In 1997, a project was launched in Kenya to introduce tissue culture (TC) technology into the national banana sector to obtain pathogen-free planting material. Although small-scale farmers could potentially profit most from TC, financial constraints might hamper technology adaptation. Therefore, additional financial and structural measures along the technology dissemination channels are required.

Banana covers around 1.7 per cent of Kenya’s total arable land. The crop is grown in various agro-ecological zones, from the coast up to an altitude of about 2000 metres in the western highlands. Cultivation takes place predominantly under rainfed conditions on comparatively small farms. Different local as well as imported varieties are grown, some of which are used exclusively for cooking purposes. Within the diversified farming systems, banana usually receives a comparatively low priority in terms of labour and input allocation. It is seen as a security crop that provides a more or less continuous income flow throughout the year, even under low-input regimes. About 25 per cent of all bananas are consumed in the same households in which they are produced. In keeping with the pattern of semi-subsistence crops belonging to the female domain, women play the dominant role in Kenyan banana production.
Average banana yields in Kenya are meagre. At 14 tonnes per hectare, they achieve less than a third of the crop’s potential under the favourable conditions of the humid tropics. Experts conclude that, besides low input levels of labour, fertilizer, water and pesticides, the oppressive infestation of banana with various pests and diseases is the main determinant for this low yield. The economically most important banana pests in Kenya are weevils and nematodes. Severe disease damage is primarily attributable to Panama disease and black sigatoka, both caused by fungi. All these pathogens are spread through infected banana suckers being used by farmers for plant propagation, due to the lack of clean or resistant planting material, and the limited knowledge of farmers on this issue. The resulting yield losses make banana a relatively expensive commodity for consumers, and reduce the cash earnings of producers as well as the potential of the crop to contribute to the food security of rural households. The situation is essentially the same in other countries of East Africa.

The tissue-culture project
The banana TC project was launched in a joint effort of the Kenya Agricultural Research Institute (KARI) and the International Service for the Acquisition of Agri-biotech Applications (ISAAA). The project is sponsored by the Rockefeller Foundation (USA) and the International Development Research Centre (IDRC) of Canada. The aim is to provide small-scale farmers with pathogen-free banana planting material through the use of TC techniques. The rapid and sterile multiplication of banana plantlets by in vitro propagation is common practice in other parts of the world, but to date has not been commercially used in tropical Africa. The advantage of the technique is that large numbers of healthy banana plantlets can be produced in the laboratory in a comparatively short period of time. This reduces pest and disease problems for banana growers and offers an ideal opportunity to introduce new and superior germplasm quickly on a large scale.
Laboratory protocols for TC in conjunction with popular international banana varieties were previously available. Now, with technical assistance from the Institute for Tropical and Subtropical Crops (ITSC) in South Africa, several new protocols for local banana cultivars are being developed. KARI has already conducted on-station and on-farm field trials with the TC material in different Kenyan provinces. In the coming years, the technology is to be disseminated on a countrywide basis.
KARI itself does not have the capacity to produce enough TC plantlets. Project funds are available to purchase the material needed for the field trials from Genetic Technology Limited (GTL), a private company in Nairobi, that recently started to produce in vitro banana plantlets on a commercial basis. Explicit involvement of the private sector in the multiplication and dissemination of this technology is regarded as an important strategy to minimize the need for external financial support in the longer term. GTL is currently producing around 20,000 banana plantlets per year. For the production of enough material in the future, after widespread adoption of the technology, there are three non-exclusive options:

•	first, GTL itself could increase its production capacity;
•	second, other national public and private organizations could begin banana TC or upgrade their facilities;
•	third, in vitro plantlets could be imported from South Africa.

The potential future benefits and distribution effects of TC technology in Kenya are analysed below first at the level of the individual farm, and then for the domestic banana market as a whole.

Potential effects of banana tissue culture (TC) technology at the farm level     Small-scale  Medium-scale Large-scale   Without TC With TC Without TC With TC Without TC With TC Production cost (US$/ha)    248    571    300    652    413    792 Yield (t/ha)     10.7     26.9     13.9     32.2     18.7     36.1 Income (US$/ha)    983   2517   1212   2967   1572   3242 Increase with TC (per cent)             Cost      130      118       92 Yield      150      132       93 Income      156      145      106 Note: Although many farmers’ holdings are smaller than 1 hectare, the figures have been extrapolated to have a common reference. Calculations are made for 1998: US$ 1 = 59.70 Kenyan Shillings.

Farm-level assessment
Although initial on-farm field trial experience has already been gained with farm banana TC technology in Kenya, long-term impacts under regular conditions cannot yet be observed. To be able to quantify the potential agronomic effects of the innovation, 25 national and international banana researchers were consulted. In addition, interviews were conducted with 46 farmers and 10 local agricultural extension officers in the major growing provinces to learn more about current banana farming practices and constraints. The data thus obtained form the basis for the assumptions made in the quantitative technology evaluation.
For the analysis of distributional effects, the banana farmers were subdivided into three groups according to their banana acreage:

•	small-scale farmers with less than 0.2 hectares;
•	medium-scale farmers with between 0.2 and 0.8 hectares;
•	large-scale farmers with more than 0.8 hectares.

