0.1 Economic aspects

Over the last two decades, world production and consumption of oil fruits and oil seeds and their products has almost doubled. In terms of value, oil fruits and oil seeds now take third place after starch plants and fruits/vegetables.

Apart from their nutritious value, oil fruits and oil seeds are of particular economic importance for the developing countries, from where a substantial part of the (statistically registered) world production originates. In Africa, Asia and Latin America, the cultivation of oil plants not only plays a major role in the provision of protein and fats, but contributes considerably to exports earnings.

Table 1 reflects the development of world production of major oil seeds as registered by the Food and Agricultural Organization of the United Nations. The figures given for the last 15 years, however, do not necessarily reflect the total for each crop, since, in many countries, there is a significant production for the growers' domestic requirements which never appears in official statistics.

Table 1: World Production of Major Oil Seeds (million tons).

Source: FAO Production Yearbooks, Vols. 35 39

- For Coconuts, production is expressed in terms of weight of the whole nuts, excluding only the fibrous outer husk, whether ripe or unripe, whether consumed fresh or processed into copra or dessicated coconut
- For Cottonseed, direct production figures are reported by countries accounting for about 60 % of world production, data for the remainder are derived from ginned cotton production
- Groundouts in shell
- For Rapeseed, figures include mustard seed for a few countries (e.g. India and Pakistan)

As can be seen in Table 2, soyabeans have established a leading position and now represent more than 40 % in volume terms of world production. If the annual increase in production is considered, rapeseed, with an average increase of more than 12% per year, has an even more positive trend than soyabeans, followed by palm kernels and sunflower seed. Palm kernels are, in fact, to be seen as by-product of palm oil production. The production of palm oil fruit, however, is not recorded, and therefore palm kernels can indicate the growth for the whole crop

Table 2: World Production of Major Vegetable Oils (million tons)

Sources: Oil World, Vol. XX. No. 13, p. 302 f Marches Tropicaux, No. 2112, p. 1167

World production of major vegetable oils, as shown in Table 2, has been steadily increasing over the past decades with average annual growth rates of more than 5 %. In volume terms, the total (recorded) production now stands at more than 60 million tons per year, which contributes more than two thirds to the world consumption of all edible oils.

Table 3: World Exports of Major Oil Seeds (million tons)

Source: FAO Trade Yearbooks, Vols. 25-39

- For Groundnuts, nuts reported in the shell are converted to shelled equivalent using a conversion factor of 70 %
- For Rape and Mustard seed, most trade statistics do not allow a distinction, and figures in many instances even include other minor oil seeds

The great increase in the production of soyabean oil has been the major development on this market in the past decades. Due to the introduction of the hydrogenation technique, which made the oil suitable for margarine, and the rapid increase in demand for soya cattle cakes, world production of soyabean oil tripled over the last 20 years.

The production of vegetable oils from palmfruits, sunflowerseed and rape/mustard has doubled over the same period. The group of oils made from cottonseed, groundnut and coconut, however, is more or less constant. Since many vegetable oils are direct competitors, the relative importance of these oils may well change in the future, but the dominance of soyabean oil is unlikely to change.

In terms of volume of world production, other vegetable oils are of minor or only regional importance. Olive oil, for example, is almost exclusively produced in the mediterranean countries where it meets a unique consumer preference. Since it has not gained acceptance in other regions, olive oil will not be dealt with in the present publication. Other oils, however, are of particular relevance for specific areas in developing countries (e.g. sheanut in West-Africa) and will therefore be described in more detail than their overall importance would appear to justify.

World trade in oil fruits and oil seeds, as shown in Table 3, is even more dominated by soyabeans than the production figures seem to indicate. With one quarter to one third of the total production entering the world market, soyabeans is in fact the only oilcrop with considerable exports in an unprocessed condition. Most prominent exporter is the USA, which has two thirds of the whole market. The soyabean exports of all developing countries together only amount to about half of this share, showing South America (Argentina and Brazil) in a leading position. The main direction of the soyabean trade is towards Europe, which imports more than half of the available quantity.

