PRUDENTIAL SECURITIES, INC.
One New York Plaza, New York, New York
212-776-1000(March 13, 2000) SOYBEANS: THE BIG PICTURE--World soybean production is down in 1999/2000, and even the most optimistic estimates for this spring's South American crop will not erase the world production decline from last year of 5.44 million metric tons (MMT). World soybean ending stocks also will decline this season, with the USDA projecting a drop of 3.41 MMT. World peanut production is also lower this season, but other oilseed crops, which have a high oil content, are expected to reach record levels. Hence, the current world balance continues to be characterized by a relatively tighter meal supply than oil supply.
U.S. soybeans reached a major cyclical low in July 1999 at $4.01-1/2 per bushel. By definition, the 1999/2000 season is the first in the bull market phase of a new price cycle, which should last at least two years. The fundamental balance confirms a turn in the cycle this year, with soybean carryover expected to decline after increasing for two consecutive seasons to a peak of 348 million bushels last season. The USDA is projecting this year's carryover at 325 million bushels, but even our lower estimate of 300 million would not represent a tight supply unless stocks were needed to augment a production shortage in 2000. As is usually the case in the first year of a new price cycle, statistics are not terribly bullish, but represent an end to the bearish phase of adding to carryover stocks.
Concerns about the size of South American crops now being harvested have abated with recent beneficial rainfall in dry areas--Rio Grande do Sul in Brazil and the central soybean-growing area of Argentina. Concerns now center on the pace of harvest progress in Brazil. Brazilian soybean harvest already was expected to be two to three weeks behind normal due to late planting and has been further delayed by rainfall, opening the door for larger-than-expected U.S. soybean exports this season. The size of South American crops has its major impact on U.S. soybean usage during the following U.S. crop year (2000/01 in this case) because it determines the amount of South American supply remaining to compete with the United States during the second half of the South American season. However, the pace of South American harvest affects U.S. usage in the spring, when shipping delays from South America might create the possibility of origin-switching to the United States.
Early indications are that if U.S. soybean yields return to trend in 2000, an addition to U.S. carryover will be inevitable. However, no estimates exist for carryover of the burdensome nature forecast at this time last season, and even the USDA's Outlook Forum estimate of 2000/01 carryover at 510 million bushels was below the 565 million bushels it projected for this season at last year's Outlook Forum. When prices declined to the $4.01-1/2 level last July, there was some talk of carryover stocks reaching 600-700 million bushels, and that is very unlikely to be repeated this season.
Because subsoil moisture levels remain below normal in much of the U.S. soybean-growing area, a return to trend yields (near 40 bushels per acre and the second highest on record) may be difficult to achieve. Yields will be the key factor determining production, with acreage secondary. Price relationships alone would indicate that soybean planting intentions decline to 72.925 million acres from 73.780 million last year. However, other considerations such as the greater drought resistance of soybeans relative to corn and their lower cost of production could play a role in farmers' planting decisions.
With our current new-crop usage estimates, it would take a crop of less than 2,780 million bushels or a yield below 38.3 bushels with unchanged harvested acreage from this season to result in another reduction in carryover in 2000/01. The cited production figure would still surpass the prior record of 2,741 million bushels; the yield would be the fourth highest on record, after that of 41.4 bushels per acre in 1994, 38.9 bushels in 1998 and 38.8 bushels in 1997. In summary, a carryover increase is not assured for the 2000/01 season, which will continue to focus the market's attention on new crop production prospects.
The key factors for the new-crop outlook appear to be: (1) a reduction in beginning stocks following an increase; and (2) a second consecutive increase in usage, assuming that it is not constrained by a production shortfall. Since 1960, only one year (1992/93) that met both factors saw production increase enough to permit a larger new-crop carryover; in two years with these factors (1970/71 and 1987/88), new-crop production declined slightly and carryover dropped. In all three years, prices rallied from a winter low into June (in 1970 the rally extended into July). The winter low in November soybeans was penetrated after June in the year with a carryover increase (1992/93), but not in the two seasons with a new-crop carryover decline. Hence, if carryover declines for a second consecutive season in 2000/01, then the December 1999 low of $4.72 in the November contract is unlikely to be penetrated, and the recent setback low of $5.14 also might hold. We have been looking for a price advance into the May/June time period (more likely June) this season, with price action subsequently dependent on growing conditions.
