Princeton University Library Catalog


Gastfriend, Daniel [Browse]
Senior thesis
Deaton, Angus [Browse]
Woodrow Wilson School of Public and International Affairs [Browse]
Class year:
134 pages
Restrictions note:
Walk-in Access. This thesis can only be viewed on computer terminals at the Mudd Manuscript Library.
Summary note:
International organizations and governments throughout the developing world have long provided agricultural extension services to farmers with the hopes of raising agricultural productivity and reducing rural poverty. These programs train farmers in modern agronomic techniques through instruction and demonstration, sometimes in conjunction with strategies to improve access to credit, agricultural inputs, supply chains and markets. However, these programs typically face a set of serious challenges which threaten their effectiveness. Key among these is the difficulty of effectively measuring program impacts, which complicates efforts to identify and learn from mistakes. This thesis first discusses the theory and history of agricultural extension, exploring its economic justifications and demonstrating factors behind previous failures. I argue that mutually reinforcing information and credit market failures can prevent the private sector from providing farmers with Pareto-improving agronomic technologies in low-income environments. In these cases, information networks may disseminate applicable techniques among farmers on their own accord, but high levels of risk aversion, difficult to identify technological impacts, and fragmented social networks can slow this process. Whether public intervention will produce a better outcome depends not only on logistical factors but also the extent to which programs take these theoretical implications into account. Failure to do so, in conjunction with the organizational challenges associated with low-income rural contexts, can explain many failures of previous approaches. I explore these themes through an in-depth evaluation of a Clinton Development Initiative program in Malawi, which uses an innovative model to provide extension services and facilitate access to credit for smallholder farmers. Applying a difference-in-differences framework, I assess whether the program positively impacted the knowledge and yields of participant farmers and consider implications for the theory, organization, and evaluation of extension services. The evidence indicates that individual participants had vastly different experiences, with some perceiving real benefits and others experiencing harm. I evaluate various sources of error and demonstrate the need for repeated follow-up surveys to determine the validity of results. I conclude by assessing implications and offering recommendations for similar programs. Critically, organizations must institutionalize methods of discerning whether trainings meet the economic criteria justifying public intervention and target the needs of farmers themselves. Furthermore, I demonstrate that logistical challenges, which are common in these environments, can harm participant farmers and must therefore be seen as core risks of extension and rural credit programs. Finally, I argue that rigorous monitoring and evaluation are essential for extension operations and recommend several models for doing so.