FEED YEARBOOK April 28, 1998 April 1998, FDS-1998 Approved by the World Agricultural Outlook Board ----------------------------------------------------------------------------- FEED YEARBOOK is published annually by the Economic Research Service, U.S. Department of Agriculture, Washington, DC 20036-5831. This release contains only the text of the FEED YEARBOOK -- tables and graphics are not included. Printed copies of this yearbook are available from the ERS-NASS order desk. Call, toll-free, 1-800-999-6779 and ask for stock #FDS-1998, $21. ERS-NASS accepts MasterCard and Visa. ---------------------------------------------------------------------------- Summary U.S. feed grain production in 1997/98 totaled 265 million metric tons, down 1 percent from 1996/97. However, total supply is expected to be up 4 percent because of larger carryin stocks. Total use of the four feed grains in 1997/98 is expected to be up 3 percent from last year. Increased numbers of livestock and poultry and increased milk production are expected to boost feed and residual use of corn and the other feed grains. Food and industrial use of corn in 1997/98 is likely to be up sharply from 1996/97. However, large exportable corn supplies in other nations and weaker Asian demand have cut expected U.S. corn exports. The reduced export prospects and increased corn supplies have led to weaker corn prices. In 1997, corn production was up about 1 percent from the 1996 crop because of larger harvested area. Yields were virtually unchanged from 1996. On March 1, 1998, producers reported they intend to plant 1 percent more acres to corn than in 1997. Grain sorghum production in 1997 was down 19 percent from 1996. The reduction was because of a decline in planted acreage as yields were up 2 bushels per acre in 1997. Prospective plantings for 1998 were reported down 11 percent from 1997, suggesting further production declines. Barley production in 1997 was down 6 percent from 1996 mainly due to lower acreage. Even with declines in production, the all-barley price received by farmers in 1997/98 has been weaker than in 1996/97. Despite the strongest exports in 5 years, feed barley prices have been weak. However, malting barley prices have been relatively strong because of planting and growing problems in the major malting barley production areas. Oats production in 1997 was up 13 percent from 1996's record low crop. Still, the 1997 crop was the third lowest on record since 1866. Yield was the third highest on record. Supplies of oats in 1997/98 are expected to be up 11 percent from 1996/97 mainly because of a 13-percent increase in imports. Global coarse grain production in 1997/98 will be down from the record high of 1996/97, but still the second highest ever. China's drought-reduced corn crop accounts for the largest proportion of the decrease in global production. Outside of China, foreign production is increasing, while coarse grain consumption is forecast mostly unchanged. World coarse grain trade is forecast down, led by declines in barley. World corn trade will not drop as much. Five special articles explore various grain-related issues. One article discusses the impact of new technology on the corn sector with a 1998 update and prospects for the future. A second article re-evaluates use of the soybeans-to- corn price ratio as an indicator of acreage changes in light of increased planting flexibility. The third discusses the complexities in estimating feed use when survey data are unavailable. The fourth article examines Southeast Asia's feed imports and the effect of the region's financial crisis. The final article presents a model that analyzes market factors and government programs affecting U.S. corn prices. Caution: Supply and use data in this publication are as of the March 12, 1998 World Agricultural Supply and Demand Estimates report and do not reflect more recent reports and changes in the outlook. Feed Grain Supply and Use Supply and Use Expected To Rise in 1997/98 Total supply of feed grains in 1997/98 will be down 1 percent from 1996/97. Total use is projected to increase 3 percent from 1996/97's 257.6 million metric tons. Feed grain production in 1997/98 totaled 265 million metric tons, down only 1 percent from 1996/97 in spite of a lot of weather worries. However, output still was well below the exceptional year in 1994/95 when production was 284 million tons. Total supply of feed grains for the 1997/98 year is expected to be 295 million metric tons, up 4 percent from the prior marketing year. Increased beginning stocks in 1997/98 account for most of the difference, because imports are expected to be only marginally higher. Total use of the four feed grains in 1997/98 is expected to be up 3 percent from 1996/97. The only use category that is expected to decline in 1997/98 is exports. Exports are forecast at 48 million metric tons, down 7 percent from 1996/97. Feed and residual use in 1997/98 may total 166 million metric tons, up from 157 million in 1996/97. Lower cost grains and more animals have led to more use. Food, seed, and industrial uses in 1997/98 are expected to total 53 million tons, up from 49 million in the prior marketing year. Corn Use To Remain Strong in 1997/98 The strong economy has kept food and industrial uses of corn well above 1996/97. Exports of corn in 1997/98 are expected to be weaker than the prior marketing year. Prices received by farmers will be down from 1996/97. The 1997 corn for grain production was estimated at 9.37 billion bushels, up less than 1 percent from the 1996 crop. The 1997 production level ranks third behind 1994 and 1992, respectively. The U.S. yield of 127.0 bushels per acre was virtually the same as the 127.1 bushels of a year earlier. Planted area totaled 80.2 million acres, 1 percent more than in 1996, and the largest planted acreage since 1985. An estimated 73.7 million acres were harvested for grain, also up nearly 1 percent from a year earlier. Despite below-normal temperatures, corn planting progressed rapidly and finished ahead of the normal pace. As of June 1, 1997, corn planting was 98 percent complete, compared with the average of 89 percent. In July, extremely dry and occasionally hot weather in the central Corn Belt highlighted a month of extremes. Corn progress took a dramatic jump in the middle of July, corresponding to the jump in planting progress earlier in the spring. Lack of moisture began to stress the crops as the month progressed, but temperatures turned cooler at month's end, moderating the stress slightly. Crop conditions remained good in areas that received moisture, but conditions declined in fields only a few miles away that missed the scattered showers. By the beginning of August, 85 percent of the corn acreage was silking, compared with the average of 75 percent. At that time, 66 percent of the corn was rated in good to excellent condition, compared with 62 percent in 1996. As of the beginning of October, the majority of the Corn Belt remained frost-free, with the exception of parts of Michigan and Wisconsin. After a mid-October killing frost and snowstorm, favorable harvest conditions prevailed in the Midwest until early November. In the western Corn Belt, grain moisture levels were low and farmers were able to make rapid harvest progress. As of November 2, 75 percent of the corn crop was harvested compared with an average of 64 percent. As of November 30, 97 percent of the acreage in the 17 major States was harvested. This compares with 96 percent in 1996 and the 5-year average of 93 percent. The 1997 corn objective yield data indicated an ear count slightly below 1996, but above the 5-year average for the seven objective yield States (Illinois, Indiana, Iowa, Minnesota, Nebraska, Ohio, and Wisconsin). The record level ears per acre was set in 1996. Exports Decline Total use of corn in 1997/98 is expected to increase 5 percent from the 8.85 billion bushels used in 1997/98. Increased livestock and poultry numbers are likely to boost feed and residual use to 5,850 million bushels, up from 5,362 million used in 1996/97. Food, seed, and industrial (FSI) use in 1997/98 is forecast at 1,835 million bushels, up from 1,692 million a year earlier. Increases in the production of corn sweeteners and ethanol in 1997/98 account for much of the rise. The strong economy has kept consumers purchasing and increasing sales for the corn processors. However, large exportable corn supplies in other nations and weaker Asian demand have cut expected U.S. corn exports. In 1997/98, corn exports may total 1,625 million bushels, down from 1,795 million in 1996/97. Corn Prices Decline The reduced export prospects and increased corn supplies have meant weaker corn prices. In 1997/98, prices received by farmers may average $2.45 to $2.65 per bushel, down from $2.71 in 1996/97. Prospective Plantings Producers indicated in March that they intend to plant 80.781 million acres of corn in 1998, up 1 percent from the 80.227 million planted in 1997. Farmers in Iowa intend to plant 12.2 million acres, the same as in 1997, and the largest acreage of any State, followed by Illinois which was down 1 percent, and Nebraska, off 2 percent from 1997. The largest increases are 300,000 acres each in Texas, North Dakota, and Minnesota. Expected plantings are also up in South Dakota, Kansas, and the Southeast. Assessment of Potential Cropland in the United States Critics of the Conservation Reserve Program (CRP) have raised the issue of the potential for future market disruptions to be exacerbated by an aggressive CRP program. The concern is that cropland available for planting will be so restrictive that the United States will be unable to respond to future growth in demand for grains and oilseeds, potentially resulting in declining export market shares over time. The 1996 farm legislation eliminated acreage reduction programs, but re-authorized the Conservation Reserve Program with enrollment of up to 36.4 million acres. This assessment of cropland availability uses acreage surveys by the U.S. Bureau of the Census, National Agricultural Statistics Service (NASS), the Farm Service Agency (FSA), and the Economic Research Service (ERS) to evaluate cropland resources available to production agriculture. Further, cropland resource inventories, conducted by the National Resource Conservation Service (NRCS), combined with NRCS soil scientists' research, provide a preliminary approximation of the productivity of the U.S. farmland inventory. Every 5 years the U.S. Bureau of the Census surveys all U.S. farmers to conduct an exhaustive inventory of U.S. agriculture, including cropland inventory and utilization. The latest agricultural census, conducted in 1992, indicated that cropland in the United States totaled 435.2 million acres. This is a decline of just under 1 million acres per year since 1982. Estimates of cropland for 1993-97 assume that the reduction in cropland continued at this rate. Thus, total cropland for 1997 is estimated at 431 million acres (table 1). Annual estimates of planted and harvested acres of crops are published by NASS based on sample surveys of farmers. These annual estimates are revised every 5 years to reflect the results of the Census of Agriculture. NASS estimates indicate that planted and harvested acres of major field crops, hay, and fruits and vegetables constitute the bulk of annual cropland use. ERS annually estimates acres of summer fallow and other crops 1/. 1/------------ Agricultural Resources and Environmental Indicators, 1996-97, Agricultural Handbook Number 712, Economic Research Service, United States Department of Agriculture, July 1997. -------------- Double-cropping activities are largely related to soybeans following a spring- harvested crop, and vegetables or other crops following vegetable crops. The annual acreage of double-cropping of soybeans is estimated from NASS survey data (table 1). The acres of double-cropping due to multiple vegetable crops are estimated based on Census of Agriculture data (table 1). How Much Cropland Remains Idle Each Year? Total cropland in 1991 was estimated at 437 million acres, of which only 337 million acres were planted to crops. Therefore, nearly 100 million acres of cropland were idled in that year, of which nearly 70 million acres were idled under annual programs (acreage reduction program [ARP}, 0-50/85-92, and idled flex acres) and the long-term CRP (Table 1). The remaining 30 million acres were nonprogram idled cropland. These estimates are similar for 1995, the last year of acreage reduction and price support programs. By 1996, total cropland had declined to an estimated 432 million acres, down about 5 million acres from 1991. However, because of relatively strong crop prices, greater planting flexibility due to the changes brought about by the 1996 farm legislation, and fewer acres enrolled in the CRP, total cropland used for all crops exceeded 345 million acres, up more than 12 million acres from the previous year. Total cropland idled declined to about 86 million acres, of which about 33 million were in the CRP. This left nearly 54 million acres of nonprogram-idled acres. Continued favorable expected returns in 1997 boosted total land used for all crops to nearly 353 million acres, up about 7 million from 1996, with no appreciable change in CRP acreage. Total idled cropland was estimated at about 79 million acres in 1997, with about 46 million acres outside the CRP. The assessment of farmland available for crop production shows that producers have responded to market signals, i.e., higher returns, with larger plantings. Since the end of annual set-aside programs, total cropland devoted to cropping activities has increased nearly 20 million acres with little change in CRP enrollment. Is Cropland in the CRP More Productive Than Other Idled Cropland? NRCS has conducted extensive surveys to inventory and classify the types, quality, and productivity of cropland in the United States. Many factors influence soil productivity, including soil chemistry, climate, and physical properties. NRCS has developed an index that uses many of these factors to provide a single-valued measure to compare the productivity of land with broadly different characteristics. The soil rating for plant growth (SRPG) is an index for estimating a soil's relative productivity for producing seed, forage, and fiber. It considers soil chemical and physical properties, climatic data, and landscape features. The numerical index values for SRPG range from 0 to 100: soil with an index of 66 would be 10 percent more productive than a soil with an index of 60. The average SRPG index for all U.S. cropland is 61. Aggregate data show that land being cropped is more productive than land in the CRP. For example, it is estimated that the 33 million acres of cropland in the CRP in 1996 had an average productivity index (based on the SRPG) of 52.8, compared with an estimated index of 62.5 for all U.S. cropped land and 56.9 for other idled cropland. The selection criteria used for consideration of land enrolled in the CRP for sign-ups 15 and 16 have assured that the less productive land will continue to be enrolled in the CRP. For sign-ups 15 and 16, the average index for selected acres was 52.2 and 52.7, respectively. Are Industry Concerns of a Fully Implemented CRP Warranted? Based on the assessment of cropland availability and its associated productivity, sufficient cropland capable of reasonable crop production remains available to satisfy short-term variation in cropping needs. In 1997, cropped land exceeded 350 million acres, up about 20 million acres from 1995 when 25 million acres were idled in annual programs. With about 30 million acres in the CRP, it is likely that over 40 million acres of cropland with average productivity exceeding the average productivity of land in the CRP will be idled by farmers. This suggests that the CRP program, as it is currently implemented, is not a major impediment to expanding planted crop acreage in the future. Rather, it is the economics of crop production that will likely determine total land in crop production in the future. If market prices are sufficient to cover variable cost per acre on the most productive idled cropland, additional cropland may be brought into production--as it has been during the previous 2 years. Box end Sorghum Production Down 19 Percent Area planted to sorghum declined sharply in 1997 from the increased acreage in 1996. The resulting lower production is expected to result in total use being down 15 percent from 1996/97. Grain sorghum production for 1997 was estimated at 653 million bushels, 19 percent below 1996. Grain yields, at 69.5 bushels per acre, were 2.0 bushels above the 1996 average. The final estimate of all sorghum planted was 10.1 million acres, down 23 percent from the year earlier. Area harvested for grain was estimated at 9.4 million acres, down 21 percent from 1996. A cool, wet spring delayed plantings in most areas. Kansas and Texas showed the largest declines in planted acreage from 1996, down 1.15 and 1.10 million acres, respectively. In 1996, a large amount of abandoned winter wheat acres were replanted to sorghum in Kansas, and Texas farmers substituted sorghum on ground where it had been too dry to plant cotton. Timely rains last August, along with mild late summer weather in the major sorghum areas, contributed to good growth and development of the U.S. sorghum crop. Harvest progressed well, with only a small area of unharvested sorghum left at the end of December. Sorghum Use Declines The reduced supply of sorghum in 1997/98 has meant a reduction in the use of sorghum. In 1997/98, feed and residual use of sorghum is expected to be down 20 percent from the 529 million bushels used in 1996/97. In addition, FSI use may be down 12 percent from the 40-million bushels estimated to be used in 1996/97. Most of the FSI use is for alcohol production because in dry mills, alcohol yields are the same as what can be obtained by using corn but at a lower price. The "spent" sorghum from alcohol production is more desirable by many feeders than corn "spent" grains. Grain Sorghum Prices Weak Prices in 1997/98 for grain sorghum are expected to range from $2.15 to $2.35 per bushel, compared with 1996/97's $2.34. In spite of a smaller crop, sorghum prices continued to follow corn. In 1997/98, sorghum prices are expected to average around 88 percent of corn, down from the 10-year average of 92 percent. Prospective Plantings for 1998 On March 1, farmers indicated they intended to plant 9 million acres of sorghum, down 11 percent from 1997. If farmers follow through with their reported intentions, Kansas will have the largest acreage again in 1998, followed by Texas and Nebraska. Texas farmers indicated they planned to cut back the number of acres of sorghum the most of any State at 700,000 acres, Kansas was second at 150,000, and Nebraska was third at 100,000 acres. Among the seven States indicating increased plantings or no change from 1997, Colorado and Louisiana were both up 20,000 acres. Barley Production and Use Decline in 1997/98 Barley production in 1997 was down 6 percent from 1996. Exports of barley in 1997/98 have remained strong, and with smaller supplies, feed and residual use is expected to decline from 1996/97. Barley production in 1997 is estimated at 374 million bushels, down 5 percent from 1996 because of lower area. The national yield averaged 58.3 bushels per acre, down slightly from 1996's 58.5. Planted acres in 1997 were down 3 percent to 6.9 million, but less acres were harvested, leaving harvested acres down 5 percent. North Dakota continued to rank as the number one barley-producing State, followed by Montana, Idaho, Washington, and Minnesota. These five States produced over three-fourths of the 1997 barley crop. In North Dakota, acres harvested for other uses or abandoned increased from the previous year due to dry conditions in the western portion of the State, head disease in the east, and a regional shortage of hay. North Dakota yields dropped 10 bushels from 1996 to 45.0 bushels per acre. Conversely, yields in Montana increased 10 bushels over 1996 as timely rains during the growing season helped set the stage for an excellent year. The eastern portion of the State experienced some drought-like conditions early in the summer, but late rains proved beneficial and boosted yields. Spring field preparation progressed at a much slower rate across Washington due to cold, wet conditions. Seedings progressed at a below-normal rate early in the season, but reached normal progress by mid-May. The crop was slower to develop due to cool temperatures, but conditions throughout most of the year were rated as good to excellent. Average yields in Washington, at 76.0 bushels per acre, surpassed the previous record of 75.0 bushels set in 1980. Strong Exports Highlight Barley Outlook Exports for 1997/98 are expected to reach 80 million bushels, up 49 million from 1996/97 and the largest in 5 years. Despite the absence of the Export Enhancement Program, there were strong sales early in the season, especially to Saudi Arabia. The European Union (EU) apparently miscalculated the strength of world barley supplies and refused to raise subsidies. This allowed some optional origin barley sales to Saudi Arabia to be supplied by U.S. exporters. While the plentiful supplies of corn were expected to lead to more corn use for feeding and less barley in 1997/98, the robust pace of exports will mean less domestic use than expected earlier. The forecast of feed and residual use, at 160 million bushels, is down 60 million from a year earlier. In 1997/98, barley used for FSI is the same as a year earlier, despite some quality concerns stemming from disease. Food and industrial demand for barley is largely inelastic since substitution possibilities are limited. However, domestic supplies will be augmented by imports, which consist largely of malting barley. Barley Prices Decline Even with declines in production, barley prices received by farmers in 1997/98 have been weaker than a year ago. In 1996/97, much of the crop was sold when corn supplies were short and prices were high. However, as corn supplies have improved, barley prices have come down. Despite the strongest exports in 5 years, feed barley prices have been weak. However, malting barley prices have been relatively strong because of planting and growing problems in the major malting barley production areas. Prospective Plantings for 1998 Farmers reported intentions to plant 6.78 million acres of barley in 1998. If realized, this will be a decrease of 2 percent from the 6.91 million planted in 1997. North Dakota has the largest intended acreage, followed by Montana and Idaho, the same top three States as in 1997. Farmers in North Dakota reported they intend to cut plantings 4 percent from a year earlier, while Montana producers plan to increase plantings 8 percent, and Idaho producers may cut back 1 percent in planted acreage. In terms of a decline in acreage, Minnesota producers plan to cut 120,000 acres from 1997, and Montana producers would increase plantings by 100,000 acres. Oats Production Increases 13 Percent In 1997/98, oats supplies are expected to be up 11 percent from 1996/97's 319 million bushels. Total use may be 272 million bushels, up 8 percent from 1996/97. Production of oats in 1997 totaled 176 million bushels, 13 percent above 1996's record low crop. Despite the increase in production, the 1997 crop is the third lowest production on record since 1866. Grain yield per harvested acre averaged 60.5 bushels, up 2.7 bushels from 1996 and the third highest yield on record. Area harvested for grain, at 2.91 million acres, was up 8 percent from a year earlier, but is the second lowest harvested acreage since estimates were first made. In North Dakota, harvested area for grain declined due to a regional hay shortage and low grain yields. Similar circumstances were reported in several other States in the northern Plains and upper Mississippi Valley. Yield was up 2.4 bushels from the August forecast. In North Dakota and surrounding States, the lowest yielding acres were apparently harvested for hay, thus increasing yields. Also, oats