SUGAR AND SWEETENERS October 20, 1999 September 1999, ERS-SSS-226 Approved by the World Agricultural Outlook Board --------------------------------------------------------------------------- SUGAR AND SWEETENERS is published three times a year (includes yearbook) by the Economic Research Service, U.S. Department of Agriculture, Washington, DC 20036- 5831. This release contains only the text of the report -- tables and graphics are not included. Subscriptions to the printed version of the report are available from the ERS-NASS order desk. Call, toll-free, 1-800-999-6779 and ask for stock # SUB-SSS-4033, $25/2 issues. ERS-NASS accepts MasterCard and Visa. --------------------------------------------------------------------------- Contents Summary U.S. Sugar Current Year, FY 1999 Prices Forecast, FY 2000 Special Article Stocks-to-Use Ratios and Sugar Pricing Relationships List of Tables Autofax Access to Sugar-Related Data Internet Access to Sugar-Related Data Report Coordinator Stephen Haley (202) 694-5247 FAX (202) 694-5884 E-mail: SHALEY@ECON.AG.GOV Principal Contributors Stephen Haley Nydia Suarez Database Coordinator/Graphics & Table Design Fannye Lockley-Jolly Layout & Text Design Wynnice Napper Approved by the World Agricultural Outlook Board. Summary released September 21, 1999. The next Sugar and Sweetener Situation and Outlook is scheduled for release on December 21, 1999. Summaries and full text of Situation and Outlook reports may be accessed electronically via the ERS web site at www.econ.ag.gov. The Sugar and Sweetener Situation and Outlook is published two times a year and supplemented by a yearbook. To order, call 1-800-999-6779 in the United States or Canada. Other areas please call (703) 605-6220. Or write ERS-NASS, 5285 Port Royal Road, Springfield, VA 22161. Summary U.S. sugar supplies, minus imports under the raw and refined tariff-rate quota (TRQ), are projected for fiscal year (FY) 2000 at 10.9 million short tons, raw value (STRV). Record high production of 8.9 million tons from sugarbeets and sugarcane add to 1.5 million tons of carryin stocks and 500,000 tons of expected imports of non-TRQ sugar. Compared with non-TRQ supplies in FY 1999, the FY 2000 total is 3.1 percent higher, while combined consumption and exports are projected up 1.2 percent. The TRQ for FY 2000 has not been announced. The National Agricultural Statistics Service (NASS) is currently forecasting FY 2000 sugarbeet acreage planted at 1.560 million acres, more than 62,000 acres, or 4.16 percent larger than the previous year. Acreage harvested is forecast at 1.526 million acres, for an increase of slightly less than 74,000 acres, or 5.1 percent above the previous year. NASS forecasts record sugarbeet production at 33.629 million tons, 3.1 percent higher than the previous year's record crop. The national average yield is forecast at 22.0 tons per acre, about half a ton lower than last year's yield. Beet sugar production is currently projected at 4.67 million short tons, raw value (STRV), which, if realized, would be a record. Current crop conditions, along with continuing technological improvements, imply a sugar yield of 3.06 tons per acre. NASS forecasts Florida sugarcane acreage harvested for sugar and seed at 456,000 acres, up 9,000 acres. Assuming about 20,000 of those acres are for seed use and based on the NASS yield forecast of 39.0 tons per acre, sugarcane for sugar production is projected at about 17.0 million tons, slightly less than FY 1999. Florida sugar production is projected at 2.125 million STRV. NASS forecasts FY 2000 Louisiana sugarcane acreage harvested for sugar and seed at 465,000 acres, up 30,000 acres over last year. Sugarcane for sugar is projected to comprise 95 percent of all Louisiana sugarcane acreage, or 442,000 acres, producing a record 14.578 million tons of sugarcane for milling. Trend technological improvements and the currently forecasted sugarcane yield imply a record sugar per acre yield of 3.66 tons, implying sugar production at 1.615 million STRV. Texas cane sugar production is expected to be 100,000 STRV, and Hawaii is projected at 360,000 STRV. Total deliveries for FY 2000 are projected at 10.250 million STRV, a 1.7-percent increase over the previous year. Total deliveries include sugar for the Sugar Containing Products Re-export Program (175,000 STRV), for use in the Polyhydric Alcohol Program (15,000), for feed use (7,000), and for human consumption (10.053 million). High-tier tariff imports from Mexico are projected at 125,000 STRV. USDA revised its projection of sugar syrup imports under HTS 1702.90.4000 from a sugar equivalent of 100,000 to 10,000 STRV after the U.S. Customs Service reclassified the product. Raw sugar imports for the Refined Sugar Re-export Program are projected at 175,000 STRV for FY 2000. Imports for the Sugar Containing Products Re-export Program are also projected at 175,000 STRV, and for the Polyhydric Alcohol Program at 15,000 STRV. Beet sugar production for FY 1999 is estimated at 4.375 million STRV. The 1998 sugarbeet crop was 32.606 million tons. Acreage harvested was 1.452 million acres, and yield averaged 22.5 tons per acre. Sugarbeet yield and production are record highs. Cane sugar production is estimated at 3.954 million STRV, with production records in both Florida and Louisiana. Raw and refined TRQ imports for FY 1999 are estimated at 1.247 million STRV. All three raw sugar TRQ tranche allocations were canceled this year. Other imports are estimated at 560,000 STRV. High-tier tariff imports from Mexico are estimated at 70,000 STRV. Raw sugar imports for the Refined Sugar Re-export Program are estimated at 200,000 STRV, for the Sugar Containing Products Re-export Program at 175,000 STRV, and for the Polyhydric Alcohol Program at 15,000 STRV. Sugar syrup entering under HTS 1702.90.4000 is estimated at 100,000 STRV of sugar. Estimated total supply for FY 1999, at 11.815 million STRV, represents 1.679 million in beginning stocks, 8.329 million from production, and 1.807 million of raw and refined sugar imports. Refined sugar exports are