Agrimoney.com - http://www.agrimoney.com/features/feature.php?id=274
One number key to negotiating big day for USDA crop reports
By Agrimoney.com - Published 27/03/2014

There is good news and bad news for grain and oilseed investors preparing for what, on Monday, may well prove one of the most momentous days of the year.

The good news is that, out of the all the important data that the US Department of Agriculture will release in a double bill of big reports one on prospective crop areas this year in the US this year and the other on quarterly grain inventories there is one that really stands out as important.

Bill Tierney, chief economist at Chicago-based AgResource, says: "You have got old crop information and information and new crop information coming at the same time.

"But what prices seem to respond to is the stocks report, and in particular for what is going on in corn.

"Even for soybeans, what appears to have more impact on prices than soybean data is what is happening in corn, which seems to be driving price response."

So investors attempting to position ahead of the report need only look at the US corn balance sheet to guide their strategy.

Price behaviour

And correct positioning is important. Quarterly stocks reports have a habit of producing big price swings.

Forecasts for US corn data, March 31 USDA reports, (change on last year)

2014 sowings: 92.748m acres, (95.365m acres)

Range of estimates: 90.50m-94.0m acres

March 1 stocks: 7.099bn bushels, (5.40bn bushels)

Range of estimates: 6.861bn-7.540bn bushels

Sources: USDA, ThomsonReuters

At the last one, on January 10, corn prices soared 5.0% on the day as the data signalled that livestock feeders had switched to the grain from wheat, encouraged by discounts of more than \$2 a bushel, to a far greater degree than investors had expected.

Soybeans actually managed that time to gain only 0.5%. (Wheat tumbled 2.6% in Chicago.)

A year ago, the row crops did perform in line, undertaking a synchronised dive, with the March stocks report sending corn down 5.4% and soybeans plunging 3.4%. Wheat tumbled 6.7%.

A year before that, in March 2012, the contracts bounced in harmony, with corn futures soaring 6.6%, soybeans 3.5% and wheat 7.9%.

Big range

The bad news is that, as can be inferred from the extent of the price moves after the reports, guessing the corn stocks figure right is a tricky business.

That looks especially true this time, given that the so-called "consensus" estimate of 7.099bn bushels for the corn stocks figure disguises a huge range of forecasts. A range of 679m bushels from highest to lowest, in fact.

The range is, in percentage terms, some twice the average of 5.2%, and is, indeed, not far short of the 13.6% record since 1984, set in 1984, according to Mr Tierney.

"I can't say I am surprised, the way the market has been beat up by previous reports," he told Agrimoney.com.

"It's understandable in those circumstances that you widen the range of estimates."

For wheat, the range is wide too, at 130m bushels around the consensus figure of 1.042bn bushels, with soybeans' at 132m bushels around a consensus of 989m bushels.

The maths, in theory

So how to guess what might come up?

That's the tricky part.

Forecasts for US soy data, March 31 USDA reports, (last year's figure)

2014 sowings: 81.075m acres, (76.533m acres)

Range of estimates: 78.50m-83.60m acres

March 1 stocks: 989m bushels, (998m bushels)

Range of estimates: 955m-1.087bn bushels

Sources: USDA, ThomsonReuters

In theory, it looks pretty straightforward. Investors know some of the mechanics which determine how much stocks are left, through ethanol production data, which gives an implied reading on how much corn is used in making the biofuel.

Exports too are closely measured while, for soybeans, there is also monthly crush data which gives an extra insight into use of the oilseed.

Plug in estimates for the size of hog, poultry and cattle herds, with an adjustment for smaller uses such as seed, and the answer should give how much of the crop has been used over the last quarter.

Not so easy

The trouble is that estimating, say, livestock feeding is not quite so easy.

At Allendale, which forecasts Monday's data showing corn stocks of 6.995bn bushels as of March 1, and soybean inventories at 981m bushels, a little below average, Rich Nelson, director of research, said that the group had relied on an estimate of a 5% increase in feed use over five years.

"But it's very hard to estimate.

"You could justify a 20% increase, or a 20% decrease, in feed use, depending what assumptions you make."

'Quickly switch grains'

After all, hopes of growth in the US hog herd are being dashed by porcine epidemic diahorrea virus (PEDv). But by how much?

Forecasts for US wheat data, March 31 USDA reports, (last year's figure)

2014 sowings: 56.277m acres, (56.156m acres)

Range of estimates: 54.80m-57.711m acres

Includes, other spring wheat: 12.27m acres, (11.596m acres)

Range of estimates: 11.00m-13.50m acres

Durum: 1.794m acres, (1.470m acres)

Range of estimates: 1.39m-2.20m acres

March 1 stocks: 989m bushels, (998m bushels)

Range of estimates: 955m-1.087bn bushels

Sources: USDA, ThomsonReuters

Estimates vary, as highlighted by a Rabobank report earlier this week, which forecast a far less benign situation than the USDA is factoring in.

And even if animal numbers are guessed correctly, "the hard bit is actually estimating how much grain is used per head," Mr Nelson told Agrimoney.com.

"Producers can quickly switch grains to ones other than corn," with distillers' grains, a byproduct of corn ethanol manufacture, another alternative to factor in.

Soybean skewballs

And there are many other potential factors which can skew the data too.

Macquarie flagged a historically strong pace of US soybean exports in the first two weeks of March, which would mean more of the oilseed in transit as of the March 1 cut-off date for the quarterly stocks data.

"Transit volumes of soybeans are likely invisible to the USDA stocks report, so this volume can be lost from the stocks estimate only to be found back in the next quarter," said Macquarie analyst Chris Gadd.

Conversely, Monday's soybean stocks number may be swollen by a trend among farmers of selling the oilseed while hoarding corn, which "means a far-greater-than-normal proportion of soybeans will likely be held off-farm".

With off-farm inventories "measured more accurately than on-farm stocks", this dynamic could cause Monday's inventory estimate "to surprise on the high side", Mr Gadd said.

'Find back' factor

So how to prepare for Monday's report?

RJ O'Brien's Richard Feltes said that the estimate for March 1 soybean stocks of 989m bushels was somewhat reassuring for bulls, in that it "given known crush and exports" for the previous three months, it "assumes that we'll 'find back' 30m bushels of soybeans".

That is, the report will reveal higher stocks than implied by other data, perhaps a reflection of the 2013 harvest being bigger than has been thought.

"Given fact that in most years during the December 1-March 1 quarter there is a loss of soybeans via residual use, and that the largest 'find back' of soybeans for the quarter was 42m bushels in 2010m we think odds of a bearish surprise on March 1 US soybean stocks is quite low," Mr Feltes said.

Corn factor

The problem for bulls is, of course, that as far as prices go, it is not soybeans that matter, but corn.

And Mr Feltes highlighted increasing jitters over the corn number, with some commentators talking of a record tonnage of distillers' grains set to be produced in the US this year, displacing corn, while other are focusing on the lower numbers of cattle and hogs on feed.

"We can say that odds are greater for a bearish surprise on corn stocks than encountering a bearish surprise on soybean stocks," Mr Feltes said.

And that could mean bearish moves for all.

© Agrimoney 2010