Of more relevance to the developing countries is the world trade in vegetable oils. As shown in Table 4, palm oil alone accounts for about one third of the trade, the major exporters being Malaysia and Indonesia, with Singapore as the major port of the region. Prior to 1945, Africa produced and exported most palm oil, but has now lost this position due to the massive planting of high yielding varieties in South East Asia. For palm oil, Asia has also the highest demand, although import statistics do not correspond to actual consumption due to substantial re-exports (Singapore).

Table 4: World Exports of Major Vegetable Oils (million tons) * no records

Source: FAO Trade Yearbooks, Vols. 25-39

- In general, figures do not reflect entire world trade since some national statistics classify all or a large part of their trade under such headings as “edible oils", “vegetable oils, nes" etc. and therefore have been omitted

Whereas coconut and palm oils are almost exclusively exported from developing countries, they are able to supply three quarters of the world market for linseed oil and groundnut oil and even realize shares of about 50 % for sunflower, cottonseed and soyabean oils. Major suppliers in these markets are currently Malaysia (palm and palm kernel oil), Philippines (coconut oil), Brazil (soyabean and groundout oil) and Argentina (linseed and sunflower oil).

The world trade in vegetable oils is, to a considerable extent influenced by the demand for by-products (oilseed cake and meal) and the prices of others chemically similar, i.e. competing vegetable oils.

As shown in Table 5, the trade in oilseed cake and meal consists to more than 70 % of the by-product of soyabean; a crop for which the meal is - due to the low oil content - relatively more important than the oil. A great advantage of the soyabean and, in fact, a reason for the production increase, is that it can be freely fed to all livestock groups. Major exporters of soyabean meal are Brazil, USA and Argentina, the major importing region is West Europe.

Table 5: World Exports of Oilseed Cake and Meal (million tons)

Source: FAO Trade Yearbooks, Vols. 25-39

As illustrated in Figure 1, world market prices for vegetable oils depend as much on the specific variety of oil as on apparently periodical fluctuations. The different characteristics of vegetable oils, which determine the relative scarcity and the possible use of the product, will be discussed in more detail in Chapter 1.1 of this publication.

Figure 1: World Prices for Selected Vegetable Oils (monthly average quotations, West European ports) in US $ per ton.

Source: E. A. Weiss, Oilseed Crops, 1983, p. 9

The price fluctuations tend to assume a cyclic pattern with a periodicity of several years. During this period, stock build-up leads to more demand and subsequently higher prices, followed by a period of stock liquidation with the opposite price effect with shorter periods of market consolidation in between. As reported in market observations of “Marches Tropicaux" (5/86), prices for specific oils have, over the last few years, shown fluctuations of more than 300 % (based on the lower price). In absolute terms, the price for coconut oil, for example, rose from about 400 $/t (CIF, Rotterdam) in 1983 to about 1400 $/t in the following year only to fall back to the 1983 low the year after. Except for soyabean, prices in the last two to three years have taken another drastic downturn and are likely to stabilize at a relatively low level.

For certain vegetable oils or specific qualities of oils, only regional or even local markets have been established. Accordingly, no representative market data are available for these products. For the African context, however, the following examples should demonstrate that these local markets can show particular consumer preferences and do not necessarily follow the same trend as the world market for the major varieties.

In nearly all oil palm growing countries in West Africa, the red palm oil is an important food ingredient. The colour and the taste from the oil of the traditional aura variety are generally more highly valued than those from the (hybrid) tenera variety. Also, the taste of the oil processed by traditional methods is preferred. In Benin, Cameroon and Nigeria about 50 % of the palm oil consumed is produced in the traditional way. In Ghana and Sierra Leone it is even 70 % and 90 % respectively.

In the Pacific Islands and along the coast of West Africa, coconut oil is also produced in a traditional way. Although this kind of oil has a particularly preferred taste its keeping quality is reported to be low. Shea nut butter is an important fat in Burkina Faso and Mali and also has a special taste that is highly regarded.

Most of these traditionally produced oil varieties are only available in a certain season of the year, which leads to low market prices at harvest time and co-nversely to high prices out of season.

(introduction...)