estimated at 225,000 STRV and total deliveries at 10.075 million STRV. Total use is the sum of exports and deliveries, or 10.3 million STRV. Ending stocks, at 1.515 million tons, are 14.7 percent of total use. U.S. Sugar On October 8, 1999, the U.S. Department of Agriculture (USDA) released its revised estimate for fiscal year (FY) 1999 and its revised projection for FY 2000. Current Year, FY 1999 Beet sugar production is estimated at 4.375 million short tons, raw value (STRV). The National Agricultural Statistics Service (NASS) estimates the sugarbeet crop at 32.606 million tons. Acreage harvested is estimated at 1.452 million acres, and yield is estimated at 22.5 tons per acre. Both sugarbeet yield and production represent records. Higher-than-average winter temperatures introduced concerns, especially in the Red River Valley, regarding sugar losses. However, sugar production remained strong from March to June when nearly 1.05 million STRV was produced, nearly 150,000 STRV higher than the previous year. Beet processing in the Red River Valley continued into June as expected. Production from stored thick juice occurred in other areas, especially Idaho, and sugar production from California contributed to the good spring production levels. Although production in July was somewhat lower than expected, pre-pile harvesting and processing of the new beet crop started in September. Because the FY 2000 beet crop is expected to be a record, sugar production in September is expected to bring the FY 1999 production level up to an estimated 4.375 million STRV. Cane sugar production is estimated at 3.954 million STRV. Current-season campaigns have ended in Louisiana, Florida, and Texas for a combined up-to-date total of 3.496 million STRV. The season in Louisiana extended well into January, with a January record just short of 95,000 STRV. The season in Florida extended with over 125,000 STRV being produced in April, also a record for so late in the season. Late-season cane growth helped Texas recover from drier-than-normal conditions during most of the year. The Louisiana production total has been revised upward by 75,000 STRV, the result of an anticipated record cane harvest for FY 2000. Because of the large crop, two mills are scheduled to start the 1999/2000 campaign on September 15. Fourteen other mills are scheduled to start on September 22. If all mills were to produce at 100 percent capacity, over 100,000 STRV of sugar could be produced in September and counted in FY 1999. Factory performance, however, may be less than optimal, and sugar recovery rates are likely to be lower than season totals. Hawaiian production is estimated at 380,000 STRV. With 2 months remaining, FY 1999 production is currently over 339,000 STRV, about 12 percent greater than the previous year. Production is likely to end up being about 30,000 STRV greater than last year. With only 1 mill operating this year, Puerto Rican production was only 3,300 STRV. The industry struggled with the impacts of Hurricane Georges from last year and continues to face major financial difficulties. Raw and refined tariff-rate quota (TRQ) imports for FY 1999 are estimated at 1.252 million STRV. All three raw sugar TRQ tranche allocations were canceled this year because the projected ending stocks-to-use ratios published in the January, March, and May World Agricultural Supply and Demand Estimates (WASDE) reports were greater than the 15.5 percent trigger level. The TRQ shortfall is estimated at 59,298 STRV. Other imports are estimated at 553,000 STRV. High-tier tariff imports from Mexico had been projected in May at 145,000 STRV but were revised downward in August to 70,000 STRV. Margins between U.S. No. 14 Contract raw sugar prices and prices at which Mexican export sugar had been valued declined during the summer, thus making it less profitable for traders to enter the sugar into the U.S. market. Mexican sugar is currently in bonded warehouses awaiting entry and could remain there at least until January 1, 2000, when the North American Free Trade Agreement's (NAFTA) raw sugar high-tier tariff decreases from 13.60 cents to 12.09 cents a pound. Yearly storage costs for sugar in bond are estimated to amount to only about 0.5 cent a pound. Raw sugar imports for the Refined Sugar Re-export Program are estimated at 200,000 STRV. Imports for the Sugar Containing Products Re-export Program are estimated at 174,000 STRV and 12,000 STRV for the Polyhydric Alcohol Program. Sugar syrup entering under HTS 1702.90.4000 is estimated to add an additional 100,000 STRV to FY 1999 imports. Total supply is estimated at 11.813 million STRV. This represents the sum of 1.679 million STRV of beginning stocks, 8.329 million STRV from production, and 1.805 million STRV of raw and refined sugar imports. Refined sugar exports are estimated at 225,000 STRV, and total deliveries are estimated at 10.075 million STRV. Total deliveries through July have been about 2.2 percent higher than the corresponding period for the previous year. Refined beet sugar prices quoted from Milling and Baking News have not decreased below 27 cents a pound, even though No. 14 Contract raw sugar prices have declined to below 21 cents a pound. Total use is the sum of exports and deliveries, or 10.3 million STRV. Ending stocks are estimated to be 1.515 million for an estimated ending stocks-to-use ratio of 14.7 percent. Prices The big news on the New York market has been the continuing decline of the No. 14 Contract domestic raw sugar contract. Within a 6-week period from mid-July to the end of August, the closing price for the nearby contract declined more than 2 cents per pound, from 22.85 cents on July 19 to 20.63 cents on August 27. The monthly average close declined 133 points, from 22.85 cents per pound in July to 21.28 cents per pound in August. After the WASDE was released on September 12, the No. 14 Contract price declined to 19.90 cents pressured by expectations for a record crop. The world market price strengthened during August, with the No. 11 spot rising above 7 cents on the last day of the month