growers in several eastern Corn Belt States realized record or near record yields. The start of oat planting was delayed by extensive flooding in the Red River Valley of North Dakota and Minnesota. Cool, wet soils in the Northwest also delayed planting. Soil moisture supplies were adequate for the growing season, except in the northern Plains. There, hot, dry weather allowed the crop to catch up to the normal pace of development but stressed the crop during critical stages of growth. During August, generally hot, dry weather across the United States provided ideal harvest conditions. Oats harvested in the nine major producing States were 91 percent complete by August 31, 5 percentage points ahead of the 5-year average. Nearly One-Third of Oats Imported Even with the increase in production in 1997, supplies are being boosted by the increase in imports of oats, primarily from Canada, with lesser amounts from Finland and Sweden. Imports in 1997/98 are expected to total 110 million bushels, up from 97 million in 1996/97. All three of the countries tend to have cooler summers that are conducive to production of the heavy white oats favored by the processing industry and many horse enthusiasts. As a result, imports are forecast to comprise about one-third of the oat supply in 1997/98. Total use of oats in 1997/98 may rebound to 272 million bushels, up 20 million from a year earlier, and near the use of 1995/96. Ending stocks may be up 21 percent from the 67 million in 1996/97. Food use is expected to remain about at the level of 1996/97, even with the ability of food processors to claim health benefits of oat consumption. It is expected that those who wanted the health benefit of oats have already increased oats in their diets and others will not. Prices Down from 1996/97 Prices received by farmers for oats in 1997/98 are expected to average between $1.55 and $1.65, down from the $1.96 received in 1996/97. Prices are expected to be weaker because of the larger supplies of oats and plentiful supplies of competing feed grains. Prospective Plantings At the first of March, farmers indicated they intended to harvest for grain 3,058,000 acres of oats, up 5 percent from 1997. Surprising yields in some areas and forecasts for a cool, wet spring again, may have prompted farmers to try oats again this year. North Dakota farmers reported intentions to harvest for grain more oats than any other State, followed by Minnesota and South Dakota. The largest producing States also had the largest acreage increases reported, with 100,000 in North Dakota, 80,000 acres in Minnesota, and 20,000 in South Dakota, an equal increase with Wisconsin. Hay Stocks Decline, Record Prices Likely Hay stocks per roughage consuming animal unit on December 1, 1997, are the same as in 1996. Hay prices in 1997/98 have been above year-earlier levels until February 1998 and will likely set a new record. Stocks of all hay on farms December 1, 1997, were down 2 percent from 1996's 105 million tons. Stock decreases occurred in 26 of the 48 contiguous States, mainly in the upper Midwest, New England States, Mid-Atlantic States, and California, Nevada, and Oregon in the West. The largest stock increases occurred in the Mountain States, eastern Corn Belt States, and most of the Southern States. Roughage consuming animal units (RCAU) in 1997/98 are estimated to be down 2 percent from 1996/97. Thus, hay stocks per RCAU on December 1, 1997, are the same as in 1996/97, 1.38 tons/RCAU. The winter has been warmer than usual and many areas have been able to utilize pasture much more than usual this year. As a result, less hay may be needed, and market year ending stocks may be above a year earlier even though stocks per RCAU would suggest similar disappearance. Hay production in 1997 totaled 152 million tons, up 2 percent from the revised 1996 total. Acreage of all hay was almost the same as a year earlier, but yields were up 2 percent at 2.5 tons per acre. Texas became the number one hay-producing State followed by California, South Dakota, Missouri, Kansas, and Nebraska. All States in the West except Nevada, which was unchanged, showed increased production. Texas and Arizona had the largest increases at 38 and 22 percent, respectively. The increases were nearly offset by declines in the New England States, Mid-Atlantic States, and the South. Alfalfa and alfalfa mixtures were down 275,000 tons, or less than 1 percent from 1996. All of the decline resulted from a decline in acreage of 2.5 percent. Yields in 1997 were up 2.4 percent, offsetting almost all of the acreage decline. The National Agricultural Statistics Service collected data on the number of acres seeded to alfalfa and alfalfa mixtures during 1997. Wisconsin was the number one State with 650,000 acres seeded, followed by Minnesota with 300,000. (See table on new seedings of alfalfa and alfalfa mixtures in Crop Production, 1997 Summary, page A-88.) Other hay production was up 4 percent from 1996's revised 70 million tons. In 1997, the area harvested of other hay was 37 million acres, up 1.5 percent from 1996. Average yields in 1997 were up 2.6 percent from the 1.91 tons per acre in 1996. Production increases were common across the Nation with the exception of the New England States, the Mid-Atlantic States, and the Southern States of Louisiana and Florida. Corn for silage in 1997 was up 10 percent from the 84 million tons produced in 1996. Acreage was up 6 percent, and yields were up 4 percent. Sorghum for silage in 1997 totaled 4 million tons, down 11 percent from 1996, with a 16-percent decline accounting for all of the drop. Total supplies of harvested roughage (hay stocks on May 1, plus hay and silage production) in 1997/98 are up 2 percent from the 258 million tons available in 1996/97. Supplies of harvested roughage per roughage consuming animal unit in 1997/98 are 3.6 tons, compared with 3.4 tons in 1996/97. Prices remain firm relative to historical levels, but all hay prices reported by farmers in February were below a year earlier. In 1996/97, prices received by farmers were record high, averaging $93 per ton for the season. The smaller crop in 1996 helped strengthen prices, especially for high quality hays. Early spring rains tended to keep prices high early in the 1997/98 marketing year, as the market worried about the potential hay crop. Prices declined as the harvesting progressed but still did not go below prior year levels until February 1998. Prices received for alfalfa hay in 1997/98 appear to be headed for a record high as prices have been above a year ago until February. Other hay prices received by farmers may average near a year earlier as the prices have been above and below last year's price. Other hay was selling below last year during July through September, then moved above a year earlier until January. Prospective Plantings for 1998 Farmers in March indicated they planned to harvest 60.735 million acres of hay in 1998, up from 60.815 million in 1997. Texas, the State with traditionally the largest hay acreage, is expected to be the largest again in 1998, followed by South Dakota and Missouri. In 1998, Texas hay producers indicated they plan to harvest the largest additional acreage at 300,000 acres, more acres than any other State, followed by Oklahoma with 110,000 acres. The State with the largest drop in acreage was South Dakota with 300,000 acres, followed by Montana with 200,000 acres. Twenty-four States reported an increase in expected harvested acreage and 25 reported no change or a decline. Feed Demand Feed Demand To Remain Strong The index of grain consuming animal units (GCAU) in 1997/98 is expected to be up 3 percent from 1996/97's 85 million units. Feed and residual use of the four feed grains plus wheat in 1997/98 is expected to increase 5 percent from the 165 million metric tons used in September 1996 - August 1997. Feed and residual use in September-November 1997 was up 9 percent from a year earlier. Corn accounted for 92 percent of feed and residual use in September-November 1997 and is expected to represent 86 percent of feed and residual use in 1997/98. The index of grain consuming animal units (GCAU's) for 1997/98 is expected to be up 3 percent from 1996/97's 85 million units. Since grain prices have dropped from the high levels in 1995/96, the animals most dependent on grain have been increasing. Cattle on feed, hogs produced, layers, and broilers have all increased, while dairy numbers have been down along with other cattle, sheep, goats, and turkeys. Dairy cows and turkeys consume grain, but special circumstances have kept their numbers lower. The feed and residual used per GCAU in 1997/98 is expected to be 1.96 tons, up 2 percent from 1996/97. Cattle and calves on feed on January 1, 1998, totaled 13.6 million head, up 3 percent from 1997. In addition, the cattle on feed on February 1 for feedlots with capacity of 1,000 head reporting monthly, were up 4 percent from a year earlier; placements were down 9 percent from a year earlier. Thus, current feed use by cattle feedlots is likely stronger than a year earlier, but placements on feed have been below 1997 since last October. Consequently, expected feed use by cattle feedlots is declining because the calf crop from which feeders are eventually drawn was down 1 percent from a year earlier in 1996 and down 3 percent in 1997. The feeder cattle supply outside feedlots on January 1, 1998, was down 4 percent. Pork production in 1998 is expected to increase 10 percent from the 17.2 billion pounds produced in 1997, which was up 1 percent from 1996. Hog farmers responding to the March 1998 survey indicated that they intend to increase the number of sows farrowing in December 1997-May 1998 by 5 percent relative to the prior year. If producers carry through with these reported intentions, feed needs by the pork sector are likely to be stronger in 1997/98. However, with increased production and the Asian financial situation resulting in declining demand for imported meats, U.S. hog prices have been low. Hog prices in the Iowa-South Minnesota market during February 1998 averaged 35 cents per pound, down from 52 cents a year earlier. Even though corn prices are lower than a year earlier, the weak hog prices are hurting profitability in the hog industry. Broiler and egg production in 1998 are expected to increase from the expected 1997 levels and continue strong demand for feed grains. Broiler production in 1998 is expected to increase 4 percent from 1997 as producers respond to strong domestic demand as well as abundant feed supplies and lower prices. In 1998, turkey producers are expected to maintain production even with lower feed prices and about equal the 1997 output. Egg producers are expected to produce 6.6 billion dozen eggs in 1998, up 3 percent from the 1997 output. With these increases in production, feed needs by the poultry sector are likely to remain strong. Dairy cow numbers on January 1, 1998, were down 1 percent from the 9.3 million head on farms on the same date in 1997. In 1997, production per cow averaged 16,954 pounds, up 3 percent from 1996, and milk production was up 1.5 percent from the prior year. Milk production in 1998 is expected to total 157.2 billion pounds, up from 156.5 billion in 1997. On January 1, 1998, grain and other concentrates fed to milk cows averaged 19.8 pounds, up .6 pound from the same date a year earlier. Thus, with the increased grain fed per cow, feed use by the dairy industry will continue strong. Food, Seed, and Industrial Use of Corn Food, Seed, and Industrial Uses of Corn to Increase in 1997/98 Forecast corn use in 1997/98 for high fructose corn syrup, starch, and ethanol are up more than the other uses. Food, seed, and industrial (FSI) use of corn in 1997/98 is expected to total 1,835 million bushels, up from 1,792 million in 1996/97. In 1997/98, FSI use would represent 20 percent of total use, up 1 percent from 1996/97 and 1995/96. FSI use in 1997/98 is expected to be up for all the categories, but forecast corn used for high fructose corn syrup (HFCS), starch, and ethanol are up more than the other uses. In 1997/98, corn used for HFCS production is expected to increase 7 percent from the 504 million bushels used in 1996/97. With the U.S. economy strong and employment high, consumers are expected to keep consumption high. This would include continued strong consumption of soft drinks and "new age" drinks, a major use for HFCS. Corn used to make glucose and dextrose in 1997/98 may increase 2 percent from 1996/97's 246 million bushels. These sweeteners are used in bakery goods to offset the reduction of fat, and this market has likely matured, slowing expected growth. In the 1997/98 corn marketing year, corn used for starch production may be up 5 percent from the same period in 1996/97. The strong economy and paper recycling are expected to keep starch sales strong, plus the mild winter in most areas of the country has kept housing starts up and likely boosted sales of starch to building materials firms. Corn used to make ethanol in September-November 1997 totaled 123 million bushels, up from 96 million in 1996 based on the monthly ethanol production reported by the Energy Information Administration of the Department of Energy. This is the largest quarterly use since the 1994/95 record year. It largely reflects higher use of capacity as the industry continues to recover from the 1995/96 cost-price squeeze. Although there has been a modest expansion of capacity in Minnesota in recent months, a few plants that previously closed have not come back into operation. Corn used to make ethanol in 1997/98 is estimated to be up 20 percent from the 429 million bushels used in 1996/97. Corn used in 1997/98 for beverage and manufacturing alcohol is expected to grow modestly, slightly better than population growth. With the manufacturing industries strong, manufacturing uses of alcohol should remain strong. Cereals and other products produced from corn in 1997/98 are expected to grow at the population growth rate or about 1 percent per year. Growth in production of corn chips and other Mexican foods seems to have tapered off, plus "soft" tacos (made from wheat flour) have appeared on the menu to cut into corn use. In 1997/98, corn prices are expected to be weaker than in 1996/97, so the input costs for corn processors will likely be less than a year earlier. However, the value of their feed by-products may be weaker because soybean meal prices are forecast at $190 to $200 per ton, down from $270.90 in 1996/97. Corn oil prices may be the exception if they follow the prices of soybean oil. Soybean oil prices in 1997/98 are expected to average 25.50 to 27.50 cents per pound, up from 22.50 in 1996/97. Even so, net corn costs to corn processors are likely to be near to slightly weaker than a year earlier. World Coarse Grain Outlook World Coarse Grain Production and Stocks Down, But Prices Declining in 1997/98 The drop in global production and stocks is concentrated in China, but China's exports are up. Outside of China, foreign production is increasing, while coarse grain consumption is forecast mostly unchanged. World trade is forecast down. China's Corn Production Drops, But Exports Increase China's 1997/98 coarse grain production is forecast to fall 19 million tons to 122 million, with almost all of the drop in corn. A hot, dry summer in 1997 sharply reduced yields. The forecast is based on the preliminary 1997 total grain production numbers released by China's government statistical bureau. China has not specifically released an official corn production number. The analysis of China's role in the global coarse grain market is further complicated by the lack of information on the size of grain stocks. Stocks are a state secret in China, and only partial data and anecdotal information are available. No statistically based data on China's feed use are available either. Even though the data and forecasts underlying the analysis of the China market have serious weaknesses, they do reflect a dramatic story. From 1994/95 until 1996/97 China, to support prices, sharply built corn stocks through government purchase programs because of back-to-back record harvests in 1995 and 1996. The record wheat harvest in 1997 also led to a huge flow of wheat into government stocks, prompting the government to unload old corn stocks. As the 1997 corn crop was harvested, additional old-crop corn was sold from stocks to make room for the new crop. Although the 1997 corn crop turned out to be down sharply, the released stocks were large enough to completely compensate, and use, including exports, continued to grow. Soft Demand for Coarse Grains Widespread Since the United States and China are the world's largest coarse grain producers, it is useful to look at the rest of the world, which, this year, accounts for about 57 percent of world production. For these countries, production is forecast up 2 percent in 1997/98. However, their consumption is projected up 0.2 percent, virtually unchanged, and if the European Union (EU) (the world's third largest producer/consumer/exporter), Eastern Europe, and the New Independent States (previously referred to as the former Soviet Union) are excluded from the group, consumption is forecast to decline year-to-year by 30 percent. So the demand that generates world trade is declining. Much of the reduced demand for feed grains is in East Asian markets where imports are down because of financial and macroeconomic problems in Indonesia, South Korea, Malaysia, and the Philippines, and because of hoof and mouth disease in Taiwan's swine. However, consumption is also expected to decline in Latin America, the Middle East, Africa, and South Asia. In Africa and South Asia where human consumption is the largest use of coarse grains, reduced production is not expected to be offset by increased imports partly because of the limited ability of some of the countries to finance imports. Stocks data in many of these countries are unreliable, so bigger crops often lead to bigger estimated use, and smaller crops result in smaller apparent consumption. Since coarse grain crops are smaller this year, consumption is expected to decline. Coarse grain consumption is forecast up in Europe and the New Independent States (NIS) as increased production has boosted availabilities. In Eastern Europe and the NIS, stocks data are also unreliable, and increased production is expected to lead to increased disappearance. In the Middle East, consumption is declining because better than normal rainfall has improved pastures, especially in Saudi Arabia, and feed wheat imports have replaced some coarse grains in Israel and Jordan. In Latin America, coarse grain consumption is forecast to decline in 1997/98, after 8 consecutive years of expansion. Brazil and Mexico are not expected to import enough to completely offset reduced production, and consumption is expected to stagnate in Mexico and drop in Brazil. Livestock sector growth and demand for feed grains are slowing in some parts of the region, but outside of Brazil and Mexico, feed grain consumption is still expected to expand. World Trade Expected To Decline in 1997/98, Barley To Drop Most Global coarse grains trade is forecast down 4 percent in 1997/98 to 89 million tons because of softening demand. Barley trade is expected to show the sharpest year-to-year drop, with corn trade declining more slowly. The EU Commission in the summer of 1997 and into the new year was not subsidizing barley exports aggressively. When Saudi Arabia (the world's dominant importer of feed barley) tendered to import, the EU barley was too high priced and Saudi Arabia purchased from alternative suppliers, especially from the United States and NIS. Then better than expected rains in Saudi Arabia provided ample pastures for the sheep and camels that normally are fed barley, so Saudi Arabia did not return to tender for additional barley as soon as the EU expected. Meanwhile, large intervention stocks of barley are being accumulated in the EU. By March 1998 the prospects for huge increased EU barley stocks led the Commission to authorize very large subsidies, driving world feed barley prices down in order to stimulate demand and move the barley. The large "special" subsidy is not enough to avoid significant stock increases and reduced trade compared with previous years. Another factor contributing to slow barley trade is a drop in China's malting barley imports. Evidently the dramatic increase in imports a year ago resulted in large malting barley stocks in China, and the pace of imports has dropped sharply. World Coarse Grain Ending Stocks To Decline in 1997/98, But Prices To Decline Also Coarse grain ending stocks are expected sharply lower in China where they are relatively isolated from the world market, but increase significantly in the EU and United States. Coarse grain stocks in the rest of the world are expected to be little changed. International coarse grain prices are more determined by supply and demand prospects for U.S. corn and the level of EU subsidies than by global stocks-to-use ratios. With prospects for increased stocks in the United States, and strong competition from Argentina's corn, it is hard to see prices responding to the tighter world supply and demand fundamentals until problems with new-crop prospects develop. U.S. Corn Exports Drop in the Face of Reduced World Trade and Tough Competition World coarse grain and corn trade are forecast down in 1997/98, but foreign exports are expected to increase, dropping U.S. exports and squeezing the U.S. share. Large corn stocks are allowing China to increase corn exports even though production dropped sharply due to drought. U.S. Corn Exports Get Off to a Slow Start in 1997/98 According to U.S. Export Sales, corn export shipments for September through mid-March are running almost 25 percent behind a year ago. Moreover, outstanding sales as of mid-March are also down 25 percent. Corn exports have been down sharply for more than half the marketing year. However, U.S. corn exports and export sales are expected to pick up in the latter part of the year. The largest year-to-year drop in U.S. corn export commitments is to South Korea. South Korea is an unusually flexible purchaser of feed grains, willing to purchase whatever is cheapest at the moment and alter feed rations to use it. While South Korea is located near China's surplus corn region, it has also bought corn from a variety of exporters, with low freight rates making purchases of corn and feed wheat from the Black Sea competitive. U.S. shipments to South Korea through mid-March are down almost 70 percent, a drop of over 2.8 million tons from a year earlier, and even more behind the 1996 pace. U.S. sales to South Korea for the rest of the year are likely to be under the GSM-102 Program. On March 17, USDA announced an increase in operational GSM-102 credit guarantees to South Korea that increased the total authorized for corn, barley, oats, sorghum, rye, and soybean meal to $350 million. At that time, almost $200 million was still available, with $116 million already used for corn. U.S. corn exports to Japan are up so far this year, but outstanding sales are down significantly. There are reports of sharply increased corn purchases from Argentina and there have been purchases from Eastern Europe and China. U.S. sales to Taiwan are lagging notably, with shipments to date down over 15 percent, while outstanding sales are only 20 percent of a year ago. For a period of time during the negotiation of improved access of U.S. meat products to the Taiwan market, the major Taiwan buyers of feed ingredients announced they would not buy corn from the United States. This happened to coincide with a period of favorable price quotes from Argentina, and Taiwan increased its purchases from Argentina. Taiwan's corn imports are forecast down 13 percent in 1997/98 as hog numbers are reduced by the lingering impact of the foot and mouth disease outbreak. U.S. corn exports are also down sharply to several other Asian markets. Macroeconomic and exchange rate problems are reducing imports by Indonesia, Malaysia, and the Philippines (see special article) and U.S. exports to the region have