Chapter 4: The Donors’ Failures

Some of the blame for the overall failure to cope with desertification rests with those who have expressed the desire to finance and assist such work.

Clearly, they have not provided sufficient funds. But even those funds which were made available have not always been used effectively. Donors have opted too often for  the  easy way out: the financing of the simpler projects often those with possibilities of procurement of goods and services from the donor country – rather than the more difficult ones for which there is greater need. They have failed to take proper heed of recurrent costs in their projects, with the result that years of work have later proved useless. They have tended to finance individual projects without examining the details  of the programme in which they were situated with the  result that promising projects never fulfilled that promise. And they have consistently poured more aid into relatively rich countries rather than support those who need it most the rural poor in the arid and semi-arid regions of the world.

The resources needed to combat desertification can be channelled through different mechanisms and it is from this point of view that desertification should be analysed since each mechanism pursues different objectives and follows different courses. This type of action or mechanism is summarized as follows:

  • aid
  • economic and technical co-operation
  • technical assistance
  • contribution to the Plan of Action to combat desertification

Aid can be implemented bilaterally or by multilateral arrangement. Strictly speaking, aid is of a short-term character and fails to deal with the causes of the phenomenon. It is a compensatory type of action which does not attempt to change the structural causes of the phenomenon. It alleviates pressure in the short term, but if the problem is to be solved, then other measures should be taken to eliminate the causes.

The other three types of action can be short-term or long-term. What is important is to have the appropriate mixi.e. action to relieve short-term , immediate pressure but also action to tackle the structural causes of the problem.

A second aspect refers to the sectoral or subsectoral classification of the resources. The classification that lists everything under ‘Agricultural Aid’ does not show how much has been channelled to the fight against desertification. In other words, the analysis of the changes in the flow of international resources under the title of Agricultural Aid can be misleading.

An exhaustive analysis of this type is beyond the scope of the present Report which has to rely on available secondary data.

 

The Quantity of Aid

Lack of financial resources has often been mentioned as one of the main reasons for failure to carry out the UNCOD Plan of Action. The United Nations Administrative Committee on Co-ordination (ACC), for example, reported that insufficient financing was ‘seriously limiting the efforts of the United Nations system to implement the Plan of Action and that there was an urgent need for external sources to increase their assistance to anti-desertification projects.

As already mentioned, desertification does not appear as a separate item in the budgets of most bilateral and multilateral donors. This makes it difficult to analyse the funding situation. One approach is to attempt a breakdown of external aid to agriculture. This is not entirely satisfactory, for although in arid and semi-arid regions much agricultural aid is related to anti-desertification work, much of it also goes to cash-cropping schemes and poorly designed and executed irrigation projects, both of which can be part of the problem rather than part of the solution. For instance, new irrigation schemes can cost as much  as  $17,000 per hectare, and in the Sahel one-tenth of all aid goes  to irrigation. Yet during the early 1980s, for every  new hectare of land brought under irrigated cultivation, another irrigated hectare went out of cultivation because bad design and bad management had led to water-logging and salinization.

During the 1970s, external aid for agriculture increased sharply as a percentage of total aid. The terms of aid also improved, with a substantial increase in concessional commitments and a small decline in non-concessional ones (see Table 5).

From Table 5 it is clear that multilateral organizations devote a much higher proportion of their commitments to agriculture (33 %) than bilateral donors (11%). Some 57% of bilateral concessional assistance to agriculture comes from three countries (the United States, Japan, and the Federal Republic of Germany); if the Netherlands, France and Canada are included, more than 80% comes from just six countries. The United Nations Meeting on the Least Developed Countries held in Paris in 1981 tried to rectify the situation, and most donors at the meeting agreed to increase the share of agricultural aid in the poorest countries.

However, the proportion of aid going to agriculture – which is likely to have some relevance to the problem of desertication – has been stagnating since the late I 970s. According to FAO:

Recent assistance levels fall far short of what is believed to give agricultural development a n adequate boost only 45% of the need, according to one FAO study. Further analysis suggests that assistance for agriculture would need progressively to increase by at least a year in real terms in order for developing countries to double agricultural production within 20 years. Moreover, there are trends for the assistance that is provided to be less concessional and for there to be longer delays in loan disbursements.