The present sub-chapter aims at giving an outline of the existing processes for manufacturing vegetable oils without entering into complex technical details.

In the presentation, an initial distinction is made between the aspects involved in the processing of oil fruits and those involved in the processing of oil seeds. For oil fruits, the processes for the extraction of oil from oilpalmfruit is given as the most important example. For oil seeds, the principles for different crops are sketched.

In oilcrop processing, many technologies have been developed, which have their place in different economic and cultural situations. Therefore, a second distinction is made between traditional methods and modern methods in each of the above categories. Whereas traditional methods are seen as clearly reflecting their social environment, modern industrial processes are the result of high level technological experience.

Small scale oil production systems, as the main target of this publication, try to combine these two characteristics; i.e. they should be adaptable to a given social context, and they should be technically efficient and reliable (which in any case requires proper maintenance and an adequate supply of spare parts). Small scale systems, therefore, can only be improved on the basis of an understanding of traditional methods and a thorough knowledge of modern technology. The presentation of this intermediary technology level, however, is given in more detail in later chapters.

0.2.1 Processes for oil fruits

Since the oilpalm gives the economically most important tropical oilfruit, the technologies for its extraction can serve as an example in this category.

In the traditional process, the fruit is first removed from the bunches, generally after the bunches have fermented for a few days. The fruit is then cooked and pounded or trampled. The mashed mass is mixed into water. The oil and oilcontaining cell material is separated from the fibre and the nuts by rinsing with excess water and pressing by hand. The oil-containing mass, now floating on the top, is collected and boiled. In this step, the oil separates from the rest and collects on the surface. It is skimmed off and finally dried.

The actual execution of the process may vary somewhat from area to area; most traditional processes, however, have in common the superfluous use of water. Using this process, generally not more than 50% of the oil is obtained. The problems are:

- the digestion by means of pounding or trampling,
- the separation of the oil and oilcontaining material from the fibres and the nuts by means of water and
- the liberation of the oil by cooking afterwards.

The potential for improvement of this technology and thereby the development of small scale extraction equipment in principal depends on

- better cooking by means of steam,
- better digesting using a reheating step with steam and
- more effective pressing in a batch press or continuously working screw press.

The modern process of extracting palm oil, used on a larger scale, starts with the steam sterilization of the bunches. The bunches are threshed and the fruit is digested mechanically, while heated with steam. The mass is then pressed in hydraulic presses or continuously in screw presses. The oil is separated from the press fluid by heating and is finally dried.

0.2.2 Processes for oil seeds

In addition to the distinction made between traditional and modern methods, the processes for oil seeds should also be divided into so-called wet and dry extraction methods.

Of the traditional wet processes, the extraction of coconut oil from fresh coconuts is the best known. It starts with grating the meat, after which the oil as well as the proteins and impurities are extracted as a milk from the fibrous residue by pressing (by hand or foot) and rinsing with fresh water. The milk is left to stand to form an oilrich cream on top. The cream is boiled to separate the oil from water and other impurities. The oil can be skimmed off. It still contains a protein- rich residue that can be filtered off after drying and used for human consumption.

Other oil seeds, like groundnuts, palmkernels and sheanuts are roasted and crushed as fine as possible (e.g. first by pounding, followed by crushing between stones or a stone and an iron bar). The crushed mass is mixed with water, and the oil is obtained by cooking the mixture, causing the oil to float. The oil is finally skimmed off and dried by heating. Sheanut oil is often obtained by beating air into a mixture of crushed seeds with some water using a hand-operated buttermaking process. The milk or cream floating on top of the beaten mass at the end of the process is then cooked to evaporate the water and dry the oil.

The weak points of these processes are the grating or crushing steps. They are timeconsuming and exhausting work, yet crushing is generally not fine enough. Thorough crushing can improve the oil recovery considerably. In many areas, engine-driven discmills are used by women in small commercial enterprises to get their seed crushed.

All the traditional dry processes, as well as the modern dry-extraction methods, consist of three essential unit operations:

- size reduction,
- conditioning by heating and
- eparation of the oil.

