NO. 7, September 1996



Romano Pantanali


Table of Contents


1. The Machobane Approach

2. The Farming Technology

3. Long-Term Technical Sustainability

4. Research

5. Financial Results

6. Food Security

7. Implications for Sector Strategy and Policy Formulation


This paper was prepared by the Investment Centre Division of the Food and Agriculture Organization of the United Nations (FAO), Rome, Italy, in the context of a mission to Lesotho carried out under the Cooperative Programme with the International Fund for Agricultural Development (IFAD), Rome, Italy. The views, findings, interpretations and conclusions expressed in this paper are entirely those of the author, and should not be attributed in any manner to FAO or IFAD. The designations employed and the presentation of material in this publication do not imply the expression of any opinion whatsoever on the part of FAO or IFAD concerning the legal status of any country, territory, city or area, or of its authorities, or concerning the delimitation of its frontiers or boundaries.



This note is based on information gathered by an FAO Investment Centre mission which visited Lesotho in July 1996 on behalf of the Government and the International Fund for Agricultural Development (IFAD), to prepare the ground for the formulation of a Sustainable Mountain Agriculture Programme for possible IFAD financing. It is expected that such programme would be part of the forthcoming Agricultural Sector Investment Programme (ASIP) for Lesotho. Information on the Machobane system has been provided by Ms. Letla Mosenene, Director of the Agro-forestry Research Component of the IFAD funded SWaCAP project, and Mr. J.J. Machobane himself. The present note complements a previous note 1 on farming risk and its impact on the adoptability of the improved cropping practices recommended by the Ministry of Agriculture (MOA), which was produced as a result of an FAO mission undertaken in November 1995 to the Districts of Thaba T'seka and Qacha's Nek. Observations of farmers attitudes and behaviour as well as socio-economic data have been collected by a Socio Economic and Production System Study (SEPSS) 2 carried out in February 1996 in several villages of the two Districts.

2. Stochastic efficiency analysis, based on 15-year time series of district crop yield statistics, suggested that, with the exception of wheat, improved technologies based on the use of high yielding crop varieties, chemical fertilizers and pesticides are not likely to be acceptable to farmers in the mountain areas of Lesotho. Wheat, however, is not a preferred crop from the consumers' point of view. Potato production looked promising in principle. However, as potatoes are a recently introduced crop in the mountains, no stochastic efficiency profile could be estimated for potato production. As a result of the expected impact of market deregulation, cropping activities in the mountain areas did not seem to be financially viable, nor public investment in promoting agricultural development economically justified. Yet, poverty alleviation and food security are the major objectives of Government of Lesotho (GOL) policy in agriculture, and the mountain areas of Lesotho contain a large proportion of poor people, whose food security is further threatened by the declining trend of remittances of migrant labour from the Republic of South Africa (RSA).

3. It was thus necessary to deepen the search for a valid technological basis of a sustainable mountain agricultural development programme. The Machobane system has begun to spread among mountain farmers, particularly in Thaba T'seka. The IFAD funded SWaCAP project, through its agro-forestry component, has acquired some knowledge about the system, working with Machobane farmers, mostly in the lowlands, and in the applied agricultural research station in Thaba T'seka. The purpose of this note is to summarize the information available on the system, provide a consistent framework to understand and explain, to the extent possible, its basic agronomic and economic features, and to compare its effects to those of traditional and recommended improved technologies. The conclusion of the analysis is that the Machobane technology does provide an answer to sustainable farming in the mountain areas of Lesotho. The inclusion of potatoes in the cropping pattern, a standard practice of Machobane, satisfies the requirement of people's food security within the constraints of local consumers' preferences, provided sufficient support is given to the development of the potato market, essentially by means of improving transport infrastructure. Finally, by securing a significant increase in land productivity and in farm income per unit of land used, the widespread application of the Machobane system would result in a significant reduction of the land area required to achieve food security for an average household, thus providing, in theory, scope for tackling the serious problem of the landless and remittance-less households in the mountain areas.

1. The Machobane Approach

4. Unlike most extension methods, the Machobane approach starts with the statement of the basic behavioural conditions required to adopt its technical message. These are, essentially:

(a) self reliance: farmers must be convinced that they can achieve food security without external assistance: it is their will that makes the difference;

(b) appreciation of the resource base: farmers must believe that they can improve crop production by fully exploiting their global resource base (land, labour, livestock, waste and by-products), turning all its potential to their advantage;

(c) readiness to do hard work: those who want to achieve independence of basic needs must be prepared to sweat for it;

(d) learning and teaching by doing: farmers must be trained in their own fields, and farmer trainers must be ready to work along with them;

(e) spontaneous technology spreading: farmers learn from other farmers: Machobane farmers have a duty to help their neighbours.

5. Some of these behavioural conditions are implicitly assumed by other extension methods, other ones are not. In other methods, however, they are seldom emphasised as a critical part of the extension message, or are simply not mentioned at all. Often, non-compliance with some of those conditions is presented as an explanation of possible lack of success. In most other extension methodologies, the economic incentive alone is expected to trigger interest in technological innovations, and the relationship "farmers-public extension service" is seen as a permanent requirement of farmers' production improvement, with the result that extension organisations tend to become an end in themselves. The Machobane system, instead, focuses on farmers' attitudes towards independence and self reliance: the goal is to establish a network of self-reliant people, supporting each other, and capable of a direct dialogue with those who develop and sell new technological solutions to their production problems.

