Risk aversion as effort incentive: A correction and prima facie test of the moral hazard theory of share tenancy

We show that Stiglitz's (1974) principal-agency theory of share tenancy does not imply, as alleged, that the optimal tenant share is less than one for risk-averse tenants nor that the share decreases monotonically with tenant risk aversion. Tenants may self insure by working harder increasingly so for higher levels of risk aversion with the result that the more risk averse work for higher shares. When the model is parameterized based on previous studies of Philippine agriculture, it predicts a U-shaped relationship between optimal tenant''s share and risk aversion. Landlords choose rent contracts for both high and low levels of risk aversion and shares from 80-99% for intermediate levels. In contrast, actual shares in the study area ranged from 50-60%, with most farmers contracted on a 50:50 basis. We conclude that rent contracts must have additional disadvantages and/or share tenancy additional benefits that are not accounted for in the static principal-agency theory.