C.I.A. Beale
Formerly Livestock Economist, UNDP/FAO Project ML W/75/020 "Assistance to Livestock Development"
Leucaena leucocephala (Lem) de Wit is a leguminous shrub or small tree which originated in Central America and is now widely distributed in the hot humid tropics. The plant has a variety of uses as a source, for example, of timber, firewood, dyes, ornamentation of parks and gardens, etc. In agriculture, it is commonly used as a source of protein-rich feed for browsing or for cutting and feeding to livestock in pens and yards.
Leucaena's chief disadvantage in livestock feeding is the presence of a free amino acid, mimosine, which when fed in excessive quantities causes lack of thrift, reproductive problems and other side-effects. With due care however, these effects can readily be avoided. Moreover, new low-mimosine cultivars are expected to be released by plant breeders in Australia in the near future.
Leucaena has been grown in Malawi for ten years. Evaluation at a number of sites has shown it to be very productive and best adapted to altitudes of less than 1,000 m above msl. The FAO project began agronomy studies in the early 1970's at Mpemba Veterinary Station near Blantyre. Meanwhile, Leucaena growing was promoted under an extension programme as a protein concentrate for dairy cows. The leaves were harvested and dried in the wet season. The pinnules and other tender parts were stripped and bagged, and the dry material was fed to cows towards the end of the dry season.
It was soon realized that the protein concentrate was marketable and enquires were made regarding potential outlets. Chemical assays showed that Leucaena contained exceptionally high levels of carotenoids, including the xanthophyll group which provides the pigments giving a bright yellow or orange colour to egg-yolks, and beta-carotene, the precursor of vitamin A. The finer parts of the leaf were also low in fibre compared to the commercially available green meals such as Lucerne. It was therefore apparent that Leucaena was also marketable as a source of carotenoids for inclusion in poultry layers' rations.
In 1974, a pre-investment study of the commercial prospects for Leucaena-growing on an industrial scale was proposed and approved by the project's sponsors. The area chosen for the study was the Lower Shire Valley, where the altitude for the most part was less than 200 m above msl.
The agronomy trials at Mpemba were continued and expanded, and further trials were laid down at Zunde and Ngabu in the Valley1. From 1975/76 onwards, farm production trials were carried out. Some were made in cooperation with smallholders, and some by the project using its own labour force. During the study a few estates, both privately and publicly owned, became interested in the work and established small trial areas of Leucaena.
1 The agronomy trials are described by my colleague, Mr. R. Savory, above, p. 411.
Meanwhile, feeding trials with Leucaena leaf had been initiated by the Agricultural Research Department of the Ministry of Agriculture and Natural Resources. Stall feeding experiments with cattle had shown live weight gains over a six-month period of up to 1.17 kg/head/day with a 4:1 mixture of maize bran and Leucaena meal.
Experiments by the project and the University of Edinburgh had shown that Peru was unsuitable for feeding to chicks, which showed a marked lack of thrift. On the other hand, fresh pellets fed to layers increased the intensity of egg yolk colour to Roche Index values of more than 11 initially, and to values of 7 to 8 after storage for more than six months. These figures may be compared to the common commercial target value of 9.
On the marketing side, enquiries into the availability of export outlets were continued. From the start, it was evident that Leucaena leaf was little known as a trading commodity and that a market would have to be created. A close study of the requirements of manufacturers of livestock feed would have to be made, and development on the production side would have to be market-oriented.
In keeping with the priorities thus established, this paper presents the market findings of the study first, followed by those for processing. Production comes last, and the profitability and future prospects of the crop are assessed in the marketing context.
The sole commercial outlet for Leucaena during the study was provided by a local company producing milled products and livestock feed compounds for sale in Malawi. The company bought Leucaena leaf delivered to its works in Limbre. To accommodate the firm's requirements and to specify the nature of the product, a grading system was devised. Grade A leaf was defined as the dried pinnules of good colour and a crude protein content of 25% to 30%. Up to 10% of rachillae was accepted in the sample. Grade B consisted of rachillae, rachises, twigs, pods and flowers, with a crude protein content of 10% to 15%. Grade C described leaf which had become discoloured as a result of wetting, drying and the formation of moulds, but which still had a crude protein content varying between 10% and 25%.