for the first time in over 2 months. The average spot price for August was 6.39 cents per pound, with a maximum of 7.05 cents on August 31 and a minimum of 6.17 cents on August 12. Forecast for FY 2000 Beet Sugar Low producer returns on alternative crops have led to greater acreage commitments for the sugar crops. NASS is currently forecasting sugarbeet acreage planted at 1.560 million acres, more than 62,000 acres, or 4.16 percent larger than the previous year. Acreage harvested is forecast at 1.525 million acres, for an increase of 73,000 acres, or 5 percent above the previous year. The largest regional acreage harvested increase is forecast in the Great Plains region: 15.1 percent. Growth in the Great Lakes region is forecast at 8.1 percent; Upper Midwest, 3.1 percent; and Far West, 1.5 percent. The Far West rate is skewed downward because of developments in Washington State where the Moses Lake processing facility had major operational difficulties last year that lowered producer returns. Obtaining acreage commitments proved difficult, but nonetheless, the overall acreage total was higher than expected, but still lower than the previous year by about 9,000 acres. Excluding Washington State, the Far West acreage harvested growth forecast would be 4.4 percent. NASS forecasts sugarbeet production at 34.009 million tons, about 4.3 percent higher than the previous year's record crop. The national yield is forecast at 22.3 tons per acre, about 0.2 a ton lower than last year's yield. For States where it breaks out beet crop conditions on a scale of poor to excellent, NASS indicates crop conditions are better this year in four of the five States (Colorado, Montana, North Dakota, and Wyoming) than the previous 2 years. Only Minnesota is rated worse because of excessive soil moisture. Beet sugar production is currently projected at 4.7 million STRV, which, if realized, would be a record. Current crop conditions, along with historical technological improvements, imply a sugar yield of 3.08 tons per acre. The projected recovery rate is 276.4 pounds raw value per beet ton, about 8 pounds greater than the previous year. Sugar from desugared molasses should be greater than that estimated for FY 1999, which through July 1999 is only 78 percent of a year earlier (FY 1998). Cane Sugar NASS forecasts Florida sugarcane acreage harvested for sugar and seed at 456,000 acres, an increase of 9,000 acres. Assuming about 20,000 of those acres are for seed use, and based on the NASS yield forecast of 39.0 tons per acre, sugarcane for sugar production is projected at about 17.0 million tons, slightly less than last year. Technological improvements and the currently forecasted sugarcane yield imply a sugar per acre yield of 4.87 tons. This level is less than last year's 5.00 tons per acre but, if realized, would be the second highest level ever attained. Florida sugar production is, therefore, projected at 2.125 million STRV. The projected recovery rate is 250 pounds raw value per cane ton. NASS forecasts Louisiana sugarcane acreage harvested for sugar and seed at 465,000 acres, an increase of 30,000 acres over last year. This expansion has occurred in northern and western producing areas and has been induced by low prices of competing crops of soybeans, corn, and rice, as well as cattle. Also, the adaption of the high-yielding variety LCP85-384 has allowed an additional ratoon crop, which reduces fallow acreage and permits more acreage to be allocated to sugarcane for sugar instead of for seed. Louisiana now has more sugarcane acreage than Florida, the first time this has occurred since the 1977 crop year. NASS forecasts Louisiana sugarcane yields of 33.0 tons per acre. Crop growing conditions have been very good throughout all producing areas, unlike last year when lack of adequate moisture was a problem. Sources indicate that the LCP85-384 variety is planted on about 60 percent of all planted acreage, which helps improve overall yields, as continuing adaption of new harvesting equipment optimizes returns from LCP85-384. In particular, the new equipment harvests lodged cane, which could otherwise be a problem with the LCP85-384. Sugarcane for sugar is projected to comprise 95 percent (442,000 acres) of all Louisiana sugarcane acreage. The sugarcane for sugar crop is projected at 14,578 million tons, a record. Technological improvements and the currently forecasted sugarcane yield imply a record sugar yield of 3.66 tons per acre, implying sugar production at 1.615 million STRV. The sugar recovery rate is projected to be 221.6 pounds per cane ton. A significant risk factor, however, is that the harvest season will likely run well into January. A freeze could reduce the size of the crop and the resulting sugar recovery. NASS forecasts Texas sugarcane acreage at 31,200 acres, down 1,400 acres from last year. Yield is forecast at 33.6 tons per acre. If sugarcane for seed comprises the same percentage of total acreage as last year (1.5 percent), the sugarcane for sugar crop would be 1.048 million tons. Sugar production is expected to be 100,000 STRV, which would imply a recovery rate of 193.8 pounds per cane ton, slightly less than last year. Hawaii cane sugar production is projected at 360,000 STRV. Unlike the other cane producing regions, the Hawaii crop year is the calendar year, meaning that the current NASS sugarcane forecasts are relevant for FY 1999 and not for FY 2000. Generally, the previous year's production history is examined quite closely to make projections for the coming year. For FY 2000, the most significant event is the closing of the AMFAC/JMB facility on Maui. The facility produced between 24,000 and 25,000 STRV in 1999. With a current-year projection of 380,000 STRV, loss of the plant and its acreage would imply production in FY 2000 of about 355,000 STRV. However, the other sugar-producing operation on Maui is making serious investments in increasing yields, containing costs, and improving the return from sugar byproducts. It is expected that increased production will partially offset the decline implied by the