The Adoption and Impacts of Improved Storage Technology Among Smallholder Farm Households in Uganda

Oluwatoba J Omotilewa, Purdue University


This dissertation comprises of three essays that address the adoption and impacts of grain hermetic (airtight) storage bags to reduce on-farm grain storage losses in Uganda. To understand adoption of the technology and its impacts, I conducted two levels of randomized field experiments in 2015: First, village-level cluster-randomized controlled trial; and second, household-level individual randomized controlled trial. Prior to these randomized controlled trials (RCTs), I conducted a baseline survey to understand current agricultural production and post-production practices in 2014 among smallholders, using a multi-level stratified sampling strategy. In 2016, I conducted a follow-up post-intervention survey to estimate the impacts from these randomized experiments. Broadly, each essay stands as an independent study addressing different research questions, and sometimes, different methodologies to address these questions. In essay 1, I focus on the role of information through large-scale extension efforts in the awareness and adoption of an improved storage technology—hermetic (airtight) storage bags—at the early stage in the diffusion process. Using a cluster-randomized controlled experiment implemented among 1,200 smallholders in 48 villages, I test whether the following all affect the likelihood of awareness and adoption among smallholders: 1) living in a village randomly assigned to receive extension activities, 2) participation in such extension activities conditional on being in a village randomly assigned extension activities, and 3) non-participation in an extension activity by eligible households. Overall, I find that on average, households in clusters that received extension activities are significantly more likely to be aware and adopt the technology. The likelihood of adoption further increases significantly if households participate in an extension activity. However, with regards to adoption of the technology among non-participating but eligible households, I find little evidence of informal diffusion. This finding suggests that more efforts on social learning may be needed to sustain adoption and continued diffusion of the technology beyond the extension activities. I conclude it might be cost effective to use alternative mode of information, such as combining extension activities with social learning, to increase both adoption and sustained diffusion of the technology. In essay 2, I used household-level RCT to estimate if access to an improved storage technology affects smallholder households’ use of modern inputs such as higher-yielding maize varieties and inorganic fertilizer. I further examined if access to hermetic storage bags 1) improves food security for smallholders through increased duration of storage for grains and reduction of on-farm storage losses, and 2) reduces the use of storage chemicals to preserve grains. After two seasons, results show that households exposed to the technology were 10 percentage points more likely to plant higher-yielding hybrid maize varieties that are known to be more susceptible to insect pests in storage than traditional lower-yielding varieties. Treated smallholders also stored maize for a longer period, reported a substantial drop in storage losses, and were less likely to use storage chemicals than untreated cohorts. These results indicate that policies to promote softer kernel higher-yielding hybrid maize varieties in sub-Saharan Africa should consider an improvement in post-harvest storage as a complementary intervention to increase adoption of these varieties. In essay3, I estimate the effects of a one-time subsidy in the form of free hermetic bags on commercial market participation for the bags among smallholders. I further estimate spillover effects of this one-time subsidy on market participation outcomes for exposed households within treatment villages, but who did not receive the subsidy intervention themselves. The empirical results show that on average, subsidized households are 5.2 percentage points more likely to buy an additional bag at commercial prices relative to the households with no subsidy who are equally aware of the technology. This suggests that under certain circumstances, such as when there is uncertainty about the effectiveness of a new agricultural technology, and the private sector market for the technology is weak or nascent, a one-time use of subsidy as a means to build awareness and reduce risk could help generate demand for the new technology. In this context, a subsidy can allow farmers to experiment with the technology and learn from the experience before investing in it. Lastly, I find a positive and significant 3.4 percentage points spillover effects in commercial participation of the technology among households who lived within the intervention villages but did not receive the subsidy treatment.




Ricker-Gilbert, Purdue University.

Subject Area

Agricultural economics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server