The company paid 9.3 cents/kg at the start of the study, but in 1978 raised the price to 11.9 cents/kg. Transport costs did not rise and it was possible to raise the farm gate price from 5.3 cents/kg to 8.0 cents/kg. The Government of Malawi required the domestic demand to be satisfied before any export sales were made. Although production remained on an experimental scale during the study, at less than 50 tonnes/ year, it was estimated that the company's requirements were no more than 600 tonnes/year and that no commercial demand was likely to arise anywhere else in the country in the short term. Taking a conservative average yield of 2 tonnes/ha, only 300 ha of crop would be required to meet that demand, and it was clear that any future Leucaena industry would cater essentially for an export market.
Reports were received (private communication) of markets for Leucaena leaf in Japan and Singapore, drawing on supplies from the Philippines, China, Taiwan and Malaysia. The Japanese market was said to require 20,000 tonnes/year, but this was largely unsatisfied. No firm data covering the size of either market was obtained, but one milling company in the Philippines reported regular export of Leucaena leaf in pellet form to Japan. The Singapore market was apparently an informal one, forming part of a general trade in a variety of green materials for livestock feeding.
The general market prospects for Leucaena were favourable in the medium to long term. While the domestic market was very small, export prospects were good. Leucaena enjoyed a distinct comparative advantage from the point of view of quality over other green meals, notably its chief competitor, Lucerne. It had a lower fibre content and a higher content of protein and carotene, the components for which it was chiefly marketable.
From a quantitative point of view also, the competitive position of Leucaena offered favourable prospects. Lucerne production in Europe was reported to have remained static over the five years from 1974 to 1978, while production in the USA had declined over several years (Agra Europe, 1979). Furthermore, drying of the Lucerne crop was dependant on oil fired drying systems, which were likely to become more expensive with the rapidly mounting world price of oil. For Leucaena, on the other hand, there were good prospects of developing an effective drying method using woody by-products of the crop.
Fish meal also, for which Leucaena was a partial substitute, was becoming more expensive. This was due to over-fishing and improvements in the technology of fish processing, which had resulted in lower availability for the cheaper by-product.
Interest in Leucaena leaf in response to the project's market enquiries was shown by importers in Zambia and South Africa, but the main interest came from Europe, notably the Netherlands. There an immediate market for 6,000 tonnes/ year was available (private communication). The potential market was estimated to be around 18,000 tonnes/year.
Marketing in Europe ,however, involved high freight and handling costs. The exact amount of these costs was not obtained because Leucaena, as a new crop, had no established tariff classification for either rail or ocean freight. Estimates were therefore made of both transport costs and landed prices in Europe. It was concluded from market trends that by 1982, when the earliest exports from Malawi could be expected, the value of Grade A leaf, c. i. f. Rotterdam, would be 25 US cents/kg. Transport costs from railheads in Malawi, which were calculated from existing freight rates for similar commodities, were expected to rise by 1(W/10% and to equal about half the landed value. An f. o. b. value of 12.5 cents/kg would still allow a small increase on the current farm gate price of 8.0 cents/kg.
However, the projections were based on the assumption that Leucaena would be available for consignment in commercial quantities. This was hardly likely by 1982 and it was concluded that development of trade in Leucaena would require a special development price, concession freight rates or both, if supplies were to be forthcoming and demand was to respond. Thereafter, economies of scale should ensure profitable disposal of the crop without artificial support.
No work was done on the marketing and utilization of Grade B material. But experience showed that l kg of Grade B was produced for every 5 kg of Grade A. Given a crude protein content of 10% or 15%, it was held that Grade B leaf would be suitable material for stall feeding cattle. The unit value would be small compared to that of Grade A, but the marginal effect of sales on the profitability of Leucaena growing could be considerable.
