Those forecasts are based on a large survey of producers and objective yield data gathered in the largest production states.

The first production forecasts are always very important and anticipated with a high degree of interest. Actual production can deviate substantially from the August forecasts, but those forecasts provide important benchmarks for judging the actual size of the crops. They are starting points from which the market evaluates the impact of subsequent weather and crop developments. While always important, the August forecasts take on additional importance this year.

For corn, the size of the 2010 crop has important price implications for two reasons. First, inventories of old crop corn at the beginning of the 2010-11 marketing year are expected to be relatively small. The USDA’s June stocks estimate created some uncertainty about the size of the 2009 crop, the rate of domestic consumption, and prospects for year ending stocks. Still, those stocks are likely to be less than 1.5 billion bushels, or about 11 per cent of projected consumption. Second, it appears that demand for US corn will be very strong during the 2010-11 marketing year. Domestically, ethanol production is expected to continue to expand. At a minimum, growth will be in line with the mandated levels of biofuels production. Growth beyond that will depend on the fate of the blenders tax credit that expires at the end of 2010 and on the EPA decision about increasing the blend rate. Export demand for US corn may be enhanced by a much smaller wheat crop in the rest of world, led by declines in Canada, Russia, and Kazakhstan. While current stocks of wheat are large, smaller crops this year and the resulting higher prices of wheat may result in an increase in demand for corn. Exports may get a further boost if China needs to import more corn.

For soybeans, the size of the 2010 US crop may have slightly less importance than the size of the corn crop. Stocks at the beginning of the 2010-11 marketing year will be small, currently forecast at 175 million bushels or 5 per cent of projected consumption. However, there would be opportunity for South American producers to adjust their planting decisions to the likely size of the US crop. Still, price might need to adjust, at least in the short run, to motivate either an increase or decrease in area planted in South America.

Expectations about the average yield of the 2010 crops are in a wide range, particularly for corn. Those private firms that tend to be most closely followed have projections in the 166 to 167 bushel range, well above the record yield of 164.7 bushels last year. There are few, if any, forecasts below 160 bushels. Our expectations for 2010 average yields are guided by yield models that reflect the impact of trend, planting date, growing season weather in Iowa, Illinois, and Indiana, and US crop condition ratings. The details of model specification and forecasting procedures were reported last year. Those models have been updated to include 2009 results.

Our forecast of the 2010 average corn yield is based on crop condition ratings as of 1 August, weather through July, and equal chances that August weather will replicate the August weather of each of the past 50 years. The result is a forecast of 158.1 bushels per acre. Unlike last year when the crop yielded better than it looked because of very favorable summer weather (cool and wet), our analysis suggests that the 2010 yield will be lower than implied by crop condition ratings alone. The expected shortfall is due to excessive June precipitation and above average summer temperature. For soybeans, our modeling points to a 2010 US average yield of 43.7 bushels, very near the record 44 bushels of last year. That forecast is within the range of expectations being reported.

While the yield models performed well in 2009, it should be recognized that the models do not capture all of the factors that influence yield and that actual August weather is yet to be determined. The result is that the models have relatively large forecast errors so that confidence in the specific forecasts is limited. Still, the results suggest that the final 2010 yield estimate for corn could be below current market expectations.