It seems like almost everyone associated with the market believes the old adage that, "big crops get bigger and small crops get smaller." I think belief in that old saying explains most of why the reaction in the corn futures market today was so small in the face of what should have been very bullish news from the USDA crop report. So, it could be pretty important to know whether this adage really should be used to heavily discount today's USDA yield estimate. Darrel Good, myself, and a former post-doc researcher, Olga Isengeldina, wrote a paper on this exact topic that can be found here(free access!): http://ageconsearch.umn.edu/handle/143639.

A quick summary:

1. Looking back from the vantage point of the January estimate each year from 1970-2010 there is indeed a tendency for what turn out to be big crops to get bigger from August through the final January estimate. However, this is an ex post analysis that uses information not available in August each year.

2. A different question is whether one can predict future changes in USDA yield estimates using only data available each August. This a true predictive or ex ante exercise (what you have to conduct in order to trade). In other words, not knowing whether the final yield estimate really does turn out to be "big" or "small" can you reliably predict the direction and magnitude of future changes to USDA yield forecasts? At least based on the tests we ran using forecast trend deviations and crop conditions ratings, the answer is an unequivocal no.

3. We concluded that the widespread belief in the adage "big crops get bigger and small crops get smaller" is based on a hindsight bias, and in fact, it is very difficult to predict in real-time.

The bottom-line is that our research suggests that today's USDA corn yield forecast should be viewed as unbiased and having an equal probability of increasing or decreasing going forward.

Let the arguments begin!

Scott Irwin

140.45