The processing of Leucanea produced in the course of the production trials is described below.
The crop was harvested by hand, using pangas or knives to cut off the whole branch to a stubble height of 30 cm. Various methods of sun/air drying were tried. However, spreading the cut branches over the stubble of the adjoining row was the only workable one for the larger areas. Spreading along the fence or drying in open sided sheds gave satisfactory results for small areas. Artificial drying with fuel oil and electricity was also attempted, but with the methods used, and at prevailing costs of energy, proved to be expensive.
The most effective technique found for threshing the dried branches was to heap them on a mat and to beat the topmost ones on those lower down. Loss of pinnules by shattering was minimized by gathering dry branches early in the morning while the dew was still on them. Good separation of Grade A and Grade B leaf was achieved by hand sieving. Milling of Grade A leaf doubled its density, but the product proved too dusty for either transport in jute sacks or the pelleting process.
The marketing studies had shown that pelleting of the product would be an essential requirement of export markets. Pelleting was found to decrease the dust problems which occurred even when unmilled leaf was handled, increase ease of handling, and increase the density of the material.
The low density of Grade A leaf was a problem for the pelleting process itself. The nominal output of the pilot plant used by the project was 500 kg/hr, but the throughput of unmilled leaf without binders rarely exceeded 200 kg/hr. Towards the end of the project, pelleting trials were conducted on the effects of adding water, molasses and vegetable oil, separately and in varying combinations. All additives reduced losses from dust and tended to increase throughput and compaction rations.
Tests carried out at Bvumbe Research Station showed no damage to Leucaena during storage by most common pests. The main storage problem was the oxidation of carotenoids which occurred over time and reduced egg yolk pigmenting ability. Preliminary studies of the effect of incorporating antioxidants in pelleting additives carried out by the project showed promise. More detailed studies have been initiated by the University of Edinburgh and the Tropical Products Institute.
Leucaena leaf is a bulky commodity, expensive to transport and low in value per unit of weight or volume compared to other commercial products. Pelleting is therefore essential to minimize transport costs. For the same reason, processing plant should be located close to growing areas.
Drying is a critical feature of commercial Leucaena production. Leuceana is a perennial crop which grows all the year round in suitable locations. However, it grows in the wet season when drying conditions are at their worst. For larger scale holdings, using mechanized systems of production and processing, it will be necessary to devise artificial drying methods. It was concluded that flash-drying methods would be necessary, similar to those used for Lucerne or grass. This would be so because the drying of whole branches by sun/air methods involved drying out the woody parts as well as the softer parts of the leaf. Rapid drying combined with automatic threshing would dry out the latter first, without runback of moisture from the stem. It was calculated that pinnules could be dried from a moisture content of 70% in the green state to a desirable 15% in three minutes with a sufficient volume of air at a temperature of 500 °C.
It was further concluded that the woody parts of the plant could provide the energy necessary for drying in the above manner. Economic evaluation of course, would have to await appropriate engineering studies.
Pelleting of Leucaena, by the time the project ended, remained a slow and expensive process compared to that of other products, but scope clearly existed for improvements. The use of additives required further investigation. It was considered also that wilting might increase pelleting efficiency in large-scale processing units. Milling within a closed system would increase the density of the input material and thus increase the output of the pelleting machines.
i ) Organization: Farm production studies were carried out using a variety of forms of organization. The production units included plots established on traditional smallholdings, plots allocated to tenants on larger units established by the project and larger units staffed and managed exclusively by project personnel.
The encouragement of smallholders to establish Leucaena on their traditional holdings ceased after the first two years of the study. It was found that the cost of goat-proof fencing, which was essential for Leucaena growing in the Lower Shire Valley, was too high for economic production on the 0.2 ha plots which had been established. Moreover, logistic difficulties were experienced in assisting growers on widely scattered farms with credit, extension and marketing services.
