Evaluating Farm Policy Changes Across Commodities: A Stochastic Simulation Approach

Since the 1930’s, agricultural policy has played a major role in stabilizing the United States agricultural economy. Agricultural policy is a minor player in terms of overall federal budgetary impact, but the programs remain significant for US producers. Understanding how changes to farm safety net programs would affect government expenditures would be beneficial in maintaining and refining agricultural support programs into the future. The objective of the project was to create a model that uses stochastic simulation to estimate expenditures or score agricultural policy changes, for 9 major agricultural commodities. The study used many of the factors that the federal government uses to accomplish the same goal including calculating payments based on a national scale. This was intentional because estimates for this project were expected to approximate the scores achieved by the Congressional Budget Office. As part of the analysis, a State based model was created as an alternative approach to scoring the main agricultural support programs. This alternative was compared to the original approach to determine differences between the two in terms of effectiveness and detail provided when estimating future program payments. This research provided models that score the programs as they are currently, but allows for modifications to be made on various parameters. This has provided – and will continue to provide – the ability to evaluate effects of proposed program changes on projected program payments. The results of the study indicate the national model scored projected program payments more accurately than the state model structure. However, after both models were adjusted to limit the difference in the PLC projected payments and CBO, the state model preformed more closely to CBO. The models are a useful tool for current analysis of ARC and PLC program expectations. Even more valuable though, is the potential to analyze future proposals to these programs. With that in mind, the national model is far more seamless to adjust program parameters, and the state model enables the regional effects of program changes to be seen