other plant's closure; therefore, the projection is adjusted upward to 360,000 STRV. Puerto Rican sugar production is currently projected at 15,000 STRV. There is considerable uncertainty regarding this projection, however. An optimistic scenario would be for the operation of the two remaining mills (one mill did not operate in 1999 because of hurricane damage) and a return of sugarcane acreage to 1998 levels. Additionally, the mills are dependent on precarious external financing that only serves to increase the uncertainty of the mills' continued operations. Deliveries and Exports Total deliveries are projected at 10.250 million STRV, a 1.7-percent increase over the previous year. Total deliveries include sugar included in products for the Sugar Containing Products Re-export Program (175,000 STRV), deliveries for use in the Polyhydric Alcohol Program (15,000 STRV), and for feed use (7,000 STRV). Deliveries for human consumption are calculated as a residual at 10.053 million STRV. Additionally, exports for the Refined Sugar Re-export Program are projected at 175,000 STRV. Projected total use for FY 2000 is 10.425 million STRV, a growth of 125,000 STRV over FY 1999. Non-TRQ Imports High-tier tariff imports from Mexico are projected at 125,000 STRV. This represents known shipment commitments and sugar currently in bonded warehouses awaiting possible entry after January 1, 2000, when the NAFTA raw sugar high-tier tariff decreases from 13.60 cents to 12.09 cents a pound. It is not expected that margins between U.S. No. 14 Contract raw sugar prices and prices at which Mexican export sugar is valued will widen sufficiently to justify additional high-tier tariff imports from Mexico. Additionally, sources indicate that Mexico may hold back the issuance of certificates of origin for sugar destined for the U.S. market unless there is a sufficient price premium that would make the shipment profitable to Mexican sugar interests. There has been some concern expressed by Mexico that high-tier tariff exports benefit traders and are unnecessarily unsettling to the U.S. sugar industry. On September 8, 1999, the U.S. Customs Service reclassified a product that had been entering into the United States under a classification of a sugar syrup containing sugar solids, water, and more than 6 percent of soluble non-sugar solids under HTS subheading 1702.90.4000. Imports under this subheading are not subject to the sugar TRQ and can enter at a low duty with no quantitative limitation. In its ruling, the Customs Service noted that molasses with its impurities is added to sugar in sufficient quantities to assure that the syrup exceeds the 6 percent limitation so that the syrup is not subject to the sugar TRQ. The molasses is removed after importation and the sugar is used in the same way as sugar subject to the TRQ. There appears to be no commercial use for the syrup in the state in which it is imported. USDA forecasts that these sugar syrup imports will add an equivalent of 100,000 STRV of sugar into the United States market during FY 1999. The ruling takes effect 60 days after the publication date in early November. Accordingly, USDA has revised its projection of sugar syrup imports from a sugar equivalent of 100,000 STRV to 10,000 STRV. This amount is projected to enter only during October and the very first part of November. Raw sugar imports for the Refined Sugar Re-export Program are projected at 175,000 STRV for FY 2000. Imports for the Sugar Containing Products Re-export Program are projected at 175,000 STRV and 15,000 STRV for the Polyhydric Alcohol Program. In all, FY 2000 non-TRQ imports are projected at 500,000 STRV. TRQ Imports On July 22, 1999, USDA issued press release No. 0301.99 soliciting comments regarding the FY 2000 administration of the sugar TRQs (see Box). The administrative approach in place since FY 1997 has used supply and utilization projections published in the September WASDE to determine the level of the sugar TRQ for the fiscal year set to begin in the following October. The TRQ level was set so that the projected ending stocks-to-use ratio would match a target level of about 14.5 percent. The raw and refined sugar TRQs, however, would have to be at least equal to, or greater than, a minimum level of 1.256 million STRV set under terms of the Uruguay Round General Agreement on Tariffs and Trade (GATT). The established TRQ level was included in the October WASDE, although only about 70 percent of the TRQ was typically made available to the U.S. Trade Representative (USTR) for immediate allocation in September. The remainder was divided into three equally-sized tranches, each of which was allocated in January, March, and May if the projected ending stocks-to-use ratio published in the WASDE for the corresponding months were 15.5 percent or less; otherwise, the tranche for that month was canceled. FY 2000 projections complicated implementing the current administrative approach. In order to achieve an ending stocks-to-use ratio of 14.5 percent, ending stocks would have to equal 1.512 million STRV (total use of 10.425 million STRV multiplied by 0.145). Total supply excluding the TRQ is projected at 10.9 million STRV (the sum of projected beginning stocks, production, and non-TRQ imports), which is 450,000 STRV above projected total use. The difference between target ending stocks of 1.512 million STRV and 450,000 STRV would be the level of the sugar TRQ at 1.062 million STRV. This level is below the GATT minimum of 1.256 million STRV. If the TRQ level were increased to be at or modestly above the GATT-minimum, the USDA would lose the flexibility it had achieved with the tranche allocation system, making it more difficult to handle a situation where more sugar might be needed due to unforeseen production disruptions. Another complication is that the Federal Agriculture Improvement and Reform Act of 1996 requires that processor loans made under the U.S. Sugar Program during any fiscal year be recourse if the fiscal year raw sugar TRQ is established at or below 1.5 million STRV. Given these difficulties, the