Accordingly, the tenant system was introduced in the 1977/78 crop year. Comparatively large blocks (up to 28 ha) were cleared, ploughed, cultivated, planted and fenced by the project. All operations except clearing were done with machinery. The blocks were subdivided into 0.4 ha plots and let to selected farmers. The cost of developing the blocks was to be recovered by a levy on production over the first two years of the scheme. No rent was charged. Several blocks, varying in size up to 14 ha, were planted and managed by the project.
Fencing requirements per hectare fell sharply with the change in organization. Separate 0.2 ha plots had required 2 rolls each of "Veldspan" mesh wire fencing, or 8 rolls/ha. On the larger holding less than 1 roll/ha was required.
ii) Data Collection and Analysis. Due to the changes in technique and production organization which occurred as the study progressed, it was not possible to carry out a fully satisfactory input/output analysis. Nevertheless, data were collected and analysed with a view to identifying trends and quantifying the relationships.
a) Yields. The last crop year for which a full year's figures are available is 1977/78. By that time there was little information concerning the new tenant-type holdings. Yield information covering a number of years is therefore confined to results from nine smallholders who had planted in 1975 and consistently harvested these Leucaena plots and their traditional holdings.
Table 1 gives yield data on the nine smallholdings for the three years 1975/76 to 1977/78. Some reservations about the data must be made with regard to the small size of the sample and its bias towards the better and more successful farmers. Also, the plots were small and varied between only 0.2 ha and 0.3 ha in size, whereas experience showed that yields per hectare tended to fall with size of plot. Furthermore, yield differences were often attributable to site differences and the sample was in no way a random one.
Table 1. Leucaena yield data from nine small holdings, 1975/76 to 1977/78 (Kg/ha)
1st Year |
2nd Year |
3rd Year | |
Average yield |
587 |
1,656 |
2,168 |
Highest individual yield |
975 |
2,140 |
4,420 |
Lowest individual yield |
438 |
544 |
975 |
Even so, it is considered that the results give some indication of the yields which are possible in the future as technology improves and farmers become more skilled in growing the crop. Particularly notable in Table 1 is the high rate of growth in yields which occurred year by year. This was another factor which inhibited economic analysis. The study covered too short a period for plateau levels of output to be identified on even the oldest Leucaena stands, and thus for long-term average yields per hectare to be identified.
b) Establishment costs. Details of the costs of establishing the larger blocks in 1977/78, at current prices, are shown in Table 2.
Table 2. Establishment costs on larger blocks.
Site Area |
Goma 14 ha |
Tomali 28 ha |
Costs US $/hal |
– US $/ha– | |
Fencing | ||
Materials |
143 |
88 |
Labour |
4 |
1 |
Tractor |
4 |
4 |
Clearing (labour only) |
39 |
27 |
Ploughing and cultivation (tractor costs) |
|
|
Planting (tractor costs) |
39 |
19 |
Weeding, replanting, supplying and spraying (labour only) |
39 |
35 |
Insecticides |
8 |
53 |
318 |
272 | |
1 Any errors in addition are due to rounding in the conversion of Malawi kwachas to US dollars.
The dominance of fencing costs among total costs should be noted. Attention is also drawn to the smaller fencing cost per hectare in the larger block, illustrating the economy of scale already mentioned. No charge was made for seed, which was supplied free to growers. No fertilizer was used on any of the production units in the study. Weeding was mainly carried out in the establishment years. Some weeding was also carried out in the second year, but thereafter the vigorous growth of the crop suppressed all competition and weeding was no longer required. It was therefore treated as an establishment cost only.
c) Operating Costs. Operating costs consisted entirely of labour costs. Once the crop was established, the only activities were harvesting, threshing, sieving and bagging. Bags were supplied by the project and their cost allowed for in determining the farm-gate price for Grade A leaf.
Details of labour use in these activities were recorded for the project's own block at Ngabu throughout 1978, as shown in Table 3. Costs were calculated at the current minimum daily rate for rural workers of 30.0 US cents for a 7.5 hour working day.