A prospective increase in new-crop carryover would permit a normal, seasonal, harvest low, but an early low would occur if carryover were projected to decline. In 1992, when carryover increased, the harvest low occurred in October; in the other two years it was early (September 1, 1970 and August 12, 1987) as one would expect in a year with constructive fundamentals. Prices reached a cyclical peak in two of these seasons, but the rally in 1970/71 extended into 1973 because new-crop production, while higher, was not keeping pace with demand. In the 1992/93 season, when carryover increased and prices reached a new low at harvest, prices rallied into July 1993, pushed higher by heavy rainfall in the Midwest that caused a production decline in 1993. In 1987/88, prices worked up to a high in June 1988, which has not been surpassed, on the combination of declining stocks and a drought that reduced new-crop production. The idea of a high in the 2000/01 season that surpasses that of the current season also fits with our analysis that the current season is the first of at least a two-year bull market phase in the soybean market.
SUPPLY/DEMAND--The USDA issued its revised old-crop balance sheets last Friday, increasing its export projection 20 million bushels, but leaving its crush estimate unchanged at 1,600 million bushels and reducing its old-crop carryover 20 million to 325 million. The slow pace of Brazilian harvest and recent aggressive buying by China contributed to the larger export projection. Cumulative shipments through March 2 (for the first half of the crop year) were up 77 million bushels according to export inspections and up 85 million bushels according to the Export Sales report; unshipped export bookings were 29 million bushels above last year's level. Hence, current indications would project U.S. soybean exports this season up 106-114 million bushels versus last year's level, in line with the USDA's new estimate calling for a 109-million-bushel increase.
There was talk last week that China had bought one or two additional cargoes of U.S. soybeans and that it had extended purchases of South American soybeans by at least two cargoes. Last week's Export Sales report showed total U.S. sales to China through March 2 at 2.964 MMT, of which 845,000 tonnes remain to be shipped. (The year-earlier levels were 1.630 MMT commitments, of which 20,300 tonnes remained to be shipped.) China accounts for most of the increase in unshipped export bookings this season, with 825,000 tonnes. As of March 2, the following major destinations showed increases in unshipped bookings: Western Europe, 94,000 tonnes; Indonesia, 140,000 tonnes; and Mexico, 69,000 tonnes. Unshipped export bookings were down from last year's comparable levels in the following countries: Taiwan, 185,000 tonnes; South Korea, 97,000 tonnes; Thailand, 68,000 tonnes; and Japan, 58,000 tonnes. These latter four countries are large U.S. customers that could switch origins to the United States from South America on shipping delays.
Board crush margins, which have been low this season, have been edging higher since February 4. A recent upward trend in U.S. crush margins subsequently may reverse the year-to-year decline in processing expected in the March statistics. We are projecting that crush in March will fall 2 million bushels below last year's level of 140 million. However, we expect crush for the remainder of the crop year to slightly exceed last year's level, rising to 1,611 million bushels, possibly a conservatively low estimate, even though it is above the USDA's projection.
The USDA will report March 1 stocks in all positions on March 31. We are currently projecting this figure at 1,417.5 million bushels, down 40 million bushels from the year-earlier level of 1,456.3 million, even though supplies at the beginning of this season were up 50 million bushels. Our stocks estimate is based on December-February quarterly exports at 318.5 million and crush for the quarter at 410.8 million bushels. The average cumulative residual "usage" for the first half of the crop year has been 121.7 million bushels over the last five years; we are using a figure of 120 million bushels in our balance sheet.
The USDA also will release its Prospective Plantings report on March 31. As detailed in last week's comments, price relationships alone would indicate a decline in soybean plantings to 72.925 million versus 73.780 million last year. However, other considerations may play a part, chief among them the fact that soybeans cost less to plant, which could be an important figure this year given the high cost of energy and nitrogen fertilizer. Also, soybeans are more drought-resistant than corn, which might spur an acreage increase this season because of the dry soils in many U.S. soybean growing areas. However, crop rotation considerations would work against soybean plantings this year. Also, the reduction in winter wheat plantings this season would reduce acreage available for double-cropped soybeans, a factor beneficial for yields, not acreage.
