This article uses Nordhaus' framework to determine the efficiency of the revision process for USDA corn and soybean production forecasts over the 1970/1971 through 2004/2005 marketing years. Positive autocorrelation and consistency of directional changes in forecast revisions suggest these forecasts are “smoothed.” Evidence is provided that the loss in forecast accuracy due to smoothing is statistically and economically significant in several cases. A conservative bias in farm operators' assessments of yield potential and in the procedure for translating enumerator's information about plant fruit counts into objective yield estimates are identified as plausible sources of smoothing. Copyright 2006, Oxford University Press.