faced increased competition from China and Argentina. In the Middle East, competition from Eastern Europe's and Russia's exports of feed wheat, barley, and corn has cut into U.S. corn exports, especially to Israel, Jordan, Lebanon, and Egypt. The declines to these countries are not offset by small increases in exports to Saudi Arabia and Algeria. U.S. corn exports to the Western Hemisphere have lagged a year ago, but that is offset by increased outstanding sales. In Canada, strong demand for corn for processing into ethanol has boosted imports. In Latin America, exports and sales are down to Colombia, Ecuador, Honduras, and Venezuela as purchases have not met expectations. Increased sales by Argentina explain the drop in U.S. sales to Chile, Peru, Guatemala, and El Salvador. Mexico is the largest market in the region for U.S. corn, and although shipments have lagged slightly, outstanding sales are up enough to be offsetting. U.S. corn exports to Europe and the New Independent States (NIS) are also down. Increased production in the region has reduced the need for imports, and so far during 1997/98, imports by the EU under market access agreements have not come from the United States. Several factors are likely to support the pace of U.S. corn exports during the last months of 1997/98. China has been selling smaller quantities of corn in recent months and while many expect China to continue to export in coming months, the United States is likely to face reduced competition from China and Eastern Europe in the closing months of 1997/98. Moreover, competition from South African corn is expected to be down sharply. Argentina's corn crop has been marketed more rapidly than a year ago. With Argentina's crop expected to be up to record levels, Argentina will likely be a strong competitor during most of the remainder of 1997/98. Another increase in competition is expected from low feed barley prices, but the number of import markets willing to switch between corn and barley is limited. Export shipments in the final months of 1996/97 were seasonably low, so maintaining U.S. exports in coming months at the pace set earlier in the year would be enough to boost corn exports in the summer of 1998 above those of a year ago. Box China's Agricultural Commodity Information System Is Improving But Lags the Development of Its Economic System Despite progress in liberalizing its economy, China's Government retains a major role in the grain sector. More statistics and agricultural data are becoming available, but central government decisions on stockholding and trade remain internal matters and are not transparent to analysts. China is a critical player in the world corn market but its role is far from stable. In 1994/95, China's transition from a major corn exporter to a major importer boosted U.S. exports and contributed to the steep rise in corn prices during 1995/96. China resumed exporting significant amounts of corn in 1996/97 and is forecast to continue doing so during 1997/98. The U. S. Department of Agriculture's (USDA) forecasts of China's 1997/98 exports have actually increased over recent months while production forecasts were revised down. This perplexes analysts and illustrates the need for better information on stocks, consumption, and China's policy making framework. Inadequate information can lead to inefficiencies both within China's agricultural sector and the global corn market. Information Needs and Changing Economic Structure China is in transition to an economy with more of a market focus, but the government retains a prominent role. The present information system is fragmented across a number of government ministries and research institutes and is based on a system set up for a centrally controlled economic structure. A centrally planned government focuses on production and stock levels since consumption is directed by the government. The role of prices in supply and demand decisions is only beginning to receive attention. Thus, transition between economic systems often gives rise to problems in understanding conditions in grain markets because the centrally planned information system may under- or over-report consumption that arises in a fledgling market economy. Role of Government in the Economy Because of tradition and long experience dealing with famines, a political culture developed within China that accepts the government's prominent role in the grain economy. Even with recent liberalization, the government continues to play a dominant role in China's grain economy by setting prices at numerous points in the purchasing, milling, transporting, storing, importing, and retailing of grains. In addition, it manages the wholesale grain markets and holds strategic grain stocks. It regards the size of these stocks as secret information. The central government also controls the trade of key grains such as rice, wheat, and corn. Central government decisions on stockholding and trade are an internal matter and not transparent to analysts. Another matter complicating the understanding of China's grain market is the changing relationship between the central government and local governments. A few decades ago, central authorities could transfer corn from a surplus producing province like Jilin to a deficit province like Sichuan. Provincial authorities now have more autonomy as they purchase and store corn. In the late 1980s, the central government prohibited provinces from exporting corn, but provinces in China's northeast cracked their corn and exported it under a different classification to nearby Asian markets. More recently, when the world price of corn rose in 1996, Jilin authorities wanted to export corn to earn foreign exchange but the central government did not grant authority for this transaction, thereby creating tension between the central and provincial governments. Outlook for China's Economic Information System The outlook for improvement in China's economic information system is encouraging. However, China is a modernizing country with many demands on a government with a limited resource base. Since reforms were initiated, China's Ministry of Agriculture and State Statistical Bureau have published annual statistical yearbooks that have greatly aided decisionmakers in China and the rest of the world. China's statistical authorities have responded to the statistical requirements for a modern market economy, i.e. the addition of data according to standard gross national product and national income accounting measures. In addition, statistical agencies such as the State Statistical Bureau, Ministry of Agriculture, and Ministry of Internal Trade have begun to collect and publish retail and market price information. Also, they recognized the need for a Census of Agriculture. China's first agricultural census was conducted in January 1997. Enumerators collected data such as (1) the characteristics of rural and non-rural households; (2) inventories of arable land and livestock; (3) number and flow of the rural labor force; (4) stock and flow of capital in the rural sector, and (5) individual household and community living conditions. Official results of the census have yet to be made public. Unofficial statements suggest that arable land in China may be under-reported by 30 to 40 percent, in line with USDA analysis that identified this issue a few years ago.1/ ------------ 1/ Crook, Frederick. "Under-reporting of China's Cultural Land Area: Implications for World Agricultural Trade," (China: Situation and Outlook Series. International Agriculture and Trade Reports). U.S. Department of Agriculture, Economic Research Service. RS-93-4. July 1993, pp. 33-39. ------------- The private sector is increasing its contribution to the supply of information. In the 1990s, a number of major private news agencies began to provide commodity information and market analysis. In addition, several private consulting groups publish periodic commodity reports on a fee basis. For example, these reports discuss production, consumption, stocks, trade, and food prices on a national or provincial level. After 1978, when the USDA and the Ministry of Agriculture signed a formal memorandum establishing annual meetings to manage the Science and Technical Exchange program, contacts expanded. In 1996, USDA's Economic Research Service (ERS) began working with China on a project to build an institutional capacity to incorporate market-oriented economics into government and private decisionmaking in China. ERS is also assisting in the publication of regular reports assessing agricultural commodity market fundamentals and in database development necessary for analysis of commodity markets or policy issues. Commodities covered include cotton, rice, corn, and hogs. This project will also help U.S. analysts understand China's production, consumption, and trade. USDA's National Agricultural Statistics Service (NASS) has a number of programs with China's State Statistical Bureau, such as survey design, data processing, and data dissemination. Teams have been exchanged, and NASS personnel participated in the agricultural census as observers. Box end The Impact of New Technology on the Corn Sector: 1998 Update and Prospects for the Future by Peter A. Riley 1/ ----------- 1/ Agricultural Economist, Market and Trade Economics Division, Economic Research Service. ----------- Abstract: Expansion of new technology such as Bt corn is proceeding quickly and is likely to increase for the next few years. Incentives such as cost savings, reductions in input use, and yield advantages that can increase net returns are driving farmer adoption. The spread of corn with value-enhanced traits is at an earlier stage of development than corn with crop protection features, but will also expand if there are economic incentives. Growth of value-enhanced corn could lead to more extensive changes in pricing and marketing. Keywords: Corn, Bt, genetically modified, value-enhanced. Acreage of Bt corn and other new corn varieties is expected to increase sharply in 1998, following strong growth in 1997. Farmers' widespread interest in new technology and expansion of new products signal the beginning of a new era for the corn sector in which biotechnology and value-enhanced traits are expected to play prominent roles.2/ ------------ 2/ Biotechnology as used here refers to a set of tools that use living organisms to solve problems or make useful products. Organisms that are genetically engineered or modified are often referred to as genetically modified organisms (GMOs) or transgenic. Value-enhanced refers to particular quality charactereistics that add value to corn for one or more types of end-use. ---------- This is still a very early stage for the new technology, and there will be continued change and, for many products, improvements along the way. There has been a tremendous investment in corn research by private companies recently, suggesting the brisk pace of innovation is expected to continue. Over the next few years, more new corn seed technology will reach the marketplace which could have a wide range of economic impacts on production, marketing, trade, and pricing. If some of the developments identified by industry come true, eventually it will become more difficult to just call corn "corn" because of product differentiation. The seed technology for corn can be broadly characterized in two categories. First, there is technology that generally reduces input use or leads to more effective input use and that is mainly developed through biotechnology. The major product to date is Bt corn that is resistant to an insect pest, the European corn borer. Herbicide-resistant corn is also on the market now, but this will be probably used on a smaller scale than Bt corn or Roundup Ready soybeans in 1998. In addition, there is some herbicide-resistant corn available that was developed through conventional breeding. Second, there is corn with enhanced-value traits aimed at specific end uses, such as high oil corn, hard endosperm corn, waxy corn, and white corn. These have been developed through conventional breeding, and some of these types are already established on a relatively small scale. This distinction between categories is expected to become blurred over time, however, as genes are stacked to enable the use of genetically modified seed in conjunction with high- value traits. Limited quantities of stacked hybrids are available in 1998: some combine Bt protection and herbicide resistance, and others combine resistance to two kinds of herbicide. Acreage Outlook for 1998 Industry sources indicate Bt corn could be planted on as much as 15-18 million acres in 1998, up from under 5 million in 1997. Depending upon total plantings, Bt corn could amount to close to 20 percent of all corn in 1998. It is probably more difficult to pin down the range for herbicide-resistant corn. It appears that seed will be available for around 13-15 million acres, up from perhaps 4 million acres planted in 1997. (This category includes corn resistant to a number of different herbicides, see next page for details.) Thus, if farmers' use approaches the supply of seed, acreage of genetically modified corn and other corn based on new technology could exceed 25 million acres. This would be in the neighborhood of transgenic soybeans plantings, with forecasts for Roundup Ready soybeans mostly running around 20 million acres or higher. Expansion of value-enhanced corn is likely to be less than Bt or herbicide- resistant corn in 1998. Acreage of the major new product, high oil corn, is targeted by the providing company to reach over 1.5 million acres. If realized, this would about double the estimated 700,000-750,000 acres planted in 1997. However, early evidence from seed companies indicates acreage will probably not increase that much this year. Acreage increases are also expected in 1998 for some, but not all, of the other special trait corn that is based on existing technology. A fairly strong gain in white corn plantings is expected because of higher price premiums. In 1997, industry sources estimated white corn at 550,000-575,000 acres. Waxy corn is also likely to increase from nearly 600,000 acres grown in 1997. Acreage of hard endosperm or food grade corn is expected to be stable, remaining around the 1997 range of about 750,000 to 850,000 acres. This category is somewhat loosely defined, according to the U.S. Feed Grains Council, since many hybrids with food grade characteristics are grown because of good yield potential and are not processed for food. Finally, little or no change appears in store for high amylose corn, at around 35,000 acres, and high lysine corn, around 30,000 acres. USDA does not collect separate acreage, production, or other data on these specialty or new varieties, but they are included in total corn data. Industry sources were consulted to develop estimates and forecasts cited here, and thus any numbers referring to specific types of corn are not official USDA data. Leading Products Bt Corn Bt corn is aimed at resisting damage from the European corn borer (ECB), a major insect pest that is widespread in the Corn Belt, particularly in the western and northern regions. Bt corn is enhanced with a gene from a naturally occurring soil bacterium (Bacillius thuringiensis) that produces proteins that selectively kill specific groups of insects, such as the corn borer, but have no direct effect on others, including beneficial insects. Corn borers disrupt the corn plant's growth and reduce yields. However, because the borer tunnels inside the stalk, the impact is not always readily apparent until damage has occurred. Farmers pay a premium for the Bt corn seed, which is available from many seed companies and incorporated in an increasing number of hybrids. For those farmers who spray against borer, the higher cost of the seed is offset by savings on chemicals. Because of the difficulty in predicting infestation and in properly timing treatment, the effectiveness of spraying has been mixed, and not all the farmers who grow Bt corn treated their fields previously. Given favorable yield indications, many farmers are apparently planting Bt corn as a form of insurance just in case there is bad infestation, and higher yields can more than offset the added seed costs. Bt corn was first approved for sale in 1996, but availability and use was limited. In addition, ECB infestation was generally light in 1996, and did not provide a strong test of the technology. In 1997, use expanded greatly and corn borer infestation was relatively high. Results were generally very positive in terms of protection from borer damage, as judged by looking at adjacent non-Bt corn. For the most part, the Bt technology worked well, but the yield performance was dependent on the particular hybrid, and tended to be better in northern parts of the Corn Belt. In side-by-side comparisons, yields of Bt corn varieties in some areas were higher than non-Bt corn and dramatically higher where infestation was very heavy.3/ ---------- 3/ This is based on industry, media, and extension sources, not any official USDA data. ---------- However, much of the early Bt corn is attached to somewhat dated genetics, and the industry expects yield results to improve quickly as Bt becomes available with more elite germplasm. Herbicide-Resistant Corn Some herbicide-resistant corn is also on the market now, with varieties resistant to popular herbicides based on glyphosate (Roundup) sold as Roundup Ready corn, glufosinate ammonium (Liberty) sold as Liberty Link corn, and imidazolinone (such as Pursuit, Lightning) known as IMI corn, and also as IT(imidazolinone tolerant) or IR(imidazolinone resistant) corn. There is also some herbicide-resistant corn developed through conventional breeding, including corn resistant to sethoxydim (Poast). For 1998, seed is available for more than 7 million acres of IMI corn, over 6 million acres of Liberty Link corn, and possibly as much as 900,000 for Roundup Ready corn. One seed company official noted that the outlook for farmers' response to herbicide-resistant corn is more complicated than for insect control such as Bt corn, suggesting a somewhat slower expansion. Weed problems tend to be more varied, both by geography and by year, than insects. Like all the different corn products, the usefulness and performance will vary by region and management practices. In areas where crop rotations and conventional tillage are more common, weed control may be less dependent on herbicide use. Combining herbicide tolerance with insect resistance will probably accelerate their adoption. High-Oil Corn High-oil corn (HOC) has seen rapid growth since its commercial introduction in 1994. Acreage doubled in 1996 and again in 1997. Research efforts have been led by a major chemical company, and the technology made available through many seed companies. There has been a major advertising campaign mounted, more marketing opportunities established relative to older generations of specialty corn, and even use of the Internet to post information, such as participating elevators, the premium schedule, and sample contracts. These efforts contributed to sharply raising interest in this crop. Conventional corn typically contains 3.5 to 4 percent oil, whereas high-oil corn can contain 7 percent or even more. It is mainly attractive because of its good feeding characteristics, rather than for the oil, per se. The extra oil boosts the energy content of the feed, eliminating or reducing the need to add fat to the ration, as well as reduces the need for supplemental protein or some amino acids. Added benefits are less dust and good mixing and grinding traits. As part of an identity-preserved delivery system, 4/ ----------- 4/ Identity preserved means the product is segregated to avoid commingling with other products all the way from harvest to delivery to the end user. ----------- HOC also should have less variability for end users than normal purchases of corn. Although HOC hybrids that are grown like conventional corn are available, most growers are using the Top Cross system because the potential oil content is higher. In the Top Cross system, a portion of the field (about 8-10 percent) is planted with very high oil pollinators, while the rest is planted with male- sterile hybrids. The pollinators are low yielding, but offer a longer pollen shed than traditional hybrids, and boost the oil in the other plants. The recommended plant population is higher than normal corn to offset yield loss from the pollinators. Larger fields and/or buffer strips are recommended to prevent stray pollen from normal corn, reducing the oil content of the corn. While many test results show comparable HOC yields to regular corn, some industry officials identify the risk of slightly lower yields than normal corn, mainly reflecting more risk at the pollination stage. In addition, the oil content is variable, depending on growing conditions, soil types, and other factors. These risk issues appear to be an obstacle blocking faster adoption by farmers, along with concern about the price premium. Premiums for high-oil corn are offered on a sliding scale based on the oil content of the corn delivered, and, if large enough, the premiums will cover the additional costs of producing and segregating the corn, plus cover any yield drag if this were to occur. Demand for high-oil corn is expected to increase as users gain familiarity with it and the volume available becomes larger. Trials by many large users are underway. The earliest gains have been in export markets and in on-farm feeding use by growers who can avoid the need to purchase and store additional fats. Most of the HOC exports are going to developing, tropical countries, where supplies of fat are expensive or unreliable, including markets in Asia, Latin America, and the Caribbean. Expansion of HOC corn to larger U.S. feed users, such as the big integrators, is expected, but the timing is difficult to predict because of the need for large quantities. Despite the numerous feeding advantages identified, a key economic issue that will shape demand is the price of competing feed ingredients. In the United States, the huge supply of grease and fats from the fast food industry and other sources, and the high cost of their disposal, will likely provide strong competition for high-oil corn. Like other aspects of new corn technology, combining the high-oil feature with other traits will increase the market potential. Future Developments Seeds that are stacked with the Bt trait and resistance to various herbicides are expected to increase substantially in 1999 and continue to grow in the future. The next major insect resistance feature is one to deal with the corn rootworm. This will be introduced in the next 2 or 3 years, and market prospects look good. Research is underway to develop resistance to more insect pests in the years ahead. Another focus of research is better resistance to diseases. Like many other desirable features, disease resistance has been a goal of conventional breeding and selection efforts. The initial emphasis of genetically modifying corn has been more toward insect and weed control because of the greater expected economic impact. For enhanced-end traits, there will be both new products introduced and improvements in existing traits, along with stacking to combine various traits with insect and/or herbicide resistance. In addition to high-oil corn, there has been considerable interest in the development of low phytate corn that is expected to reach the market soon, in 2 or 3 years. This corn is low in phytic acid and enables poultry and hogs to better utilize phosphorous, thereby reducing the amount of phosphorous excreted, and reducing odor and runoff problems. Many other traits to improve feeding value or tailor the corn to special feed needs will be emphasized based on higher protein, higher content of amino acids such as lysine or methionine, or altered fatty acid profiles. These traits are sometimes categorized as nutritionally dense corn. In addition to improved feeding traits, research is underway to provide corn with desirable traits for various food and industrial products from wet