6. Conventional extension methods have been presented as "ideologically neutral" methods, though in practice they are not. Mass communication/organization techniques have been refined to achieve "optimum" conveyance of exogenous messages. J.J. Machobane emphasises a solid ideological basis as the explicit starting point for the development of endogenous potentialities, in which human attitudes and natural resource development are interlocked and interdependent. The emphasis on the human factor (values/behaviour) has led technicians to view the Machobane system with some scepticism, as a sort of "religious" movement, possibly good in itself, but of limited general applicability. The technology, being very labour-intensive, has been judged, a priori, as a type of "gardening" of limited interest to most farmers, providing no solution to the major problems of Lesotho agriculture or at best "a good example of the ingenuity by which an intelligent farmer adapts his practices to the environmental challenge". Furthermore, Machobane's refusal to adopt "modern" inputs except as a last resort, has been viewed as a conservative approach to farming, de facto hampering modernisation. Yet, seen in the light of the most recent approaches to modern extension practice and theory, Machobane appears as a precursor of many features, both agronomic and promotional, now advocated as a result of critical reviews of the experience of half a century of agricultural development efforts in the LDCs.

7. J.J. Machobane (now 82 years old) first conceived his "system" over 40 years ago, when he developed an interest in Lesotho farming problems, questioned the approach of the Department of Agriculture, and decided that answers were not to come from foreign cultures. He then quit his job at the Church of Lesotho's book depot, burned all his school certificates, and started experimenting on his own land. Machobane worked on the technology for 13 years before attempting to launch it amongst his fellow farmers and persuading the GOL that this was the way to go. It is interesting to follow the official reactions to Machobane's proposals. His attempts at persuading the British Colonial Administration were encouraged: he received letters of appreciation and support by the highest authorities. After Independence, he was ignored by Government and donors alike: the emphasis was put on "tractors, chemical fertilisers, and the army of Village Extension Workers". Many years passed by until the German GTZ and IFAD decided that the Machobane system was something worth a trial. Nevertheless, neither IFAD (through SWaCAP) nor GTZ (through the Mafeteng RDP) tried to critically analyse the agronomic and economic characteristics of the system, or to systematically collect technical and economic data on the performance of Machobane farmers. Furthermore, the macro-economic and social implications of widespread introduction of the technology have been totally ignored in sector strategy and policy formulation.

8. A typical example of this attitude is to be found in the crop matrix appended to the aide-memoire of the last WB ASIP mission to Maseru, dated March 1996: a 10 hectare pilot project on the Machobane system is proposed. Yet, the number of registered Machobane farmers actually applying the Machobane system in Lesotho currently totals 2,008 3, equivalent to some 1,000 hectares of practical farming, some over several years by now. New enrolments by interested farmers have been between 350 and 400 per annum in recent years. This suggests there is nothing to be gained by undertaking a pilot project on a technology which is rapidly spreading without official recognition and significant support. Concurrently, in the same matrix it is also proposed to undertake soil analyses over 100,000 ha, with a view to launching a massive "soil liming" programme. Yet, the intensive localised application of ash and farm-yard manure (FYM) by Machobane farmers may well make expensive liming unnecessary.

9. J.J. Machobane has designed a training programme for farmers wanting to acquire the necessary attitudes and skills, and ready to accept his behavioural conditions, including the commitment to educate at least twenty other farmers. The curriculum is spread over 5 years. It involves farming one acre of their own land, with advice from the trainers, combined with short courses (essentially conversational) on the philosophy of the system and the theory underlying the farming method. At the end of the 5 years, Machobane issues a diploma to participants signifying the mastering of the system, while every year a certificate is issued in recognition of the individual's self-development.

10. Farmer selection is an essential feature of the Machobane system: only those farmers who convincingly show that they accept the philosophical approach summarised in paragraph 2 are enrolled, the principle being that a few keen people with a sense of responsibility towards their community are enormously more valuable than a large number of lukewarm fellow travellers, perhaps only interested in some sort of Government subsidy. The numbers will come, as a result of farmers learning from other farmers the advantages of a superior technology which offers large benefits over the traditional practice, and more benefits and less risk than improved practices based on exogenous models.

11. The mission met some Machobane farmers and was impressed with their attitude to experimenting with the use and mobilisation of their own resources, their capacity to discuss the agronomic problems they encounter, to compare the results of different procedures experimented with in the past, and with their pride in extending their knowledge to others. Farmers report three main advantages of the system: (i) intensification = much higher land productivity, (ii) potato intercropping = large cash revenue, and (iii) their fields are green even when non-Machobane fields are dry in case of drought.

12. Machobane has also begun to train GOL extension workers at Subject Matter Specialists (SMS) level, on the condition that they accept the principle of training by doing, i.e. sweating in the fields.

13. With his own financial resources, Machobane set up the Machobane Foundation, with the objective of spreading the philosophy and the farming method. An American volunteer organisation, VOCA, assisted in drawing up the Constitution. As he received no further support, either in the form of further grants or contracts, the Foundation is today hardly more than a legal document, a few dedicated people engaged in promoting the system under SWaCAP and the Mafeteng RDP and a few young graduates of the Lesotho Agricultural College, who currently help without pay.

2. The Farming Technology

14. There are many conventional views of Lesotho's agriculture which continue to be recalled a-critically. One of these is that there is a declining trend in land productivity. There is, however, little statistically significant evidence for this, over the last 15 years for which crop yield data are available 4. Another view is that excessive population pressure on the land prevents farmers from achieving household food security: the land scarcity allegation 5. Machobane never believed in that: he has strived to show how an average household (about 5 people) could achieve food security by farming 1 acre of land, 1/3 of the area conventionally thought necessary. A third widely-held view is that improved land productivity can only be brought about by using purchased inputs, chemicals, tractor services, and seeds of high yielding varieties. Connected to this is the view that agriculture and livestock are essentially two separate production systems. None of these views can be supported by a thoughtful and objective observation of the reality.