Soybean prices have a potential setback low at $5.00-1/2, basis May, and $5.08-1/2 in the July. Although we think there is downside risk to the $4.95 and $5.02 levels, respectively, the longer prices are able to hold above the recent lows, the less likely it is that these risk lows will be seen. An upside breakout would occur with the penetration of the $5.38 level in the May and $5.47 in the July. We still expect to see prices reach a spring high in May or, more likely, June with the July contract rallying to the $5.85 level. Based on price action in fundamentally similar years, there is still a chance for the July contract to rally to the $6.50 level if there are renewed concerns about the size of South American crops (which is unlikely) and/or concerns about the potential for U.S. yields (which is still likely).
SOYBEAN MEAL--THE BIG PICTURE--Soybean meal demand is the most price-sensitive of the soy complex, and, as a result, meal tends to be the first member of the complex to reach major highs and lows. Soybean meal prices reached a cyclical low in February 1999, five months ahead of soybeans, and meal prices are well into the bullish phase of the price cycle that began at that low.
Potential world meal production this season is being held in check by reduced production of soybeans. However, world meal supplies are not tight due to larger supplies of other oilseed meals as well as fish meal. The world demand picture is mixed. Stronger economies in Asia are boosting imports with the important exception of China, which has a duty structure that encourages the import of soybeans in lieu of meal. China's absence from the market has caused a decline in world meal trade this season. EU meal imports, which account for about 50% of world trade, will decline slightly this season, according to the USDA. However, U.S. meal exports have been supported by reduced export supplies from Brazil and India. Shipments from Argentina, the world's largest meal exporter, are likely to approach or be slightly larger than last year's level during the current world crop year.
Exports are the swing factor for U.S. soybean meal usage, but the bulk of U.S. production is consumed domestically. In addition, the domestic market acts as a repository for any excess meal production. Because the increase in broiler numbers is slowing and hog numbers are still down from last year's levels, the outlook for domestic meal offtake is not rosy. Broiler/feed price ratios have been declining, but still do not look low on an historical basis. However, increases in administration and selling expenses have hurt, and in some cases removed, profitability for broiler producers. Hog feeding remains profitable, but not enough to encourage much expansion in animal numbers. Hence, if increasing feed costs further erode profitability, domestic usage of meal could be rationed. This is the type of scenario we would expect to see at the top of the current price cycle. We are not there yet, however.
Soybean meal prices are very sensitive to the soybean supply outlook, and it still looks likely that uncertainty about the size of the U.S. soybean crop, and to a lesser extent, the South American crop, will push meal prices higher into the spring. Soybean meal prices remain fairly low historically, trading in the bottom quartile of their range over the last 20 years. However, as soybean meal prices move higher, it is important to be aware that high meal prices contain the seeds of their own demise.
SUPPLY/DEMAND--The size of this year's South American soybean crops will have more impact on the U.S. soybean meal balance in 2000/01 than in the current season because the effect of a short or surplus South American crop tends to be felt in exportable supplies the second half of their season (the first half of the new U.S. crop year). In its March report, the USDA left its Brazilian crop estimate at 30.5 MMT, but raised its Argentine crop size to a record 20.0 MMT. However, combined production is still expected to be below last year's level of 50.9 MMT.
There was a rumor last week that the Brazilian crushers group ABIOVE might be successful in persuading the Brazilian government to charge export taxes on soybeans to allow crushers to better compete for supplies. Since the differential export taxes were eliminated beginning with the 1997 crop year, there has been an increase in soybean exports at the expense of crush and meal exports in Brazil. As a result of this factor and rapidly expanding production, Argentina overtook Brazil as the world's largest soybean meal exporter. A Brazilian Agricultural Ministry official denied that any changes in the tax structure were being considered; if such a change were made, it would be detrimental to U.S. soybean crushers and soybean meal exports. The state of Parana is scheduled to raise road tolls by 116% beginning March 27. If that measure takes effect, it will raise the cost of transporting both soybeans and meal by truck to the main soy exporting port of Paranagua, which is located in Parana.
An annual U.S. agricultural attache report released last week forecast an increase in China's oilseed production and a reversal in this year's trend toward imports of oilseeds in lieu of products. China's soybean production was forecast to increase to 15.3 MMT from 13.9 MMT this year on a 10.2% increase in plantings after an 8% decline last year. Soybean imports were projected to decline to 2.75 MMT from 4.4 MMT this year, a figure that looks low. However, soybean meal imports were forecast to reach 2.155 MMT versus only 750,000 tonnes this year and 1.408 MMT in 1998. This would be below the level of 3.6 MMT imported in 1997/98. In soybean meal-equivalent, the forecast calls for Chinese imports of 4.34 MMT next season versus 4.25 MMT this season, a very conservative increase, but one that assumes a 1.11 MMT soybean meal-equivalent increase in production.