In I98I, partly as a result of a worldwide economic recession, official commitments of external aid for agriculture fell by 7.5% – the first fall for many years. The decline was largely due to a 15% reduction in bilateral aid, which fell for the second consecutive year. OPEC aid was cut back particularly sharply as a result of declining oil prices. Oil-importing developing countries suffer whichever way the oil pendulum swings. In the early 1970s their needs for foreign exchange rapidly increased when the price of oil quadrupled. Five years later, their levels of ‘oil aid’ fell as oil prices stagnated and declined.

It is difficult to assess what might have been done if bilateral aid had continued to expand . The point can at least be illustrated by one example. FAO’s Food Security Assistance Scheme has been helping to provide warehouses for grain storage; these provide an emergency supply for time s of need , and help to stabilize and control grain prices. In Mali, Niger and Burkina Faso, a total of7 3,500 tonnes of storage have already been provided and 9,500 tonnes have been provided for Cape Verde. But, as of late 1983, FAO had $200 million worth of grain storage projects in 15 countries still unfunded. Gaps such as these take on huge significance during times of emergency such as those now threatening much of the Sahel again.

Multilateral aid also suffered from the  recession. For example, the World Bank’s ‘soft loan’ division, the International Development Association (IDA), had to reduce its commitments to all sectors from $3.5 billion in fiscal 1981 to $2.7 billion in 1982. The United Nations Development Programme was even more severely affected by declining contributions. In 1982, it announced that it could authorize budget expenditures at only 55% of the level originally planned for the period 1982- 86. One direct result of this was to reduce the number of FAO field experts working on UNDP projects from 1,313 in 1981 to 1,037 in 1982. In fact, at this level, FAO had only about two-thirds the number of UNDP-supported field experts of six years previously (in 1974).

This bodes ill for the future. UNDP is a major funding source for United Nations specialized agencies such as UNESCO and FAO – in fact, it provides about half of the external funds used in FAO field projects, and in the past provided substantially more. Many of these projects are essentially pilot studies and demonstrations, designed primarily to develop effective techniques. When successful, they can lead to major investment projects which are typically funded by institutions such as the World Bank and the regional banks. These are the projects which could begin to stop desertification. There is thus a double threat to the future of desertification control: seed money from UNDP is hard to come by, and this is inevitably reducing the number of investment projects which can follow up on technical successes. And the banks themselves are short of the funds which are needed to support investment projects. The second problem comes at a critical time, because many pilot projects have now evolved successful schemes particularly, for example, in community forestry and the time is ripe for major investment. It may be slow to materialize .

Also, organizations providing non-concessional aid, such as the World Bank and the regional development banks, may have a hard time justifying investment in anti­ desertification work, because the work takes such a long time and the returns are often uncertain. This can be an especially important consideration if the focus of the project in question is to improve a nation’s food self-sufficiency. In countries where food prices are maintained at low levels, these projects may never pay for themselves. Thus, the banks are under pressure, for financial reasons, to invest in other areas.

 

The Focus of Aid

While the recession played an important role in halting the expansion of agricultural aid , other issues were also involved. For example, agricultural projects in general – and certainly anti-desertification projects in  particular – tend to be more difficult to implement than projects in other sectors, and donors tend to shy away from  them, particularly in periods of declining aid. An attitude which flies directly in the face of the maxim that aid should  be directed where there is most need. As FAO comments, ‘…it  is unthinkable that aid donors should be deterred by such problems from contributing to the sector which at least for the foreseeable future holds the key to economic and social development  in the countries in greatest need of aid. The difficulties need to be confronted, not shirked.

One example of the way this works is the tendency for donors to use their aid to finance the supply of agricultural inputs, which avoids most of the difficulties associated with agricultural projects in general, and also has the advantage, from the donors’ point of view, of creating a demand for their own productive systems. This tendency, therefore, contributes to maintaining the level of economic activity in donor countries – an element that can be of great importance in periods of economic recession. How the inputs are then used remains an open quest ion. Nor does aid of this kind do anything to encourage recipient countries to produce their own agricultural inputs more cheaply or more efficiently.

Agricultural projects – and particularly those connected with desertification – also tend to have a higher proportion of local and recurrent costs than projects in other sectors. The chances of success in their implementation also depend critically on the way individual projects are related to programme development in general. For instance, tree nurseries for reforestation projects may be pointless when there is no system or funding for growing the trees planted in the right places and cared for and protected while they grow. Failure to co-ordinate, integrate and follow up desertification projects has been a major cause of project failure. The WCARRD Plan of Action addressed these and other related issues in 1979: 51

In the least developed and most seriously affected countries in particular, donors should be ready to finance: (a) local and foreign capital exchange costs, as appropriate; (b) recurrent as well as capital expenditures; and (c) an increasing share of programme and sectoral supports as compared to the project approach.