milling, dry milling, and alkaline processing. Some of these traits include starch content, susceptibility to cracks, and various milling characteristics. The vision of the seed industry is that corn will become more specialized over time to fit the particular needs of end users. Thus, rather than just buying number 2 or 3 yellow dent corn, a buyer will specify various traits required for the type of animals being fed or for the specific industrial or food product. In some cases, a particular hybrid will be specified. Based on trends that are beginning to emerge, end-use corn specialization and marketing are no longer a far-fetched idea. While the technology is getting closer, however, it will only develop if the economics are favorable. Economic Implications and Issues Historical perspective on specialty corn: Yellow dent corn dominates in the United States, and the various specialty corns, excluding high-oil corn, collectively account for only about 5 percent of the market. As a preface to looking ahead, a brief glimpse backward might be instructive. The track record of white corn, one of the leading specialty types, has been mixed. Plantings reached nearly 700,000 acres in the mid-1970's, (at that time data were tracked by USDA), but then sunk as low as 300,000 acres by the early 1980's. The market was relatively thin, and included a very volatile export component. In addition to market risk, the average yield of white corn is lower than yellow corn, and many growers were discouraged, despite the potential for premiums. By the 1990's, white corn acres began to rebound, spurred by solid domestic demand for white corn for snacks and Mexican-style foods and increasing exports. Most production is under contract, and premiums can exceed 40 cents per bushel, sufficient to allow for reduced yields and other higher costs. However, the seed industry expects some recently released hybrids to match yields of yellow corn. The focus of the most successful specialty corn has been food use and to a lesser extent, industrial use. Much hard endosperm and other food grade corn goes into snack foods. Waxy corn, for example, is used for food starch because it contains no amylose which is a less digestible part of ordinary corn starch. Relatively mature demand accounts for recent modest growth in these categories. The market for feed corn has been almost untouched, however. Some specialty corn for feeding has been available for years, but growth has been stagnant. For example, high lysine corn was introduced about 30 years ago. Lysine is an important amino acid for hogs, but it is limited in corn. The combination of low yields and the lack of a large enough premium to cover higher costs has constrained growth. Users have alternative means of improving the amino acid balance, such as synthetic lysine. Production that is linked to an assured buyer--and thus reduced marketing risk -- is one of the keys to success in specialty crops. One major snack food company has developed a successful supply channel with Nebraska farmers, who produce specified varieties of white and yellow food grade corn. There have been many similar developments like this in recent years on a localized basis, not just for corn, but for soybeans and other field crops. Another critical need is a price that is sufficiently high to cover additional costs such as lower yields or special handling. Adoption U.S. farmers are highly receptive to new technology, particularly given their widespread interest in gaining more value to their products or reducing costs. Adoption of some of the first new biotech seeds, such as Bt corn and Roundup Ready soybeans, has been quick despite the extra cost of the seed. Farmers will readily pay premiums for the technology if the benefits are perceived to outweigh the costs. Still, with so much new technology coming on stream, and limited test data from public sources, many farmers will wait to plant some new varieties until performance can be proven on the local level. Early indications of the effects of many of the new technologies are favorable, although adoption is not without risk. Along with the sustained performance over time, performance of the new technologies under stress conditions, such as a drought, is an unknown that could influence future adoption rates. For value-enhanced corn, the size of the price premium will be a critical factor, and prices of competing substitutes will be a critical factor in shaping demand. In addition, changing market conditions could be a factor, if higher production were to lead to a lower price premium for some end traits, for example, especially in thin markets. Environmental benefits: Additional pressures from the environmental side -- at the farm and for end users--could help to promote adoption of new technology. Much of the new corn will mean less chemical use, adding to its appeal. Corn with enhanced feeding value can improve the digestibility of certain nutrients, so less nutrients will end up as animal waste. This should reinforce the willingness of end users to pay premiums for the corn's improved feed efficiency. Management requirements: In addition to cost savings, one of the biggest incentives to adopt some new technology is convenience and, in some cases, its compatibility with conservation tillage practices. For example, Bt corn can reduce the management load on growers by potentially reducing scouting needs and eliminating some insecticide use. Incentives to use herbicide-resistant crops are also strong, as growers can simplify herbicide use and often reduce applications. The promise of stacked traits could similarly contribute to reducing management requirements. The proliferation of new herbicide-resistant crops could cause confusion, however, as it becomes more difficult to keep straight which herbicides can or cannot be applied to a particular crop, especially when replanting or dealing with any potential residual carryover. In addition, knowledge of the optimum times to apply the herbicides with these products is still evolving. Aspects of some new corn varieties will clearly increase management demands. This would be most obvious for corn with special production needs. For example, high-oil corn, like seed corn, needs additional attention to prevent cross pollination, which could reduce the oil content. Handling needs will also increase for value-enhanced crops as a particular crop's special characteristics must be maintained from the producer to the end user. Yield effects: Most of the new technology introduced so far could be considered yield "neutral" to the extent it is not explicitly aimed at increasing yields. However, the technology may have yield effects, both positive and negative, not just from developments in genetics, but from changes in management demands. In general, elite germplasm will still be the underlying driving force in future productivity gains, regardless of the new technology attached. As discussed in the section on Bt corn, in the rush to bring new products to market, some of the genetics used are dated, and yields of conventional hybrids may be higher, but this is expected to diminish quickly. Some of the new products will effectively boost yields by cutting losses to pests or weeds, protecting whatever yield potential is already present in the particular hybrid. Benefits will vary from year to year and over different locations depending on environmental factors such as the amount of pest infestation. There is no solid estimate on yield loss at the national level due to pest damage. Thus, it is hard to judge the impact on aggregate yields from adoption of Bt corn. But if adopted widely enough, and if yield advantages are sustained, then it could "bump" the average U.S. yield above the long-term trend. Despite attention to numerous other factors, increasing yield has been the traditional focus of the corn seed industry. The focus of most enhanced-value crops is an attribute for end use, however, and, in some cases, yields may be compromised. Farmers are very reluctant to grow corn if they perceive lower yield potential unless there is a clear price premium to compensate. Over time, gene stacking and more research may be able to overcome any yield penalties of the enhanced trait corns. Biotechnology increases the tools available to scientists, and its use should speed the pace of research, with positive implications for yields. Pest and weed resistance: Although industry is taking preventative measures, some critics are concerned that insects or weeds may develop resistance to the technology intended to suppress them, and they have expressed vocal opposition to the use of Bt in corn and cotton. They fear resistance could reduce its effectiveness for other uses as well. Traditional use of Bt foliar sprays was less worrisome because the Bt was quickly broken down by sunlight, and insects had very short exposure, compared with Bt corn where it persists in the crop throughout the season. The companies selling Bt seed have a very strong economic incentive to prevent resistance to preserve their markets, and they acknowledge the danger and even the likelihood that resistance will eventually develop. Producers using Bt seed must agree to certain production practices designed to slow or prevent the development of resistance, such as not planting 100 percent of their fields in the Bt variety in order to provide a refuge for the survival of non-resistant insects. In many cotton growing areas where Bt cotton may be grown, use of Bt corn has been restricted to prevent resistance. Another aspect that may help prevent or delay resistance is the introduction of additional Bt genes that have different modes of action, and one is expected to be approved momentarily by the Environmental Protection Agency (EPA) and USDA. It employs a protein that attaches to a different part of the insect than the protein most widely used now. If resistance were to develop, the seed industry is prepared to tap different strains and versions of Bt, and offer new generations of product, similar to what is done with some antibiotics. The effectiveness of these measures will need to be evaluated over time. Continued use of a single herbicide raises fear of weed resistance as well by some critics and by some farmers. Again, the companies involved have very strong commercial incentives to prevent this. Some extension agents point out that weed resistance may be less of a concern than weed shift, when the species most susceptible to the herbicide decline over time, while less susceptible species build up. Marketing: The growing emphasis on end traits implies changes in the marketing system, with more identity preservation changing the traditional bulk commodity focus of corn, the largest field crop. If taken far enough, this trend to more emphasis on end use traits will reduce--not eliminate--the traditional bulk focus of the commodity markets, which emphasize large volumes and blending. With more trait orientation, testing and certification will become critical. The extent of growth partly hinges on technical factors--the complexity and quality of end-quality testing along with the expense. Industry sources are optimistic about near infra-red reflectance spectroscopy testing, used to measure the oil content of corn when the farmer delivers the corn. It gives a quick (about 45 seconds) and reliable result. This testing can be used for certain other traits, as well, such as protein and starch content. Movement to this type of testing is a significant change from current practices, however. The existing system of grades and standards does not identify the inherent traits of the crops. While some feeders run assays on corn and other feed ingredients, it is not routinely done at the country elevator. Contracting as a means of coordination from the farm to the consumer will likely expand, as well as the role of niche marketers who link growers with buyers. Another possible outcome is more integration where successive stages of the production and marketing chain are linked together, for example, under direct ownership or through cooperative arrangements. Among the key issues is who will capture the value created by new technology. There has to be some allocation for all parties in the marketing chain to provide incentives to develop the seed, and then to grow, elevate, handle, and transport new products. Exactly how the value is shared along the chain will evolve over the coming years. The providers of new technology have started to devise arrangements that respect the intellectual property rights that are critical in providing incentives to invest and develop products while allowing successful commercialization. An analysis done a few years ago raised the possibility that diffusion of biotech- derived products would be slowed by monopoly pricing opportunities afforded by patent protection [Fleisher, 1989]. Despite numerous legal battles involving things such as patent rights, however, pricing issues have largely been avoided as indicated by the very rapid spread of new technology like Bt corn and Roundup Ready soybeans. In the seed and chemical industries, there has been an incredible wave of alliances, mergers and acquisitions, joint ventures, and licensing arrangements undertaken in the last few years that could lead to further change elsewhere in the corn industry if more links are established to processors and users. Trade: In late 1997 and early 1998, there was friction in the European Union (EU) about acceptance of U.S. corn and corn product exports because particular Bt varieties had not yet been approved under the EU's approval process. Some varieties were recently approved by an EU scientific advisory panel and an EU regulatory committee, but there are still more hurdles to clear before final approval. The EU also has enacted a broad labeling requirement. The rapid introduction of new genetically modified varieties and a slow approval process in the EU suggests delays could occur again under the prevailing regulatory system. If difficulties persist, some U.S. corn processors who export to these markets may prefer to avoid purchase of Bt or other genetically modified corn for their operations. Most countries have not placed any restrictions on imports from the United States, and expansion of genetically modified corn should not disrupt trade. Furthermore, other corn exporting countries are likely to grow genetically modified corn in the future, limiting possible alternatives. Niche markets for corn that does not contain GMOs may develop in some foreign markets if consumers are willing to pay more. Because of interest in value-enhanced products, expansion of identity-preserved corn trade is likely. Recent growth in U.S. exports of high-oil corn indicates this is a realistic possibility. Pricing: Most of the new technology is proprietary, and premiums are charged for the seed. For some products, the premium covers technology fees or gene fees that go to the originating company. Through various licensing agreements, payment of fees, and numerous corporate alliances, the technology is generally available from many different seed companies. On the output side, the crops that are genetically modified should have little or no direct impact on prices received by farmers, assuming the varieties are approved under the regulatory process and are accepted by consumers and other countries. This is because the products are basically indistinguishable from conventional crops. For a relatively small group of consumers, a niche market for non-GMO products may develop, similar to the present market for organic foods, that will involve separate identity-preserved marketing and thus premium prices. (The seed industry has requested in some cases that growers segregate specific genetically modified varieties that have not been approved for import by some foreign countries.) Enhanced-value products do imply changes, however. Product prices will be higher: first, to reflect the value of the end-use trait, and second, to cover the higher costs involved in keeping and transporting the crops separately. Segregation or identity preservation will be required at all points in the marketing chain, starting at the farm. More contracting is expected as a means to assure a guaranteed market for farmers and minimize risk, similar to the way many vegetables and some specialty corn are currently contracted. However, contracting does not necessarily eliminate risk. If a producer fails to meet contract specifications due to quality problems, there may be no premium earned. Evaluating the economic effects on overall prices and returns from more value- enhanced crops will hinge on the extent these products are grown. If the products remain as specialty crops only, there will be a fairly small impact, but if the products become very popular, larger changes will occur. As more farmers grow enhanced-value crops, the size of the premium needed as an incentive to farmers could change. Economies of scale could also reduce marketing and transportation costs if adoption is widespread. From the user's point of view, the price of substitutes, such as grease that competes with high -oil corn, will also adjust, and in turn influence the price of the corn. The main commodity market for corn is likely to remain an important reference point in the price determination process for value-enhanced corn. Users of specialty corn typically offer a premium relative to a spot or futures price. This has basically occurred at the local level adjusted for basis with a limited number of elevators buying a particular product. Sometimes farmers can deal directly with a processor. As production expands, the range of marketing opportunities should expand and more elevators will buy value-enhanced corn. As the first of this new wave of products, the case of high-oil corn will be instructive with the attempt to build a large marketing network, along with the use of the Internet as a means of price discovery. Conclusions The seed corn business in the United States is one of the most dynamic industries in the country, as judged by its innovations that have been an essential part of strong productivity growth. The results reflect a long tradition of research at the public and private level that is now backed by massive investment from chemical companies and others. Despite impressive new soybean products from the same companies, corn technology is probably leading the pack relative to other crops with a wider range of developments. This reflects corn's status as the largest U.S. crop and the fact that virtually all corn seed is purchased, providing large potential returns to investment. The seed companies continue to turn out new hybrids, and they get seed to the market quickly by supplementing U.S. operations with winter production of seed in Argentina and Chile in the Southern Hemisphere. Expansion of new technology in the corn sector is proceeding quickly and is likely to increase for the next few years. Strong incentives related to cost savings, reductions in input use, and yield advantages that can increase net returns are driving farmers' adoption of corn seed derived from biotechnology. The distinction between different seed varieties developed with new technology will become increasingly blurred in the future through the use of gene stacking. The spread of corn with value-enhanced traits is at an earlier stage of development, but its expansion could lead to more extensive changes in pricing and marketing. The economic relationships that allocate the value added by new products across the various participants and share the risk are just starting to evolve. References American Seed Trade Association. (1996) "Proceedings of the Fifty-First Annual Corn and Sorghum Research Conference 1996," Publication No. 51. Caswell, Margriet F., Keith O. Fuglie, and Cassandra A. Klotz. (1994) Agricultural Biotechnology, An Economic Perspective. Agricultural Economic Report No. 687. USDA, Economic Research Service. Fleisher, Beverly. (1989) "The Evolving Biotechnology Industry and Its Effect on Farming," Agricultural-Food Policy Review, Agricultural Economic Report No. 620. USDA, Economic Research Service. Hill, Lowell D. (1990) Grain Grades and Standards, Historical Issues Shaping the Future. University of Illinois Press. Union of Concerned Scientists (1997). "The Gene Exchange." Fall 1997. University of Minnesota Extension Service (1997). "Bt Corn and European Corn Borer, Long-Term Success Through Resistance Management." U.S. Feed Grains Council. (1997) "1996-1997 Value-Enhanced Corn Quality Report." Rethinking the Soybeans-to-Corn Price Ratio: Is Is Still A Good Indicator For Planting Decisions? by William Lin and Peter A. Riley 1/ ---------- 1/ Agricultural economists, Market and Trade Economics Division, Economic Research Service. ---------- Abstract: The soybeans-to-corn price ratio is a simple measure of relative returns for corn and soybeans. This article examines the ratio in terms of its role in explaining the switch between corn and soybean plantings in historical context and the evolution of the soybeans-to-corn breakeven price ratio. In addition, this article points out deficiencies of the ratio as an indicator for planting decisions, particularly after the enactment of the 1996 farm legislation. Keywords: Soybeans-to-corn price ratio, breakeven price ratio, corn, soybeans. Analysts have long used the soybeans-to-corn price ratio as an indicator to determine whether there would be an acreage switch from corn to soybeans, or vice versa. If producers expect the soybeans-to-corn price ratio to exceed the breakeven price ratio (BEPR), there would be a tendency to switch from corn to soybean plantings. Conversely, if the expected soybeans-to-corn price ratio is below the BEPR, the reverse would be true. The BEPR is the ratio of expected soybeans-to-corn prices which equates the expected net returns of producing corn and soybeans, given trend yields of corn and soybeans, the expected price of corn, the variable costs of corn and soybean production, the expected program payments, and other program expenses. The soybeans-to-corn price ratio is a simple measure of relative returns for corn and soybeans. When producers expect the price ratio to exceed the BEPR, it suggests that expected net returns for soybeans are greater than those for corn. In contrast, if the expected price ratio is below the BEPR, it suggests the opposite is true. However, producers do not simply look at the price of soybeans versus the price of corn in deciding what to plant. Other than crop rotation concerns and restraints, the use of the soybeans-to-corn price ratio assumes that the market prices for corn and soybeans accurately reflect their relative net returns. This may be a heroic assumption in the current climate. This article examines the ratio in terms of its role in explaining the switch between corn and soybean plantings in historical context and the evolution of the soybeans-to-corn breakeven price ratio. In addition, this article points out deficiencies of the ratio as an indicator for planting decisions, particularly after the enactment of the 1996 farm legislation. Given more planting flexibility, States outside of the main Corn Belt have accounted for much of the change in corn and soybean acreage in the last 2 years. The soybeans-to-corn price ratio presents a less complete picture of crop choices than in the past as (1) corn and soybean production continues to expand to the Central and Northern Plains region and, to a lesser extent, the Delta and Southeast regions over the last 2 years, and (2) plantings became more flexible under the 1996 Act. Analysis of other indicators, such as the soybeans-to-cotton price ratio and the corn-to-cotton price ratio for the Delta and Southeast, becomes necessary to more fully capture the competition for cropland use among major field crops. In addition, due to planting flexibility and a faster yield gain in soybean production since the early 1990's, the BEPR, which used to be around 2.6 at the national level in the early 1990's, hovers around 2.5 for this year, offering producers greater incentives to switch from corn to soybean plantings. 