15. On the other hand, the official literature on Lesotho agriculture almost totally disregards the impact of crop yield fluctuations on farmers' income and decision making: years of drought are mentioned only as part of the background descriptive material in sector analyses. The integration of livestock in the farming system is seldom analysed. The issue of the landless, by no means marginal in terms of the number of households affected, is overridden by discussions on how to formalise the current land tenure system, which is not felt as a major problem by the rural people.

16. In Lesotho mountain areas, crops are grown on table land, permanently terraced wherever the slope of the terrain requires terracing, except for illegal cropping 6. Most households own livestock, sheep, goats, and cattle, the latter being essential to secure tillage of the land. When not grazing on summer "cattle posts", away from the village, livestock is kept at night in kraals, next to the farmers' houses. Kraals are the most important source of energy for the rural people, who burn dung briquettes for cooking, lighting, heating, and fumigation. Not all animal dung is turned into briquettes: the "powder dung" left at the bottom of the kraal, after making briquettes, is an important source of organic manure. Moreover, all households burn dung briquettes, and, to a lesser extent, wood and brush, and in the process produce ash.

17. The starting points for the technical approach of Machobane are that Lesotho land productivity is low and yields are subject to wide fluctuations due to extensive farming methods, inadequate nutrients in the soil to support better yields, and poor soil structure resulting in poor moisture retention capacity. These agronomic constraints can be overcome by a rational exploitation of the full resource basis available to rural families, reducing the need for purchased inputs, and thereby farmers' financial risk, to the minimum. The technical paradigms upon which the system is based are:

(a) farmers' resource base (land, livestock, labour, household by-products) in Lesotho can be used to establish intensive inter-cropping by growing several crops simultaneously or in relay in the same field, provided:

(b) adequate soil fertility and moisture retention capacity is achieved though localised placement of ash and manure, combined with adequate weeding;

(c) the household "waste" produced by a typical family (mostly ash, up to 2 tons per annum) and FYM (1 - 2 tons, after allowing for briquette making) is sufficient for 1 acre of land using localised placement techniques;

(d) intensive cropping of one acre of land is sufficient to grow enough, for home consumption and sale, to ensure sufficient food for the family (grown or purchased with the earnings of surplus crops, sold so as to meet consumers' preferences);

(e) continuous intensive cropping, with appropriate incorporation of organic matter and ash in the soil, offers a further, synergistic, means of enhancing soil and moisture conservation;

(f) multiple cropping will substantially reduce farm income fluctuations, through a combination of lowering yield fluctuations of individual crops, and spreading risk of fluctuations in yields and prices by planting a range of crops simultaneously;

(g) pest control is best carried out by traditional methods, including local concoctions, local practices, winter ploughing, and understanding of pest biology: this would protect land productivity, avoid the cost of chemical applications, preserve natural predators, and "prevent the risk of worse pests developing as a result of using chemical pesticides".

18. It is unfortunate that data have not yet been systematically collected to measure the impact of the Machobane system in terms of production, gross income and net returns to labour, compared with traditional cropping methods and with recommended "modern" improved technology practices based on monocropping, mechanical power, chemical fertilisers, and pesticides. An attempt at constructing some models based on the limited information available is presented in the attached tables.

19. From an agronomic point of view it is also unfortunate that no data have been collected to relate farm output to the basic characteristics of the soil where it has been obtained. No information is available regarding plant nutrients in the soil (N, K, soluble P2O5, calcium), nutrients added by way of ash and manure, the overall balance between plant nutrient requirements, crop yield and nutrient availability, rainfall, frost, soil moisture retention capacity, and the impact of organic manuring on this and on soil pH.

20. The number of plants recommended by Machobane is compared to those recommended for monocropping in the attached tables M.1 A and M.1 B, for the lowlands and highlands of Lesotho, respectively. Essentially, the Machobane system of planting involves leaving a space between the rows of all crops, except for wheat and pumpkins 7, of about 2 metres, twice as much as in monocropping, with the plant spacing along the rows unchanged. In the lowlands several crops (maize, Pinto beans, sorghum, squash, peas) are planted in relay, at different times of the year, along the same row in which the soil has been enriched by ash and manure. Between one row of cereal and pulses and another, a row of potatoes is also planted. Figure 1 illustrates the arrangement.

21. As a result, half the plant population used for each crop in monocropping is sown, but the total plant population is aggregated: this means that the plant population of one acre of Machobane adds together the monocropping equivalent of 1/2 acre of maize, 1/2 acre of potatoes, 1/2 acre of sorghum, 1/2 acre of beans and 1/2 acre of peas. Machobane technology claims that two rows of wheat, grown on residual moisture, can be also added, one to each side of the intercropping row (spaced 15 cm from the main row). This would add the equivalent of about 15% of the plant population on a monocropped acre of wheat to the total. However, Machobane farmers apparently do not to grow wheat: wheat is not a preferred crop in Lesotho anyway, despite the agronomic advantages it offers. At higher altitudes in the mountains wheat is seldom grown.

22. The number of crops that can be planted in practice depends on local conditions: in the lowlands, continuous year-round cropping is possible, and actually set as an objective for soil and moisture conservation purposes. In the highlands, only one cropping season is possible: maize, beans and potatoes are planted simultaneously at higher altitudes. From year to year a rotation is practiced between rows within the same field: the cereals and pulses are planted in the row where potatoes were grown the previous year and vice versa.

