Adoption of agricultural innovations by smallholder farmers in the context of HIV/AIDS: the case of tissue-cultured banana in Kenya

Agricultural technology is known to be a catalyst for agricultural development and rural poverty reduction through increases in food production and/or reduction in production costs. Various government policy and strategic documents emphasize the important role of the agricultural sector as the leading driver for development in Kenya. The recently launched Economic Recovery Strategy for Wealth and Employment Creation (ERS Republic of Kenya 2003) identifies the adoption of appropriate agricultural technologies and practices as one of the imperative themes that will generate the surplus needed to feed the increasing population and to propel economic growth. However, the role of agricultural technology in poverty reduction is currently being played out in an increasingly multifaceted environment, featuring the growing complexity of farming rural households' livelihood strategies and the effects of HIV/AIDS on household assets and livelihood strategies. The current challenge facing agricultural research in countries threatened by HIV/AIDS is to develop technologies that meet the evolving challenges of HIV/AIDS-affected farming households without compromising productivity and sustainability of their livelihoods.This study investigated the suitability of the tissue-cultured banana technology in rural farming households in Central Kenya in the context of HIV/AIDS. The study adopts a livelihood approach and provides detailed information on farming household assets and livelihood strategies. It examines the effects of HIV/AIDS on household assets and compares the consequent livelihood strategies undertaken in HIV/AIDS-affected and non-affected farming households. The study evaluates in detail banana farming as one of the livelihood strategies and assesses the significance of the tissue-cultured banana technology for the livelihood of the farming households. The effects of HIV/AIDS on the local extension services and how this influences the adoption of tissue cultured-banana technology adoption is investigated. A gender perspective of access to assets, HIV/AIDS impacts, livelihood activities and outcomes is integrated. The sample population was selected on the basis of use or non-use of the tissue-cultured banana among banana farming households. Within each of these samples, both HIV/AIDS-affected and non-affected households were selected. The data used in the analysis were collected through a mixed-method approach incorporating quantitative and qualitative data. The quantitative data was collected through a formal survey whereas qualitative data was collected through in-depth interviews, focus group discussions and key informant interviews. The study was conducted in Maragua district, which is a banana-growing region in Central Kenya where the tissue-cultured banana has been introduced. The study elaborates the following research questions:1. How do assets available to farming households influence their livelihood options, activities and outcomes?2. What are the impacts of HIV/AIDS on farming households' assets and consequently livelihood options, activities and outcomes?3. What is the role of the tissue-cultured banana in the livelihood activities and outcomes of farming households?Because farming households differ in their asset endowment, they undertake differing livelihood strategies and respond differently to both shocks and development interventions. To capture this heterogeneity, by means of factor and cluster analysis, a typology according to resource endowment was constructed yielding three types of farming households: low-, medium- and high-resource endowment farming households using factor and cluster analysis. The classification was based on households' asset endowment in terms of human capital (age, sex, education level of the household head and household size); natural capital (size of land); financial capital (savings and access to credit); physical capital (livestock, farm equipment and personal household item value); and social capital (membership in community organization).The low resource endowment category is composed of a high proportion of elderly (65+) female-headed households with a low level of education of the household head and a high demographic dependency ratio. Households in this category have small pieces of land inherited from their parents, and half of them have no title deeds for the land. The households have no access to formal credit and no savings. They have low physical capital in terms of livestock, farm equipment and household items. They are producing mainly for own consumption and earn extra income by selling their own labour in agricultural activities.The medium resource endowment group is composed almost entirely of male-headed households who have relative high education level. They have large pieces of land (on average>0.8 ha) with title deeds, have some savings and access to both formal and informal credit. They have more livestock, farm equipment and personal household items than the type one households. They mostly produce for own consumption and also for the market. They also earn extra income through formal and non-formal employment and are engaged in trading, service provision. They receive remittances from migrant household members.The high resource endowment farming households have of educated male household heads with large land size which is registered with title deeds. They have access to informal credit but do not have savings. They have the highest livestock and farm equipment value. The majority of households in this group do not take on other income diversification activities and are mostly engaged in production of banana for the market and maize and beans for home consumption. They have the largest proportion of migrants and receive more remittances than the other two categories of households.Generally, there is a high level of social capital in