|
|
| |
| |
Good of IL – Soybeans: Large U. S. crop; what about South America?
|
Grain Price Outlook issued by Darrel Good
Extension Economist, University of Illinois
10/24/2005 -- USDA reports provided two fundamental surprises for the market over the past three weeks. First, the September Grain Stocks report showed September 1, 2005, inventories of soybeans of 256 million bushels, about 35 million bushels less than expected. Even though the estimate of the 2004 crop was reduced, the current estimate of 3.124 billion bushels still seems too high.
Second, the October Crop Production report reduced the estimate of 2005 planted and harvested acreage by about 900,000 acres. The yield forecast was increased by 2 bushels, to 41.6 bushels per acre, about as expected.
However, the resulting production forecast of 2.967 billion bushels was about 40 million less than expected. Still, supplies for the 2005-06 marketing year appear ample, particularly if the 2006 South American crop is near the projected level of 3.97 billion bushels. That projection is 425 million larger than the 2005 crop that was adversely impacted by drought in southern Brazil.
November 2005 soybean futures declined nearly $2 from the peak in late June to the low in late September, and then recovered about $.40 following the USDA reports. Basis levels remain extremely weak, but cash prices rebounded about $.30 in the second week of October.
The seasonal low in the soybean market may have occurred, depending on the size of the November production forecast, but a significant rally is not anticipated unless the South American crop encounters difficulty. Current cost/price relationships point to an increase in U.S. soybean acreage in 2006. A marketing year average U.S. farm price near $5.60 is suggested by the current projection of the year endings stocks-to-use ratio.
Soybean supplies remain record large
The USDA's estimate of September 1, 2005 soybean stocks of 256 million bushels was surprisingly small, about 40 million less than projected in early September, but was the largest in 5 years (Table 1). The small estimate resulted in a 17 million bushels reduction in the estimated size of the 2004 crop, but that estimate of 3.124 billion bushels still appears to be too large. Seed, feed, and residual use of soybeans during the 2004-05 marketing year totaled about 187 million bushels, 15 to 25 million more than is typically expected. Large residual use, 107 million bushels, is evidence that the 2004 crop was smaller than estimated. The same scenario was observed in 1998-99, but the crop estimate was not revised that year (see Table 1).).
The USDA’s October Crop Production report forecast the 2005 U.S. soybean crop at 2.967 billion bushels, second in size to the record 2004 crop (Table 2). [Editor’s note: All tables will be found in the 11-page PDF October Soybean Tables linked at the bottom of the page, same as the link for Table 1 references above.] The forecast was 111 million bushels larger than the September forecast and 176 million larger than the August 2005 forecast. Based on Farm Service Agency (FSA) data on certified planted acreage, the USDA lowered the estimate of planted acreage of soybeans by about 900,000 acres, to a seven-year low of 72.2 million acres (Table 3).
Prior to 2003, FSA acreage data was typically reflected in the USDA’s January Crop Production report, but is now routinely reflected in the October report, due to earlier access to the data. The planted acreage estimate was reduced in June, September and October with the current estimate 1.71 million below the report of planting intentions in March (Table 3). The reduction in planted acreage in 2005 was generally distributed across the major producing regions, with southern growing areas showing slightly larger declines as a percentage of total planted acres (Table 4), partially as a result of the threat of Asian soybean rust.
Harvested acreage of soybeans is projected at 71.27 million acres, only 930,000 acres less than planted acreage. The magnitude of abandoned acres is near the low end of historic values, contrary to early expectations that adverse weather would result in more than normal abandonment. Still, harvested acreage is at the lowest level in seven years (Table 3).
The U.S. average soybean yield is projected at 41.6 bushels per acre, two bushels above the September forecast and 2.9 bushels above the August forecast (Table 5). The projection is only 0.6 bushels below the record yield of 2004, when growing conditions were considered almost ideal. Average yields in 2005 are expected to exceed the 2004 average in the northern tier of states and in Iowa and Nebraska.