In its world balance sheet released last week, the USDA reduced its world meal export projection for 1999/2000 by 200,000 tonnes. The USDA retained its export estimates for Argentina and Brazil at 14.07 and 9.74 MMT, respectively. (By comparison, U.S. exports were projected at 6.35 MMT). The USDA lowered its estimate of Indian soybean meal shipments to 2.08 MMT from 2.30 MMT last month and 2.80 MMT in 1998/99. The USDA reduced its estimate of China's soybean meal imports this season to 500,000 tonnes from 620,000 tonnes estimated last month and 1.40 MMT shipped last year. EU imports were reduced to 19.17 MMT versus 19.22 MMT shown last month and 19.33 MMT last season.
The USDA made no changes in its U.S. soybean meal balance sheet, other than increasing its forecast of annual average prices for 48% soy meal at Decatur by $5 on each end of the range, to $150-$170. We have reduced our soybean meal export projection (and our soybean crush estimate) to 7.536 million tons, but it remains above the USDA's forecast of 7.1 million. As of March 2, cumulative meal exports were 97% of the year-earlier level, lagging by 90,000 tons, while unshipped export bookings had edged higher to 99% of a year earlier, or only 15,000 tons behind last year's pace, after good weekly sales were reported. The slow pace of the South American harvest this year is likely to result in exports during the March-June period exceeding last year's level. During the first five months of the season (for which official monthly reports or weekly figures are now complete), soybean meal exports appear to have increased 6% versus a year ago, largely because officially reported figures to date have exceeded indications from weekly shipment data.
Domestic disappearance during the first quarter of the season was up 1.2% versus the year-ago level. For the first five months of the season, we estimate it to be up 1.6%. For the season, we project usage will be up only 1% as higher prices and waning feeding profitability take a toll on consumption. The USDA's estimate of 31.15 million tons looks slightly optimistic.
Soybean meal prices have support in the nearby contract at $158 per ton, but it appears that the May, which will be the nearby contract on Wednesday, made a setback low at $160.40. and the July contract reached a low at $162.50. We expect prices to work higher into the May/June period toward $198 in the July contract, with a chance of reaching the $220 level if there is concern about U.S. yield prospects at planting time. Price patterns beyond June will hinge on growing weather.
SOYBEAN OIL--THE BIG PICTURE--Exports and stocks are the two factors with the most influence on soybean oil prices, and this year's decline in exports and increase in stocks are both factors that argue for lower prices. The world supply/demand balance sheet for fats and oils in 1999/2000 features record production led by record rapeseed and sunflower seed crops and record production of palm oil. In the United States, soybean meal offtake still exceeds that for soybean oil on a soybean-equivalent basis. Thus, crush to meet meal consumption has overproduced oil, adding to stocks. The main culprit in the oil stocks increase has been the decline in soybean oil exports, a reflection of: (1) reduced imports by China; and (2) heavy export competition resulting from abundant supplies of cheaper palm oil and large supplies of competitively priced South American soybean oil. U.S. soybean oil ending stocks are expected to increase for the second consecutive season this year, and stocks at the end of each quarter of the crop year should be above the year-earlier levels.
Nearby soybean oil prices reached a low of 14.65 cents per pound in July 1999, which is a candidate for a possible cyclical low. However, the price level was not as extreme as that seen in 1986, when prices dipped below 13.00 cents, while the cyclical lows in soybeans and soybean meal (protein-percent adjusted) seen in 1999 were lower than their comparable lows of the mid-1980's. Cyclical price patterns indicate that soybean oil has another chance to make a cyclical low in the August/September period, but that probably would hinge on the outlook for favorable U.S. soybean production this year. We still expect to see the oil share set a low in the 30%-31% range this spring in conjunction with a high in the soybean meal price.
The next major turn in soybean oil prices will be upward, and even though we do not expect that to occur until late summer, soybean oil is a highly anticipatory market and tends to react quickly to changes in fundamentals. The following factors could trigger a change in the psychology and price pattern for soybean oil:
--The increase in Malaysian palm oil production is expected to slow in 2000. During the April-July period, Malaysian palm oil production is likely to fall below last year's level.