the study area as indicated by the large number of households involved in community organizations. The community groups existing in the area are: farming groups, rotating savings-and-credit associations (ROSCAs), burial societies, village committees, clan-based groups, religious groups and HIV/AIDS-associated groups. In general, households in medium and high resource endowment categories have higher number of household members engaged in groups than in the low resource endowment category. However, these groups may not necessarily lead to economic prosperity as most of them are homogenous and hardly connected to outside resources.Despite the differences in assets endowment among the farming households, no significant differences were observed in their incidence of HIV/AIDS. However, differences were observed in asset endowment and household characteristics between HIV/AIDS-affected and non-affected farming households. Comparing affected and non- affected farming households showed that HIV/AIDS-affected households are mostly female-headed, have a significantly higher dependency ratio, and experience a greater shortage of labour despite their larger household size. Land sale, a commonly quoted strategy for household labour loss related to HIV/AIDS is rare in the study area. This is possibly due to cultural beliefs associated with land which deter the sale of ancestral land. Lack of transferability rights (as more than half of the households in sample population have no title deeds to their land) could also be a deterrence factor of land sale. Affected households have adopted various labour coping strategies such as labour re-allocation, hiring labour and bringing in relatives. Some affected households have stopped growing labour-intensive vegetable crops while some have altogether abandoned their land. To cater for the high direct and indirect costs, HIV/AIDS-affected households have various sources of income. These include borrowing from informal sources (ROSCAS, relatives and extended family members) and they also receive remittances from migrant household members. Membership in ROSCAS is crucial as it enables households to access informal credit. In addition to farming, HIV/AIDS-affected households have diversified income sources in the form of rent from land and they engage in labour migration.The adoption of tissue-cultured banana is positively related to financial and physical capitals. The high and medium resource endowment households that have financial capital in form of savings and access to credit are more likely to grow tissue-cultured banana. This is primarily because of the high cost of the tissue-cultured plantlets. Furthermore, the results suggest that affected households are most likely not to adopt the tissue-cultured banana technology. Thus, for the technology to benefit resource-poor farming households who constitute the majority of the sample population, efforts must be made to increase their access to financial capital or lower the cost of tissue-cultured plantlets.Increased farm output resulting from technology adoption is expected to result in various outcomes such as increased household food security, increased income through the sale of extra produce, and reinvestment into household activities. Both HIV/AIDS-affected and non-affected households growing tissue-cultured banana reported an increase in banana production, income and food, which they attributed to the adoption of the technology. However, HIV/AIDS-affected reported higher increases in production, income and food supply as a result of growing tissue-cultured banana.Although the area extension services have not directly suffered the attrition caused by HIV/AIDS, they are lacking skills, finances and human capacity in providing appropriate services to farming households and in particular the HIV/AIDS-affected households. Not only will more extension workers be needed, but contents of extension services also need to change in order to be responsive to the AIDS epidemic. Extension services need to cater to the knowledge needs of women, the elderly and the young.Female-headed households were found to be disadvantaged in several ways. The majority of them are in the low resource endowment farming household category. Firstly, female household heads have significantly lower education level than male-headed households. Secondly, their financial capital base is also low because they have no savings and their access to informal credit provided by community organisation is limited, having few household members engaged in these groups. In addition, female-headed households have significantly lower physical capital in terms of household assets which they could sell and get cash in times of crisis. Slightly over half of the female-headed households are HIV/AIDS-affected. These households have a significantly higher dependency ratio and a higher incidence of labour shortage than non-affected households despite their larger size. Renting out of land a strategy undertaken by HIV/AIDS-affected households is not common in female-headed households due to lack of transferability rights. Some of the household members have therefore opted to migrate to urban areas in search of employment, a strategy that further puts the households at risk of HIV/AIDS.In conclusion, policy-makers and development agents targeting agricultural technology development for food security and poverty reduction should take into account the diversity of farming households. Farming household capabilities, assets and activities should play a major role in shaping policy on agricultural technology development. Labour-saving technologies may indeed be appropriate for many households, especially female-headed HIV/AID-affected households that lack cash for hiring labour. However, agricultural research should focus on developing low risk technologies in terms of financial requirement that can assure farming household food security, as well as cash income to pay for school fees and basic necessities.