Table 3. Labour use and operating costs of leucaena production, Ngabu plot, 1978.
Man-hours/ha |
US $/ha | |
Harvesting |
600 |
34 |
Threshing |
627 |
25 |
Sieving and bagging |
418a |
17 |
Total |
1,645 |
66 |
a Man-hours for sieving and bagging were calculated using the yield figure for Ngabu plot for 1978 of 2,842 kg/ha and a throughput rate of 6.8 kg/hr.
It should be stressed that these data were obtained using government-employed, daily-paid labourers. Other studies had indicated that output per hour by similar labourers on task work, or by self-employed smallholders, was generally more than double the figures quoted in Table 3. The Ngabu figures, however, provided the most comprehensive set of labour input data collected. Furthermore, they referred to a pattern of harvests per year, which was expected to be the regular pattern for the future. A total yield per hectare of 2,842 kg was attained from four harvests in 1978 which took place respectively in January/February, April/May, July and October/November. The crop had been established in 1974.
A daily record of all labour inputs was kept for the 1977/78 cropping year for ten households which had established Leucaena plots on their own holdings in 1975/76. The householders were selected for their nearness to the project office, rather than on a random basis, because of labour constraints on the project itself. The farmers gave Leucaena growing a low priority in allocating labour time. The average farmer devoted only 4% of total working hours to the crop, compared to 62% for cotton and 28% for food crops. Leucaena was given most attention in the dry season, when the requirements of other crops were low.
Yields per man-day of labour applied to Leucaena growing varied widely. The data from one holding could not be used for input/output comparison, while two others did not harvest their crops at all. Yields per man-day for the remaining seven were, in ascending order: 6.7, 7.9, 11.9, 28.8, 41.4, and 43.2 kg.
Apart from matters of efficiency and the low priority which the farmers attached to Leucaena growing, the explanation of the wide divergence in the returns to labour may have been associated with the fact that they rarely harvest the crop completely in one continuous operation. It was therefore cropped at differing stages of growth; different amounts of leaf loss occurred and the yield of marketable leaf obtainable from a day's work varied accordingly.
The wide variation in returns to labour was confirmed by project experience on holdings under the direct control of its own staff. Variations associated with the existence or otherwise of incentives to complete the task were illustrated by separate observations of the results obtained when the work was done on a piecework basis and when it was done for a daily wage. The mean amount of time taken to cut one hectare of Leucaena on task work was 67 man-hours (10 observations). The same task under similar conditions, but done by day labour, took 149 man-hours (22 observations).
Output per unit of labour was also found to depend on the stage of maturity of the crop. This occurred in two respects. The more mature crops took longer to cut and, being cut less frequently, suffered greater leaf loss between harvests.
1) Yields. Yields obtained by the project and by smallholders were variable, but as noted, could be expected to increase as existing plantings became older and more strongly established. They would also rise as techniques of growing were improved and, above all, as farmers gave greater priority to Leucaena growing compared to other crops. In this connection, a marked increase in supplies of leaf and in activity, on Leucaena plots generally, occurred when the farm gate price was raised from 5.3 US cents/kg to 8.0 US cents/kg early in 1979.
For purposes of planning and rough analysis of the economic prospects of Leucaena growing, it is necessary to make assumptions regarding likely yields. It was assumed that smallholders could achieve in the long term an average yield of Grade A leaf of 2 tones/ha. At the current farm gate price, this would provide a gross return of $ 160/ha.
2) Size of holdings. The economies of scale achieved by concentrating smallholders' plots into tenant blocks, could be increased by the establishment of full-scale settlement schemes. The tenants would reside on the scheme area and, apart from growing food crops, would earn their incomes entirely from Leucaena. Processing facilities would be provided on the site and further economies thus made in handling and transport.
