Imperfect information and learning: Evidence from cotton cultivation in Pakistan

Information problems are pervasive in developing economies and can hinder productivity growth. This paper studies how much rural producers in developing countries can learn from their own experience to redress important information gaps. It builds a model of learning from experience and applies it using a rich dataset on cotton farmers in Pakistan. I test whether farmers learn from cultivation experience about the pest resistance of their seeds and use this information to improve selection and productivity. I find no such learning effect and this conclusion is robust to several parameters that could signal learning. The findings document the difficulty of parsing out and processing information from cultivation experience alone and point to the importance of information provision to producers by the government or external agencies.