Complementarities of Training, Technology, and Credit in Smallholder Agriculture: Impact, Sustainability, and Policy for Scaling-up in Uganda
How to cite
SMITH, STEPHEN C; Bobic, Vida. 2019. "Complementarities of Training, Technology, and Credit in Smallholder Agriculture: Impact, Sustainability, and Policy for Scaling-up in Uganda." AEA RCT Registry. AEARCTR-0004022. https://doi.org/10.1257/rct.4022-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
Many development programs are short-term interventions, either because of external funding constraints or an assumption of impact sustainability. Using a novel randomized phase-out research method, we provide experimental tests program phase-out effects of an extension program designed for women smallholder farmers in Uganda. We find that program phase-out does not diminish demand for improved seeds, as farmers shift purchases from NGO-sponsored village supply networks to market sources, indicating persistent learning effects. We find no evidence of declines in improved cultivation practices taught by the program. These results have implications for both efficient program design and for models of technology adoption.
Intervention
Many development programs are short-term interventions, either because of external funding constraints or an assumption of impact sustainability. Using a novel randomized phase-out research method, we provide experimental tests program phase-out effects of an extension program designed for women smallholder farmers in Uganda. In particular we examine whether program phase-out diminishes demand for improved seeds, and or leads to changes in the source of seeds such as from the NGO-sponsored village supply networks to other market sources. In this, we look for the presence of persistent learning effects. We also study whether the phase-out leads to declines in improved cultivation techniques and practices taught by the program.
Experimental design
Four years after the start of the agricultural extension and input supply program, it was discontinued in a randomly selected subsample of the treatment population. There are three treatment groups: Continuation (which continued to receive treatment), Model Farmer phaseout (where the Model Farmer component of the program was discontinued first, and the CAP component one year later), and CAP phaseout (where the Community Agriculture Promoter component was discontinued first, followed by the MF component a year later). The experiment covers 15 BRAC branches (broadly corresponding to Ugandan districts) in the Eastern Region of Uganda; the experiment was stratified at the branch level. We use a clustered design, where a cluster is one village or a geographically proximate pair of villages in which both components of the program were active in the year preceding phaseout (2013). There are 99 clusters in total: 32 in Continuation, 33 in Model Farmer Phaseout, and 34 in CAP Phaseout. In addition to a baseline survey in 2013, there were three follow-up surveys: one agricultural season after phaseout started (while both Phaseout treatment arms still had one active component of the program), three seasons (1.5 years) after phaseout (after both Phaseout treatment arms had both types of treatment discontinued), and finally six seasons (three years) after phaseout.
Design details
Cited by
No citation data available for this registration yet.