23. The sequence of cultural operations is illustrated in Figure 2 for the lowland and highland areas. The probability of a minimum rainfall of 30 mm/month is also shown in the figure. Recommended practices involve several weedings and hand cultivations along the rows, with the result that crops are actually grown in a sort of a continuous "pot", filled with a good mixture of soil, ash and manure (which has suggested the image of "gardening" for the Machobane system), the moisture retention capacity of which is much higher than in the neighbouring soils, and may go a long way to explaining the fact that "Machobane fields are green when other fields are dry" in poor rainfall years, as reported earlier.

24. As Figure 1 and 2 show, continuous all-year-round cropping is practised in the lowlands, with very little time during which the land is not occupied by one or several crops, or being tilled. In the highlands, however, the land is not used during June, July and August, the most severe winter months during which the fields are snow-covered at times. Winter ploughing is recommended by Machobane, with a view to breaking some important pests' biological cycle. The practice, however, is seldom followed: oxen are too weak in winter, due to lack of any stored fodder. Growing a fodder crop seems the obvious solution, now impeded by the inadequate arable land area available when land productivity is constrained by farming with traditional technologies.

25. Experimentation with fodder crops in the mountain areas of Lesotho has been very limited. Successful introduction of a high nitrogen fixing leguminous fodder crop would be clearly of considerable interest, not only to supply much-needed winter feed for the animals, but also with a view to adding nitrogen to the soil. The SWaCAP Agroforestry component has succeeded in growing one crop of red clover in the adaptive research centre at Thaba T'seka, obtaining a "good" crop. The red clover, however, did not survive the severe winter and must be replanted 8. Further experimentation is greatly needed: an ideal solution would involve rotating the Machobane fields with a leguminous fodder crop, or a combination of fodder and peas, a common crop in the mountain areas, but which does not fit the Machobane system at high altitudes, and must be grown in separate fields.

26. A major analytical issue is what yields the cropping intensity of the Machobane system can sustain, given the recommended application of ash and manure, compared to those obtained under monocropping on the equivalent land area and climatic conditions.

27. No data have been collected regarding the plant nutrient value of a ton of manure and a ton of ash in Lesotho. All one can say is that, at the Machobane recommended rate of 1:1 for potatoes, beans, sorghum, and maize (combined), the localized combined application of 4 tons of ash and manure per acre would be roughly equivalent to at least twice as much applied on the open field (almost 20 t on a per ha basis), which is no small amount, and ought to explain the sustainability of the cropping intensity 9. More should be known, however, about the nutrients balance, particularly with regard to N, possibly P2O5, and the effect on the soil pH.

28. Another major analytical issue concerns the impact of the "system" on the soil's capacity for retaining moisture and making it available for plant growth. Here again no data have been collected to explain why "Machobane fields are green when other fields are dry in years of poor rainfall". Yet, as discussed at length in the paper dealing with the stochastic efficiency profiles of cropping with different technologies in Lesotho, reducing crop yield fluctuations is the major challenge in improving farming in the country.

29. The Machobane system does not envisage any particular technique to convey moisture into the cultivated rows: explanations of what happens have to be confined to guessing, until a research effort is applied to collect information systematically and to provide adequate answers. One feature is of course evident; namely: that the soil structure along the crop rows, as a result of continuous cultivation and of massive addition of ash and organic manure, is very different of that between the rows, and that difference makes for much higher moisture retention and infiltration capacity. Fields, whether on table land or on terraces, are never perfectly horizontal. Presumably, this causes some of the rain falling between the rows to drain into the rows where the infiltration is much more, with a moisture accumulation effect.

30. The cumulative effect of added organic plant nutrients and improved moisture retention is reported as impressive. By way of illustration, the information provided by one Machobane farmer of Maseru District, who actually measured the production of maize, sorghum and beans obtained from one acre of land in 1995 and 1996 was as follows:

31. The Machobane yields to be compared with the District monocropping average yields are the total of the 3 crops (it is likely that other crops should be added, such as pumpkins, whose production was not recorded). 1995 was a very dry year, and 1996 an unusually wet one. Notice that the total yield fluctuation in Machobane (840 kg in 1995 vs 1,050 kg in 1996) was limited to about 25%. Comparable figures for the District averages in 1995 and 1996 are not yet available; however, the 15-year time series of District yields data show a gap between minimum and maximum yields of about 1:3 or even 4, depending on the crop. The Machobane system production compares with 15-year average yields of monocropping in the District of 344 kg/acre for maize, or 334 kg for sorghum, or 195 kg for beans, suggesting that land productivity obtained by this particular farmer, in terms of total grain equivalent produced per acre, was almost 3 times the District average for sorghum or maize, and that the yields of each crop grown under Machobane are about as much as in traditional monocropping, despite the much lower plant population.

32. Two year results from one single farmer do not lend themselves to drawing general conclusions. The fact that yields in the Machobane system are significantly better than corresponding yields in monocropping is, however, confirmed by observations about the size of the maize cob harvested compared with those harvested by the best monocropping farmers participating in Agricultural Fairs, and by the richness and multiplicity of sorghum heads observed in Machobane fields (only one head is normally obtained). In tables M.1 A and M.1 B, therefore, as a preliminary assumption, a conservative coefficient of 1.7 has been used to estimate the impact of the Machobane system on yields compared with the corresponding traditional monocropping: the adjustment is undoubtedly rough, and additional more reliable information ought to be collected to refine this aspect of the analysis.

33. The following analysis deals with a highland, high altitude, scenario, where sorghum is not grown. At lower altitudes, the Machobane system would offer the advantage of growing yet one additional crop.

