Determinants of low loan repayment on cotton inputs in Bariadi district, Tanzania

Input loan plays a vital role on economic transformations to cotton farming households. However, their acquisition and repayment are characterized by numerous challenges, including high defaults among beneficiaries. This study intended to examine factors contributing to low loan repayment on cotton inputs; and sought to improve its supply to farmers in Bariadi district. Primary and secondary data were collected from farmers and ginners respectively. A purposive and two stage stratified sampling was used to select four wards which received input loans and get a total of 112 respondents, with 70 defaulter and 42 non-defaulter farmers depending on proportions of default for 2011/12 cropping season. Descriptive t- and chi-square statistics were used to describe demographic, farming, socioeconomic, inputs credit requisition, use and repayment characteristics. Results showed significant differences between defaulters and non- defaulters with respect to age, marital status, education, family size, experience in farming, land area under cotton, access to extension, input supplier company, sufficiency of inputs, other sources of loan, perception, training and conditions imposed on loan repayment. Probit model was adopted to determine confounding factors to low input loan repayment among cotton farmers. Results showed that seed cotton harvested (-), type of input (+), input supplier company (+), timeliness delivery of inputs (+), perception on input loan (-), training on loan repayment (-), interval on loan collection (-) and conditions on input loan repayment (-) were statistically significant at p<0.001, p<0.01 and p<0.1. To enhance loan recovery among cotton farmers’; policies efficient on increasing productivity, reviewing beneficiary groups, effective training, close monitoring and development of risk mitigation strategies on repayment conditions should be advocated. However, further disbursement of loan should be targeted to farmers who areiii more likely to moderate the desired levels for each significant variable to attain sustainability on input supply.