--It appears highly likely that plantings of rapeseed/canola and cottonseed will decline in most major producing countries this season, tightening potential world oilseed supplies.
--China is likely to import more vegetable oil in lieu of oilseeds next season. A recent agricultural attache report forecast China's imports of palm oil, soybean oil and rapeseed oil combined at 3.035 MMT in 2000/01 versus 2.380 MMT this season. Oil World is forecasting combined imports of the three oils at 2.630 MMT in 2000/01 versus 2.430 MMT this season. However, if China is admitted to the World Trade Organization (WTO), it would have to raise its import quota for soybean oil to 1.7 MMT versus forecast imports this year of 800,000 tonnes to 1.0 MMT. Although palm oil import quotas under the WTO have not been negotiated, it is reasonable to expect a minimum import quota initially in the range of 1.5-2.0 MMT, with an additional quantity for rapeseed oil.
--India's domestic consumption of edible oils is likely to continue increasing because of its growing economy and population pressures, even if (as seems likely) domestic oilseed production recovers in 2000.
--In the United States, lifestyle changes--including more meals consumed in and taken out from restaurants as well as diets that are switching the emphasis from reducing fat to reducing sugar and carbohydrates--should keep domestic oil consumption growing.
--Finally, in the event that adverse weather affects U.S. soybean production, soybean meal would be the stronger product. An advance in soybean meal prices would further hurt already fragile feeding profitability and would ration domestic meal usage and soybean crush. A reduction in soybean crush would then tighten soybean oil supplies. This factor is conjecture at this stage of the growing season, but bears mentioning.
SUPPLY/DEMAND--The USDA reduced its soybean oil export projection 100 million pounds in last week's Supply/Demand report and increased its domestic usage figure by 100 million pounds. Thus, ending stocks were unchanged at 2,130 million pounds, up from 1,520 million last year and the second highest on record after the 2,239-million-pound figure seen in 1991/92. We essentially agree with the USDA's export projection, but remain more optimistic on domestic disappearance.
As of March 2, soybean oil shipments, as reported by the Export Sales report, were only 40% of the year-earlier level, and unshipped bookings stood at 53% of the level seen one year ago. With Malaysian palm oil stocks likely to stay above 1.0 MMT for the remainder of the season, and with an increase forecast for this year's Argentine sunflowerseed and soybean crops (Argentina is the largest exporter of both sun oil and soybean oil), there is no reason to expect a recovery in the pace of U.S. soybean oil exports this season. Palm oil prices continue to trade at a discount to soybean oil.
The domestic usage outlook is more positive than the export outlook. Consumption of soybean oil in salad and cooking oil, baking and frying fats and margarine was up 2.7% for the first four months of the season, even though usage in margarine declined about 8%. With the notable exception of cottonseed oil and tropical oils, domestic production and imports of competing fats and oils are stagnating or declining this season. However, it should be noted that prices of canola oil, corn oil and palm oil are declining relative to soybean oil. Cottonseed oil prices are trading at a lower premium to soybean oil than for the comparable period last year, but this premium remains unusually high. We estimate that apparent domestic usage in the first five months of the season rose 7.6%, probably including some increases in invisible stocks. We look for domestic consumption this season to rise 3.7% versus last year's level. A surge in offtake that occurred in the last four months of the 1998/99 crop year will make this year's consumption look lower by comparison during that period.
Our ending stocks projection of 2.022 billion pounds is lower than the USDA's figure, but still ample. In addition, we are tentatively projecting another increase in stocks next season, as did the USDA at its February Outlook Forum. As a result, the fundamental picture still looks bearish for soybean oil. However, much of the bearish outlook may have been discounted for the time being. Despite a brief short-covering rally, managed funds probably remain heavily net short, which means there is buying power in abeyance. We think the May contract can rally to the 17.10- to 17.35-cent range before retesting the recent contract low of 15.60 cents. We expect the nearby contract to trade in a range of 15.00-17.50 cents into the spring. Our spring high target on the July contract is 17.50-18.00 cents, with an outside target of 19.00 cents if soybean and meal prices hit our extended targets of $6.50 and $220, respectively.
Anne Frick
Hosted by:
CONSENSUS, INC. AND INVESTORS
CO-OP
1737 McGee
Kansas City, MO 64108
(816) 471-3862
editor@consensus-inc.com