The issue of recurrent costs is important in desertification control and land recovery, which by its nature is a long-term activity. Seeds may be sown, pasture rested, tree seedlings planted, and sand dunes stabilized. But unless projects to organize such activities also ensure that provision is made for the future, the work is wasted. Because donors have not examined in sufficient detail the issue of recurrent costs, a  number of desertification projects have floundered: stabilized sand dunes have become unstabilized, forestry plantations have been left untended, and reclaimed cropland has reverted to scrub. As Anne de Lattre of the Club de Sahel puts it:

New projects generate very heavy recurrent budget costs, which cannot be met , and so the project deteriorates. Phase 2  and phase 3 of the project then have to be rehab ili tat ion, not expansion. Since 1976, there has been a 355% increase in real aid flows to the Sahel. But there is actually massive disinvestment taking place, because of those rehabilitation costs.

Donors have often also ignored the programme implica­tions of their projects. For example, it is easier to mount a project aimed at improving subsistence agriculture techniques in an arid region than it is to assist  a  government  in instituting new pricing policies. Yet the former may be useless without the latter, because the farmer has little incentive to adopt the new practices unless he can get a reasonable price for his crop. Mistakes of this kind are now rarer among multilateral agencies but still common with bilateral donors, who often do not know enough about agricultural constraints in drylands.

Many of the community forestry programmes in the Sahel have come unstuck for this reason. As already mentioned, land reform is often an essential prerequisite to forestry schemes if local people are to identify with, and benefit from, the tree crop. In one project in Niger, seedlings were actually destroyed by local people because they had been planted against their will on traditional grazing grounds.

There remains one overriding reason why funds for desertification control have not been forthcoming, and have not been sensibly directed into the areas where the need is greatest; donors still tend to direct their agricultural aid at areas with low poverty rather than those with high poverty (see Table 6).

Unfortunately high poverty areas  are  precisely those worst affected by desertification. A recent study by the Italian Dipartimento per la Cooperazione on international aid to Sahelian countries shows that, despite concentratioof international aid on irrigation development schemes in the area as a whole, in the three countries with the lowest GNP, international aid goes to rural development and water supply projects whereas, in the countries with higher GNP, it goes to specific areas like livestock or irrigation schemes.

No doubt the returns are safer, the investment more easily justified. In the Sudan, for example, investment in the expansion of irrigated farming along the Nile looks like a much better prospect, and donors are likely to favour it. But such a priority cannot serve well the causes of the fight against desertification, whatever its apparent economic justification.

Specific  examples of desertification-affected countries which receive small amounts of agricultural aid per head of agricultural population include  Bangladesh  ($2.8), Brazil ($4.1), Ethiopia ($1.6) and India ($1.4). By contrast, countries such as Argent i na , Costa Rica,  Honduras, Nicaragua and Panama which are not affected by this problem receive more than $20 of aid per  head  of agricultural  population. It is also significant that while figures for Bangladesh, Brazil and India improved greatly in 1980-81 compared with the  average  for   1974-81  (increases of 59%, 31% and 36% respectively)the situation in Ethiopia failed  to  show  the  slightest  improvement. These  figures should  be  analysed  more  carefully  in  order  to  avoid important conceptual errors. For example, Bangladesh has a very large rural population and high population density. Its agriculture is labour intensive and mainly  geared  to  local mar kets. In contrast, Argentina’s smaller population is predominantly urban, its population density is rather low, agricultural production is more mechanized requiring a relatively low labour input, and production is  oriented mainly to international markets.

India is also mainly a rural country with a population about 20 times that of Argentina and its agricultural productive structure more diversified and characterized by the co-existence of different systems.

The conclusion is inescapable: donors tend to put their money where the action is, not where it should be.

The problem of Ethiopia raises the issue of political allegiance. Ethiopia is aligned with the Soviet Union, which gives it very little of what Western nations refer to as official development aid. But because of that alliance, Western nations also tend to give Ethiopia relatively little consider ing the needs of the country. Africa receives on average about $19 aid per head, Ethiopia a mere $6. In 1984, about half the United States’ total economic aid for Sub-Saharan Africa went to five close allies: Kenya, Liberia, Somalia, Sudan and Zaire.