2/ ---------- 2/ The soybeans-to-corn price ratio in this article, unless indicated otherwise, refers to the ratio that is computed from the new crop corn and soybean futures price in mid-March after being adjusted by basis, difference between new crop futures prices in the month preceding delivery, and corresponding monthly cash prices received by farmers. ---------- Due to greater planting flexibility, corn and soybean plantings in future years are likely to be increasingly affected by factors other than the soybeans-to-corn price ratio. In the Great Plains region, these other factors could include the soybeans-to-wheat price ratio and the corn-to-wheat price ratio. In the Delta and Southeast regions, these price relationships include the soybeans-to- cotton price ratio, the corn-to-cotton price ratio, and the soybeans-to-wheat price ratio. In the spring wheat areas, the soybeans-to-wheat price ratio could be relevant to forecasting soybean acreage in future years. Much of the other price relationships need further investigation before the exact relationships are established. U.S. Soybeans-to-Corn Price Ratios The soybeans-to-corn price ratio is calculated by dividing November soybean futures prices by December corn futures prices in mid-March--the time when most corn planting decisions are made by producers--at the Chicago Board of Trade after the futures prices are adjusted to a U.S. farm-level equivalent by a basis. The basis is computed by subtracting the monthly average futures price for the month preceding delivery of the contract---November for December corn futures prices and October for November soybean futures prices--from the corresponding monthly price received by farmers. The expected basis for each year during 1986-97 is the previous 8-year moving average of bases (tables B-1 and B-2). 3/ --------- 3/ An 8-year moving average is chosen based on a research finding that expected basis calculated from the previous 8-year moving average has the least forecasting error (Heifner). --------- The expected basis has been trending downward (in absolute value) for corn and soybeans, with an average value of -$0.26/bu for soybeans and -$0.22/bu for corn at the national level. Since the late 1970's, the basis for corn and soybeans has been negative, reflecting transportation costs of shipping corn and soybeans from the farm gate to terminal markets at the delivery point of the futures contract. The basis, as shown in tables B-1 and B-2, also refers to an average at the national level, which may not reflect the actual farm price received by producers in specific localities. For example, if new crushing plants are being built further west and north, then the price of soybeans (net of delivery costs) received by farmers would increase as a result of the decline in delivery costs to these new plants. Since the mid-1980's, U.S. soybeans-to-corn price ratios (as used here) have fluctuated between 2.39 in 1996 and 3.27 in 1988 (table B-3 ). Based on new crop futures prices in mid-March 1998, the ratio is around 2.35. Over the last few years, the ratio hovered around the range of 2.4-2.6. In general, the price ratio adjusted by basis is higher than the one computed directly by dividing new crop soybean futures prices by new crop corn futures prices. Over the last 3 years, that difference in the two price ratios was generally around 0.1. The soybeans-to-corn price ratio, as estimated in this article, generally worked well in indicating the direction of the switch between U.S. corn and soybean plantings. For example, in 1987, the corn share of combined corn and soybean intended plantings was 54.3 percent, down from 55.7 percent in 1986, largely because the price ratio increased from 2.70 to 3.13. The increase in the Acreage Reduction Program (ARP) level from 17.5 percent in 1986 to 20 percent in 1987 also contributed to the decline in the corn share of plantings. Similarly, in 1990, the corn share of plantings was 55.7 percent, up from 54.3 percent in 1989, largely because the price ratio declined from 2.86 to 2.53. The ARP was not a factor in the switch that year because it was set at 10 percent for 1989 and 1990. Over the last 2 years, factors other than the soybeans-to-corn price ratio and ARP levels became increasingly important in explaining producers' acreage choices between corn and soybeans. In 1996, the increase of 4.6 million acres in U.S. corn planting intentions in early March was largely caused by strong corn prices along with the elimination of the annual acreage reduction and 0/85-92 programs under the 1996 Act, which had idled 7.7 million acres in 1995 (table B-3). 4/ ---------- 4/ In 1995, corn planting intentions in March totaled 75.3 million acres, which are 4.1 million more than the 71.2 million actual plantings due to very wet conditions at planting time. ---------- However, soybean acreage also expanded mainly in the Central and Northern Plains, and Delta and Southeast. In 1997, the sharp rise in U.S. soybean plantings was only partly explained by the increase in the price ratio, from 2.39 in 1996 to 2.59 in 1997. Other important factors include the reduced seedings of winter wheat in the fall of 1996 and a decline of about 1 million acres altogether in corn and cotton plantings in the Delta and Southeast. 5/ ---------- 5/ For producers in these regions that planted soybeans instead of corn or cotton, many of them were not looking at the November soybean futures price relationships. Instead, they were looking at the high, late 1996/97 season prices. A November/December soybeans-to-corn price ratio would not capture this. -------- The reduced seeding of winter wheat had little to do with price. It was because farmers in the Midwest had suffered from second years of low-quality crops due to disease and fungus. Also, in some areas it was too wet to plant in the fall of 1996. The recent introduction and fast adoption of new crop technologies will add uncertainty to the soybeans-to-corn price ratio in future years. At this point, these crop technologies tend to achieve more cost-savings in input use for soybeans than for corn. For example, Roundup Ready varieties of soybeans reportedly could achieve a cost-saving of $15 to $20 per acre in herbicide use. On assumption that Roundup Ready varieties will account for about 25 percent of all soybean acreage in 1998, adoption of these varieties would lower the soybeans-to-corn breakeven price ratio by about 0.03. 6/ ---------- 6/ This estimate does not factor in any yield change expectations. ---------- A faster rate of adoption of the crop technology in the future could lower the price ratio even more, assuming no dramatic changes in corn costs. Bt corn, which was developed in recent years primarily to prevent yield losses caused by insect damage, could also affect the breakeven price ratio (see special article on "The Impact of New Technology on the Corn Sector"). Similarly, the use of high plant population in recent years, facilitated by new seed varieties and herbicides contributed to faster yield gains for soybeans since the early 1990's, relative to the historic yield trend. The yield gains in soybeans have been astounding in recent years, while the yield pattern for corn has been more erratic, and for the last 3 years, at or below trend. Over the last 5 years, higher relative yield growth led to an average decline of 0.04 in the soybeans-to-corn breakeven price ratio. Locational Shifts in Corn and Soybean Production Corn and soybean plantings have greatly expanded in the Central and Northern Plains region (including the Dakotas, Nebraska, Kansas, Montana, Wyoming, and Colorado) in recent years (Schwartz). In 1996, corn plantings totaled 17.1 million acres in this region, up 10 percent from its 1993-95 average. More significantly, soybean plantings in 1997 totaled 10.7 million acres, up 15 percent from the average. The expansion of corn and soybean production in this region follows the shift in feedlot operations from the Corn Belt region to the Western States, a long-term trend. However, nearly full planting flexibility under the 1996 Act, coupled with strong corn and soybean prices and adequate precipitation over the last 2 years, enticed producers to switch from wheat, sorghum, barley, and oats to more profitable crops, such as corn and soybeans. Also, the 1996 Act facilitated crop rotations, which were constrained by the previous farm legislation. In 1997, corn plantings in the Central and Northern Plains regions totaled 17.8 million acres, up from 13.9 million in 1986. The regions' share of U.S. corn plantings increased from 18.2 percent in 1986 to 22.1 percent in 1997 (fig. B-1). Similarly, soybean plantings totaled 10.7 million acres, up from 6.2 million in 1986. The regions' share of U.S. soybean plantings increased from 10.2 percent to 15.0 percent during this period. The long-term increasing trends in corn and soybean plantings appeared to be steady. Expansion in corn and soybean plantings in the Delta and Southeast regions occurred only 2 years ago. In 1996, corn plantings totaled 6.0 million acres, up 20 percent from the 1993-95 average. Similarly, 1997 soybean plantings totaled 12.4 million acres, up 8 percent from the 1993-95 average. For corn in 1996 and soybeans in 1997, the expansion had largely to do with huge farm price advantages for the regions' early harvested crops. Many producers planned to capture the strength of late 1995/96 corn prices in 1996 and late 1996/97 soybean prices in 1997. Despite the growth of corn and soybean plantings, the regions' share of U.S. plantings remained virtually unchanged in recent years -- 7.1 percent for corn and 17.5 percent for soybeans in 1997. In the mid-1980's, the share of U.S. corn and soybean plantings was even higher--9.3 percent for corn and 24.5 percent for soybeans. Corn and soybean acreage actually declined, from 7.1 million acres of corn and 14.8 million acres of soybeans in the mid- 1980's to 5.2 million of corn and 11.3 million of soybeans in the mid-1990's. In contrast, cotton acreage expanded during this period. Evolution of the Soybeans-to-Corn Breakeven Price Ratio The BEPR has gone through changes over time. At the national level, the BEPR hovers about 2.54 this year, down from 2.60 in the early 1990's and 2.95 in the late 1980's. In the North Central region (including Iowa, Illinois, Ohio, Indiana, Missouri, Minnesota, Michigan, and Wisconsin), the BEPR hovers around 2.50, down from 2.55 in the early 1990's. 7/ -------- 7/ At the regional level, basis is computed as the differential between new crop futures prices in the month preceding delivery and corresponding monthly cash prices received by farmers in the region. -------- In the Delta and Southeast, the BEPR is about 2.55 this year, mainly because the region is a feed-deficit area where corn prices are considerably higher than in the North Central region. Soybean prices in the South are also higher than in the North Central, but not as much as the difference for corn. The BEPR is lower this year because program payments are no longer a factor in producers' planting decisions, and a faster yield gain for soybeans continues. 8/ -------- 8/ Considerable differences in the reported BEPR's for 1998 exist among analysts, ranging from 2.0-2.1 to 2.4-2.5 (Good; Kohlmeyer; Smetana). Although a part of the differences is probably attributed to the mechanics of computing the price ratio (i.e., whether the futures prices are adjusted for basis or for non-fundamental influence on futures prices due to the trends), the remaining differences are attributed to other reasons yet unexplained. -------- If corn prices expected by producers were at the same, lower level (around $2.40 per bushel) as in the early 1990's, the U.S. BEPR would be even lower, at about 2.48. Thus, over the last decade, the BEPR has shown a decline as a result of changes in the corn program, a faster yield gain for soybeans since the early 1990's, and changing corn market conditions. From 1986 to 1990 Producers had a tendency to continue growing corn, and were deterred from making a switch from corn to soybean plantings during this period. The potential for large corn deficiency payments during 1986-87, as high as over $1 per bushel, greatly deterred producers from making a switch from corn to soybean plantings. Also, maximum payment acreage during this period was at the highest level in recent history, equaling base acreage less ARP acres. As a result, the BEPR averaged about 2.95 at the national level and 2.80 in the North Central region. However, the same magnitude of increase in the soybeans-to-corn price ratio expected by producers then had a smaller effect on producers' decisions in switching corn to soybean plantings than it would have since 1996. This is because, in addition to the economic barrier to planting flexibility, protecting corn base acreage represented an institutional barrier which is not reflected in the estimated soybeans-to-corn price ratio (Westcott). The BEPR for this period is defined as: BEPR = (1/Pc*Ysoy)[(1-ARP)(Pc*Yc-VCc)+(1-ARP)(DPc*PYc)-VCarp+VCsoy] where Pc is the expected price of corn, Ysoy is trend yield of soybeans, ARP is the Acreage Reduction Program idled acres, Yc is trend yield of corn, DPc is per-bushel deficiency payment for corn, PYc is program yield of corn, Vc, VC soy, and VCarp are variable costs of growing corn and soybeans, and of maintaining conserving-use of the ARP idled acre, respectively. 9/ --------- 9/ This BEPR applies to producers who participated in the feed grains program, about 85 percent of corn base. Nonparticipants would have a BEPR equation identical to the one for the 1996-98 period. --------- The expected price is the December corn futures at Chicago Board of Trade in mid-March, further adjusted by the basis to allow for price differentials across States. From 1991 to 1995 Producers had more planting flexibility to switch between corn and soybeans during this period as a result of the evolution of the farm program towards market orientation. The BEPR declined to around 2.60 at the national level and 2.55 in the North Central region during 1991-95. There are three factors that have contributed to the decline: (1) a significant decline in per-bushel corn deficiency payments due to lower target prices, (2) a decline in the maximum payment acres by 15 percent because of normal flex acreage (NFA), and (3) a greater yield gain for soybeans since the early 1990's. The U.S. BEPR would have been even lower, at around 2.55, if program payments had not been a factor in producers' planting decisions. On the other hand, smaller ARP's during 1991-95, relative to those during 1986-90, had kept the BEPR lower. The deficiency payment for corn declined to $0.10-$0.35 per bushel due to lower target prices lowering program incentives for corn production. In the meantime, producers were given limited planting flexibility through NFA, which amounted to 15 percent of the base, but received no deficiency payments. To many producers, the 15-percent NFA was sufficient flexibility to switch plantings from corn to soybeans, or other competing crops. This greatly reduced the institutional barrier to planting flexibility. The faster yield gain in soybean production started to show an apparent effect on lowering BEPR since the early 1990's. For example, trend yields for U.S. soybeans increased from 34.7 bushels per acre in 1991 to 38.6 bushels by 1997, a 35-percent increase in the annual yield increment during 1986-90. The faster yield gain lessened the soybean prices in order to equate the expected net returns between corn and soybean production. The BEPR for this period is defined as: BEPR = (1/Pc*Ysoy)[(1-ARP)(Pc*Yc-VCc)+(1-ARP-0.15) (DPc*PYc)-VCarp+VCsoy] All variables have the same meanings as defined earlier. From 1996 to 1998 The breakeven price ratio continued to decline during this period, from 2.60 during 1991-95 to around 2.54 this year--about 2.50 in the North Central region and 2.55 in the Delta and Southeast regions, but could be lowered to around 2.48 if expected corn prices were at the early 1990's level. The 1996 Act offers producers nearly full planting flexibility by removing institutional and economic barriers. With a few exceptions, producers can plant any crop on their entire base acreage. Also, program payments are no longer a factor in producers' planting decisions. In the meantime, the rising yield gain in soybean production continues in force. The BEPR for this period is defined as: BEPR = (1/Pc*Ysoy)[(Pc*Yc-VCc)+VCsoy] All variables were defined earlier. Conclusions While the soybeans-to-corn price ratio may have indicated well the direction of the switch between corn and soybean plantings in the past, it will provide only part of the explanation for producers' acreage choices between corn and soybeans in the future. Producers will increasingly pay closer attention to other price ratios beyond the soybeans-to-corn price ratio as plantings become more flexible under the 1996 Act and as corn and soybean production expands outside the Corn Belt. Related to this, is the viewpoint that forecasting future corn and soybean plantings can best be done at the regional level, where other price ratios, such as the soybeans-to-cotton and the corn-to-cotton price ratios in the Southeast and the soybeans-to-wheat and the corn-to-wheat price ratio in the Central and Northern Plains, can be effectively addressed. These other price ratios are difficult to address at the national level. This also means that the soybeans-to-corn price ratio would be more meaningful at the regional level than the national level, including regions such as the Central and Northern Plains, and Delta and Southeast. Even regional price ratios are only rough proxies for producer decisions. The individual producer's cost and expected return may differ substantially from farm to farm. For example, a corn/soybean farmer in part of the Corn Belt may plant soybeans (no matter what the price ratio) if he/she cannot get Bt corn to plant. Those farmers who faced severe corn borer infestation last year saw dramatically lower yields. Finally, corn and soybean plantings would still be affected by weather conditions even if all the relevant price relationships are captured in the analysis. Persistent wet conditions in the springtime can delay corn plantings and cause a switch from corn to soybeans regardless of the soybeans-to-corn price ratio. Similarly, plant disease for wheat can lead to switches to soybeans, other oilseeds, or other alternatives, such as summer fallow. References Good, Darrel. "Corn and Soybeans at Mid-Year," Weekly Outlook 3/2/98. University of Illinois, March 2, 1998. Heifner, Richard. "Price Analysis for Determining Revenue Insurance Indemnities and Premiums," Report to the Risk Management Agency. ERS-USDA. Nov. 6, 1996. Kohlmeyer, Bob. "Evolving Significance of Corn/Soy Ratio," World Perspectives, Inc., March 3, 1998. Schwartz, Sara. "An Examination of U.S. Corn Area by Region, 1975-94," Feed Situation and Outlook Yearbook, FDS-1995, ERS-USDA, Nov. 1995. Smetana, Dick. "1998 Coarse Grains and Wheat Topics," Agricultural Outlook Forum 98, U.S. Dept. Agri., Feb. 23, 1998. Westcott, Paul. "Policy and Modeling Issues Affecting the Estimation of Supply Elasticities," Seminar presented at the FCB-MTED seminar series on supply response, ERS-USDA, April 1997. Estimating Feed Use: Background and Issues by Allen J. Baker 1/ --------- 1/ Agricultural economist, Market and Trade Economics Division, Economic Research Service. --------- Abstract: Feed use is an important part of total use of feed grains, but there are no feed survey data published by the U. S. Department Agriculture (USDA). USDA calculates this category as a residual, equal to total supply minus the sum of other known uses. But for forecasting feed and residual, an index of animal units is used as an indicator of changes in feed and residual. The animal index does not provide an estimate of feed need or use. Feed use estimates would need to incorporate animal numbers, rates of gain, weather patterns, and regional, and seasonal feeding practices. In addition, periodic updates to include technological advancements, genetic improvements, and new feed additives or regulations would need to be incorporated. Keywords: Feed and residual, feed grains, animal units, feed units. Introduction Feed and residual use of feed grains (corn, sorghum, barley, and oats) plus wheat during 1992/93 through 1996/97, accounted for an average of 68 percent of domestic use and 49 percent of total use. Determining the important variables affecting feed use could help in forecasting total use and prices of grains. As noted by a leading analyst at the 1998 Agricultural Outlook Forum, the grain industry would benefit from a national survey on feeding, which is not done. Feed use is important because it is grouped with a residual component which occurs throughout the production and marketing process. The Economic Research Service's (ERS) estimates of feed consumption by type of livestock based on typical rations were discontinued in 1985. At that time, the rations used to allocate feed use were out of date, and feeding had become more complex. Animal scientists had learned much more about the nutritional needs of animals. Also, computers have allowed feed formulation to be a quick and easy process. Grain prices and market factors are critical determinants of demand for feed, like other uses, but livestock feeders tend to deal with a wider array of substitution choices among the grains and competing feedstuffs. Meat prices and animal market forces not only shape total feed demand, but also the type of feed ingredients used. However, reactions to economic signals may be limited at least in the short run by biological constraints, as manifested in livestock cycles. Some knowledge of animal science and nutritional considerations is also useful to understand the dynamics of feed conversion, impact of different breeds, and the interaction of environmental influences such as weather. On the commercial level, rations are typically based on a least-cost mix of ingredients that meet the nutritional needs of the particular animals. The amount of any one grain fed to particular types of livestock is hard to estimate because of the substitution potential. Although grains are not perfect substitutes for one another and other energy sources, they are enough so that prices tend to move together. For example, the fact that some feeders will include wheat as part of the ration if wheat prices are low tends to lower demand for corn or sorghum that "usually" is fed. However, during a period of drought-caused high grain prices, feeders are forced to find ways to conserve feed and use non-traditional feeds. Cattle and other herbivores can be maintained on grass alone. However, to fatten or produce more milk requires a ration that provides more nutrients (energy), which usually includes grains but could include other sources of nutrition. Cattle can be placed on feed as light as under 400 pounds and many weigh over 800 pounds, greatly affecting grain use. Growth in the food processing industry has provided a wide range of byproducts, and even returned or stale products are often included as other alternative feeds, such as stale donuts, gummy bears, and candy bars. Since feeders use least-cost rations that meet nutritional needs, often the most economical grain is the one produced nearby. For example, the Pacific Northwest grows barley, and the dairies in that area tend to feed barley in their rations. On the other hand, more animals are being fed in grain deficit areas of the country. For example, broilers grown in the Southeast are fed locally grown grains which are supplemented by grains from other regions, usually corn. Feed and Residual Derived - Measurement Issues In calculating feed use of the individual grains, feed and residual is estimated by subtracting other uses from total use. The National Agricultural Statistics Service (NASS) surveys farmers and others who produce, store, and use grains and report production and/or stocks. Stocks at the beginning of a year (or a quarter) plus production and imports yield total