The goal of this thesis is to determine which agronomic and market interventions maximize the welfare of cocoa farmers in West Africa, using Ghana as the key case study. First I look at methods to expand yields using econometric analysis of data on 200 cocoa farming households. Results suggest that the best way to increase cocoa production is by promoting fertilizer use. The Ghanaian government's CODAPEC spraying program, and access to extension services were also found to have a positive effect on yields. However, due to cocoa pricing trends and market structures, increasing yields alone is not likely a sustainable way to improve farmer incomes, and it is important to consider other measures, like vertical integration into cocoa processing. Simulations run on a model of total Ghanaian welfare derived from cocoa surprisingly suggest that under current conditions Ghana should export 100% of beans in raw form. However, if a higher percentage of the industry were in the hands of Ghanaian interests, then it would be welfare optimizing to process more beans domestically. Potential policies to promote such a situation include differential incentives for Ghanaian versus foreign processing firms and a marketing for Ghanaian processed cocoa. Another option is to change the law on cocoa purchasing by the government such that Kuapa Kokoo, Ghana's largest cooperative and the only one truly owned by farmers, can process and export their own cocoa directly. Such a move would likely have the highest and longest-lasting impact on the welfare of Ghanaian cocoa farmers.