An analysis of capital commitments over the years to agriculture in developing countries, and to  various  sectors within them, provides a key to the effect the Plan of Action has had on desertification spending. Had the plan been effective in increasing funds and altering priorities, changes should be apparent in the overall pattern of sector allocation, and in particular within desertification-affected countries. The funds involved should have been large enough to effect major changes of this type. The cost of the Plan in external assistance was  estimated at $2.4 billion a year. Over the two year period 1980-81, total external commitments to agriculture were in the region of $10 billion annually.

The figures in Table 7 provide no convincing evidence that the Plan has had any major effect on the pattern of spending within agriculture. On the contrary, in the one area where an increase would be expected – land and water development – there was a 2% decline. The proportion  of spending on livestock and forestry also declined. It is possible that some spending on desertification is concealed within the 4% rise in the rural development and infrastructure sector, the 3% rise in research, training and extension, and even the 2% rise in crop production. Possible but improbable, as is shown by a more detailed analysis of similar figures which compare the situation in desertification-affected countries for con­venience sake, those in the Sudano-Sahelian region will be used – with the pattern in Africa as a whole, other regions and developing countries as a bloc (see Table 8).

The figures in Table 8 are more revealing; they also highlight a few instances of concerted action in a few nations. In Cape Verde , for example, donors channelled their efforts to improve land-use planning and to a large­ scale reforestation programme, aimed at planting one million trees a year. Chad devoted nearly three quarters of its pitifully small volume of agricultural aid – the lowest level of aid within the Sudano-Sahelian region – to crop  production. In contrast, Mauritania, with equally little aid, was devoting most of it to land and water conservation projects. Benin was planning to spend nearly all its agricultural aid on rural development, in keeping with its aggressive land reform policies. Niger was the only other country to devote  more than half its   aid to rural development, a result of its determined policy to increase food production and raise the standard of living of small farmers. Land and water figures for Somalia and Sudan were also relatively high; both countries have attempted to make national desertification plans and set up desertification units.

A comparison of figures for the Sudano-Sahelian region with those for the whole of Africa reveals that a higher proportion of aid is being spent in the affected region on land and water projects than in the continent as a whole. Proportionally more money is being spent on forestry in the region (3.7% as against 2.8%). However, the proportion going to livestock is not significantly higher and the proportion spent on crops is actually significantly lower in the desertification-affected countries  than in the continent as a whole. Perhaps the most striking fact is how small a percentage of agricultural aid is spent on crops (10.2%), livestock (3.9%) and forestry (3.7%).

Obviously, it is impossible to decide whether donors or recipients are most responsible for the priorities reflected in these figures. But it is clear that donors are not getting their money into the desertified countrysides of the poor nations’ drylands.

Table 8 which has not been published elsewhere and which is based on FAO computerized data on agricultural aid to every nation in the world provides a clear summary of desertification action in the early 1980s. The picture that emerges from it confirms the general impression. Country figures reveal isolated pockets of action where substantial amounts of agricultural aid have been directed towards specific desertification related goals for example, forestry in Cape Verde, rural development in Niger, anti-desertifica­tion planning in Somalia and Sudan.

But a few pockets of action do  not add  up to a major 80 international or even regional programme. If the figures are analysed in terms of larger geographical areas, these pockets of action become less significant and eventually disappear from the picture altogether. Thus figures for the Sudano­-Sahelian region reveal some differences from figures for Africa as a whole, and even from the global figures. But when the global figures are analysed over time (Table 7), it is clear that there has been no significant redirection of global agricultural aid towards desertification-related activities. The co-ordinated international offensive against desertifica­tion has simply never materialized, and donors have not responded in any significant way to the calls to devote more resources to desertification.

For example, Canadian international development aid to the Sudano-Sahelian countries rose from 4.6% of total CIDA aid to developing countries in 1978-79 (out of a total amount ofUS$53 million) to about 9.4% of the total amount ofUS$139 million in 1981-82. But out of this percentage less than 1% was directed to anti-desertification activities.

The Conditions of Aid

According to FAO: ‘Conditions of aid have been deteriorat­ing. Some bilateral programmes have hardened their loan terms, bringing them closer to commercial borrowing. There has also been an increase in the proportion of aid which is tied to the procurement of goods and services in the donor country, even if they can be obtained elsewhere at lower cost; this tends to lower the real value of aid.’

For instance, external concessional assistance to agricul­ture fell by 6.9% over 1980-81 while non-concessional assistance increased by 21.4%. This has undoubtedly made life harder for desertification-affected countries. The high interest rates which prevailed in the 1970s and 1980s had already forced developing countries into an impossible strait-jacket. What they needed to pay, to service their loans, quickly became exorbitant in terms of their declining incomes, which derived mostly from the sale of agricultural commodities of decreasing value. For the first time, in 1983 developing countries disbursed more to service their medium-long-term debts than they received. According to a recent report of the World Bank, the net outflows from these countries reached in 1983 the unprecedented level of $11 billion.’