USDA issued its press release describing two alternatives to the present system and solicited comments from interested parties. As of early October, the raw sugar TRQ had not been announced. The refined sugar TRQ for FY 2000 was announced on October 1, 1999. It was set at 66,139 STRV. The United States Trade Representative (USTR) allocated 27,558 STRV of the refined sugar TRQ to Mexico to fulfill obligations under the NAFTA. A total of 11,354 STRV was allocated to Canada pursuant to an agreement reached with Canada. Separate from NAFTA, an additional 3,256 STRV of refined sugar were allocated to Mexico. The remainder of the refined sugar TRQ is available on a first-come, first-served basis. This amount includes 16,155 STRV reserved for specialty sugars. Box - Proposed Alternatives to TRQ Administration On July 22, USDA Secretary Dan Glickman invited public comments on the administration of the sugar TRQs in FY 2000, which begins October 1, 1999. Forty respondents sent in their comments regarding the two options being considered by USDA to preserve the nonrecourse provision of the sugar loan program. Over half of the respondents were from the four cane growing States, including 15 from Louisiana. Five responses were from TRQ countries, four from industrial users, and three each from refiners and the beet sector. USDA proposed two options. Option No.1 -- Sufficiently increase the stocks-to-use trigger (currently 15.5 percent) in half-point increments until USDA can establish a raw sugar TRQ greater than 1.5 million short tons, raw value (STRV). Once the raw sugar TRQ level is established, USDA would make available to the U.S. Trade Representative's (USTR) office for allocation the minimum quantity required by current trade agreements. The difference between the established TRQ and the quantity allocated by the USTR would be put in reserve and allocated using the newly established stocks-to-use trigger in equal quantities in March and May of fiscal year 2000. Option No. 2 -- Establish the size of the raw sugar TRQ at a level 496,035 tons greater than needed to achieve the current stocks-to-use ratio target consistent with current practice. An initial amount would be allocated sufficient to achieve the stocks to-use-target, but not less than required by current trade agreements. The monthly WASDE estimates would record only the allocated TRQ rather than the entire amount as is currently done. USDA would use the current reserve system, with 165,347-ton quantities to be made available for allocation or cancellation in January, March, and May 2000. USDA would add the relevant quantity for possible allocation with the WASDE import forecast, once the WASDE report is released in each of those months. When the resulting stocks-to-use ratio (after adding the 165,347 tons to the sugar supply in WASDE) is 15.5 percent or less, the 165,347 tons would be allocated. When the resulting stocks-to-use ratio is above 15.5 percent, the tranche would be canceled. USDA is also proposing increasing the TRQ for specialty refined sugar by 10,000 metric tons to 14,656 metric tons, raw value, primarily to accommodate increasing demand for organic sugar. The total TRQ for refined sugar in fiscal 2000 would rise accordingly to 60,000 metric tons, raw value. Special Article Stocks-to-Use Ratios and Sugar Pricing Relationships: Implications for U.S. Sugar Policy Stephen L. Haley 1/ ---------- 1/ Agricultural economist, Specialty Crops Branch, Market and Trade Economics Division, Economic Research Service. --------- Abstract: The General Accounting Office (GAO) has argued that the U.S. Department of Agriculture (USDA) has managed the raw sugar tariff-rate quota (TRQ) program such that raw and refined sugar prices are higher than necessary to prevent forfeiture of sugar to the USDA's Credit Commodity Corporation. Because since FY 1997 the USDA has incorporated projected ending stocks-to-use ratios in determining the size of the raw sugar TRQ, the GAO argues that statistical relationships between actual ending stocks-to-use ratios and sugar prices are useful for evaluating the USDA's performance in preventing sugar forfeitures. This article analyzes raw and refined sugar pricing relationships with stocks-to-use ratios more closely. It shows that refined beet sugar pricing is much less predictable than raw sugar pricing. Because USDA's primary policy instrument is the control of the raw sugar TRQ that does not directly affect the pricing of refined sugar, achieving a refined sugar price high enough to avoid beet sugar forfeitures is difficult. The article concludes that a credible hypothesis is that "higher-than-necessary" raw sugar prices to prevent forfeiture may be a reasonable policy goal given statistically less reliable relationships between sugar stocks, end uses, and refined sugar prices. Keywords: sugar, stocks-to-use, U.S. Sugar Policy, sugar prices. The basis of the raw sugar tariff-rate quota (TRQ) administrative plan established in fiscal year (FY) 1997 is to provide adequate supplies of sugar at reasonable prices by targeting sugar stocks-to-use levels and not sugar prices. Recently, the General Accounting Office (GAO, 1999) has argued that the U.S. Department of Agriculture (USDA) should interpret reasonable prices as those just high enough to prevent forfeiture of sugar to the Credit Commodity Corporation (CCC). They argue that under this criterion, the USDA could reasonably increase the size of the raw sugar TRQ to a level high enough to cause the price of raw sugar to decrease by 2 cents a pound without the risk of sugar forfeitures. This article examines the GAO argument in greater detail. Specifically, it analyzes the relationship between the ending stocks-to-use ratio and the sugar price series important to USDA in assessing the likelihood of sugar loan forfeitures. It reviews historical raw and refined sugar price patterns and presents statistical relationships between stocks-to-use ratios and sugar prices through the marketing year. The U.S. Sugar Program and TRQ Administration Governing provisions of the