This study compares the effectiveness of two crop insurance and two disaster assistance program designs used in conjunction with a government commodity program and a linked crop insurance/government commodity program design. Stochastic dominance analysis of farm‐level net return distributions is used to select the preferred design(s). The results indicate that the disaster assistance programs are preferred over the alternatives. The results also suggest that individual crop insurance is preferred to area crop insurance. A subsidy is required for risk‐averse managers to prefer area crop insurance to individual crop insurance.

Stochastic dominance analysis of two tillage systems, conventional tillage and no-tillage, for five crop rotations, wheat-fallow, grain sorghum-fallow, continuous wheat, continuous grain sorghum, and wheat-grain sorghum-fallow, shows that riskaverse managers prefer a conventional tillage wheat-sorghum-fallow system. Small changes in production costs or yields lead to indifference between this system and the no-tillage wheat-sorghum-fallow and no-till and conventional wheat-fallow systems. Participation in the basic government commodity program generally increases average net returns and lowers variation of returns. Government commodity program payments calculated under a variety of scenarios do not generally encourage the use of no-till practices for grain sorghum and wheat in the central Great Plains.

This study evaluates the Adjusted Gross Revenue-Lite (AGR-Lite) whole-farm adjusted gross revenue insurance program on net farm income risk using panel data from 49 southeast Kansas beef farms. On average for the group, but not each individual farm, AGR-Lite reduces the mean and standard deviation of net farm income, raises the average minimum, and lowers the average maximum observations of the net income distribution. Thirty-four farms (69%) received at least one indemnity payment. Stochastic efficiency with respect to a function reveals that AGR-Lite is preferred by 18 of the farm managers (37%) when an upper bound on the risk-aversion coefficient is used.