Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Agricultural technologies typically spread as farmers learn about profitability through social networks. This process can be nuanced, however, when net returns for some farmers may not be positive. We investigate how social learning influences demand for a resource-conserving technology in eastern Uttar Pradesh, India. We identify potential adopters through an experimental auction and randomly select a subset to adopt. We exploit this variation in adoption across networks to estimate network effects on demand for the technology one year later using a second auction. Technology benefits vary, and network effects are completely conditional on benefits. Having a benefiting adopter in one's network increased demand by over 50%, whereas having a non-benefiting adopter had no effect. These effects are strong enough to bring average demand in line with expected benefits. For many farmers, however, demand remains below the market price, suggesting that network effects will lead to increased—but not rapid widespread—adoption.