How accurate is USDA's prediction of corn and soybean yield in September?

Introduction: U.S Department of agriculture's September production report was planned to be released on September 12 A series of private production forecasts have been released, and there are countless yield forecasts based on USDA's weekly crop status rating report All of these forecasts are interesting to read, but it's the USDA forecast that the market really has to respond to This has always been the case The methodology of USDA is the most objective and comprehensive method so far compared with other expected yield methods Even so, the final yield estimate in January after harvest often differs from the forecast in September, because the final yield estimate reflects more information, including the impact of late season weather, actual harvest results and the prediction of yield loss In 22 September reports since 1979, the difference between the prediction of corn yield in September and that in August was 0.1 bushel five times; the difference between the prediction of corn yield in September and that in August was more than 0.1 bushel ten times; the difference between the prediction of corn yield in August and that in August was more than 0.1 bushel seven times From August to September, there were only two times when the expected per unit yield decreased by more than three bushels, namely in 1983 (14.8 bushels) and 1995 (4.5 bushels) The average reduction in the other eight years was 1.6 bushels Between August and September, the expected growth range for unit production ranged from 0.6 to 4 bushels per acre, with an average of 2.3 bushels This year's consensus forecast for September is lower than that for August History will show that the reduction will be relatively small Based on the prediction of the actual harvest area in August, the predicted yield per unit area needs to be reduced by more than 3.8 bushels to make the corn yield lower than 9 billion bushels In the past 22 years, the estimated corn yield in January after harvest is different from that in September In 14 cases, January's forecast was higher than September's, and 8 times lower than September's forecast The gap between January and September is slightly larger than the gap between August and September reports The decline in yield per unit ranged from 0.4 bushels to 12.4 bushels (1993), with an average of 4 bushels Excluding the extremes of 1993 and 1995, the average reduction was 220 million bushels The range of yield increase was from 0.3 to 10 bushels, with an average of 4.2 bushels For the past 22 years, the difference between the single output value in January and the predicted value in September has occurred five times in one bushel, six times in two bushels, nine times in three bushels Interestingly, in the past 10 years, the predicted per unit yield in September decreased by more than 0.1 bushels compared with that in August, the number of times in January was 6 times lower than that in September, and the number of times higher than that in September was 4 times In terms of soybeans, in most years, the soybean yield forecast in September has been very close to that in August In 22 years, there were only six years when the difference was more than one bushel Three of the six occurred in 1998, 1999 and 2000 The biggest difference occurred in 1983, a decrease of 4.8 bushels The largest increase in yield was 1.7 bushels in 1985 In the past 22 years, September's forecast has been lower than August's forecast 12 times This year, the market's outlook on unit production is mixed Based on the estimated harvest area in August, the yield forecast in September needs to be lower than that in August by more than one bushel, so that the soybean yield forecast is lower than 2.8 billion bushels In the past 22 years, the estimated yield per unit in January has been lower than that in September, 11 times in total, 6 times in the past 8 years, and decreased annually in the past four years In the 11 years when the yield estimates were lower than September, the decline range was from 0.3 bushels to 2.1 bushels, with an average of 1.2 bushels The average decline over the past three years has been relatively large, from 1.4 to 1.7 bushels In the 11 years in which the yield estimates increased in January, the increase ranged from 0.4 bushels to 3.7 bushels, with an average of 1.6 bushels Therefore, from September to January, the increase of per unit yield tends to be higher than the decrease of per unit yield In a word, from August to September, the change range of soybean yield prediction value is relatively small The change also tends to be smaller than many private forecasts In particular, the decline in yield forecasts is not as large as that predicted by private forecasters This year is likely to be no exception The change of unit yield from September to January tends to be larger than that from August to September, but the situation of large change is rare Considering the maturity of crops this year, especially corn, the unit yield forecast in September should well reflect the final unit yield value.