34. By way of illustration, in table a, the total estimated production of grain equivalent by Machobane system is compared to the expected production under traditional and "low input" improved monocropping technologies. Traditional technology involves land preparation (ploughing and harrowing) by oxen, sowing in line, often with the help of an oxen drawn drill, conventional seeds, some weeding, using sometimes small doses of pesticides. Improved "low input' technologies as recommended by GOL normally add quality seeds, hybrid seeds in the case of maize (composite maize varieties seeds are also available), chemical fertilizers, and heavier doses of pesticides. As indicted earlier (paragraph 2) most improved technologies are too risky for farmers to adopt under conditions of serious uncertainty about crop yields, and hence several projects for their promotion have failed to achieve their objectives.

35. In table a the total output of the Machobane system is compared with the individual crop yields indicated for the monocropping system. In all cases Machobane technology production is several times higher than that obtainable with monocropping, with the exception of potatoes. This general conclusion would not change when the assumption of a yield increase factor of 1.7 is dropped 10. For the sake of simplicity, average yields are used for this comparison, although available data on crop yields suggests that no statistically significant average can be calculated, because of the range of inter-annual fluctuations.

36. An important theme of the Machobane system relates to Integrated Pest Management (IPM). In this field, however, much work needs to be done to respond to farmers' declared needs and to their interest in Machobane recommendations, by applying a more systematic approach to monitoring the pests' biological cycles, identifying predators and understanding their role, as well as in understanding the effectiveness of some of the local practices. This is an area where much needs to be done with the full support of a re-oriented research effort.

37. Except for potato production, the Machobane system is much more labour intensive than traditional or "low input" improved technologies. The labour requirements have been estimated in Table 7 B, and plotted along the operations calendar chart in Figure 2. The maximum monthly requirement of about 14 labour/days per acre seems to be well within the reach of the average Basotho household in the mountain areas. Nevertheless, the need to hire some help during periods of peak demand for labour is reported by some Machobane farmers.

3. Long-Term Technical Sustainability

38. The Machobane technology is a farming systems approach: it pivots around the integration of cropping and livestock rearing activities, and requires a good understanding of land and crop management requirements. Its widespread application raises the issue of the availability of sufficient organic manure of animal origin in the long term. The sustainability problem may differ between the lowlands and in the mountain areas of Lesotho.

39. In the mountains, large areas of pasture land are available, albeit some in serious state of degradation due to overgrazing 11. On average, a household would own some 30 to 50 small ruminants (sheep and goats, reared for the wool and mohair and occasional slaughtering for meet consumption), and 3 to 6 cattle (essentially kept for animal traction and reproduction). Well over half of the dung collected in the kraals is consumed as fuel. To sustain a large scale application of the Machobane system, more dung should be produced and collected, and less dung should be burned.

40. More dung production (and collection) does not required more animals, it requires healthier animals, better nourished in winter when they are kept in the kraals. Short of purchasing supplementary feed (an expensive solution, at the moment out of reach of the local people), better winter feeding can be provided by growing a fodder crop: space for that would be provided by the higher land productivity obtained as a result of applying the Machobane farming method.

41. Burning less dung requires substituting an alternative source of energy for dung briquettes. Coal imported from the RSA - the cheapest alternative - is still much too costly a solution. Fuelwood can be grown, however, and Machobane farmers in the mountains are aware that this has become an imperative for them to survive. Burning planted wood - a renewable source of energy - would produce the required amount of ash, while increasing the amount of animal manure available for use in the fields.

42. Thus the long-term technical sustainability of the Machobane technology is a distinct possibility in the mountain areas: it will require, however, the close integration of all activities in the farming system: crops, livestock, and forestry.

4. Research

43. In future, the development of the Machobane system ought to be supported by a minimum of research activities. These fall broadly in two main categories:

44. With respect to the first category, a representative sample of Machobane fields should be selected in each ecological area of the country, the required soil tests and analyses undertaken in each field, the plant nutrient content of added ash and manure established, farmers practices recorded, and crop productivity measured. Correlation of basic field agronomic features, cropping practices, and detailed meteorological data from the nearest station would be studied. The monitoring programme should be for a minimum of 5 years. Economic data (costs and benefits) would also be collected.

45. With respect to the second category, a preliminary list of research themes would include:

(a) integrated pest management: this would be the first priority starting with the collection of information and the study of current traditional practices, going on to the classification of important pests and diseases, the farmer participative observation and study of their biology, the identification of important predators, etc., and ending up with specific recommendations as to improved practices, including minimum pesticide dosages;

(b) fodder crops, with a view to identifying the best high nitrogen-fixing legumes suitable for the local conditions;

(c) reduced tillage: the cost of land tillage would be reduced by limiting land ploughing to the crop-growing rows only: the impact on production by comparison with the standard Machobane practice should be monitored;

(d) mulching of inter-row spaces with weeds after weeding, on regular Machobane fields and in combination with reduced-tillage experimentation: a possible increase in the moisture retained in the cropping rows should be ascertained;

(e) high altitude maize varieties, with emphasis on non-hybrid varieties.

46. In keeping with Machobane's philosophy, all research activities must be undertaken with the active participation of farmers, and as much as possible on their own fields. The above list obviously represents only a preliminary suggestion, which needs to be completed with the contribution of specialists and farmers alike: Machobane farmers, in particular, would have the advantage of the experience and knowledge necessary to suggest specific research themes which may help solving the problems they meet with in the application of the system.

