soil conservation programme in the Far East, the project proposal failed to appreciate the importance of language difficulties which are now seriously inhibiting progress. This could have been avoided by including in the project design a large component of language training at the beginning of the project.
Although in the past the need for institution-building has not always been recognized, there is nowadays a strong trend, particularly in FAO projects but also in other agencies, towards more projects with institution-building as a main objective (FAO 1987). All projects, whether mainly institution-building or not, should aim to reduce the dependency on international experts, for otherwise they can never become self-sustaining.
Some other potential problems of inputs by local agencies are funds for local expenditure, local research inputs, and cooperation between departments. Few of the projects which have been evaluated do not record some problems with local funding. The FAO review also finds this a frequent weakness, and one which is increasing, especially in Africa. Sometimes the problem arises from unpredictable external factors such as the price of oil or commodities, or exchange rates. But at the design stage it is important to try to identify and anticipate any possible financial shortfalls since unforeseen budget cuts can mean reducing counterparts, other local staff and transport to the point where the project becomes ineffective. Few agencies are able to follow the example of the Swedish authority SIDA which can in emergency provide funds for local recurrent expenditure.
The ability of local research institutions should be considered at project design, and is discussed in Section 4. It will be seen that in agriculture generally, and soil conservation in particular, there is a history of inappropriate technology. Even if local facilities are not strong enough for innovative research, it must be able to test technology which originated elsewhere through on-farm research, pilot studies, market research and social surveys.
The lack of cooperation between departments and agencies is a universal problem and by no means confined to developing countries. In project design the important thing is to identify potential problems and try to bypass them. In the past a popular solution has been to set up a new and separate Project Management Unit (PMU), but experience now shows that this should be avoided if possible.
Experience also suggests that the design of institutional arrangements relies too much on discussions at national level, probably with token visits to the field and some superficial contact there with the local officials. But experience shows few projects get down to discussions allowing a real input of ideas from village management and farmer associations. Of course this partly results from the inability of the visiting project planners to communicate directly at this level, but state-of-the-art thinking is that involving the people who are going to carry out the project is essential at the stage of project identification and preparation.
Project duration
Projects of short duration are not as effective as long-term projects and programmes. In the past three-year projects were quite common, with World Bank having an average of four or five years, but evaluation and assessment reports point out what should have been recognized in the beginning, that agricultural systems change slowly, and a ten year horizon should be considered as the norm. The uptake of a new variety can be quick, but anything that affects the system or the social structure is slow.
The problems of short projects are well documented. There is usually a slow start-up on both sides. Donor staff are not in post as soon as expected, tendering and procurement of donor-provided hardware always takes longer than expected. So does clearing through customs and delivery.
On the borrower side there are always delays in finding and making available counterparts and other local staff, in setting up the administrative procedures for the project, and creating a distribution mechanism for inputs. Together these lead to under-disbursement, cost overrun and general inefficiency.
Training is particularly constrained in short projects. In a typical five-year project which plans to send local staff for overseas training, it takes one year to locate and select the candidates, anything up to three years for the overseas training, and the officer returns to the project in time to help it close down.
The common practice of setting up projects with the possibility of extension is seldom satisfactory, particularly the approach, "Let us look again after three years and then decide what to do". The uncertainty undermines the confidence both of the project and the cooperating farmers. Where donor procedures do not permit the unequivocal promise of long-term support, it should still be possible for donors to buy a three-year or five-year time-slice of a longer programme. This has a better chance of continuity than an isolated fixed-term project.
Successful programmes which have been running for a long time invariably refer to the length of the programme as one of the factors helping towards its success. An outstanding example is the SIDA soil conservation programme in Kenya which has been operating for fifteen years and there is a clear understanding that the assistance will continue to be available as long as it is welcomed by the Government of Kenya.
At project design there should be consideration of sustainability and what will happen at the end of the project. Too many of the early projects simply faded away and after a few years the only remaining trace is the buildings. There should be plans, not necessarily detailed but carefully thought out, for the project eventually being institutionalized, i.e. incorporated into government agencies or institutions, or becoming independent and self-sustaining, which is often the desired end-product of cooperatives or credit schemes. In soil conservation programmes one would hope that they become absorbed into the national agencies, but an even better plan is that the whole project should be carried out from within an existing department, as in the SIDA Kenya project.
One other factor related to time is the idea of going back to some form of assessment or evaluation a number of years after the project has been completed. So far only the World Bank has been observed to do this on a regular basis, but it does seem to be a worthwhile exercise. The conclusion of the Bank survey is that in many cases a modest intervention of support and assistance after the project would be well worthwhile. A problem is that the budgeting and disbursement procedures of most agencies would find difficulty in accommodating this idea.
Training and education
During the last ten or fifteen years since evaluation became a serious activity, there have been two shifts in the approach to training and education. The first is that ten years ago it was commonly assumed that the most important way to achieve soil conservation was to educate the farmer, tell him about erosion and erosion control, and its importance to him and his grandchildren and the country. This was probably a legacy from the extension approach of North America where over the last forty years a very high level of awareness of erosion and conservation has been built up. However, even in America the weaknesses and deficiencies of this approach are now beginning to show, and in development projects they have shown up much more quickly.
It is now appreciated that on the whole the small farmer is well aware of the cause, the process, and the results of erosion, but he is not interested in doing anything about it because there are more important problems to attend to.
The other new realization is that the farmer's limited resources and attention are focused on short-term planning and activities. He may have anxieties about the future for his grandchildren, but he cannot afford to let his attention and efforts be sidetracked away from the immediate problem of feeding his family this week or this year.
A third point is that planners now appreciate that activities on farmers' fields are not going to be influenced by what the downstream effects might be, such as flooding and sedimentation. But land use planners with their feet on the ground have known this for a long time.
At the same time there has been a strongly increasing trend towards the realization that technical assistance should pay less attention to showing developing countries what to do (or what the planners think they ought to do) and put more effort towards institution-building, that is increasing the local capacity for being able to solve the problems themselves. So there has been a steady increase in the number and the size of training components, and this trend is still continuing. There has also been a sharp move away from what is now seen as a rather old-fashioned approach of sending people on long academic training courses to universities and research institutions in the developed countries. Instead there has been an increasing realisation that the value of this type of training is mainly in universities and research, and that the more pressing need in most developing countries is training in management, and a broadening of experience, preferably acquired in neighbouring countries.