Cost and income calculations of current banana production practices have been carried out for each group (without TC). Since banana is a perennial crop, mean annual figures for the total length of a plantation cycle, which is 14 years on average, have been derived. The results are compared with a situation in which the use of TC plantlets is assumed (with TC, see table).
The current production costs of the individual farm types without TC differ to a noteworthy extent. This is particularly due to the higher labour intensity provided by both household and hired labour and the greater amounts of inputs, notably farmyard manure, used on the larger farms. Accordingly, the yields and incomes obtained per hectare of banana production are also higher on these farms. The table shows that adopting TC could bring about substantial yield increases for all three farm types, given that reinfection would not occur immediately due to high pest and disease pressure. In relative terms, the potential gains are most pronounced for the smallholders. This is not surprising because, as a result of the limited care and the small amounts of inputs applied, the small-scale farmers currently suffer the biggest stress-induced banana losses. However, using in vitro plantlets not only raises the yields but is also associated with additional costs. This is attributable to two factors.

•

Firstly, the TC planting material itself is expensive if compared with the cost that farmers incur when using traditional suckers for banana propagation from their own old plantations or from neighbours. The calculations in table 1 have been carried out using a price of approximately US$ 1.3 per TC plantlet; this is the current cost-effective level at which in vitro material can be produced in Kenyan laboratories. This price is about five times the estimated price of traditional suckers.

•

Secondly, the in vitro material is quite delicate, especially in the first months after field transplantation, and it demands good growing conditions to produce satisfactory yields. This implies that prevailing banana cultivation practices would need to be intensified to some extent. Of particular importance are operations for better field sanitation, such as weeding, and improving the availability of plant nutrients through the regular application of manure. Not following these rules could seriously depress the technology’s yield advantages and it is therefore important to combine the delivery of the technology with extension services on how to use it successfully.

It is anticipated that differences in farming practices between the farm types would remain similar after taking over the new technology. The input intensity on small farms would still be lower than on large farms. At the same time, the relative cost increase is highest for the smallholders. Although this cost increase could be more than offset by the rising revenues, additional monetary outlays can lead to seasonal liquidity problems. This might cause additional economic risk, especially for the resource-poor producers with little buffer capacity, if plants are lost due to mismanagement or unfavourable weather conditions. So, in spite of the considerable potential profits, technology adoption is likely to be slower for the small-scale farmers than for the larger ones. These potential constraints are explicitly taken into account in the following market model considerations.

Market-level assessment
This market-level assessment simulates the likely effects of TC technology for the Kenyan banana sector as a whole until the year 2020. Of course, the aggregate impact crucially hinges on the banana area that will eventually be cultivated with TC bananas. Technology adoption is usually a gradual process over time, and it has already been explained that behavioural differences between the farm groups have to be expected. Group-specific adoption profiles have been predicted in collaboration with the interviewed researchers and extension officers. The resulting adoption rates for each single year of the considered period were multiplied by the potential productivity increases at the farm level to derive the technology’s nationwide production effects. The benefits were quantified by calculating the respective changes in producer and consumer surplus. These are the standard welfare measures used in economic analyses. The average annual results are shown in the table for two different scenarios.
Scenario (1) builds upon the TC price assumption of the previous section, which is based on approximately US$ 1.3 per in vitro plantlet. The total benefit would be approximately US$ 1.6 million per year. Of this amount, some 44 per cent is gained by the consumers, for whom banana purchases will become cheaper owing to productivity increases in production. The remainder is attributable to income gains on the banana-producing farms, including subsistence consumption. However, in scenario (1) the benefit share of the smallholders is only marginal and far below their initial production share. Driven by their individual risk perception, this group of farmers might be restrained in adopting the technology, and income disparities might increase.
Scenario (2), on the other hand, assumes a TC price of approximately US$ 0.6 per plantlet. Price reductions are likely in the future because of growing experience and competition in the production of in vitro banana plantlets. Furthermore, the cost of planting material could shrink if farmers were able to use first generation suckers from once acquired in vitro plants for further propagation, without losing the TC health and vigour. This option, which also depends on appropriate crop management, is currently being tested in KARI field trials. The calculations in scenario (2) suggest that lower TC prices would bring about huge additional benefits on account of higher productivity increases and predicted faster technology adoption. The aggregate welfare gains in scenario (2) reach a level that is more than eight times the benefits under the higher price assumption. Moreover, such a TC price reduction would render the greatest relative benefit increase for small-scale farmers.

Projected annual benefits for banana producers and consumers     Producers  Consumers   Scenario (1): Higher price for TC plants    Annual benefits (Million US$)                       0.9     0.7   Small-scale Medium-scale Large-scale   Benefit share (per cent)       1       66      33   Production share (per cent)      37       41      22     Scenario (2): Lower price for TC plants    Annual benefits (Million US$)                       7.4     5.4   Small-scale Medium-scale Large-scale   Benefit share (per cent)      29       51      20   Production share (per cent)      37       41      22   Note: The figures are annual averages from the 1999 to 2020 model simulations.