5. Financial Results

47. A major problem encountered in undertaking the financial evaluation of Machobane versus other technologies in the mountain areas of Lesotho is which crop prices to use. The MOA collects price information from the major shops in the main cities and district headquarters, which refer to processed and packed products at retail level, and are of marginal relevance at the farm gate. The information collected from farmers during the SEPSS reflected essentially four major specific factors: (i) a situation of severe local scarcity: 1995 was a very dry year, (ii) the isolation of the mountain villages where the SEPSS was conducted, (iii) the GOL price/import control policy was still being enforced and (iv) with respect to potatoes, the novelty of the crop. As a result, for example, the price of maize was reported by the SEPSS as M 1/kg, twice as much as the likely equilibrium price after the import liberalization enforced in mid-1996; the price of sorghum at M 0.85/kg, also an extremely high price. The price of potatoes in the lowlands was reported in July 1996 as double price of maize, the latter having dropped to about M 0.5/kg, after deregulation.

48. Similar problems are encountered with respect to the cost of labour. The wage paid by the Highland Water Development Fund for unskilled labour employed in construction work, M 17/day, is universally recognised as much to high. In the lowlands, Machobane farmers report a wage of M 10/day, often paid on an hourly basis. For the purpose of the analysis, a nominal rate of M 7/day has been used, although the SEPSS reports much lower remuneration, of the order of 3 to 4 M/day, in the most isolated areas.

49. The prices used for the analysis 12 are intended to reflect the long term equilibrium prices expected when the full impact of the deregulation policy will be felt throughout the economy, account being taken of the extra costs of distribution in and out of the mountain area.

50. In table b, the gross value of production, the net farm income, and the returns to labour obtainable from one acre of land under Machobane, under monocropping of different crops with traditional technologies, and monocropping of potatoes with "low input" improved technology are compared. Returns to labour are defined as the algebraic sum of the imputed cost of labour (valued at M 7/day) and the net farm income. Only the monocropping of potatoes results in a higher value of production: however, returns per labour day of monocropped potatoes are only slightly higher than for Machobane plots: in terms of returns to the family, they may well be less, since an average household may have to hire a fair amount of help to handle one acre of potatoes cropping at the peak of labour input requirement.

51. It is interesting to notice that, in the case of the marginal crops grown at high altitudes, the gross returns per labour day provided by the estimates are not out of line with the reported wage rates paid in those areas (see paragraph 47, above). The figure for beans reflects the very high price of the crop and the limited labour requirement. The returns per day of labour estimated for Machobane, all crops combined, however, is twice as high as that for beans, and several times higher than that for the other crops, except for monocropped potatoes.

6. Food Security

52. By way of illustration, table c shows a hypothetical food balance model based on cropping one acre of land at high altitude with the Machobane technology. The illustration assumes that all the potato crop is sold and the income (net of all production costs, except for labour) is used to buy maize meal13. Calorie intake requirements are assumed to be 2,000 per person per day, and the family composed of 5 people including small children.

53. The model of table c shows that, in order to achieve total food security at high altitudes, a typical family ought to farm about one and a half acres of land with Machobane technology, plus about one acre of peas and do some vegetable gardening, in addition to consuming most of the livestock products, except of course for the wool and mohair. Two and a half acres represent about half the area necessary to feed a family with traditional technologies. At lower altitudes, with sorghum also grown, and perhaps some wheat, less land would be needed. In the lowland, crop yields are higher, and possibly the thesis that one acre is enough could be confirmed in theory.

54. There are, of course, many other technical solutions to the food security problem of the average Basotho family in the mountain areas. For example: the returns to labour from selling the entire output of half an acre of potato monocropping (about M 2,900) would cover almost 100% of the calorie requirements in the form of purchased maize meal. The Machobane solution, however, would be preferable because, in addition to offering several agronomic advantages, and a more diversified production, would require fewer purchased inputs 14. It can be shown that adequate food supplies could be obtained also by farming about 3 acres under different crops by using improved "low input" technologies, but at the risk of incurring serious financial losses in years of poor rainfall; or by farming 5 acres using traditional technologies. In all the latter solutions, the risk of not meeting the requirements in case of drought is likely to be much higher than with the Machobane system.


7. Implications for Sector Strategy and Policy Formulation

55. Several important implications for the formulation of a strategy for Sustainable Mountain Agriculture (and the related formulation of specific policies) result from the possibility of a widespread application of the Machobane system: some are complementary to, others are in contrast with, the approach taken so far in elaborating the ASIP for the country.

56. The combination of population increase and of reduced remittances from the RSA is singled out as a major threat to the survival strategies of the Basothos, and a cause for great concern. Like all coins, these have also another side: they provide a much-needed incentive to intensify crop production, and to do so at the lowest possible cost in terms of purchased inputs. Fewer remittances imply more labour available in the family, higher incentives to substitute labour for capital, and an economic value attached to useful "waste" which up to now has been accorded either no importance or a nuisance value. These are the conditions which have led to agricultural intensification and to irrigation development in other countries. However, scope for irrigation development is very limited the mountain areas of Lesotho, and in most of the lowlands as well. It is not the avenue for solving the survival problems of the people of the mountains: at best irrigation can provide some supplementary income and help improve the basic diet of some households. On the other hand, the integration of livestock rearing in the farming system, a common feature of the country's agriculture so poorly appreciated in the current sector policy literature, does provide an essential ingredient for increasing land productivity by a large number of rural Basotho households. This situation explains farmers' interest in the Machobane system.

57. Another positive factor is the successful introduction of potatoes in the mountain cropping patterns. Past emphasis on wheat is hard to understand in the face of very marked consumer preferences in favour of sorghum and maize, despite the agronomic advantages of wheat over maize and sorghum. Potatoes are much more suitable mountain crop than wheat, have a ready market, and can provide all the cash required for a family to buy the preferred maize meal by selling the potatoes grown over a fraction of the land required to produce the required amount of maize 15. A serious effort to multiply seed potatoes in the mountain areas of Lesotho has only recently been initiated by the Rural Development Programme of The Lesotho Highland Water Authority (LHWA), albeit based on high input use technologies, which may not be well adapted to small farmers conditions 16.