The fact that conditions of aid are hardening has other implications for anti-desertification work. Bilateral donors now require more feedback from their aid in terms of contracts for home-based industries or employment of their nationals on projects in developing countries. Italy, for example, is currently entering the bilateral field in a big way, with specific contributions for food production and rural development in the Sahel. But it is also insisting on substantial sales of its own equipment as part of its packages, even though heavy mechanization may be inappropriate in desertified countries. And, in general, this means that donors shy away from desertification-related  projects, which rarely contain any significant kick-back for the donors; they concentrate rather on projects which may provide valuable returns for example, building and  road construction, and the setting up of industrial plants.

Similarly, agricultural aid is also often unwisely tied to cash crop production. For example, under the first 1975 EEC Lome Convention, 40% of rural development funds went to the development of crops for export.

 

The Research Gaps

One of the causes of desertification is the pressure to expand agriculture in marginal land, thus accelerating soil deteriora­tion. This pressure is particularly strong in Africa. The joint report by the Club du Sahel and CILSS stated: ‘ Rained cereals production…has continued to develop by extending cultivated areas (instead of by intensification of produc­tion).’

This development is frequently associated with the technological gap between Africa and other regions and the idea that desertification is an exclusively  African  pheno­menon. The hasty conclusion is that  the reason is that there is no Green Revolution in Africa and that there is no attempt by the international research community to focus attention on the crops that Africa grows. But the assertion is only partly false.

The expansion of agricultural land into marginal areas has occurred not only in Africa but also in Green Revolution territory, i. e. in Latin America and Asia where agriculture has generally followed during the last ten years a process of intensification. Thus the absence of a Green Revolution is not necessarily a reason for expansion into marginal areas. The problem should be examined from two different perspectives: first, the type of research and technological development and implementation likely to reduce the pressure of agriculture on desertification-prone areas; and secondly, the type of research and technological development necessary for the optimization of the biomass of arid and semi-arid areas and for desertification control.

The staple crops of African drylands are different from those of other continents. Millet and sorghum are the two main staples. Cassava, though low in protein, is popular as it is both drought resistant and requires relatively little labour. In the wetter regions a number of other  roots and tubers  (yams, etc.) play an important part in the diet (roots and tubers supply an estimated 20% of calories consumed  in Africa).

The World Bank pointed out that: ‘No major break­ through has been achieved in genetic improvement of rain­ fed millet and sorghum which account for 80% of the cultivated land in the Sahel and other areas of low rainfall.’ Instead, work has concentrated on the more common cereals – particularly rice. wheat and maize – which are consumed by more than 3.5 billion people in more than 80 different countries. Millet and sorghum are the staple foods of 13 countries with a combined population of only about 200 million people.

In this context a distinction must be drawn between lack of technological development and the low productivity of certain areas or crops because of poor application of new scientific development. International financial resources for scientific and technological development go to those areas and products most favoured by international markets. Thus some types of grain, especially wheat, are in high demand on world markets not only to feed the human but also the animal population. In 1975-77, 493 million tonnes of grain were used to feed cattle in developed countries whereas developing countries, with much larger human populations, consumed 538 million tonnes.

Sorghum and millet are typical cereals of semi-arid tropical areas. Africa’s share of world sorghum production is only 14% whereas its share of sorghum cultivated area is 3 2%. With the scientific and technological development of the past 20 years or more, yields of hybrid sorghum are now as high as 8 to 10 tonnes per hectare in the United States, while yields of ICRISAT African varieties range between 1,970 and 7,753 kilos per hectare. But the average yield in developing countries is still only 960 kg/ha and as low as 700 kg/ha in Sub-Saharan Africa.

This does not mean that massive distribution of genetically improved seeds is necessarily a solution. The problem is far more complex: adaptation to different environmental conditions and varying tastes also have to be taken into account. Each variety will have to contend with local pests, diseases, soil and moisture conditions, weeds, etc.

But the main incentive for technological innovation is that of quick returns in monetary terms. And as Dr. Swaminathan pointed out, ‘the terrible irony is that wherever  these crops are feed crops, they are improving, but in those places where they are food crops, they are stagnating’.

