The end of the 2017-18 season is still more than a year away, but if USDA’s new estimates are close, world corn use will exceed production for the first time in seven seasons by 28.64 million metric tons or 1.13 billion bushels.
USDA’s lower estimate of world ending corn stocks at 195.3 mmt doesn’t justify $5.00 a bushel corn, but if true, does offer an easing in the bearish dynamic that has afflicted corn prices since 2013.
Given this modest bit of bullish hope, it seemed like a good time to look at the record of USDA’s previous May new-crop estimates. A look at 16 years of USDA estimates, from 1999-00 to 2014-15, showed estimates of world corn production were off by an average of 3.5% a year.
The largest error came in 2004-05 when USDA underestimated world production by 11.2%. For 2017-18, a 3.5% average miss represents an error of plus or minus 36 mmt on USDA’s production estimate of 1,033.66 mmt (40.7 bb).
It also helps to know that USDA’s first May estimates of world corn production were lower than final production estimates in 12 of the 16 years or 75% of the time. Among the major corn-producing regions, production was most likely to be underestimated in Ukraine, Brazil, and Europe. The largest estimate errors were consistently for Ukraine, Argentina and Brazil.
What about the demand side of USDA’s May estimates? As we might expect, USDA has had a little better luck guessing world demand with an average error of 2.3% per year.
While the errors were generally smaller, USDA underestimated world corn demand in 13 of the 16 years or 81% of the time. A 2.3% error in 2017-18, if it happened, would represent a swing of plus or minus 24 mmt from USDA’s May estimate of 1,062.3 mmt.
In other words, USDA’s estimate that world corn demand will exceed production by 28.64 mmt in 2017-18 may not survive the season, given an average error on the production side of 36 mmt and an average demand error of 24 mmt.
Among crop estimates, the 163 mmt total counted for Ukraine, Argentina, and Brazil is the least reliable and has typically been too low in the past.
On the other hand, the argument for lower corn supplies in 2017-18 has two points in its favor. One is that this spring’s excessive wet conditions from Arkansas and Missouri to Indiana is making it difficult for producers to reach USDA’s goal of 82.4 million harvested acres this fall.
The other point is that world corn demand averaged a little over 3% annual growth over the 16 years studied and USDA is only estimating an increase of .8% in 2017-18.
With the world economy expected to do a little better in 2017, livestock inventories increasing, and cheap gasoline prices supporting ethanol demand, USDA’s initial estimate of .8% growth looks cautiously low.
There is plenty of uncertainty in the season ahead, but if USDA is close to correct, corn prices have a chance to make a modest improvement on the $3.40 a bushel average that USDA is estimating for 2016-17.
Todd Hultman can be reached at Todd.Hultman@dtn.com
Follow Todd Hultman on Twitter @ToddHultman1