U.S. Sugar Program are contained in Section 156 of the Federal Agriculture Improvement and Reform Act of 1996 ("1996 Farm Act"). Under the 1996 Farm Act, the U.S. Sugar Program provides for USDA to make loans available to processors of domestically grown sugarcane at a rate of 18 cents per pound and to processors of domestically grown sugarbeets at a rate of 22.9 cents per pound for refined beet sugar. Sugar loans are issued as nonrecourse loans as long as the raw sugar TRQ is set higher than 1.5 million short tons, raw value (STRV). The nonrecourse aspect means that when the loan matures, the USDA must accept sugar pledged as collateral as payment in full in lieu of cash at the discretion of the processor. Nonrecourse loans effectively support the price of sugar in the market. Under provisions of the 1985 Farm Act, as amended, the USDA was effectively mandated to operate the Sugar Program established under Section 206 of the Agricultural Act of 1949 at no cost to the Federal Government. The "no-cost provision" meant that the CCC was barred from accumulating sugar acquired under loan rate operations. The means of accomplishing the mandate were to adjust sugar import quotas or use domestic marketing quotas. Because section 206 of the 1949 Act was repealed by section 171 of the 1996 Farm Act, the "no-cost provision" is no longer part of the sugar program. The implication for TRQ management is that there is no longer any legal requirement that sugar prices must be high enough to prevent sugar forfeitures to the CCC. The USDA authority to set the level of the raw and refined TRQs is contained in "Additional U.S. Note 5 of the Harmonized Tariff Schedule of the United States" and is not part of the U.S. Sugar Program. The basis of the TRQ administrative plan established in fiscal year (FY) 1997 is to provide adequate supplies of sugar at reasonable prices by targeting sugar stocks-to-use levels and not sugar prices. For FY 1997 through 1999, the initial targeted stocks-to-use levels have been about 14.5 percent. To accommodate supply and demand changes during the year, a portion of the TRQ is held in reserve and allowed into the market at specific intervals if the projected ending stocks-to-use ratio at those intervals is equal to or less than 15.5 percent. Although formal linkages between the Sugar Program and TRQ management have been reduced due to the elimination of the "no-cost provision," some observers, including the GAO, still expect the raw sugar TRQ to be set so that raw and refined prices are just high enough to forestall forfeiture to CCC. The GAO cites research that shows correlations between the ending stocks-to-use ratio and the fourth quarter FY No.14 raw sugar price. Their argument is that initially targeting an ending stocks-to-use ratio of 14.5 percent means an expected fourth-quarter No. 14 raw sugar price of about 22.6 cents a pound. They conclude that this level is over 2 cents per pound higher than the minimum raw sugar price needed to prevent forfeiture of sugar to the CCC. Sugar Loan Rates and Minimum Prices To Avoid Forfeiture The USDA adjusts national raw cane and refined beet sugar loan rates to levels for regional cane and beet producing areas. The resulting regional loan rates reflect regional freight costs between the production regions and their respective normal destinations. The goal is to equalize the probability of forfeiture across regions. The 1996 Farm Act required that processors who forfeit sugar pledged as collateral for a nonrecourse loan face a penalty of 1 cent a pound for raw cane sugar and 1.07 cents a pound for refined beet sugar. Processors would have to consider these penalties when deciding whether to forfeit sugar to the CCC. Loans are taken for a maximum term of 9 months and are repaid along with interest charges before September 30. In order to forestall forfeiture, the resulting sugar price must be high enough to cover the interest expenses. Cane processors share interest expenses with their growers, but beet processors do not and must therefore recover the entire interest expense of loan repayment in their share of the sugar's selling price. Cane processors incur transportation and distribution costs in moving sugar to the refiner and also face location discounts required by some refiners. These additional costs must be included in the minimum price to avoid forfeiture calculation. Because beet sugar is refined sugar requiring no further processing, the minimum price does not include transport adjustments. However, because beet sugar is normally sold subject to a 2-percent cash discount, this amount must be added to arrive at the minimum price. Given these adjustments, the minimum price to avoid forfeiture is some amount greater than the national loan rate. These amounts vary by region, both across and within cane and beet producing areas, and by time period. In FY 1998, the average amount above the national loan rate for beet sugar was 2.21 cents a pound. For domestic cane sugar, the amount was 1.96 cents a pound. For demonstration purposes in the analysis below, it is assumed that the minimum price to avoid forfeiture is 2 cents a pound higher than the national loan rate. This adjustment may be low if one applies the now-outdated "no-cost provision" to the adjustment. In that case, one would be forced to focus on the region where the upward adjustment is the greatest and apply it as the standard in all regions. Raw and Refined Sugar Pricing Relationships The fiscal year ending stock-to-use ratio can be expected to explain much of the variation in sugar prices throughout the year because the ending stock-to-use ratio represents the availability of sugar throughout the annual marketing period. Sugar is supplied to the domestic market on an annual basis by government policy and the nature of sugar production. The quantity of sugar permitted to be imported at the low tariff rate is established on a federal fiscal year basis. Most domestic sugarcane and sugarbeets are harvested during a limited period of time although the sugar is sold throughout the year. The