supply during that time period. Total supply less ending stocks give the total use during the year (or quarter). Data are available for exports, and Census reports and other sources give information on food and industrial uses. There are, however, no surveys on feed use of the various grains, and this category represents a true residual that is not measured directly. The feed and residual category includes any reporting or estimating errors in the surveys conducted on production, stocks, food, industrial use, and trade. Several factors may contribute to this measurement error. Shrinkage due to changes in moisture content contributes to measurement problems. A farmer may report that he or she has so many bushels of corn in his or her on-farm storage. Yet, when the farmer takes the grain from the on-farm location to an elevator to store or sell, the grain is weighed. Since grain usually loses moisture in storage, and also "shrinks" when it is handled because of losses in the form of dust and kernels left in the augers or elevators, the farmer may have "lost" bushels during the period it was stored. This "shrinkage" would not be consistent from year to year due to the moisture content of the grain at harvest and weather conditions while the grain is stored. In addition, the residual could include grain in transit, especially on river barges where the grain could be uncounted. Finally, grain to be fed may be cracked, put in a silo, and because the grain has changed forms, it is not counted as grain in the stocks report. Thus, use may be overstated in the first quarter of the marketing year for grains (e.g. September-August for corn) because it is not going into storage as "corn" grain. Residual would be expected to have a positive bias because shrinkage, grain in transit, etc. would add to the total and be allocated to feed and residual. Despite these factors, there is no particular reason to believe that errors in the surveys are biased either upward or downward. In addition, there is no particular reason to believe that more or less grain is unreported or in transit one year over another. Variables That Complicate Feed Analysis Many types of animals consume feed. We usually think of the animals on farms and ranches that are grown for meat, milk, or fiber. In addition, there are pets of all types and much of their food is made with grain or grain byproducts. Although a small part of the total, horses, mules and donkeys are another complicated group that may be pets or working animals, such as race horses or pack animals. Depending on their role, animals need different feeds. Working animals need more feed than those that lead sedentary lives. Animals that are nursing their young also have different nutrient needs than others. Even within a specific group of animals, feed needs can vary widely. For instance, dairy cows are fed a ration to supply their needs for body maintenance and milk production. Smaller breeds of dairy cows need less feed for maintenance than larger breeds, so to estimate feed needs, some assumptions are made as to the proportion of the inventory in the various size categories. Similar types of problems are associated with feeding cattle, hogs, chickens, and fish. Rations change at different ages and weights as the animals develop. Just as animal nutrition research brings a better understanding of nutrient needs for animals, cereal scientists and other researchers have found ways to make feed ingredients more substitutable in the ration. Steam treating and flaking grains make them more digestible for some animals. Enzyme treating feather meal can make it useful in some animal rations. The discovery and use of micro ingredients has allowed more substitution of feeds in rations. However, feed mills have to be especially careful to get micro nutrients thoroughly mixed in the feed so that all the animals get what they need. Even though it costs more to pellet feed, pelleting assures the animal gets all the feed and keeps them from sorting out the parts that have a better appeal (taste better). This also means that the ration can be changed without the animals noticing as much, or "good" stuff can be added to keep intake up for greater consumption and gains (shorter feeding periods). (Ingredients such as molasses, whey, and sugars enhance feed desirability.) Another complicating factor in determining feed use of particular feeds is feeding value. Grains and roughage vary in quality and feeding value both between types of grains or roughage, or even within a grain or roughage type. Many feeders test the incoming grains to know how to balance the rations for their animals. (Also, if the incoming grain has a toxin in it, feeders want to know.) When farmers used to grow the grains they fed, they had some idea if the grain was immature when harvested or other potential problems. Many animal nutritionists would say that they knew less about their feed than they thought they did. Now that is not the case, as most feed grain is purchased. The feed value, as measured by various criterion, such as starch or protein level, can vary seasonally, by location, and even among types of hybrid used. Changes in the Livestock Sector High performing animals have been around for a long time, as farmers selected the best animals for increased performance. Historically, farmers were limited to their herds or neighboring herds for high performing animals. Now, animal data banks are available to help in selecting for high performance from all over the world. Soon, animal husbandry professionals may be able to use cloning and gene transplanting to increase performance. Modern broilers are a good example of breeders developing higher performing birds, so that each year it takes less time to raise the broiler to market weight. This is a combination of both genetics, improved feed rations, and often, improved management. 2/ --------- 2/ Dudley-Cash, W. A., "Poultry Nutritionist's Role Is Not To Improve Bird Performance," Feedstuffs, April 1, 1996, page 11 & 23. --------- Livestock operations are getting larger over time as more groups shift to attain economies of size in production, purchasing, and marketing. Large hog operations are changing the way pork is produced and some of these operations have breeding herds, feeding facilities, and slaughter plants. Some are contract operations that raise hogs for packers and some are offshoots of the broiler operations. Larger operations often have one or more nutritionists on staff (or a consulting nutritionist) to formulate their feeds. As the dairy operations have gotten larger, cows are kept on dry lots and not put on pasture, except when they are not producing milk. This practice began in California and has spread across the country. Cattle on feed will gain faster in the feedlot where they don't have pasture, but depending upon grain prices, they may be grown to heavier weights before they go on intensive feed. Few hogs or chickens are allowed to fatten on free range systems as in the past because they will grow faster if they are confined. Hogs and chickens are single stomach animals that do not convert roughage to meat very efficiently. They also need a high quality balanced protein source. Some specialty birds are raised on free range, but they are the exception. With rations geared to specific needs of the livestock, an average ration to estimate feed use would likely oversimplify the real world. For all the different types of livestock, rations would have to be designed the same as the nutritionists are presently doing in their job or as consultants. The rations would have to be adjusted over time to reflect changing production patterns and genetic improvements in the animals and birds. Historical Perspective Estimating feed needs was originally started as an effort to provide supplies to war-ravaged countries during and after the Second World War. The estimates obtained gave policymakers the information on potential "extra" supplies of feed in the various States. The work was based on the State level and then aggregated to the United States. Data were available from farm management studies on feed use by various livestock categories and supplemented with information from the experiment stations on feeding practices. Inventories of animals were available from the surveys done by the Statistical Reporting Service (now the National Agricultural Statistics Service). This information allowed indexing animal numbers to show consistent changes in total grain consuming animal units (GCAU's) and roughage consuming animal units (RCAU's). Efforts were made to re-estimate average feeding rates about every 10 years, starting in 1940. The last time the rates were re-estimated and a new index published was in 1974, using feed units based on 1969-71. Animal units were based on the annual feeding rate for dairy cows and were estimated to be 11,093 feed units, of which 4,293 were concentrates and 6,800 were roughage. Feed units were defined as equivalent to 1 pound of corn containing 78.6 percent total digestible nutrients. Feeding rates were based on the minimum feed requirements, published by the National Academy of Sciences, Council of Animal Nutrition, increased by an estimate of losses from harvest to ingestion for both concentrates and harvested roughage. Rations change over time as more is learned about the nutrient needs of the animals. In the 1959-61 feeding rates, dairy cows were estimated to consume 2,495 pounds of concentrate feed units, and 4,365 pounds of roughage units. Average milk production in 1959-61 was 7,045 pounds per cow, and in 1969-71, milk production was 9,733 pounds. Milk production in 1997 was 16,954 pounds per cow. More milk production requires increases in feed units. For example, using information from McCullough, 3/ ---------- 3/ McMullough, M.E. Total Mixed Rations and Supercows, W.D. Hoard & Sons Company, Fort Atkinson, Wisconsin, 1994. ---------- roughly comparable feeding units today are estimated to be 11,382 pounds from concentrates and 4,787 pounds from roughage for a total of 16,169 pounds, an increase of 46 percent since 1969-71. The herd-average milk production for the ration was 28,900 pounds. Keep in mind that we are talking in pounds of feed units, not actual feed the animal consumes on a dry matter basis, which remains constant at 2-4 percent of body weight per day. This is because the animals will only consume so many pounds of feed per day and the daily consumption has not changed, but the composition of the consumption has changed. If we looked at other classes of livestock, we would expect to see feed per head decline rather than increase. In broilers, for example, producers are using less feed to produce more meat than they were a few years ago. The decrease in feed use comes from both changes in feed and genetics of the birds. These increases in production of meat per unit of feed are likely to be true of cattle on feed as well as hogs, although slaughter weights for both continue to trend higher. So, if we recalculated the GCAU indexing factors based on the dairy cow, the dairy cow would still be equal to one but the factors for the other animals would likely be smaller. To get the indexing factors, we divide the feed units needed by the dairy units, so if the concentrates for dairy were 11,382 and the feed units declined 4 percent for broilers, the new factor for broilers would be .0007, down from .002 in 1969-71. The factors in the past were calculated on a State basis then aggregated to the national level. This was to account for variations in feeding practices and use of roughage across different States. For example, in the past, dairy producers in the South could utilize pasture more months of the year than those in Northern States and would use less hay during the year. Also, some States grow feed barley and little or no corn, so livestock in the barley-producing States would be more likely to be fed barley than corn. Thus, feedstuffs produced in the State were allocated to the livestock produced there. Any extra feed needed was allocated from another State. Otherwise, a typical ration might leave out certain types of feed. One advantage of this approach is that when the amount of barley fed to certain classes of livestock are needed, one only has to estimate the within-State allocation of barley to various kinds of animals. Suppose you had a State with dairy operations, layers, beef cattle herds, hogs, and horses. The State produces barley, corn, oats, wheat, corn silage, and hay. Now suppose the State is grain deficit and feed grain is brought in. How much barley is fed to dairy and how much to layers, etc? One doesn't have enough information to definitely allocate grains to animal-type categories and would have to make some estimates. In the mid-to-late 1970's, ERS initiated an effort to update the feed use coefficients. Iowa State University cooperated with ERS, developing a new procedure for estimating feed use that was based on animal needs and current feeding practices. It did not, however, allow for shifts based on crop prices, and the data were developed by region rather than State. The study made many simplifying assumptions because livestock data to estimate feed needs would require not only livestock numbers but ages, weights, stage of production, and sex, well beyond the available data. Within the Iowa State framework, feed needs were based on net energy for maintenance and net energy for weight gain or milk production. Also, factors were included to reflect weather variables on a monthly basis because when it is very cold, animals use more feed for heat production, and when it is hot animals do not eat as much. Dry matter intake was estimated for different classes of animals based on weather, sex, and size. For cattle on feed, a rate of gain was also needed because, depending on relative prices of grain and roughage, feeders may reduce grain and slow rate of gain. The bottom line is that to use present day nutrient variables, a lot of information is needed to duplicate the decisions made by livestock feeders. Data were pieced together in the Iowa State study from many sources to develop estimates. In general, NASS data provide some of the numbers needed but not all. NASS does not report the stage of production cycle for dairy cows nor weights of the animals, but rather numbers of cows and milk production. Some State data for cattle on feed give placement weights, but not numbers on feed by weight group which is needed to be able to estimate the rate of gain in the feedlots. For hogs, data are available for total sows and gilts farrowing but not their weights. In crop years 1977 through 1984, aggregate feed use in the Iowa State study resulted in an estimate of less corn and oats used, but more barley, than reported by USDA in the feed and residual numbers. Iowa State reported that the differences were the result of holding the ration composition constant and not reflecting substitution of grains. Ration composition was established based on 1984-85 diets that had a relatively high percentage of wheat due to favorable prices for wheat relative to other feed grains in those years. As a result of holding rations constant, feed use in the Iowa study did not reflect the variation that the calculated feed and residual numbers did during the 7-year comparison. Since there was such variation between the Iowa State study's results and USDA's estimated feed and residual use and because of the data/labor intensive nature of that study's procedure, the results were not incorporated into USDA's feed estimates. Current Approach The indexing procedure using the 1969-71 weights is currently used to convert animal numbers to GCAU's and feed per GCAU is used to determine if long-term feed and residual forecasts seem reasonable for the baseline estimates. The weighting factors developed in 1969-71 are also used to calculate an index for high protein animal units (HPAU's), roughage consuming animal units (RCAU's), and grain and roughage consuming animal units (G&RCAU's). In calculating the GCAU's, only data on regularly reported animal numbers and an estimate of horses and mules are used. The animal numbers are from the annual inventory reports and are multiplied by the weighting factors. For example, the 9.2 million dairy cows on hand January 1, 1998, are multiplied by 1.0474 to give 9.6 million GCAU's. A similar procedure is used for other dairy cattle, cattle on feed, other beef cattle, sheep, and goats. The poultry numbers use poultry raised--one quarter of 1997 and three quarters of 1998 to put the inventory on a non-calendar year. GCAU's for hogs are based on spring and fall pig crops, using 20 percent of the spring 1997 pig crop, 100 percent of the fall 1997 pig crop, and 80 percent of the spring 1998 pig crop to reflect number of hogs consuming grain from the 1997 crops. These are multiplied by .2285 to arrive at the GCAU's from hogs. Cattle on feed may be undercounted, because cattle on feed used to be a sideline of grain farmers in the Corn Belt, where farmers would feed steers during the winter when they had extra time. Now cattle feeding has moved to areas that get less rain and is a year-around operation, with cattle being placed continuously. Various alternative approaches have been used or proposed to estimate feed and residual use. For example, feed and residual use could be expressed as a function of meat and egg production. However, meat can be produced by various methods, some using more grain than others. Also, when grain prices increase, animal production has declined when farmers could cut production. Breeding animals can be sold right away when grain prices make production unprofitable, but market animals usually are fed until they are big enough to be purchased by packers. Also, if meat supplies are small and prices high, feeders may increase the weight of animals before sale. This increases the supply of meat even with the same number of animals, and increases the feed per animal, but may not increase the total feed required if more animals are raised and sold at lighter weights. Still, there usually is a close relationship between total feed and residual and meat and egg production. Another source of information on feed use is the annual article by R.A. Schoeff, PhD. and D.J. Castaldo, PhD. 4/ ---------- 4/ Schoeff, R.A., and D.J. Castaldo, "Livestock and Poultry Inventories and Feed Use Potentials: U.S. Feed Market," Feed Management, Watt Publishing Co., Mt. Morris, Ill. September 1997, Volume 48, Number 9. ---------- They use NASS livestock and poultry animal production numbers and feed needed to feed these animals and birds to estimate the potential feed use for the United States. Their estimate of feed needs for 1996 was 160.8 million tons, down 2.4 percent from 1995. In contrast, USDA estimates of total concentrate feeds fed in October 1996 through September 1997 was 234 million tons, including 52.2 million tons of byproduct feeds. In 1995/96, USDA estimated total concentrate feeding of 203.4 million tons, including 49.3 million tons of byproduct feeds. USDA's estimates do not include distillers' and brewers' grains because there is no official production report for these feeds. Future Developments Many people would like to have estimates of grain and other feed consumption by livestock class to be able to estimate total use of the grains. In an era of tight resources, a new survey would probably mean dropping something else. Also, it is not known if the livestock feeding industry would cooperate even if USDA has the resources to compile the data. References: Allen, George C. and Margaret Devers, National and State Livestock-Feed Relationships, USDA, Economic Research Service, Statistical Bulletin No. 446, February 1970. Allen, George C., Earl F. Hodges, and Margaret Devers, Livestock-Feed Relationships National and State, USDA, Economic Research Service, Statistical Bulletin No. 530, June 1974. Jennings, Ralph D., Consumption of Feed by Livestock, 1909-56, Relation Between Feed, Livestock, and Food at the National Level, USDA, Agricultural Research Service, Production Research Report No. 21, November 1958. Schoeff, R.A., and D.J. Castaldo, "Livestock and Poultry Inventories and Feed Use Potentials: U.S. Feed Market," Feed Management, Watt Publishing Co., Mt. Morris, Ill. September 1997, Volume 48, Number 9. Jennings, R. D., Feed Consumption by Livestock 1910-41 Relations Between Feed, Livestock, and Food, at the National Level, USDA, Bureau of Agricultural Economics, Circular 670, April 1943. Jennings, R. D., Feed Consumption by Livestock 1910-47 Relations Between Feed, Livestock, and Food, at the National Level, USDA, Bureau of Agricultural Economics, Circular 836, December 1949. Havenstein, G. B., P. R. Ferket, S. E. Scheideler, and B. T. Larson, "Growth, Livability, and Feed Conversion of 1957 Versus 1991 Broilers When Fed "Typical" 1957 and 1991 Broiler Diets." Poultry Science 73:1785-1794. Dudley-Cash, W. A., "Poultry Nutritionist's Role Is Not To Improve Bird Performance," Feedstuffs, April 1, 1996, page 11 & 23. McMullough, M. E., Total Mixed Rations and Supercows, W.D. Hoard & Sons Company, Fort Atkinson, Wisconsin, 1994. , Red Meats Yearbook 1995, Supplement to Livestock, Dairy, and Poultry Situation and Outlook, USDA, Economic Research Service, Statistical Bulletin 921, 1995. , Livestock and Meat Statistics, 1984-88, USDA, Economic Research Service, Statistical Bulletin 784, September 1989. Madison, M., Poultry Yearbook, USDA, Economic Research Service, March 1997, available from Internet "http://www.mannlib.cornell.edu/data-sets/livestock/89007/." Lawrence, J. D., M. L. Hayenga, and M. H. Jurgens, Feed Utilization Estimates for Livestock and Poultry in the United States, Unpublished paper, USDA, Economic Research Service and Iowa State University, 1986. Southeast Asian Feed Imports and the Financial Crisis by Gary Vocke 1/ ---------- 1/ Agricultural economist, Market and Trade Economics Division, Economic Research Service. ---------- Abstract. The feed demands of the fast-growing livestock sectors in Southeast Asia were rapidly outpacing domestic production, creating rapidly growing import markets for U.S. exporters of feedstuffs. Now, currency devaluation and economic slowdown in the region have raised feed prices and reduced livestock product consumption. With less demand for livestock products, feedstuff imports are declining. However, the fundamentals that led to growing feedstuff imports during the 1990s should again lead to growing imports once these countries get their economies turned around and economic growth resumes. Key words: Southeast Asia, feed imports, financial crisis, livestock. Country after country in Southeast Asia has been forced to devalue its currency since the region's crisis began in mid-1997. Economic growth in Southeast Asia has subsequently taken a sudden dip in the last few months. This financial market disarray and reduced rates of economic growth have hurt U.S. export prospects to the region. The region's financial crisis is currently projected by the U.S. Department of Agriculture (USDA) to be a temporary phenomenon, with expectations for recovery of economic growth in a few years. But the slump in 1998 is deep; in the worst situations there will be negative economic growth. The Southeast Asian economies were some of the fastest growing economies in the world and provided emerging markets for a wide range of agricultural imports from the United States. Together, the Philippines, Indonesia, Thailand, and Malaysia increased their imports of U.S. agricultural products at an annual rate of 17 percent from 1990 to 1996. This growth accounted for 10 percent of the expansion of U.S. agricultural exports over this period. In contrast, for 1997, the level of U.S. agricultural exports to these four countries declined 9 percent from 1996. As the crisis extended into 1998, U.S. agricultural exports to the region have declined further. Imports from the United States by Southeast Asia for January 1998 are only 52 percent of January 1997's level. Key to the bright prospects for U.S. exports to the region before the crisis was the feed demands of the fast-growing livestock