Over the past 20 years, there has been a strong relationship between the percentage of the soybean crop rated good or excellent in the last report of the season and the U.S. average trend-adjusted yield. That relationship is: yield = 29.79 + .194 (percent rated good or excellent) and it has explained 87 percent of the annual variation in the trend adjusted yield. The 2005 season ended with 57 percent of the crop rated good or excellent, suggesting an average yield of 40.8 bushels per acre, only 0.8 bushel (1.9 percent) below the USDA’s October forecast. Since 1979, the October soybean yield forecast exceeded the September forecast 13 times. In those 13 years, the November yield forecast exceeded the October forecast 11 times (Table 5).
Based on the October production forecast, the supply of U.S. soybeans for the 2005 marketing year is projected at 3.227 billion bushels, only 15 million less than the supply of a year ago. With the apparent overestimate of the 2004 crop, and some chance the 2005 U.S. yield forecast will increase in November, supplies are probably record large this year. A crop forecast of 2.985 billion bushels, reflecting a yield of 41.9 bushels, is used here (Table 6).
Soybean use to remain large
Typically, the most uncertainty regarding the forecasting of annual soybean oil and meal consumption centers on the export market. When supplies are ample and prices are at modest levels, domestic oil consumption increases at roughly the rate of population growth. Similarly, domestic meal consumption grows at the rate of growth in livestock production, mostly driven by domestic human population growth. Exports, while a fairly small percentage of total consumption of U.S. product, demonstrate no trend in use.
Exports of U.S. soybean meal have varied from about 5.1 million to 7.7 million tons over the last 16 years, with both extremes occurring in the past five years (Table 7). Annual exports of U.S. soybean oil have ranged from 936 million to about 3.1 billion pounds in the same 16 year period (Table 8). For the current year, the USDA forecasts U.S. soybean oil exports at 1.3 billion pounds, just 50 million less than exports of a year ago.
While foreign consumption of soybean oil is expected to grow by nearly 8.5 percent, and trade is expected to grow by nearly 9.5 percent, the U.S. is expected to garner a smaller share of the world market. The U.S. had only 4.8 percent of the export market in 2003-04 and 6.7 percent in 2004-05. That share is projected at 5.9 percent in 2005-06, as both Argentina and Brazil are expected to capture a larger percentage of the market.
U.S. soybean meal exports during the current year are projected at 6.55 million tons, down from 7.3 million last year (Table 7). The U.S. has a slightly larger share of the world soybean meal export market, but that share is expected to decline from 14.3 last year to 12.5 percent this year, even as total world trade increases by nearly 3 percent. Exports from Argentina and Brazil are expected to grow by 4 percent on the basis of larger crops.
The USDA forecasts domestic soybean meal consumption at 34 million tons, 1.8 percent more than used last year. The rather small increase reflects expectations of only a modest increase in domestic livestock production and increased feeding of dried distillers’ grain. Similarly, the USDA projects a trend increase of 2 percent in domestic use of soybean oil to a total of 17.85 billion pounds.
The market, however, seems to think that high petroleum prices will result in a more rapid rate of increase in bio-diesel production, resulting in a larger increase in soybean oil consumption. This expectation extends beyond U.S. soybean oil to include increased consumption of vegetable oils world wide. In fact, reports of plans to increase bio-diesel production emerge on a regular basis. The USDA may be understating demand for soybean oil. A projection of domestic use of 17.9 billion pounds is used here.
For now, it appears that the domestic crush of soybeans will continue to be driven by soybean meal consumption. Based on the projected use of 40.55 million tons and an average yield of 47.65 pounds of meal and hull meal per bushel, the domestic crush would be near 1.7 billion bushels. Allowing for imports of 165,000 tons and a small draw-down in ending stocks, the crush is projected at 1.695 billion bushels. If 1.695 billion bushels of soybeans are crushed, about 19.155 billion pounds of soybean oil will be produced if the average oil yield is near 11.3 pounds per bushel. Based on the projected level of consumption, year-ending oil inventories will be very near the level of stocks at the beginning of the year (Table 8).