The appropriate size of tenant holding had been suggested by the labour utilization studies. The study of ten households indicated that the average household devoted a total of 5,089 man-hours to farm work in the 1977/78 production year. Of this figure, 3,365 man-hours was given to production of cash crops - cotton and Leucaena. At the same time, the data from the Ngabu plot, as shown in Table 3, indicated that annual Leucaena operations required 1,645 man-hours/ha. On this basis, the average household could look after 2 ha in addition to subsistence crops, and the gross return from Leucaena would be $ 320 a year.
3) Establishment costs. The figures in Table 2 suggested that a settlement scheme of perhaps 500 ha, requiring boundary fencing to keep out goats, with further scale economies in fencing, might cost between $120/ha and $180/ha to establish. Outside the Valley in areas where goats or other stock were not a problem, the cost would be considerably less.
Leucaena was a perennial, or orchard crop and stands older than 40 years were common in other countries. If establishment costs of say $140 were written off over a period of 20 years, the annual charge would be only $ 7/ha, or $14 per tenant unit.
4) Net returns. In the absence of cash costs and without putting a value on labour, the annual net return to the tenant would be $ 306 or $ 153/ha.
However, Leucaena production at the smallholder level may not always be free of cash costs. For instance, the project found response to fertilizers at a number of sites. As Leucaena becomes more widely grown, pests and disease may develop and require expenditure on chemicals and spraying equipment. Also considerable expenditure could arise if irrigation proves worth while.
The labour requirement will increase as yields increase. It may also rise with changes in management systems. For example, it may be found more productive to reduce the height to which the crop is allowed to grow, and thus to harvest more frequently. This would reduce competition by the crop and could encourage weed growth. Weeding might thus become an annual labour cost.
5) Opportunity cost. The opportunity cost of settling smallholders on 2.5 ha of land as proposed, was measured by reference to the current model budget for the Makande and Tomali plains under the Shire Valley Agricultural Consolidation Project (IBRD, 1978). This provided for a mixed farm matched to average household labour resources, of 1.94 ha. The net farm benefit to a farmer adopting Stage 1 improved techniques was given as $ 218, after deducting $ 43 of cash costs. Leucaena growing with a return of $ 306 plus the value of subsistence would thus have a clear lead over the best alternative in the area.
Although the goal of the investment study, in accordance with Malawi's national development priorities, was a new industry based on smallholder production, it was concluded that specialized Leucaena estates would have to lead the way and provide the basic structure. It was envisaged that any future industry would consist of nucleus estates and satellite smallholdings. The essential functions of the estates were seen to be as follows:
a) To mobilize capital
b) To provide technical and managerial skills.
c) To provide the infrastructure for production processing and marketing on a national scale.
d) To facilitate economies of scale in all operations.
e) To provide processing facilities near to production sites.
f) To build up rapidly, and maintain, the minimum quantity of Leucaena required to interest an overseas manufacturer of livestock feeds.
g) To provide supplies and services to smallholders, including fencing materials and seed at discount prices, a machinery pool, a credit scheme, a production advisory service and a ready market for smallholders' output of Leucaena leaf.
By the end of 1979, when the investment study ceased, private interests were considering proposals for the establishment of a Leucaena estate.
Agra Europe. Issue 829. Brussels. June 1979.
Beale C.I.A. et al. (1979). 'A Pilot Leucaena Estate. Report of a mission to appraise a proposal for a pilot estate to grow Leucaena in Malawi'. Lilongwe, Malawi.
Savory R., Breen J. A. and Beale C. I. A. (1979). 'The production of Leucaena leucocephala in Malawi'. Series of working papers:
No. 15. Cultival Evaluation. N. 16. Establishment Studies. No. 17. Plant Population Studies. No. 18. Plant Nutrient Studies. No. 19. Harvesting Studies. No. 20. Farm Production in the Lower Shire Valley. No. 21. Processing. No. 22. Utilisation.
UNDP/FAO Project MLW/75/020. Lilongwe, Malawi.
International Bank for Reconstruction and Development. (1978). 'Staff Appraisal Report: Malawi'. Shire Valley Agricultural Consolidation Project. Report No. 1945. MAI. Washington.