sectors in Southeast Asia. These feed demands were rapidly outpacing domestic production, creating rapidly growing import markets for U.S. exporters of feedstuffs. Now, currency devaluation and economic slowdown in the region have raised feed prices and reduced livestock product consumption. With less demand for livestock products, the region's feedstuff imports are reduced. The purpose of this article is to provide a background on the trends in the region before the current crisis began. These trends provide a picture of what might take place when these countries get their economies turned around and economic growth resumes. Non-ruminant Production Was Expanding Rapidly While high rates of economic growth occurred, diets in these countries were changing in predictable ways. Higher-income, urban residents were diversifying their rice-based diets and consuming more livestock products. From 1985 to 1996, the share of calories from animal products rose from 10 to 12 percent. The dietary shifts changed the agricultural sectors in the region. Increased consumption of livestock products was supplied through increased domestic production and some additional imports. Most of the increased production of livestock products was for domestic consumption. The clear exception is poultry production in Thailand. Thailand's poultry production grew an average of 7 percent annually over the past decade, and poultry meat exports accounted for 20 percent of the increase in production. Starting in the 1980s, the index for livestock production increased rapidly, outpacing the rest of the agricultural sector (Figure D-1). The expansion of livestock production was driven by increased poultry and pork production (Figure D-2), not cattle for beef or milk. Except for the Philippines, poultry is the biggest livestock sector. Pork is larger in the Philippines. For the region, the expansion of poultry and pork production has been mostly in commercial farms and intensive livestock operations. The technology for such livestock operations is easily transferable from the more developed, temperate climate agricultures where the practices originated. Poultry and pork consume about 95 percent of all the manufactured feed in Southeast Asia [U.S. Feed Grains Council, 1998]. These countries import a very small proportion of their meat consumption, only 3 percent in 1996 [Food and Agriculture Organization, 1998]. Importing livestock products is highly regulated and/or taxed by governments to protect domestic production. Among the meats, beef is the largest import item, and Malaysia and the Philippines are the largest importers. Both of these countries import low- cost buffalo meat from India for the lower end markets. All the countries import beef for the hotel, restaurant, and institutional trade and higher end supermarkets. Most of the cattle in the region are draft animals owned by smallholders. They are typically slaughtered only after their usefulness has passed. These animals are sometimes fed a short time to add some extra weight. This short-term feeding of spent cattle was the extent of the feedlot feeding in Southeast Asia until recently. Across these Asian countries, the terrain and the land-use patterns provide only limited pasture lands to produce feeder cattle. In recent years, however, in response to the rising consumer demand for beef, Southeast Asia had become the primary export destination for Australian range cattle, with the two largest markets being Indonesia and the Philippines. These imported cattle, more than 880,000 head in 1997, are for relatively short-term intensive feedlot feeding. Much of the feed consumed by these feeders is byproducts -- pineapple wastes, for example. Tariff policies in both countries favor imported live cattle over meat imports. In Indonesia, for example, imported feeder cattle (less than 350 kilograms) are not subject to tariffs, unlike slaughter cattle (15 percent tariff) and beef (20 percent tariff) [Kaus, 1997]. Indonesia's imports of live cattle have now come to a virtual halt because of the financial crisis. The Southeast Asian countries have only small dairy industries, largely because of their tropical climates. About 85 percent of dairy product consumption is imported. Australia has the largest share of these dairy imports, which are largely powders and milk fats for reconstitution into fluid milk and other dairy products. The Governments of Indonesia, Thailand, and Malaysia have intervened through import restrictions to ensure that milk prices for domestic producers are not undermined by competition from imports. The Philippines depends almost entirely on dairy imports. Its dairy-product tariffs are primarily for raising revenue for the government. Feed Demands Outpaced Domestic Supply Diversifying traditional, grain-based diets with livestock products rapidly increases crop input requirements; one kilogram of livestock product produced in intensive production systems requires 2-6 kilograms of feedstuffs, depending on the species. Figure D3 shows the increased use of feedstuffs in the four countries discussed here; and the slowing of use in 1997 as the crisis in the region began to take effect. These countries were not able to supply all of their feed requirements from domestically grown crops. Rapidly increasing imports were needed to support the rapid expansion of livestock production (Figure D-4). This trend of increasing dependency on imported feeds was sharply re-enforced with the switch of the region's only major corn exporter, Thailand, from an exporter to an importer. Feedstuff production in tropical Southeast Asia has been limited in the past by the lack of suitable varieties of corn and soybeans. Corn and soybean yields in 1997 averaged 1.93 and 1.26 tons per hectare, respectively. These yields are well below temperate climate yields. For example, 1997 U.S. corn and soybean yields are 7.97 and 2.63 tons per hectare, respectively. The low yields limit the competitiveness of these two crops for existing cropland. In addition, cropland expansion in the region has slowed since the 1980s due to environmental constraints and the rising cost of new land development. Corn production, in total, leveled off in the mid-1980s as harvest area declined. While corn area has recently increased in Indonesia, the expansion was not enough to offset declines in Thailand and the Philippines as farmers switch to more profitable crops. Corn yields have increased steadily in the region, but not enough to offset acreage losses. Corn has always been primarily grown for livestock feed in Thailand, and now also increasingly for feed in the Philippines and Indonesia. Before the take-off of livestock production in the latter two countries, corn was primarily a food crop. Malaysia produces very little corn since it is not competitive with tropical crops in there. Soybean products are an important source of protein for people in Southeast Asia, particularly in Indonesia. As with corn area, soybean area has been declining in the region as a whole. Malaysia does not produce soybeans. Yields in the region have been trending up, but not nearly enough to offset the reduced area. Production peaked in the early 1990s at about 2.2 million tons, and in 1997 the total was 1.8 million tons. Large imports of soybean meal and soybeans for crushing are needed in Southeast Asia despite the large supplies of palm kernel meal and copra meal. More than 70 percent of meal fed in the four countries is soybean meal. Malaysia and Indonesia are the world's largest producers of palm oil, and coconut production is important in the Philippines and Indonesia. However, meals made from palm kernel and copra are high-fiber, low protein meals better suited for ruminants. Because there is relatively little ruminant feeding in Southeast Asia, these abundant supplies of palm kernel meal and copra meal are exported. In 1997, these countries exported slightly more than 2 million tons of palm kernel and copra meal. By comparison, the region imported slightly more than 3.4 million tons of soybean meal. Feedstuff Imports Were Becoming less Regulated Governments in Thailand, Indonesia, and the Philippines have tended to limit the competition against their own farmers from lower-cost producers outside the region. Malaysia, in contrast, has allowed its private sector to import feedstuffs easily. The protectionist policies of the Governments of Thailand, Indonesia, and the Philippines have come under increasing challenges from the local feed manufacturers and livestock producers looking for cheap feedstuffs for their rapidly expanding sectors. These pressures and compliance with World Trade Organization (WTO) agreements have led to a progressive freeing of feed imports. Thailand. Even though Thailand's imports of corn, soybeans, and soybean meal are highly regulated to protect domestic crop farmers, the country's imports are substantial. Thailand's imports over the past 3 marketing years have averaged 865,000 tons of soybean meal, 353,000 tons of corn, and 525,000 tons of soybeans. Government policies limiting market access to low-cost foreign supplies of feedstuffs have resulted in relatively high feed costs. The government simultaneously protects its domestic livestock producers against livestock- product imports. Problems arise, however, for the country's poultry meat exporters. In Thailand, poultry meat is an important export product, mostly going to Japan and the European Union. High feed costs place the exports at a disadvantage. In response, the government offers rebates on the duties paid on the imported feedstuffs used to produce poultry meat for export. The government uses a tariff-rate quota system to regulate corn imports. For 1998, the government has raised the quantity of corn that can be imported free of tariffs from 200,000 tons to 300,000 tons. Thailand has been dropping its tariff rate on the in-quota imports. The rate was 7.5 percent in 1995, 3 percent in 1996, and 0 percent in 1997. The imports must arrive during a March 1 to June 30 window. If there is still a shortage after the 300,000 tons are imported, the import quota can be increased. However, this additional quota can be made only if the Bangkok corn price exceeds 4.80 baht/kg. Through such administrative rules, the Thai Government is able to tightly regulate the import of corn. Any non-quota imports will be charged a 77.8-percent import tariff, plus an additional surcharge of 380 baht/ton. The quota import allowance is allocated among five associations of feed manufacturers and livestock producers. For soybean and soybean meal imports, no quota or import window is applied. Soybean imports are duty free. However, the eligible importers are responsible for purchasing domestic soybean meal from local vegetable oil crushers at a minimum Bangkok price of 8.50 baht/kg. Soybean meal imports are subject to a 5-percent tariff rate, the same as in 1997. Wheat is imported for raising shrimp for export. These imports are not regulated. Even though wheat is not a suitable crop to produce in this tropical country, there is still an import tax of 1 baht/kg. Philippines. Market access is also restricted in the Philippines, but the country's imports are still substantial. The Philippine's imports over the past 3 marketing years have averaged 921,000 tons of soybean meal, 267,000 tons of wheat for feeding, 404,000 tons of corn, and 160,000 tons of soybeans. The import of corn is tightly controlled by the government to protect domestic corn producers. Corn has traditionaly been an important food crop and the corn sector, along with the rice sector, is highly regulated. Domestic corn prices are controlled by the National Food Authority (NFA) using support and ceiling price schemes. To maintain domestic price policy, the NFA regulates the local supply of corn by purchasing directly in the open market and managing the disbursement of buffer stocks. Corn imports have been a trade policy decision of the government. In the early 1990s, corn imports were briefly banned. However, as a signatory to the WTO, the government was forced to agree to a minimum access quota for the corn market. The in-quota corn imports have a tariff of 35 percent, while the out-of-quota tariff rate is 80 percent. The quota level is 155,000 tons for 1998. The NFA will import in-quota corn and distribute it via an auction [U.S. Feed Grains Council, 1998]. This auction has a minimum price that covers the import costs, tariff, and a service fee. Only end users are allowed to bid on the auction lots, with the quantity they were allowed to bid determined by each company's registered production capacity. The NFA can license qualified processors and producers to import above the quota level, but they will have to pay the much higher tariff of 80 percent. The end result of such policies is that Philippine livestock producers are disadvantaged by relatively high corn prices. When corn supplies in the past have been inadequate, livestock producers have fed substantial quantities of imported wheat instead of imported corn to avoid paying the high out-of-quota corn-tariff rate. Wheat imported by the private sector is taxed at an applied rate of 10 percent. Wheat is not grown in the Philippines. Soybean production is very limited in the Philippines and the government is not involved as it is with the corn sector. Soybean and soybean meal imports are handled by the private sector. There is a 3-percent tariff on the import of soybeans and soybean meal. Indonesia. Over the past few years, the import of feedstuffs was liberalized to reduce feed costs for the country's livestock producers. Indonesia's imports over the past 3 marketing years averaged 851,000 tons of soybean meal and 773,000 tons of corn. The imports of corn and soybean meal are by the private sector and are not taxed. The import of feedstuffs was highly regulated by the National Logistics Agency (BULOG). The liberalization of corn trade began in 1989 when BULOG's trading in corn was ended. Later, in 1994, the corn import duty was lowered from 10 percent to 5 percent. Then, in the deregulation package of 1995, the government reduced the import duty on corn to zero. Soybeans were imported by BULOG, crushed for BULOG, and then sold to feed manufacturers at administered prices above world soybean meal prices. BULOG also controlled the import of soybean meal. Deregulation of the soybean meal sector began in 1994 when the government reduced the domestic manufactured feed composition requirement from 40 percent to 30 percent. Also, the import duty on soybean meal was eliminated. In early 1995, the government first reduced the domestic source requirement to 20 percent, and then later dropped the domestic source requirement entirely. Feedmillers could then use imported soybean meal for up to 100 percent of their soybean meal requirements. In 1996, feed manufacturers could begin directly importing soybean meal. BULOG gave up its control. With this deregulation of soybean meal, feed manufacturers re-formulated their rations, primarily for poultry, to include more soybean meal and less corn. These re-formulated rations improved the efficiency of the poultry sector. The financial crisis that the country is now facing may bring some trade policy changes for soybean meal. The Indonesian Government is continuing to negotiate with the International Monetary Fund about the role of BULOG for the import of soybean meal during the crisis. There are also discussions about whether or not these imports can be temporarily subsidized to provide some relief for the country's poultry producers. Indonesia also imports soybeans, an average of 795,000 tons over the past 3 years, but these imports were for human consumption only, not for crushing to make meal and oil. Soybeans are a very important foodstuff in Indonesia. Soybeans are the dominant source of protein for about 65 percent of the population in traditional preparations of tofu, tempe, and soy sauces. Malaysia. Malaysia's imports over the past 3 marketing years have averaged 543,000 tons of soybean meal, 2,314,000 tons of corn, and 570,000 tons of soybeans. The story underlying the imports of feedstuffs to expand livestock production is quite simple. The country produces very little corn and no soybeans. Malaysia's feedstuff imports are handled by the private sector. Corn and soybeans enter the country without a tariff. There is, however, a 10-precent duty on imported soybeans. However, Malaysia's government does control the consumer price of unprocessed poultry meat. To protect consumers at the onset of the financial crisis in 1997, the government did not allow this price to be raised. By the time the government allowed the price to increase, some poultry producers had already been forced out of business as the local price of imported feeds rose dramatically with the devaluation of the Malaysian ringgit. U.S. Competitiveness Varies by Commodity The U.S. share of the feedstuff markets varies widely across the region (Table D1). U.S. exporters have a very high share of the soybean imports, more than 80 percent. U.S. shares of the corn and soybean meal import markets are much less. The principal competitors for the United States are: India and Brazil in the soybean meal markets, and China and Argentina in the corn markets. The fundamentals underlying the competitiveness of these suppliers in the region include lower prices, smaller sized shipments, and seasonal differences with the United States. India, with its freight advantage, can often export soybean meal to the region at a lower price than the United States. These Indian exports, however, are not directly comparable with the U.S. high-protein product because hulls are added back after processing the beans. The Indian exporters offer their product at a price low enough to offset the disadvantage in feeding this high-fiber meal to poultry. India, and also Brazil, supply shipments smaller than the 50,000-ton, U.S. Panamax shipments. The three major types of vessels are Panamax, Handymax, and Handysize. Panamax vessels range in cargo deadweight (dwt) from 48,000 to 66,000 tons [Schroeder, 1998). Panamax shipping reduces the per-ton cost compared with smaller vessels. This savings is important for U.S. exporters competing with nearby suppliers to the region. However, the draft of Panamax vessels excludes them from operating fully loaded in many Asian ports. The Handymax vessel's cargo dwt is 25,695 to 45,000 tons and the Handysize vessel's cargo dwt is 12,000 to 25,000 tons. The savings spread between the Panamax and Handysize vessels is $6-$10 per metric ton [U.S. Feed Grains Council, 1998]. Small-vessel shipments have advantages in addition to being able to deliver in more ports. Smaller shipments allow medium-sized mills to be able to handle a whole purchase logistically and financially. Otherwise, they have to rely on trading companies or group purchases. China is a low-cost producer of corn with the freight advantage of being in the region. Many importers in the region also favor buying from China because of the availability of smaller shipment sizes than Panamax. Argentina is a price-competitive supplier of corn to the region, and its Southern Hemisphere harvest cycle matches well with the region's import window. Besides competitive prices, some livestock producers prefer the corn from Argentina because of its high carotene levels. Devaluation Raises the Cost of Imported Feed The financial crisis in Southeast Asia has led to the devaluation of the Thai baht, the Indonesian rupiah, the Malaysian ringgit, and the Philippine peso. The devaluations vary across the region. Using average exchange rates for February 1997 with February 1998, the nominal devaluations are: 44 percent for the baht; 75 percent for the rupiah; 35 percent for the ringgit; and 35 percent for the peso. These devaluations sharply raised the export price of U.S. exports in domestic currencies (Table D-2). Compared with a year ago, the U.S. Gulf port prices for corn, soybeans, and soybean meal are down. However, for the Southeast Asian livestock producers, the import price of these feedstuffs from the United States has greatly increased because of the devaluations, especially in Indonesia. Incomes are falling for many consumers with the slowing of economic growth. Livestock product prices are rising, pricing many consumers out of the market for these products. The poultry meat sector has been particularly hard hit. In the near term, many livestock producers in the region are scaling back their operations, sometimes shutting down entirely. In Indonesia, for example, the poultry breeding industry that produced 14-15 million day-old chicks per week is now operating at only 30 percent of capacity [Foreign Agriculture Service, 1998]. The reduced feed requirements of this temporarily downsized Southeast Asian livestock sector led to lower feedstuff imports. The decline is so great that Indonesia is now exporting some corn. The Thai baht devaluation has made Thai poultry meat more competitive in world markets. After steadily expanding at an average annual rate of 25 percent from 1983 to 1992, Thailand's poultry meat exports had leveled off as the country's exporters lost their competitive edge because of high feed costs and rising labor costs for deboning the chicken meat. Now, with the devaluation, the Thai exporters have a cheaper product. This advantage has been offset to some extent by a rise in feed costs. The cost of raising broilers rose from about 23-25 baht/kg in mid-1997 to about 30 baht/kg in Feburary 1998. The USDA is projecting that Thailand will export about 227,000 tons of poultry meat in 1998, up from 197,000 in 1997 and 169,000 in 1996. The 1997 exports were boosted by a substantial expansion of exports in the last quarter of 1997 after the Thai baht had devalued. Conclusion The rapid expansion of domestic livestock production, and hence feed demand, has been reversed as consumers respond to the slowing of income growth and rising livestock prices. Feed costs have risen sharply. At the same time, consumer purchasing power has dropped. Many livestock producers are now caught in the middle. They can not pass to the consumers a price high enough to cover the increased costs of feed ingredients, and many are going out of business. However, the projected recovery of economic growth should once again boost imports of feedstuffs by Southeast Asia. When income growth resumes, livestock product consumption should again start to rise. Protectionist government policies will likely ensure that these livestock products will be supplied from domestic production. Feedstuff imports should then again grow because the agricultural fundamentals will likely remain the same: a lack of comparative advantage in producing corn and soybeans and livestock sectors based on non- ruminants for which high-fiber, palm kernel and copra meal is not a suitable protein supplement. References Food and Agriculture Organization (1998). FAOSTAT. Rome. Foreign Agriculture Service (1998). Economic Upheaval in Asian Poultry Markets. Livestock and Poultry World Markets and Trade. March. Kaus, Roger (1997). Marketing Cattle to South-East Asia. Queensland. Department of Primary Industries U.S. Feed Grains Council (1998). Private Communication. Washington, D.C. Schroeder Eric (1998). Port Development and Grain Transportation in Southeast Asia. World Grain. February. Market Factors and Government Programs Affecting U.S. Corn Prices Paul C. Westcott 1/ ---------- 1/ Agricultural economist, Market and Trade Economics Division, Economic Research Service. ---------- Abstract: An annual model for the U.S. farm price for corn is based on market factors as well as Governmental price support programs in selected years. The U.S. price support program affected corn prices, particularly in the late 1970's through the mid-1980s. Changes in the price support program since 1986 have resulted in less interference of that program with price determination. The model presented includes a shift variable to account for years when the corn price support program affected market prices. With the declining role of Government commodity programs in agriculture, price determination for