58. In theory, the combination of the Machobane system, rotation with a leguminous fodder crop/pure stand peas crop, potatoes production (as part of the Machobane system or in monocropping), marketing of potatoes in exchange for maize meal 17, and a limited amount of fuelwood planting, can go a long way to easing the food and fuel security problem of the mountain people, and increase animal productivity as well, while making use of only 2/3 of the area currently deemed necessary for achieving food security. Fuelwood production is important to increase the organic manure available for the crops; fodder crops are a means to strengthen animals in winter, reduce mortality, and provide a necessary condition for effectively de-stocking the overburdened ranges. The positive effects on the environment do not need to be emphasized. The possibility of realising such a vision in practice over a relatively limited period of time is perfectly legitimate, provided there is the strongest back up for the approach from GOL and donors organisations alike.

59. The considerations of paragraph 57 suggest that there might be scope for attempting to alleviate the problems of the landless (and remittance-less) people in the mountain areas: in theory, an average of something like one third of the area currently under cultivation could be "saved" by those who make use of it now, and re-distributed to the poorest people in the area with demonstrable willingness to work for independence in food supplies. Something like one third of the rural families in the mountain areas seems to fall in the category of "landless and remittance-less" people, and possibly half of those may qualify for re-distribution.

60. Land re-distribution is of course an extremely complex subject: an effective, though non-transparent, flexible mechanism for it existed in the traditional functions of the local chiefs, and in the principles of customary law to be applied by them in land allocation. The chiefs' power has now been vested in the recently elected VDCs which, however, may disappear as a result of the proposals of the recent White Paper on Local Government 18. This may jeopardise the possibility of linking technological progress with more equitable distribution of income and resources.

61. The emphasis on formalising the present land tenure system in current ASIP documents is not based on any technical appreciation of the potential of the land (a dynamic view of the interrelationships between technology development, economic forces, and the land "carrying capacity" is missing), nor on any future vision/plan for its rational exploitation. Despite the general plea for poverty alleviation measures, no assessment of the likely implications for income distribution has been attempted. In fact, any system to issue titles on currently allocated land will crystallise existing claims, and, eventually generate, in the expectation of those who propose the system, the growth of a land market, which will certainly not facilitate re-distribution in favour of the poor.

62. There is of course need for encouraging private investment in land improvement. An instrument for that already exists in Lesotho: the Form C title issued to those who invest in land reclamation. This instrument can be expanded in scope to make it respond to the legitimate goal of formally securing tenure to those who invest heavily on land, including not only reclamation works, but certainly other types of fixed capital formation, such as irrigation facilities and planting trees as well.

63. Another inconsistent feature is the emphasis put on "diversification and export promotion". Although the policy papers have refrained from explaining what this means in concrete terms (which crops, which markets?), the policy statements are generating many (costly) imaginative efforts to launch ideas about "new" products (even more costly to promote, when finally identified) with potential markets which are more in the realm of fiction than in that of the real world 19. Given the pattern of population settlement and the state of transport infrastructure, efforts in these directions, while laudable per se, are likely to show only marginal results for a long time. If there are potentialities, the private sector is much better suited to identify and develop them. Public attention ought to be concentrated essentially on providing the minimum of infrastructure required for the marketing of potatoes to realise the considerable production potential of this crop from the mountain areas. Other possibilities, if they exist, will eventually follow as a result of private initiative.

64. A third point which bears on the choice between the Machobane system and alternative technologies concerns the availability of human resources to implement development programmes. ASIP policy, with its emphasis on deregulation and divestiture, leading to less responsibility to be retained by Government, is obviously on the right track. The issue is whether, even after deregulation, divestiture, and a reasonable degree of decentralisation, the investment programme which will ultimately emerge will be sufficiently prioritised, so that the limited managerial and technical resources are not diluted over too many different programmes and effective implementation will be secured.

65. A word of caution ought to be spent as well with regard to the dangers inherent in attempting to push the diffusion of the Machobane system too far and too fast, with the result of altering its basic motivational and socio-cultural features. Donors, prompted by their own "requirement" to lend large amounts of money, could capture the technology, add a number of exogenously determined artificial "embellishments" to justify their versions of it to their Review Committees/Boards of Directors, and quickly transform the Machobane system in yet another extension programme of the type with which foreign decision makers are normally conversant. Such catastrophic development ought to be carefully avoided by Donors, and strongly resisted by a responsible GOL. Large sums of money can easily be spent by Donors to develop the transport infrastructure of the rural areas of Lesotho, an essential pre-requisite of any production and social development. The diffusion of the Machobane system must be supported, cajoled, nursed, help to develop along its present configuration, not overwhelmed with exogenous costly features which do not belong to its culture.

66. The limited implementation capacity of the GOL calls for concentrating on few attainable objectives: for the mountain areas, effective support to spreading the Machobane system is of paramount importance. This should be done by sub-contracting the establishment of a strong Machobane Farmers' Network to the Machobane Foundation, including measures to strengthen the Foundation itself, and should be accompanied by an all-out effort to spread seed potato production. An effective link between the Machobane Farmer Network and the applied research stations, based on adaptive research programmes, decided upon and implemented jointly with the farmers, would be an essential support measure. At the same time, encouragement of fuelwood seedlings production, and of small village/homestead woodlot planting, could lead to important results, particularly with a view to encouraging more FYM application in farming. All that would go a long way to help Basotho mountain farmers to help themselves, and to prevent further deterioration of the environment.

