BY LINDA McGURK
Indiana Correspondent

WASHINGTON, D.C. — The USDA is predicting record production for soybeans and near-record production for corn this year, but at least one analyst believes the agency’s numbers are a little too optimistic.

Lanworth, a Chicago-based firm that specializes in providing data on the food, feed, fuel and fiber supply to hedge funds and financial institutions, thinks a downward revision from USDA is likely. “We do believe that our models combined with our acreage models are showing an overall production that’s substantially below what USDA is reporting,” said David Levine, the company’s vice president of resource solutions.

In its Aug. 12 Crop Production report, the USDA’s National Agricultural Statistics Service (NASS) projected that U.S farmers will produce 12.8 billion bushels of corn and a record-breaking 3.2 billion bushels of soybeans this year. Yield is forecast at 159.5 bushels per acre for corn and 41.7 bushels per acre for soybeans. The agency also expects record ear counts per acre in several Corn Belt states.

But according to comments to investors from Lanworth, “the USDA appears to have vastly over-estimated the yield potential” for the corn crop in Illinois, Iowa, Ohio and South Dakota, and the soybean crop in Iowa, Minnesota, Ohio, North Dakota and South Dakota. Lanworth noted that these states are at or near record delays in flowering and are among those most susceptible to an early freeze.
Stalk counts and ear density may be high, the company stated, but the late flowering likely means fewer kernels per ear and lower kernel weight.

“The primary factor is the late plantings. There’s just a certain amount of time that the plants need to mature,” said Levine about Lanworth’s projections. “We’ve gone back and looked at the historic production and we’ve never seen the record-high production that they (the USDA-NASS) are talking about in some states when planting has been done this late in the season.”

Lanworth uses satellite imagery and advanced computer modeling as well as sampling straight from the field to come up with its yield and production estimates, which are released to investors every two weeks. Currently, the reports are not made public, but Levine offered that the company’s projections are “substantially below” USDA’s in production for corn and soybeans.

USDA numbers debated

But Chris Hurt, Purdue University agricultural economist, does not think USDA’s estimates are out of line. On the contrary, he’s prepared to go even higher and predicts this year’s corn crop will beat the 2004 record of 160.4 bushels per acre.

“The corn (yield) number seems aggressive given the late planting, but I’d probably add a couple of bushels to it,” he said. “I think this one could potentially reach into the mid-160s. History tells me this crop has a lot of potential to be very good.”

Hurt noted that the August crop conditions ratings are far ahead of average and said previous years with similar ratings have produced a crop that’s 10 percent bigger than normal for corn and 6-7 percent above trend for soybeans.

“Particularly on corn the cool weather is often overlooked, but that makes for very favorable conditions for pollination. And we did have a lot of cool, moist weather the last two weeks of July,” Hurt said. “Pollination gives you a lot of kernels.”

For soybeans Hurt called USDA’s current estimate of 41.7 bushels per acre “conservative.” Hurt is predicting a yield of 42.8 bushels per acre, but he added that this soybean crop will likely beat the 43-bushel-per-acre record from 2005 when all is said and done.
Looking at data from the past three years, Hurt said USDA’s August yield projections for corn have only been off by 1-1.7 bushels when compared with the final tally in January.

“They do a very good job,” Hurt said about the agency’s estimates. “My perception is that they’ve done better in the last few years when yields have been close to normal or moving up with trend. Forecasting is a lot easier to do when things are stable than when they’re bouncing around.”

Lanworth too has been right on the money in the past three years, predicting yields that were within one percent of USDA’s final tally, according to Levine. “We don’t generally compare ourselves to other sources,” he said, “but some of our major clients even think we get a more accurate tally than the USDA, because we include some production sources others might not find.”

Lanworth’s projections have been at odds with the USDA’s before. Earlier this year Lanworth stuck its neck out predicting over 85 million planted acres of corn, nearly 1 million acres above the average trade estimate.

In June, the USDA amended its prediction to 87 million planted acres, although it had initially predicted a 1-percent decrease. In fact, a Forbes Magazine article published on Aug. 19 even alleged that Lanworth is “beginning to displace the USDA as the authority in predicting planted acreage and crop yields.”