corn now is based on market supply and demand factors. The stocks-to-use ratio provides a summary measure of supply and demand and is a useful indicator of corn price movements. Model estimates for 1997/98 corn prices are compared with USDA projections. Implications of U.S. agricultural policy changes for potential inter-year, annual corn price variability are discussed. Keywords: Corn, farm price, price determination, stocks-to-use ratio, price supports, commodity programs, forecasts. The U.S. corn crop plays a major role in the agricultural sector. As a source of income to farmers, corn cash receipts are the largest among crops. Over the last 5 years, corn cash receipts have averaged more than $17 billion, accounting for nearly 18 percent of total crop cash receipts. Corn also has an important role in linkages within the agricultural sector among various crops and between crops and livestock. Corn competes with other crops for land in farmers' production decisions, particularly soybeans. Corn is also the largest feed grain used by the livestock sector. Further, the United States is the largest exporter of corn, accounting for over 70 percent of global corn trade thus far in the 1990's. Consequently, events which affect the corn sector and corn prices are carefully watched by many subsectors within agriculture. Agricultural legislation enacted in 1996 fundamentally changed the nature of farm commodity programs in the United States, furthering trends toward market orientation in the sector. In particular, changes in the income support program shifted much of the risk of price volatility from the Government to producers (see Young and Westcott). As a result, market information affecting corn prices is particularly important under the 1996 Farm Act as farmers seek to make informed farm management decisions to manage risk and other market participants work within a more market-oriented agricultural sector. To provide market information regarding the agricultural sector, each month the U.S. Department of Agriculture (USDA) analyzes major agricultural commodity markets and publishes annual supply, demand, and price projections for the current year. Additionally, once a year USDA publishes longer term, 10-year baseline projections for the agricultural sector that includes commodity supply, demand, and prices. This paper examines some of the factors that affect farm-level U.S. corn prices. An annual framework is employed to develop a corn price model designed to be used in USDA's projections activities in conjunction with ongoing commodity market analysis of supply and demand factors. The model builds on two types of factors that influence prices--market supply and demand conditions, and Government price support programs. Market forces, as measured by supply and demand, influence prices. Year-ending stocks of an annually produced commodity, such as corn, summarize the effects of both supply and demand factors during the year, and are a useful indicator of price movements for the commodity. Annual prices for corn tend to have a strong negative correlation with their ending stocks. High corn stocks typically result in lower prices, while low stocks tend to push prices up. Historically, Government programs have also been important in influencing farm- level corn prices. Some programs, such as acreage reduction and set aside programs, have influenced prices indirectly by placing restrictions on the use of land for agricultural production, thereby affecting the supply of agricultural commodities. The nonrecourse commodity loan program has directly affected prices by providing support to farm-level prices and affecting market equilibrium in some periods. The key policy variable used in the price modeling effort in this article is the price support loan rate. However, the role of the loan rate in influencing prices has differed historically as the nature of the commodity loan program has changed under different farm legislation. Previous Research Many corn price models have employed the stocks-to-use ratio to represent market conditions in explaining movements in corn prices. The stocks-to-use ratio is defined as stocks of the commodity at the end of a particular time period divided by use of the commodity during that time period. As such, market conditions of supply and demand are summarized in this measure. Van Meir, and Baker and Menzie used stocks-to-use ratios in annual frameworks analyzing corn prices, while Westcott, Hull, and Green used such an approach in a quarterly model for corn prices. Numerous other unpublished annual corn price models using stocks-to-use ratios have been used internally within USDA in its forecasting activities. In each model, the stocks-to-use variable is negatively related to corn prices and provides a downward sloping, nonlinear curve of prices plotted against ending stocks-to-use. To represent the effects of Governmental price support programs on prices, many corn price models have been estimated with the dependent variable of price minus loan rate. The Baker and Menzie annual corn price model and part of the Van Meir analysis of corn prices and stocks used this approach, as did most of the unpublished USDA models. The U.S. price support program affected corn prices, particularly in the late 1970's through the mid-1980's. During this period, the support program's loan rate for corn was generally high enough to influence market prices. However, changes in the price support program since 1986 have resulted in less interference of that program with price determination. Price Support and Commodity Storage Programs for Corn The commodity price support program for corn allows producers to receive a loan from the Government at a designated loan rate per unit of production by pledging some of their corn production as loan collateral. Following harvest of the corn crop, a farmer who has enrolled in the corn program may obtain a loan for some portion of the new crop. For each bushel put under loan and pledged as loan collateral, the farmer receives a per-bushel amount equal to that year's loan rate. Under the loan program, the producer must keep the crop designated as loan collateral in approved storage to preserve the crop's quality. The producer may repay the loan at any time during the length of the loan, usually 9 months, paying back the loan principal plus accrued interest charges. However, at the end of the 9-month loan period, the farmer may choose instead to default on the loan rather than repay it, keeping the loan money and forfeiting ownership of the loan collateral (the corn) to the Government. Defaulting on the loan would make economic sense for the producer if the market prices were below the loan rate (plus interest), because the producer would effectively have received the loan rate for the crop rather than the lower market price. Historically, loan rates were set high relative to market prices in the late- 1970's through the mid-1980s (figure E-1). Loan program defaults resulted in the acquisition of corn by the Government, and Government stocks of corn reached over 1.1 billion bushels in 1982, or 15 percent of annual use (figure E-2). Also, a multi-year Farmer-Owned Reserve (FOR) program was begun in the late 1970's, which provided storage subsidies to farmers to store grain under loan for 3 to 5 years. Additional price support was provided under the FOR program in 1980-1982, with a higher reserve loan rate than available under the regular, 9-month loan program. The long duration of the FOR program, combined with high release prices needed for grain to exit the reserve, effectively isolated a large amount of grain from the marketplace. By 1982, corn held in the FOR rose to almost 1.9 billion bushels, about 26 percent of annual use. The combination of high price supports along with stocks being isolated from the marketplace in the FOR resulted in a significant policy effect on corn prices. Changes in the price support program since 1986 have resulted in less interference of that program with price determination. Three important policy features of farm programs under legislation enacted in 1985 significantly changed the loan program and the effect of price supports on market prices, starting in 1986. First, price support levels for grains were sharply reduced. The loan rate for corn was lowered from $2.55 per bushel in 1985 to $1.92 per bushel in 1986. Second, corn produced in 1986-1990 was not permitted to enter the FOR. And third, corn in the reserve was more accessible to the marketplace as a new policy instrument introduced under 1985 farm law, generic certificates, allowed early access to grain in the reserve before its contract expired. These policy changes facilitated a reduction in corn stocks in the late 1980's that was accelerated in 1988 when a major drought in the Corn Belt region of the United States sharply lowered corn production. Corn stocks fell from over 4 billion bushels to 1.5 billion bushels at the end of the 1990/91 season. Government-owned and FOR stocks fell from nearly 3 billion bushels to under 400 million by the end of 1990/91. Importantly, the combination of lower price supports, no further FOR entry, and generic certificates allowing access to FOR stocks eliminated the strong policy effect on price determination for corn. Essentially, the loan program continued to provide producers a source of short- term liquidity, but it no longer supported corn prices. Policy changes since 1990 have continued to keep the price supporting aspects of the loan program at a minimum. Since 1986, the corn loan rate has ranged from $1.57 per bushel to $1.92 per bushel, well below market prices for corn in most years. Implementation of marketing loans for corn starting in 1993, which allow repayment of loans at less than the original loan rate, further reduced the loan program's potential effect on market prices. Although the availability of generic certificates declined in the early 1990s, FOR stocks continued to be accessible to the marketplace as a new FOR release policy allowed farmers to repay their FOR loans and reacquire the loan collateral at any time rather than when prices reached specific FOR release levels. Later, the 1996 Farm Act suspended the FOR. As a consequence, since 1986, price determination for corn has largely been based on market supply and demand conditions without the influence of the Government price support program. The Model The general framework used here relating prices to ending stocks derives from an equilibrium model. In its simplest form, without the Government price support program, supply, demand, and stocks are each a function of price, with the market-clearing, equilibrium condition of determining the price at which supply equals demand plus stocks (equations 1-4). (1) S = f (p) (Supply function) (2) D = g (p) (Demand function) (3) K = h (p) (Stocks function) (4) S - D - K = 0 (Equilibrium condition) S is supply, D is demand, K is ending stocks, and p is market price. Supply is positively related to price, while demand and stocks are negatively related to price. In equilibrium, prices can be determined from the inverse of the supply, demand, or stocks function. Taking the inverse of the stocks function provides a price determination equation, with prices negatively related to stocks. (5) p = h-1(K) (Price equation; inverse stocks function) Introducing the Government price support loan program adds to the stocks function by incorporating the commodity loan rate to the function, as represented in equation 3a. (3a) K' = h (p; LR) (Stocks function with Government loan program) K' is the revised stocks function and LR represents the commodity loan rate. The Government loan program provides an additional feature to stockholding behavior that depends on the loan rate incentive to use the loan program. With this alternative stocks function, the inverse stocks function gives the following price determination equation. (5a) p = h-1(K'; LR) Prices would be expected to be negatively related to stocks. Prices would be expected to be positively related to the loan rate, particularly in those years that loan rates were set high relative to market clearing price levels and the Farmer-Owned Reserve isolated stocks from the marketplace. Model Implementation The functional form used to estimate equation 5a for annual corn prices is logarithmic. Semi-log and exponential functional forms can alternatively be used and provide similar estimation results to those presented here. (6) Ln (p) = a + b Ln (K'/U) + c Ln (LR) * Dum7985 U represents annual corn utilization, Dum7985 represents a dummy variable equal to 1 in 1979-1985 and equal to 0 in other years, and a, b, and c are parameters to be estimated. In equation 6, stocks (K') are measured relative to an indicator of the "scale of activity" in the corn sector, represented by the realized level of demand, actual utilization (U). This adjustment is needed because of growth in the corn sector over the last 20 years, so a particular level of stocks today represents a smaller portion of total use (or realized industry demand) than the same level of stocks in 1975. The result is a stocks-to-use variable commonly used in price models, providing a summary measure of market supply and demand conditions and an indicator of relative market tightness for the commodity. The expected sign of the stocks-to-use coefficient (b) is negative. The interaction term of the loan rate (LR) times the dummy variable (Dum7985) represents the effects of the loan program on corn prices from the late 1970s through the mid-1980s. The loan rate variable used in the model includes the higher FOR loan rate available to corn producers in 1980-1982. The years 1979-1985 chosen for the interaction term were when the commodity loan program, in conjunction with the structure of the Farmer-Owned Reserve program, had a significant influence on price levels in the sector. Loan rates were relatively high in those years and the multi-year Farmer-Owned Reserve program, with high release prices, isolated those reserve stocks from the market. The price supporting aspects of the loan program in the late 1970s through the mid-1980's imply that the expected sign for the coefficient (c) for the loan rate interaction term is positive. This specification contrasts with the approach frequently used in the past of defining the dependent variable as corn price minus loan rate. For many of those earlier models, the years 1979-1985, (when high loan rates and the structure of the FOR program affected price determination), were a larger part of the sample period used for model estimation. Here, those years represent only 7 of the 22 observations, so a separate policy shift variable seems more appropriate, with the dependent variable being the corn price. The specification of the interaction term represents an intercept shift related to the loan rate rather than a slope shift related to the stocks-to-use variable. An alternative specification that also included a slope shift adjustment for 1979-1985 produced a coefficient estimate for the slope shift variable that was not statistically different from 0. Farm-level prices used to estimate the model are season average prices collected by the U.S. Department of Agriculture's National Agricultural Statistics Service and republished by the Economic Research Service elsewhere in this Feed Situation and Outlook Yearbook (April 1998). Stocks, utilization, and loan rate data also are from this Feed Situation and Outlook Yearbook. FOR loan rate data for 1980-1982 are from Lin, Glauber, Hoffman, Collins, and Evans. Model Results The model was estimated using ordinary least squares regression, with annual data from 1975 through 1996. The estimated logarithmic regression equation is: (7) Ln (p) = 1.534 - 0.2418 Ln (K'/U) + 0.2828 Ln (LR) * Dum7985 (20.1) (9.6) (7.8) R2 = 0.857 F-value = 56.829 Standard error of the estimate = 0.0687 Durbin-Watson statistic = 1.921 Numbers shown in parentheses under each coefficient are t-statistics. Over 85 percent of the variation in the logarithm of annual corn prices is explained by estimated equation 7. Transforming the equation to price levels, about 87 percent of the variation of annual corn prices is explained. Each coefficient has the expected sign, with a negative sign for the stocks-to-use variable and a positive sign for the loan rate shift variable. Each coefficient is significant at the 1 percent level. A graph of the regression equation results is shown in figure E-3, adjusting from logarithms to levels of each variable. Corn prices are plotted against ending stocks-to-use ratios. The circles in figure E-3 represent the historical observations for the 1975-1996 estimation period. The lower price curve applies for all years except 1979-1985 and represents the equation that would currently be used for forecasting corn prices. The higher price curve represents the years 1979-1985, which incorporates the average price supporting effect of high loan rates in those years. The average difference between the two price curves for 1979-1985 is about 60 cents a bushel. Model Evaluation Figure E-4 shows a graph of historical corn prices along with the predicted values derived from estimated equation 7. In general, the price model tracks actual corn prices well. Most differences between the model estimate and the actual corn price are less than 15 cents a bushel. The largest difference is in 1988, a year of a major drought in the Corn Belt region of the United States. Table E-1 shows mean absolute errors and mean absolute percentage errors for the full estimation period, 1975-1996, and for a selected subsample of recent years covering 1990-1996. For the full sample, the mean absolute error is about 11 cents a bushel, with a mean absolute percentage error under 5 percent. Importantly, for price forecasting applications, model performance is better in recent years (the 1990s), with a mean absolute error of about 7 cents a bushel and a mean absolute percentage error of 2.8 percent. These statistical measures indicate good performance for the corn price model. Corn Price Forecasts The USDA corn sector projections in March 1998, as published in the World Agricultural Supply and Demand Estimates, imply a 1997/98 ending stocks-to-use ratio of 10.2 percent. Using this stocks-to-use ratio, the corn price model's forecast for 1997/98 is $2.64 per bushel. This price forecast is at the upper end of USDA's March 1998 corn price projection range of $2.45 to $2.65 per bushel. With the standard error of the model estimate of 0.0687 (in logarithms), statistical properties of the model imply that there is a two-thirds probability that the 1997/98 corn price will fall within the confidence interval of $2.47 to $2.83 per bushel for the projected ending stocks-to-use ratio of 10.2 percent. The lower half of this confidence interval corresponds closely with the USDA corn price projections range in March 1998. Table E-2 shows the estimated model's corn price forecasts for different stocks- to-use ratios, as well as corresponding forecast ranges within one standard error of the estimate. Implications of U.S. Agricultural Policy Changes for Potential Annual Price Variability Changes in commodity stocks policies and supply management programs since the mid-1980's have significantly changed U.S. agriculture from the highly managed sector of the early 1980s to a more market-oriented sector today. These policy changes have implications for potential inter-year, annual price variability. A corn sector model that includes the price equation presented in this article was used to compare price variability in different policy settings when the sector faces shocks. Short-run impacts of shocks on prices, when only demand and prices adjust, as well as longer run price impacts over a multi-year, post-shock period when supply also adjusts, were analyzed. Under current policy, carryover stocks of corn and corresponding stocks-to-use ratios are lower than they typically have been over the last 30 years. Thus, price determination is in the steeper parts of the price function estimated in this article and shown in figure E-3. As a consequence, prices in the short run, when only demand and prices adjust, are more responsive to shocks than if stocks were larger and price determination occurred in a flatter portion of the pricing function. In the longer run, supply also adjusts. Importantly, a higher supply response elasticity associated with policy shifts to full planting flexibility allows a larger response to market price movements. In combination, the interaction of increased supply responsiveness with initially greater short-run price impacts can accelerate adjustments to shocks and mitigate longer run, annual price volatility. Conclusions The corn price model presented in this paper uses a stocks-to-use ratio formulation. The model also addresses issues regarding the historical influence of Government commodity loan and storage programs on corn price determination. Loan programs are shown to have had an effect on corn prices in the late 1970's through mid-1980's. However, with farm program changes under 1985 farm legislation, Government commodity loan and storage programs have not had as much influence on prices. Price determination is now based on market supply and demand factors without the influence of the Government price support program. The stocks-to-use ratio used in the model captures these market effects. Statistical model evaluation measures as well as the graph of actual prices and model estimates indicate good performance for the corn price model. This is particularly the case given the large range of corn prices over the sample period used to estimate the model (1975-1996) as well as the changing nature of the influence of Government programs on corn price determination. Changes in U.S. agricultural commodity policies over the last 10-15 years have implications for potential inter-year, annual price variability. With smaller carryover stocks of corn than typically held over the past 30 years, price determination is in the steeper parts of the price function, so prices are initially more responsive to shocks. In the longer run, with full planting flexibility, greater supply responsiveness facing initially greater price impacts can accelerate corn sector adjustments to shocks and mitigate longer run, annual price variability. The relatively simple structure of the estimated reduced form model for corn prices and the model's minimal data requirements lend themselves to easy use in corn price forecasting applications in conjunction with market analysis of supply and demand conditions. In particular, the model is used within USDA as part of the Department's short-term market analysis and long-term baseline projections activities. References Baker, Allen and Keith Menzie. "Drought Effects on Corn Price Forecasts," Feed Situation and Outlook Report, FdS-307, USDA-ERS, August 1988, pp. 25-28. Lin, William, Joseph Glauber, Linwood Hoffman, Keith Collins, and Sam Evans. The Farmer-Owned Reserve Release Mechanism and State Grain Prices. U.S. Department of Agriculture. Economic Research Service. ERS Staff Report AGES850717, August 1985. Lin, William, Peter Riley, and Sam Evans. Feed Grains: Background for 1995 Farm Legislation. AER-714, USDA-ERS, April 1995. U.S. Department of Agriculture. Feed Situation and Outlook Yearbook. Economic Research Service. FDS-1998, April 1998. U.S. Department of Agriculture. Storage Subsidy Programs. Economic Research Service. Staff Report AGES-9075, December 1990. U.S. Department of Agriculture. World Agricultural Supply and Demand Estimates. World Agricultural Outlook Board. WASDE-336, March 1998. Van Meir, Lawrence W. "Relationships Among Ending Stocks, Prices, and Loan Rates for Corn," Feed Outlook and Situation Report, FDS-290, USDA-ERS, August 1983, pp. 9-13. Westcott, Paul C., David B. Hull, and Robert C. Green. "Relationships Between Quarterly Corn Prices and Stocks," Agricultural Economics Research, Vol. 37, No. 1, Winter 1985, pp. 1-7. Young, C. Edwin and Paul C. Westcott. The 1996 U.S. Farm Act Increases Market Orientation, Agricultural Information Bulletin 726, August 1996. END_OF_FILE