start of the new processing year, or campaign, for most domestic sugarcane and sugarbeet regions is around the start of the federal fiscal year. Domestic processors prefer to have minimum carryover stocks on hand at the start of the new campaign because sugar can lose quality over time, storage is costly, and growers are not completely paid for their sugarcane or sugarbeets until the processor sells all the sugar made during the campaign. Figure1 shows quarterly No. 14 raw sugar prices and the national raw sugar loan rate plus 2 cents (minimum raw sugar price) since 1990 2/. ----- 2/ It is important to keep in mind that processors with sugar under loan respond to local prices rather than national prices when considering forfeiture. ----- The prices have remained above the minimum in all quarters by at least 1.1 cents. The average price for the period has been 22.14 cents a pound, 2.14 cents above the minimum. The difference between the series' maximum and minimum prices has been 2.51 cents, or 11 percent relative to the average price for the period. The coefficient of variation (i.e., the standard deviation divided by the average) has been a low 0.03. Figure 2 shows the refined beet sugar price from Milling and Baking News and the national refined beet sugar loan rate plus 2 cents (minimum refined beet sugar price). In contrast to raw sugar prices, refined prices have fallen below the minimum in 5 out of the 37 quarters (fourth quarter 1991; fourth quarter 1992; first and second quarters 1993; and second quarter 1994). The average refined price has been 26.63 cents a pound, only 1.73 cents above the minimum. The difference between the series' maximum and minimum prices has been 7.42 cents, or 28 percent relative to the average price for the period. The coefficient of variation has been 0.07. Both the relative maximum-minimum difference and coefficient of variation are about 2.5 times greater than the corresponding levels for raw sugar. Tables 1 and 2 show the results from estimating a model that regresses the average quarter price of raw sugar (table 1) and of refined beet sugar (table 2) on quarter-ending (or interim) stocks-to-use ratios and the fourth quarter FY ending stocks-to-use ratio. The expectation is that the fourth quarter price would be highly correlated with the ending stocks-to-use ratio. If the program is effective, a targeted ending stocks-to-use ratio should have a strong positive effect on prices in the first, second, and third quarters as well. Interim stocks-to-use ratios are hypothesized to be important in price determination, but they should not obscure the contribution of the ending ratio. The tables contain estimated coefficients and measures that reveal the following: the statistical significance of the coefficients (T-statistics; the higher, the better), the amount of price variation explained by the model (the adjusted R2; the closer to 1, the better), and the amount of serial correlation (the Durbin-Watson statistic; the closer to 2.00, the better). An additional item, called the beta coefficient, is reported as well. The beta coefficient transforms the estimated coefficient by multiplying it by the ratio of the standard deviation of the explanatory variable (i.e., the corresponding stocks-to-use ratio) to the standard deviation of the price for the quarter. The interpretation is that a one-standard deviation change in the explanatory variable causes the price standard deviation to change on average by the amount of the beta coefficient. This transformation allows one to directly compare the effects of the interim and ending stocks-to-use ratios on sugar prices. The raw sugar price equations from table 1 show good results. The adjusted-R2s are all higher than 0.80, implying that a high proportion of the price variability can be explained by the corresponding stocks-to-use ratios. The high T-statistics (i.e., anything greater than 1.960) indicate the coefficients differ significantly from zero, and the coefficients are of the correct sign (i.e., negative), indicating the anticipated inverse relationship between price and the stocks-to-use ratios. In the second and third quarters, the beta coefficient associated with the ending stocks-to-use ratio exceeds the corresponding interim stocks-to-use ratios by over 2.4 times, implying that program operations are successful in maintaining high raw sugar prices in the second and third quarters, as well as the fourth quarter. The ending period beta coefficient in the first quarter equation is only about 80 percent of the first-quarter beta coefficient, indicating less effect but still of high magnitude. Although the coefficients from the refined beet sugar equations of table 2 have the anticipated signs and mostly differ significantly from zero, the adjusted-R2s are all lower than the corresponding raw sugar equations. The fourth-quarter equation in particular has a low adjusted-R2 of only 0.263, which compares with 0.839 for the fourth-quarter raw sugar equation. Serial correlation does not seem to pose a problem for any of the equations. The interim stocks-to-use beta coefficients in the first and second quarter equations are about of the same magnitude with the final quarter beta coefficient. In the third quarter equation, the final quarter coefficient cannot be significantly distinguished from zero (T-statistic less than 1.96) and the value of the beta coefficient is less than half the size of the third quarter beta. These beta results imply that U.S. Sugar Program operations may have been less successful in maintaining high beet sugar prices through the year than they have been for raw cane sugar prices. Figures 3 and 4 show fourth-quarter price forecasts for raw and refined beet sugar based on the fourth quarter equations in tables 1 and 2. Also shown are the minimum forfeiture prices and one-downward standard deviations from the forecasts (i.e., the forecast less the standard error of the forecast equation). The raw sugar price forecast and its downward displacement in figure 3 are both well above the minimum forfeiture price