Tanzania: A Hierarchical Cluster Analysis Approach

Abstract

 It is common for researchers and rural development policy stakeholders to describe smallholder farmers as a homogeneous group in terms of their demand for farm credit and farm investment behaviour.  Given the diversity of factors such as farm credit products (input credit in cash, input credit in kind), farming systems (extensive Vs intensive farming, food crop Vs traditional cash crop production, crop production Vs livestock keeping), asset endowment, income sources and experience in farm credit borrowing, it is obvious that the demand for farm credit and use with which it is put are also diverse among farmers. Using survey data from Kibondo district, west Tanzania, we use hierarchical cluster analysis to classify borrower farmers according to their borrowing behaviour into four distinctive clusters. The appreciation of the existence of heterogeneous farmer clusters is vital in forging credit delivery policies that are not only appropriate for particular categories of farmers but also that do provide potential for reducing supply side transaction risks and costs.