67. At the moment, more ambitious programmes, and the planning of many more activities of the MOA, may simply dilute efforts and put potentially attainable results in jeopardy20. Concurrently, however, an important public works programme to improve and develop the transport infrastructure of the mountain areas is required, for which finance is available through the Lesotho Highland Water Development Fund (LHWDF), and implementation capacity must be enhanced outside the MOA, and separately of ASIP.

Table M.1 A: Investment in plants and crop yields: Comparison of Machobane system and Monocropping

Table M.2 A: Goss Value of Production: Comparison of Machobane system and Monocropping

Table M.4 B-a: Net Value of Crop Production: Comparison of Machobane system and Monocropping

Table M.4 B-b: Gross Returns to Labour: Comparison of Machobane system and Monocropping

Table M.4 B-a: Cost of Crop Production (M/acre): Comparison of Machobane system and Monocropping

1 Assessing the Likelihood of Adoption of Improved Crop Production Technologies by Small Farmers: the Case of the Mountain Areas in Lesotho, TCI Occasional Paper Series no.6. The methodology used in both notes has been elaborated by Romano Pantanali, Senior Advisor, Economics, FAO Investment Centre, with the assistance and advice of Alice Carloni, rural sociologist of the FAO Investment Centre, who lead the two FAO missions to Lesotho. Professional review by members of the Senior Advisors Group of the FAO Investment Centre provided useful comments and suggestions for improvement of earlier drafts of both papers.

2 The study was undertaken by a team which included GOL officials from the District Agricultural Offices and staff from Sechaba Consultants, a Lesotho firm. The team operated under the leadership Sechaba's partner, Thuso Green. Alice Carloni designed the SEPSS, joined in the participative discussion of the SEPSS findings in Lesotho, and supervised closely the elaboration of data and the final SEPSS text.

3 This figure include only those officially recorded by the SWaCAP project of IFAD: the total number may be higher.

4 This point is discussed in TCI Occasional Paper Series no 6. It is interesting to recall that, although a linear regression of total crop production statistics from 1977 to 1994 shows an apparent downward trend (by about -2% per annum) the regression coefficient is not statistically significant by a large margin. The series is actually included between a MIN and a Max value in the ratio of 1:4.3. A similarly non-significant result is obtained by running a linear regression on 3-year moving averages of the same series. In all probability, adding the data for 1995, a drought year, and 1996, an exceptional bumper crop year, would modify the regression coefficient to a large extent. Data on area sown show a slightly raising trend, but the regression coefficient of this trend is also not statistically significant.

5 Despite the uncertainty about the trends on production, land productivity, and area cropped, it is of course evident that cropped area and crop production per capita have sharply declined over the last 15 years.

6 Marijuana is the only crop grown on non-terraced land tilling slopes up to 100% gradient, normally far away from the villages.

7 Pumpkins are not included in the presentation, although they are grown on Machobane fields.

8 The very high cost of seeds is reported as a possible obstacle to replanting.

9 10 t per ha of FYM would add approximately: 50kg N, 30 Kg P2O5, 60 kg K2O; 10 t per ha ash would add 40-200 kg P2O5, 50-350 kg K2O. The impression is that the recommended soil correction is short of nitrogen for crops such a maize and sorghum, but Machobane farmers who have tried report that adding chemical N has made little difference to the yield of these crops and have discontinued the extra application.

10 If a coefficient of 1, instead of 1.7 is applied, the Machobane system is still very considerably superior.

11 In the medium term, the dung produced in excess of that used for briquette making, and all the ash, is not used, so that the scope for spreading the Machobane system is considerable. In the longer term, the improvement of pastures on the ranges calls for some de-stocking, an objective which is pursued by the GOL through the Range Management Areas, and the Grazing Associations promotion policy. Short of an important fodder production development, de-stocking of the high ranges will affect the number of animals much more in the lowland and in the lower side of the mountains valleys, than at the higher altitudes. The critical factor of dung production in the mountains, however, will be the conditions of the remaining animals during the time when they are kept in the kraals.

12 See table entitled: BASIC PARAMETERS, attached.

13 This reflects farmers' observed behaviour, as reported by the SEPSS.

14 See table M.4 B-b, attached.

15 The farmgate price of potatoes is currently twice as high as that of maize, output twenty times higher; even with a long term price only slightly higher than maize and taking into account production costs, potatoes offer a very large advantage over maize, and a distinct possibility to achieve food self sufficiency and meet consumers' preferences through trade.

16 In the lowland, the technological package recommended by official technicians for potatoes is also high technology, based on tractor services, and large doses of purchased fertilisers and pesticides.

17 Farmers will continue to grow some maize as the crop has other uses besides the meal.

18 The White Paper proposes the strengthening of the District Development Councils (DDC) and the abolition of the Village Development Councils (VDC). Below the DDCs, subordinate councils would be nominated upon request by a number of villages with a combined population of at least 1,500 people, several times more than that of a typical mountain village. It is not clear which institution will eventually be vested with land allocation powers. It is clear, however, that the DDC and the subordinate councils as conceived at the moment, are much too remote from the village communities to effectively operate any land allocation policy, except for the formalization of existing claims.

19 Dry flowers, cut flowers, fresh fruits and vegetables, artisanal products of all kind, etc.

20 The above statement does no mean that GOL activities in the livestock sector, particularly support to spontaneous grazing associations (GA), and related Range Management Areas (RMA), and support to development of animal health services, as much as possible on a private basis, should not be part of the ASIP. Farmers' training in artisanal skills of various types would also be important, provided this is handled in participative, low cost basis.