Forecasts of U.S. ending stocks for corn, soybean, and wheat issued by the USDA and private analysts are evaluated. The null hypothesis that USDA forecasts are unbiased cannot be rejected. The null hypothesis that private analysts' forecasts are unbiased forecasts of USDA forecasts cannot be rejected either. However, there is strong evidence that the forecasts issued by the USDA and private analysts are inefficient. Overall, the precision of the USDA forecasts and the private analysts' forecasts is highest for wheat and lowest for soybean.