Does it Pay to be an Optimist?
This paper develops an equilibrium model of a market, where a pessimist, an optimist, and a pragmatist trade option portfolios with each other. Incomplete markets allow them to form different opinions on the underlying distributions, while fundamental prices remain within bid ask bounds. An application of the model to S&P 500 options data shows that the agents use different and surprisingly small representative state spaces of the underlying. The pessimist and the pragmatist sell variance swaps, commonly perceived to be an insurance instrument, rather than the optimist. Robustly to changes in the input parameters and variations in the setup, the pessimist is the most successful agent, while the optimist loses consistently. Belief dispersion predicts actual trading volume and open interest.