The chance to increase the production and incomes of millet and sorghum farmers has therefore been missed. Significantly, during the 1970s millet production fell in four, and sorghum production in five, of the largest producing nations south of the Sahara.

The same is true for roots and tubers, for which technological development has been slower, precisely because those staples are of no interest to donor  countries. But production could be considerably increased. The current yield of cassava is between 15 and 20 tonnes  per hectare, whereas the potential yield is put at 60 to 100 tonnes per hectare.

Technological progress does not just mean improving the seeds . The success of improved varieties depends on the improvement of such additional factors as  fertilisers, pesticides, water development, etc. For example, the importance of pest control in developing countries and specially in tropical countries can scarcely be over­ emphasized. In tropical areas crops are confronted with a larger variety of pests and diseases than in temperate zones. There are 54 rice diseases in temperate zones but 500 to 600 in the tropics; for maize, the figures are 85 and 125 and for beans 52 and 250- 280. The importance of Research and Development (R&D) in this field for tropical agriculture is therefore considerable.

Finally, the success of any technological development depends on many other factors such as training activities, adequate extension services and efficient marketing.

The second aspect mentioned is the development of science and technology for arid and semi-arid areas. In this case, very scant activities have been developed by the int er­ national community. A project proposed to UNEP in 1978 for developing technology to understand the physiology, and for use in management, of the biomass in arid and semi-arid areas was rejected as a zero priority. Curiously enough, a little later the University of Arizona initiated a project similar to the one proposed to UNEP. In this case the motivation was clear: the need to know more about a particular plant  inhabiting arid areas – the guayule – not for the development of these areas but rather because it offers an important substitute to hevea braciliensis in the production of natural rubber. Although guayule has been used for more than 70 years, almost nothing is known about its physiology, adaptation to different types of soil, require­ments for optimal production, etc.

Similar if not greater difficulties are encountered  in the  R&D oriented to understanding such unglamorous plants as the neem tree, tamarugo or even acacia. In more general terms, similar deficiencies are noted in the development of technology for desertification control. Even in the few stimulating projects undertaken it has been noted that ‘technologies pursued have not been specially innovative.

Institutes for R&D have been created in developing countries, among them IRRI, CIMMYT, the International Institute for Tropical Agriculture, etc. However, these centres follow the lines of agricultural research institutes in developed countries, and they are heavily  dependent  on them for staff training, basic scientific inputs, etc. Research methods copy those of  developed countries and, more importantly, priorities are defined not according to regional priorities but rather by the objectives and conditions  of donor countries.

‘Local people are not sufficiently qualified; governments will never spend the foreign exchange on purchasing the right equipment ; there are no spares, no transport and no petrol.’ This type of argument is used as a pretext to justify international attitudes concerning science and technology, a field over which a handful of developed countries are determined to keep control. Financial resources for science and technology development in developing countries are actually diminishing. Moreover, as the World Bank indicated:

Research financed by donors has produced some useful results, but the overall record is mixed. Although some competition in ideas and research strategy is healthy, excessive duplication of effort and conflicting focus   among donors have caused problems. By contrast, in the 1960s, successful research  in Africa mainly resulted from work promoted by or intended for users (maize research in Southern Africa, tea research in Kenya, for example).

The World Bank went on to argue:

As a result research outputs have not been adequately disseminated; local researchers are often underemployed because programmes lack focus, continuity and coherence; research management is weak and its status low, and  what  is more important is that researchers are isolated from  farmers and extension workers, so nobody can see the direct relevance of research.

This segregation between the research and productive sectors is one of the main problems in developing countries.

At the same time, it helps transnational corporations maintain control over the productive sector of developing countries and perpetuate the South’s technological dependence.

On the other hand, a recent report indicates that approximately one-third of all developing countries have virtually no agricultural R&D infrastructure or extension services. These countries are scattered throughout the Sahel, Sub-Saharan Africa, the Caribbean, Central America and the South Pacific. It is not just that they are below a critical threshold but they suffer from an almost total absence of any relevant infrastructure.

A great deal more effort is needed in this area, particularly if regional efforts are to be substantially strengthened and research expenditure doubled over the  1980-90 decade, as the World Bank recommends. As the Bank points out, ‘the adaptation of existing technology will not by  itself be enough, especially in the dry areas. Major research is needed on new crop varieties, techniques for soil moisture conservation, land use, livestock diseases and systems, and agroforestry.’ These are exactly the areas for which the UNCOD Plan of Action recommended regional research and development centres – centres which are still on the waiting list of UNCOD Projects.