The difference between the dry processes is the way by which the oil is separated. With respect to this difference, the following traditional methods can be distinguished:

- without pressure,
- with a wedge press,
- with a screw press,
- with a beam press or
- with a ghani (which combines the above unit operations).

In historical perspective, the use of pressure in the process seems to indicate a society with a higher technical level of achievement. Wedge, beam and screw presses have already been used by the Egyptians, Romans and Chinese. The beam press, which took up a lot of room, was soon superseded by wedge and screw presses, which work fairly satisfactorily. The animal-driven ghani is mainly used on the Indian sub-continent, from where it originates.

Traditional dry processes are very labour intensive and improvements seem appropriate, at least for any kind of marketoriented production. The potential for improvements would best be tapped by the introduction of simplified versions of the modern technologies (see below); e.g. by

- crushing the seed in a roller mill,
- "cooking" in a directly fired pan,
- pressing with an unsophisticated spindle press, a hydraulic press or an enginedriven oil expeller.

In India, the productivity of ghanis has been drastically improved by the introduction of motor-driven versions, which are fast replacing the animal-driven ones.

As mentioned, the modern dry processes consist of the same unit operations as the traditional extraction methods. First, the shells or hulls are separated from the nutor seed kernels to obtain a mass with a maximum oil content. Palmnuts from the African oilpalm or American palms are cracked. Seeds, such as groundnut, sunflower, cotton and kapok are decorticated. The oil-containing kernel material is then milled between rollers to obtain a wellcrushed material in

the form of flakes. The crushed mass is “cooked" in a set of steam heated pans in a humid atmosphere and subsequently dried.

The dry mass is then pressed, a process which generally is applied twice; i.e. prepressing and deep-pressing. The extraction generally takes place by means of oil expellers. Finally, the oil is filtered.

The modern processes, as opposed to the traditional methods,

- apply higher pressures and gain higher yields,
- use power-driven size-reduction equipment and therefore have a higher power consumption per kg of oil,
- are less labour intensive, but require higher initial investments,
- involve less variable, but more constant costs.

The most modern process is the solvent extraction. In this process, the reduced seeds are chemically extracted with a nonpolar solvent (usually hexane). In contrast to the modern dry processes with expellers, the solvent extraction cannot be carried out economically on a small scale.

Nevertheless, the process has a number of advantages, such as

- high extraction yield (95 - 99 %),
- high capacity continuous process.

For the purpose of a small-scale production, the main disadvantages, such as:

- large initial investment capital needed, - large maintenance costs and
- need for skilled labour

are decisive, in particular for most projects in developing countries.

Modern wet processes have been developed for coconut and groundnut processing, but are not likely to become economic. Therefore, the modern wet processes will not be further discussed in this publication.

0.3 Development potentials

The concept of rural development aims to meet the basic needs of the majority of the population in Africa, Asia and Latin America. This direct approach to basic needs has taken a variety of forms, project types and degrees of integration of individual measures, ranging from infrastructure projects (e.g. transport, energy and water supply), institution building (agricultural and other extension services, assistance to self-help groups and cooperatives), supply with credits and inputs and the promotion of local, small-scale processing of agricultural products.

Agro-industrial development, defined as processing of agricultural products on a larger scale, is of necessity, often located in central places of the producing areas or the capital of the country. The basic concept is rather a macro-economic approach of substituting imports of consumer goods or earning foreign currency with exports. Agro-industrial projects, such as the establishment of central oil mills, have often been successful in keeping an additional value-adding processirig step in the hands of a developing country and have thereby led to a more advantageous participation in the structure of postcolonial trade.

The other side of the coin is, however, that quite a number of these projects have not proved to be viable due to insufficient or badly planned rawmaterial supply, management problems and/or highly fluctuating market prices. In macro-economic terms, the result of such projects has often been an increase in foreign debts rather than any profit for the country.

Without unfair generalization, one might say that very few agro-industrial plants have made a substantial direct contribution to the basic needs of rural people. As far as oil mills are concerned, producer prices are often kept artificially low in order to be competitive in international markets. Additional employment opportunities are relatively few, and the improved supply of vegetable oil is more often geared to the urban population. Producers of vegetable oil in rural areas, especially women, often find it more difficult to compete in local markets, since their traditional techniques are very labour-intensive and relatively inefficient. Provided that the social context is considered, as sketched below, the promotion of improved small-scale equipment for oil processing could therefore close a technological gap, increase the availability of oil for personal consumption and generate income in rural areas.

