Ex-Post Equilibrium Market Recommendations: Implications for Indian Agriculture

Abstract

Problem Definition: We study the feasibility and benefits of providing ex-post equilibria recommendations in markets where sellers face uncertainty about the price they will receive for their products.

Academic/Practical Relevance: Prior research has shown that providing sellers with price information can lead to reductions in geographic price dispersion and associated improvements in welfare. Sellers' inability to coordinate, however, often leads to market failures. We provide an algorithmic solution that a market planner can use to recommend equilibria to sellers. We show that this leads to further improved seller welfare and price dispersion.

Methodology: We model sellers' market selection decision and demonstrate that a Nash equilibrium may not exist in this game, but show that such instances are exceedingly rare. We present an algorithm to compute an equilibrium and use a parameterized numerical analysis using data on the Indian agricultural market to compare our algorithm-based recommendations to a set of natural market-selection strategies.

Results: We find that our algorithm typically leads to higher farmer welfare and lower geographic price dispersion. Furthermore, in problems where the farmer welfare maximizing strategy can be calculated, our approach captures 99.85% of the optimal farmer welfare. The results are robust to changes in market elasticities and farmer volumes.

Managerial Implications: Managers can supplement price information services with ex-post equilibrium market recommendations with little overhead, leading to increased welfare and reduced price dispersion. Since the benefits are not restricted to the participating farmers, policy makers may wish to encourage farmers to connect with services that provide such recommendations.

Keywords: Service Operations; Empirical Research; Game Theory; OM-Information Technology Interface