Exports of U.S. soybeans themselves are growing
U.S. soybean exports have been large and generally increasing since 1994-95. Exports were a record 1.103 billion bushels during the 2004-05 marketing year. U.S. exports have benefited from a rapid increase in world soybean consumption, particularly in China, and by a shortfall in Brazilian production in 2004 and 2005. The increase in Chinese soybean consumption and imports has been huge over the past 7 years. Domestic consumption grew from an estimated 640 million bushels in 1997-98 to 1.475 billion bushels in 2004-05. Consumption during the current year is projected at 1.694 billion bushels. Chinese imports from all sources grew from 110 million bushels in 1997-98 to 835 million bushels in 2004-05 and are projected at 1.01 billion bushels for the current year.
The Brazilian soybean crop was a record 1.91 billion bushels in 2003. Production remained large in 2004-05, but well below potential production due to extensive drought conditions in southern Brazil (Table 9). Argentine production was a record 1.43 billion bushels in 2005. For the 2006 harvest, the USDA projects a decline of about 4 percent in Brazilian soybean acreage, but a rebound to a more normal average yield (Table 10). Soybean area is expected to increase by nearly 6 percent in Argentina, to a record 37.5 million acres (15.2 million hectares). Average yields, then, would result in 2006 production of about 2.2 billion bushels in Brazil and 1.5 billion in Argentina.
Production in all of South America, including Paraguay, Bolivia and Uruguay, is forecast at a record 3.97 billion bushels, 425 million larger than the 2005 harvest. Production outside of the U.S. is forecast at a record 5.15 billion bushels, or 140 million tons (Table 11). Production of all oilseeds outside the U.S. is forecast at a record 292.1 million tons, 3 percent larger than last year’s harvest. Declines in cottonseed and rapeseed are expected to partially offset increases in soybean and sunflower seed production.
U.S. export prospects are bolstered by expanding Chinese demand, but may be limited by increased competition from South American production. Exports near the record level of the 2004-05 marketing year are expected, with USDA forecasting shipments of 1.115 billion bushels. U.S. soybean exports are largest during the first half of the year. Excluding the short crop year of 2003-04, exports during the first half of the year averaged 72.3 percent of the marketing year total since 2000-01 (Table 1).
As a result of this strong seasonal pattern, it is difficult to gauge the progress of exports relative to the marketing year projection on a weekly basis. Having said that, shipments got off to a very slow start in 2005-06, but picked up significantly during the week, ended October 13. Through the first 43 days of the marketing year, export inspections totaled 77.64 million bushels, compared to 99.18 million during the same period last year. In addition, unshipped export sales as of October 13 stood at only 220 million bushels, compared to 320 million on the same date last year.
Last year, China started its buying and import program for U.S. soybeans much earlier than this year. Total export commitments to China now stand at 100 million bushels, compared to 150 million at this time last year. The pace of sales to China will be important, but it is too early to suggest that China will not buy the expected quantities of U.S. soybeans.
With exports of 1.115 billion bushels; crush of 1.695 billion; and seed, feed, and residual use of 160 million bushels, year-ending stocks of U.S. soybeans are projected at 275 million bushels (Table 6).
Soybean acreage to increase
Early expectations are for U.S. producers to increase soybean acreage in 2006. Reasons cited include the escalating costs of corn production, reduced concerns about Asian soybean rust, unexpectedly high yields of soybeans in 2005, and the higher price of soybeans delivered in the 2006-07 marketing year relative to prices of feed grains.
The USDA’s January report of winter wheat seedings will give some indication of how willing producers are to respond to changing economic conditions. In addition, indications of the extent of over-wintering of soybean rust will give a better indication of the potential risk of soybean rust in the Midwest in 2006. The further south that freezing temperatures occur, the lower the perceived risk of rust.
