Risk attitude and risk perception in agroforestry decisions : the case of Babati, Tanzania

Increasing risks and uncertainties related to stochastic agro-ecological and institutional factors on one hand, and deterioration of land due to unsustainable farming on the other hand, are among the major constraints to agricultural development in the developing countries. Agroforestry is proposed as a land use option and a measure to reduce the above problems mainly because farmers' have to derive their basic needs of food, fibre, shelter and fuelwood predominantly from the land they own. The main question is how and to what extent farmers use agroforestry to reduce risk and derive their major basic needs?This research was conducted in Babati district located in the southwestern part of Arusha region in the northern part of Tanzania. The research villages were Magugu, Bonga, Singe and Himiti. The main objective of the research was to set an understanding of farmers' attitude and perception towards risk in making agroforestry production decisions.The data collection procedure involved three major surveys: a preliminary survey of 20 farmers, a single visit general household questionnaire survey with a sample of 100 farmers and a detailed research approach involving 30 farmers.The data obtained were analysed using the utility theory and the latent variable categories. In addition, the estimation of probability distribution functions was done using the maximum likelihood technique, the utility functions through linear/non-linear regression analysis and the relationship between risk attitude, risk perception and preference ranking of cropping systems with household resources and characteristics was done using linear regression analysis. The main features of the results are as outlined below.Farmers' sources of risk and risk management strategies are identified both in agroforestry and in sole cropping. Generally, there were similarities in perceived risk and risk management strategies among the sampled villages.Using the strength of conviction method, the cropping systems were classified into low risky, medium risky and high risky. Heterogeneity in perceived risk was noted between males and females and among farmer categories. There is a general indication that production of food crops and a mixture of food crops and trees are perceived as less risky as compared to the others.Risk attitude measures were derived using the utility theory approach. Generally, there were variations in risk attitude measures among the farmers and the situations analyzed. Risk attitude is influenced by wealth of the farmer, years of education, household size and the age of the respondent.Risk attitude using the latent variable category was done using factor analysis. The results show that a substantial number of farmers have a positive attitude towards risk and land resource conservation. However, the attitude towards commercialisation was low, maybe due to poor transport and infrastructure. The attitude towards land resource conservation, the attitude towards commercialisation, the wealth of the respondents and their education significantly explain the positive attitude towards risk.Farmers showed a high preference for agroforestry systems namely 'Trees (timber/fuel wood all ages) + mixed food crops' (CS2), followed by food crop systems 'Mixed food crops only' (CS3). Preference ranking of cropping systems is influenced by risk attitude, expected income, risk perception and household resources and characteristicsUsing the mean-variance analysis and a quadratic utility function, cropping systems were ranked in order of preference. The ranking was more or less in conformity with the actual cropping systems practised by the farmers.Recommendations were made based on methodology, areas for further research and policy interventions. Policy recommendations were made based on the finding that farmers use agroforestry as part of their risk management strategies. As a result policy recommendations were made towards the role of agroforestry in risk management strategies such as improvement in infrastructure and socio-economic environment (roads, extension and land tenure) and recommendations on technology development in agroforestry through breeding and selection of crops and tree species for specific suitable characteristics (such as drought tolerance, short maturity, and disease resistance).