rain and soybean producers who want to hedge their crops prior to harvest must G do so without complete knowledge as to the yields that will be obtained at harvest. Hedges in excess of realized production may result in revenues different from those expected, even when basis expectations are fulfilled. As a result, rules of thumb such as "hedge only one-half to two-thirds of expected production" have become commonplace.
Previous research (e.g., has provided models which can estimate the hedging ratio (the proportion of expected production to be hedged) given expectations about prices and yields. Although these models were developed to accomodate pre-harvest hedging by producers, on at least two counts they have not been tested adequately for individual producer use.
Previous estimates of hedging ratios have generally been based on yield data aggregated beyond the individual farm level. Aggregative yields obscure the localized vagaries of weather, pests, etc. experienced by individual farmers. have shown that hedging ratios estimated using icdividual farm yield data can differ markedly from ratios estimated using county data. However, the lack of a sufficiently long yield history may preclude the use of individual farm yield data in hedging ratio estimation. The consequences of individual farmers using hedging ratio recommendations based on aggregative yield data are unknown.
Previous research has been confined to the estimation of hedging ratios over some historical time period. That research has not provided any information as to the outcomes that producers using the estimated hedging ratios would have realized in subsequent time periods. That is, the revenue levels and variability of revenue that producers would have realized using hedging ratios estimated using information available at the decision making time (i.e., ex ante) have not been evaluated.
The objective of this paper is to compare the out-of-sample outcomes realized by individual soybean producers using:
(a) hedging ratios estimated ex ante from individual farm data, (b) hedging ratios estimated ex ante from county data, (c) the rule of thumb recommedation to hedge one-half to two-thirds of expected production, and (d) no hedging.