CHICAGO (Reuters) - The prospects for less corn acres in the United States next year is not exactly something to get bullish about, especially if yield has a say.

Corn is seen in the field that belonged to the Gibson family farm businesses which was auctioned off by a court appointed receiver in Morocco, Indiana, U.S. September 6, 2016. REUTERS/Jim Young

Corn yield in the United States blew away almost all expectations in 2016 to produce the largest harvest in history of 15.2 billion bushels and as such, domestic corn supply has built up to levels not seen in 30 years.

In USDA’s annual long-term projections released Tuesday, the agency pegged 2017 U.S. corn yield at 170.8 bushels per acre. This is notably higher than the year-ago figure for the 2016 harvest of 168.1 bpa, but considerably lower than the actual harvest yield, which is estimated at 175.3 bpa.

The agency also projected U.S. corn planted area to drop from 94.5 million to 90 million acres in 2017.

Although USDA states that the long-term projections are not official forecasts for the upcoming season, the long-term projected yield comes very close to what the initial forecast will show early next year.

The first yield forecasts come each February from USDA’s World Agricultural Outlook Board and are presented at the Agricultural Outlook Forum during that month. This number, similar to the long-term projections, is a trend line yield assuming normal weather and planting pace.

WAOB’s initial corn yield does not always exactly match the long-term projected yield for the upcoming year and in fact the February number often drops just a little lower. But the two yields have never differed by more than half a percent.

Bottom line is that analysts and traders should plug in 170.8 bushels per acre with confidence to their 2017 U.S. corn balance sheet, but they may also want to make room for some upside, as WAOB seemingly has a habit of underestimating yield.

NO CEILING ON YIELD

Since 1998, final U.S. corn yield has outperformed WAOB’s trend yield with the exception of years in which widespread weather issues hurt the crop. This makes room for the argument that WAOB may be coming in about half a percentage point too low at the start. (reut.rs/2gE1LjW)

This means that 2017 corn trend yield could realistically be greater than 171 bpa. Each bushel per acre adds about 80 million bushels to the harvest volume, and given that USDA projects corn carryout to stay north of 2 billion bushels until 2023, every extra bushel adds to the burden.

The 170.8 target is relatively modest compared with earlier 10-year projections for the 2017/18 season. The highest number came in 2010 at 174.4 bpa, and the lowest was last year’s 170.1 bpa.

But USDA did make a significant bump in the year 1 projection for 2017 relative to the past few years, meaning that 170.8 may be closer to the actual mark. Year-on-year early projected yield increased by the greatest amount since 2010, as baseline yield projections got scaled back following the 2012 drought. (reut.rs/2gE0Wb2)

Last year’s crop may serve as good justification for leaning on the higher side of yield. The summer of 2016 was persistently warm, which is known to disrupt grain fill in the corn ears. So if the 175.3 bpa crop was in fact limited by the warmth, then weather nuances over the last few years may have been masking the true potential of U.S. corn yield.

Just a couple of years ago, many industry participants were skeptical early on that national corn yield could push over 170 bpa in 2014, even under good conditions. But now in 2017, corn yield will almost positively start out at or above 170, or at least it should.

And the reduction of 4.5 million planted corn acres for 2017 as projected this week by USDA may not be such a big deal as last year’s winter wheat crop can demonstrate.

One year ago, U.S. wheat traders started eyeing the year-on-year loss of 3 million winter wheat acres. But fantastic conditions sent yield to an all-time high, boosting 2016 production by 20 percent over the previous year. Next year’s corn crop could potentially play a similar hand.