For now, we might expect soybean acreage to be near 73 million acres in 2006, with harvested area near 72 million. A trend yield near 42.5 bushels would result in a 2006 crop of 3.06 billion bushels. Changing cost and price conditions over the next 6 months will require a re-evaluation of acreage and production potential.
Price prospects
November soybean futures declined by nearly $2 per bushel from the June high to the fall low, as the U.S. crop turned out to be much larger than feared. However, prices rebounded a bit on the basis of a smaller September 1 stocks estimate and a smaller than expected October production forecast. November futures traded back to near $6 in mid-October. Cash prices reflect a generally weak basis in most markets. The average overnight cash bid is central Illinois reached a low of $5.15 on October 10, but recovered to about $5.45 on October 19.
The average basis was $-0.43, much weaker than the basis of $-0.28 of last year and the five-year average basis of $-0.21. In addition, the November-July spread in the futures market of $.30 was about $.08 larger than the spread of a year ago. The large crop, increased transportation costs, and storage limitations in some areas have resulted in an unusually large carry in the market.
The USDA projects that the 2005-06 marketing year average price will be in a range of $5.00 to $5.80. Historically, the average farm price was closely correlated to the level of year ending stocks expressed as a percentage of use (stocks-to-use ratio). For the period 1989-90 through 1997-98, that
relationship was: price = 14.96 ÷ (stocks-to use)
+ 4.63 and the stocks-to-use ratio
explained nearly 90 percent of the annual
variation in the average farm price. That
relationship changed and became much more
volatile from 1998-99 through 2004-05. In
general, prices during that period were much
lower than the earlier period for a given
stocks-to-use ratio. The best fit during that
period was: price = 14.91 ÷ (stocks-to-use) +
3.38, but the stocks-to-use ratio explained
only 55 percent of the annual variation in
price.
For the past two years, the average farm price has been exactly half-way between the prices predicted by the two models. If that is the case again in 2005-06, the projected stocks-to-use ratio forecast here (9.3 percent) would result in an average farm price of $5.60.
At any given time, the futures market reflects the average price expected (or offered) for the remainder of the crop year. Those futures prices can be transformed into an expected farm price using a two-step process.
First, for each month of the marketing year the appropriate futures contract price (nearest delivery not in the current month) is adjusted by the expected difference between the futures price and the average farm price received (reported by NASS) during that month. The USDA’s Economic Research Service has calculated those average monthly differences since 1975-76. Those can be found at here. The average difference of the last five years is used as the expected difference for 2005-06.
Second, the monthly estimate of average farm price is weighted by the expected percentage of the crop to be marketed each month. Again, the five year average is used as the expected marketing weights for 2005-06. Using this methodology with futures prices on October 19, along with the average cash price received in September, results in a forecast of average farm price of about $5.80. The average price offered by the market can be compared to the average price expected for the year to evaluate pricing opportunities.
There is some consistency in forecasts of the 2005-06 marketing year average price, with the stocks-to-use models projecting a price $.20 above the mid point of the USDA’s forecast range and the market offering a price at the upper end of the range and $.40 above the mid point. The market is currently offering an opportunity to forward price some of the farm stored crop at levels well above the expected marketing year average price. The average premium for January delivery over immediate delivery in central Illinois was $.27 on October 19. A return to a more normal basis puts late spring prices about $.60 over spot bids. Considerable price uncertainty, however, will persist until well into the South American growing season.
October 2005 Soybean Tables (PDF)
Source: Grain Price Outlook October 2005 (2005 – No. 8) posted 10/21/2005, from Marketing & Outlook, a joint publication of Purdue University and the University of Illinois at Urbana-Champaign.
For an easy way to stay informed, sign up for the weekly Bulletin e-mail newsletter from www.StopSoybeanRust.com. You also will receive our RUST FLASH e-mails alerting you to important breaking news as soon as it occurs. All are archived in Bulletins.
|
| |
|
|
|
