Job Market Paper
Salience, Misperceptions and Trade-offs in Agricultural Technology Adoption: Experimental Evidence from Uganda — AEA RCT Registry
Abstract
Farmers often have imperfect information about the traits of the crop varieties they grow. This paper shows that this misidentification is a first-order constraint on the adoption of nutrient-enhanced and disease-resistant varieties. In a randomized controlled trial with 2,141 sweet potato farmers across 108 Ugandan villages, I compare two ways of addressing this information imperfection. The first teaches farmers to read visual cues as signals of varietal traits, flesh color for nutritional content (vitamin A) and symptoms on plants for disease. The second adds variety-specific trait information, using DNA fingerprinting to tell farmers which traits their own varieties actually carry. Learning to read the cues improves farmers' knowledge but does not change what they plant. Variety-specific trait information raises correct identification of yellow-fleshed varieties by two-thirds and the area planted to varieties with nutrition and disease-resistance traits by 25 to 59 percent. Further evidence suggests that reading the cues is not enough because farmers rarely recognize that desirable traits co-occur, and so wrongly believe these traits trade off against one another; information that reveals which varieties combine them shifts planting. These changes in adoption raise consumption of vitamin A--rich sweet potato among children under two and cut their diarrhea incidence by 52 percent, with positive spillovers to children of a second group of farmers who received only community-level DNA results. By linking objectively measured varietal adoption to child health, the results provide causal evidence that varietal misidentification limits the returns to agricultural R&D.Working Papers
You Reap What (You Think) You Sow? Misperceptions of Improved Maize Varieties in Ethiopia — PSE Working Paper
Abstract
Adoption of improved seed varieties has the potential to substantially increase agricultural productivity. However, only 23 percent of farmers who grow an improved maize variety report doing so in Ethiopia. This paper provides causal evidence of the impact of misperceptions about improved maize varieties on farmers' production decisions, productivity, and profitability. I employ an instrumental variable approach, taking advantage of the roll-out of a governmental program that increases transparency in the seed sector. I find that farmers who correctly classify the improved maize variety they grow experience large increases in input usage (urea, NPS, labor), yields, and profits per hectare. Using machine learning techniques, I develop a model of interpolation to predict objectively measured varietal identification from farmers' self-reported data, providing a proof-of-concept for scalable approaches to obtaining reliable measures of crop varieties.Economic, Cultural and Taste-Based Constraints to the Scaling of Climate-Resilient Biofortified Maize in Guatemala — AEA RCT Registry — IDB Working Paper with Karen Macours and Jessica Rudder
Abstract
This paper analyzes the effectiveness of a series of demand and supply side interventions to increase the reach of biofortified maize seeds among diverse smallholder populations in Guatemala. We introduce experimental variation in demand-side subsidies and information designed to encourage farmers to try the biofortified climate-resilient seeds, and supply-side subsidies for agrodealers to promote the seeds to a wider customer base. Subsidizing agrodealers can leverage market incentives, while demand-side interventions are more traditional policy tools. We study the effect of these scaling approaches on short-term uptake, longer-term willingness-to-pay and sustained adoption by different types of farmers.Large-Scale Roll-Out and Impacts of Biofortified Crop Varieties in Uganda with Karen Macours and Julius Okello
Abstract
Hidden hunger has been identified as a key threat to human capital accumulation and long-term welfare in many low- and middle-income countries. This paper provides some of the first causal evidence of vitamin-A enhanced sweet potato on children's longer-term wellbeing. Results based on national-level roll-out data of sweet potato vine distribution show that exposure to orange flesh sweet potato during the earliest years of life leads to long-term gains in height-for-age scores. Evidence along the causal chain helps interpret this finding, as distribution of vines in a subcounty leads to larger presence of yellow and orange varieties in farmers' fields, to higher consumption of vitamin-A rich foods, and to lower morbidity at the age of 5.Publications
Alemu, S., Kosmowski, F., Stevenson, J., Mallia, P., Taye, L., Macours, K. (2024). Studying inclusive innovation with the right data: An empirical illustration from Ethiopia. Agricultural Systems, 219, 103988. Link
Work in Progress
Business Training and Its Effects on Firm Growth by Gender with Paul Winters and Alejandro Estefan
Scaling Climate-Resilient Biofortified Maize through Supply- and Demand-Side Interventions in Guatemala with Karen Macours and Jessica Rudder
Mechanistic Explanations, Social Learning, and Technology Adoption: Evidence from Organic Fertilizers in Kenya *with Aniruth Sankar, Jessica Rudder, Robbie Dulin, Tomoko Harigaya, Frank Odhiambo and Abraham Salomon