Agriculture and poverty in commodity dependent African countries: a rural household perspective from the United Republic of Tanzania

This report explores how farm productivity affects poverty, and how various factor market constraints affect farm productivity. The empirical analysis draws on representative surveys of farm households in Kilimanjaro and Ruvuma, two cash crop growing regions in Tanzania. Poorer households were found not only to possess fewer assets, but also to be much less productive. Findings show that agricultural productivity directly affects household consumption and hence overall poverty and welfare. Stochastic production frontier analysis indicates that many farmers are farming well below best practice in the region. Holding inputs constant, they attain on average only 60 percent of the output obtained by their best counterparts. Analysis of allocative efficiency suggests that family labour is substantially overutilized, a sign of considerable excess labour supply. Use of intermediate inputs on the other hand is well below what is commensurate with the estimated value of their marginal productivities. An important reason for low input use is lack of credit to purchase inputs, but difficult access to the inputs themselves and being connected to the economy more broadly are also important impediments. Easy access to credit is positively associated with being a member of a savings association or being in a contractual arrangement with a cooperative or firm. Irrigation infrastructure facilitates access to credit. Together these findings support a continuing emphasis on increasing agricultural productivity in designing poverty reduction policies. Better agronomic practices and increased input use will be crucial in this strategy. Better access to inputs and improved roads and transport services will further help boost input application. Financial constraints might be relieved through fostering institutional arrangements facilitating contract enforcement (e.g. contract farming, marketing cooperatives) and institutions that facilitate saving by the households themselves. They may also be relieved by the provision of more adequate consumption safety nets. The overall results suggest that a pro-poor rural development strategy needs to be anchored around improvements in agricultural productivit