All societies in developing countries are characterized by a specific division of tasks between men and women. Not only in developing countries, but worldwide, the tasks of childrearing and housekeeping are attributed to the female members of the economic unit; i.e. the nuclear family, the extended family, the tribe. etc.

In most developing countries, especially in Africa, housekeeping comprises all family-related activities, often without the possibilities available in industrialized countries to make use of external services and institutions. In rural areas, housekeeping includes a broad spectrum of time consuming tasks, varying from the provision of water and fuel wood, the preparation of meals, washing, cleaning, most handicraft, the cultivation of vegetable gardens to the processing of basic food (such as the production of vegetable oil).

In addition to childrearing, which is a particular stress situation with every additional baby, the numerous tasks of women in rural Africa amount to average work loads of 16 hours a day. This is a considerably heavier burden than any man would normally carry, and is, in itself, a convincing argument for regarding rural women as a target group which deserves particular development efforts.

The disadvantageous division of responsibilities is, however, not limited to the work load as such: In most African countries, rural women have to take care for at least a part of the financial needs for housekeeping (for food, clothing, medicine, schoolbooks, etc.). For this purpose, cash crops have to be cultivated and marketed, handicraft articles produced and sold and other services provided. In the West African context, the production of vegetable oil plays an important role.

The necessity for independent financial resources for women also stems from the fact that men often consider any additional income should be for the head of the household as a contribution to his personal consumption (for radios, bicycles, alcohol, etc.). Furthermore, migration away from rural areas in Africa has already lead to 30% of female heads of households, who are more or less solely responsible for all needs of the family.

The improvement of traditional techniques and the transfer of appropriate technology (small scale) for the local production of vegetable oil can therefore be seen as a contribution to improve the social and economic living conditions of rural women. Depending on the social context, which varies widely, the introduction of new processes might, however, also face problems. Difficulties might arise from:

- the access to sources of finance, in particular credit, which - for one reason or another
- is more freely given to men,
- the volume of the necessary investment, in particular for power assisted technology, which poses a considerable risk and is often intimidating to women,
- the financial needs for production costs, in particular fuel for motor-driven versions which may not always be readily available in rural areas,
- the dependence on repair and maintenance services by workshops in the village or even the next city,
- the need for a minimum degree of organization beyond the family level; i.e. in self-help groups, informal precooperatives or even cooperatives with statutes and formal membership,
- the danger of men taking over after the successful introduction of the new and attractive technology, either for reasons of prestige or as a source of income.

The above mentioned potential difficulties might not be valid in specific cultural settings; in others, even one of these points could well lead to a complete failure of a project. In particular the last two points emphasize the necessity of a detailed knowledge of the social background at the village and even family level before starting to promote new technologies for local oil processing.

In West Africa, project experience has shown rather stable structures of women self-help groups. The transfer of this experience to, for example, East African countries should, however, be handled with some caution. The traditionally less autonomous status of women in this region might make more formal structures of organization (cooperatives) necessary.

A simple transfer of appropriate technology, therefore, appears to be insufficient to reach rural women as target groups of development efforts. Rather, a social approach has to be chosen, which starts with a careful identification of existing forms of organization, includes a training component to strengthen these structures and thereby develops and secures independent sources, of income for rural women.

Although the purpose of the present publication basically is to provide technical information, the social approach - after the characterization of the major oilcrops and vegetable oils in Chapter 1 - is reflected in the main parts of the booklet. Chapter 2 first identifies socio-economic units (e.g. family, village, district), then describes the technology which could be considered for each of these units. Chapter 3 gives examples from project experience introducing improved technology at the village level. In Chapter 4, the economics for the case studies are analyzed and alternative technical solutions evaluated. [Finally, the concluding chapters provide technical details, addresses of institutions and companies, a look at current research, guidelines for the identification of an oil processing project and a short list of relevant literature.