Refining the dissemination channels
Lowering the price of the in vitro material is only one option to ameliorate the technology outcome and to reduce the adoption risk for smallholders. Additional activities, such as combining technology delivery with relevant extension services, will have to be carried out. However, to provide this kind of assistance in a broader scope goes beyond KARI’s mandate, and it must also be questioned whether the official government extension service has the capacity to perform this task. Rather, it would be desirable to build on rural grassroots organizations, such as church or women’s groups and Non-Governmental Organizations (NGOs), as a link between the TC laboratories and local farms. It would be useful if such groups were involved in KARI’s on-farm field trials, instead of dealing only with individual farmers as has been the case so far.
Thus, a larger number of potential information multipliers would be reached with only little additional effort by field researchers and extension workers. When satisfied with the field trial experience, the groups should be encouraged to start small-scale banana nursery enterprises. This could effectively combine the hardening and retailing of TC plantlets with a participatory transfer of the important extension message. The initial investment required for establishing a banana nursery is comparatively low. Nonetheless, some financial and technical support should be granted to the grassroots organizations in the beginning. After an initial learning phase, both the business itself and the provision of extension services could well be financed out of product sales. The possibility of involved groups managing a small-scale credit programme, especially tailored to facilitating TC technology adoption for resource-poor producers, should also be considered.
Up to now, agricultural extension services in Kenya have mostly been male-focused, even for semi-subsistence activities, which are predominantly managed by women. The explicit involvement of women farmers’ groups in on-farm field trials and technology dissemination is especially appealing because banana is largely a women’s crop. If women’s groups were to organize nurseries for TC bananas and related extension services, it is more likely that female farmers who are responsible for the bulk of maintenance work in banana enterprises, would understand the need to adjust their farming practices.
A woman-to-woman transfer of knowledge could also reduce the risk of men taking over traditional female responsibilities in the process of technology adoption, curtailing the women’s scope and freedom of decision-making.
In spite of the apparent advantages of working with women’s groups, however, efforts to create viable biotechnology distribution channels should not focus on one single approach. Some initiatives to establish local nurseries by farmers or by church groups can already be observed, and there should be enough flexibility in the project design to allow for such participatory activities and initiatives from the farmers themselves.

Conclusions
TC technology has the potential to bring about considerable aggregate welfare gains for Kenyan banana producers and consumers. The relative potential yield and income increases for the poorest farmers are even higher than those for the relatively richer and larger farms. However, due to the high expenses for the technology itself and for complementary inputs, small farms are facing the most severe adoption constraints. Providing this group with appropriate access to the technology will require further institutional efforts.
One could argue that remarkable yield and income increases could be achieved also without biotechnology, simply by applying more inputs and improving the banana cultivation habits. Yet the potential of an appropriately introduced technological innovation to bring about broader institutional innovation and a more comprehensive modernization of farming systems should not be underestimated. In connection with a tangible technological product such as TC plantlets it will be much easier, for instance, to start a new microcredit scheme or to transfer knowledge about improved management practices than it would be without such a tool, especially in situations where public rural services are underdeveloped. Furthermore technology always has to be understood in a dynamic fashion. For instance, in the long run genetically engineered bananas with durable resistance to major pests and diseases might be much more beneficial than TC for the smallholders. Acquiring TC capacity and establishing viable biotechnology dissemination channels in Kenya will facilitate the future transfer of more sophisticated innovations as soon as they become available.
The technological and institutional experiences gained in Kenya will also produce technology spillovers to other East African countries. An initiative to extend the banana biotechnology project to the neighbouring countries has already been started by KARI and ISAAA in collaboration with the Banana Research Network for East and Southern Africa (BARNESA) and with national research organizations in Uganda and Tanzania.
Matin Qaim

Center for Development Research (ZEF), University of Bonn, Walter-Flex-Str. 3, 53113 Bonn, Germany. Phone (+49) 228 73 1841; Fax (+49) 228 73 1869; E-mail mqaim@uni-bonn.de

This article is based on the study Assessing the Impact of Banana Biotechnology in Kenya, which was carried out partly with funds from the German Research Society (DFG) and the German Technical Co-operation (GTZ).

Sources
International Service for the Acquisition of Agri-biotech Applications (1997), "Yes, We will have Bananas; Rejuvenating Banana Orchards in Eastern Africa". ISAAA Annual Report 1996: Advancing Altruism in Africa. Ithaca, NY: ISAAA, pp. 33-36.

Kenya Agricultural Research Institute (1998), Biotechnology to Benefit Small Scale Banana Producers in Kenya. Annual Progress Report 1997. Nairobi: KARI.

Qaim, M. (1999), Assessing the Impact of Banana Biotechnology in Kenya. ISAAA Briefs No. 10. Ithaca, NY: ISAAA and ZEF.

Qaim, M. and von Braun, J. (1998), Crop Biotechnology in Developing Countries: A Conceptual Framework for Ex Ante Economic Analyses. Discussion Papers on Development Policy No. 3. Bonn: ZEF.