International donors have spent billions of dollars over the past four decades in developing and/or reforming the agricultural extension service delivery arrangements in developing countries. However, many of these reforms, supported through short-term projects, became unsustainable once aid funding had ceased. The unavailability of recurrent funding has predominantly been highlighted in the literature as the key reason for this undesirable outcome, while little has been written about institutional factors. The purpose of this article is to examine the usefulness of taking an institutional perspective in explaining the unsustainability of donor-supported extension reforms and derive lessons for improvement. Using a framework drawn from the school of institutionalism in a Bangladeshi case study, we have found that a reform becomes unsustainable because of poor demands for extension information and advice; missing, weak, incongruent, and perverse institutional frameworks governing the exchange of extension goods (services); and a lack of institutional learning and change during the reform process. Accordingly, we have argued that strategies for sustainable extension reforms should move beyond financial considerations and include such measures as making extension goods (services) more tangible and monetary in nature, commissioning in-depth studies to learn about local institutions, crafting new institutions and/or reforming the weak and perverse institutions prevailing in developing countries. We emphasize the need to address three categories of institutions – regulative, normative, and cultural-cognitive – and call for an alignment among them. We further argue that, in order to be sustainable, a reform should take a systemic approach in institutional capacity building and, for this to be possible, adopt a long-term program approach, as opposed to a short-term project approach.
Keywords: agricultural extension, reform, sustainability, recurrent costs, institutional analysis, donor, project, Bangladesh