line, indicating little chance of forfeiture risk. The refined beet sugar forecast in figure 4 is mostly above the minimum forfeiture price, but its standard-error downward displacement is below the minimum price in several years, indicating that forfeiture risk could be a significant possibility. The relatively poor forecasting performance of the fourth quarter beet price equation makes the distance between the forecast line and the standard-error displacement relatively wide. Alternative Specifications of Refined Beet Sugar Forecast Equations The poor performance of the fourth quarter refined beet sugar price equation suggests that a more complicated specification might provide better forecasting results. Table 3 shows various alternatives. The last column shows a statistic called the Schwarz criterion. It is a measure of 1-step ahead out-of-sample prediction error variance: a smaller value associated with an equation indicates that the equation has better forecasting ability relative to those equations with higher values. The specifications are ordered with respect to declining Schwarz criterion values. The first equation is the specification from table 3, which means that its forecasting performance is the worst of those shown. Forecasting performance is enhanced when the third quarter refined beet sugar price is included in the equation. This result is not surprising given that the beet sugar price is a lower end estimate of Midwest beet sugar transaction prices reported by the trade journal Milling and Baking News, i.e., the series is noted by short term persistence of particular values, sometimes not changing for a number of weeks. In specification No. 2, the inclusion of the lagged beet price greatly diminishes the significance of the fourth quarter stocks-to-use ratio as an explanatory variable. In specification No. 3, the fourth quarter raw sugar price is added as an explanatory variable. Because it is highly correlated with the fourth quarter stocks-to-use ratio, the significance of the coefficients on the raw price and ending stocks-to-use ratio variables may not be reliable, but there is some improvement in forecasting as shown by a lower Schwarz criterion relative to specification No. 2. Moreover, dropping the ending stocks-to-use variable altogether (specification No. 5) improves the forecasting ability as evidenced by an even lower Schwarz criterion of 3.199, compared with the 3.364 of specification No. 3. Specification No. 4 includes an ending beet sugar stocks-to-use ratio. It has the correct negative sign, indicating an inverse relation with the beet sugar price, but the significance level is low (as evidenced by a T-statistic well below 1.960). The best performing forecast equation, specification No. 6, includes the combined raw and refined cane sugar ending stocks-to-use ratio. The coefficient is significantly positive, indicating that high cane sugar stocks for a given raw sugar price translates into a higher forecasted beet sugar price. The usefulness of specification No. 6 for policy is not clear. Although the equation's standard error is much smaller than specification No. 1 (0.639 compared with 1.757), only the raw sugar price is affected through a policy that targets an ending stocks-to-use ratio. Figure 5 shows fitted values of the equation, along with the downward one-standard deviation displacement, using raw sugar price forecasts from table 1's fourth-quarter raw sugar price equation. Actual cane sugar stocks-to-use ratios and third-quarter beet sugar prices are used. As modeled, the beet sugar price forecast drops below the minimum beet forfeiture price in 1994. The displacement line is below the minimum in 1994, 1995, and in 1998. Efficacy of Using the Refined Beet Sugar Spot Price Unlike the nearby No.14 New York Contract price for raw sugar, the beet sugar price used in comparison with a minimum beet sugar price is not a quoted futures price. It is rather a spot price that is published weekly in the Milling and Baking News. It is based on industry contacts established and maintained by the editors and correspondents of the publication. A problem is that much of the commercial activity between beet sugar purchasers and sellers is on the basis of contracting that is not necessarily reflected in spot pricing. It is not implausible to hypothesize that the beet sugar spot price may not be accurate as a measure of returns to beet sugar production and processing. One way to examine both refined and raw sugar prices is to compare them with producer price indices (PPI) maintained by the Bureau of Labor Statistics (BLS) for refined beet sugar and for raw cane sugar. The BLS reports these PPIs monthly based on data collected from the raw cane and refined beet sugar industries. The indices are based on actual sales transactions. If the beet sugar spot price and the No.14 Contract raw sugar price are truly reflective of market activity and unit returns, they should be highly correlated with the respective PPIs. Table 4 shows results of estimating a model that links the PPI with the price series. For each quarter, the logarithmic difference of the PPIs of the current and previous period are regressed on a constant, and on the same and previous period logarithmic differences of the corresponding, similarly defined price series. If the price series are truly reflective of a large proportion of real market activity, the adjusted-R2 should be close to one, and the same-period coefficients or the sum of the same and previous period coefficients should be close to one. Table 4 shows a much closer correlation between the raw sugar PPI and No.14 Contract raw price than between the beet sugar PPI and the refined beet sugar spot price. The raw sugar adjusted-R2s are higher: all over 0.830, whereas the beet sugar adjusted-R2s are below 0.75, and the second quarter adjusted-R2 is only 0.292. The raw sugar coefficients, either the same quarter "B" coefficients or the sum of the same and lagged quarters, i.e. "B+C," are all closer to one than the coefficients of the corresponding beet sugar equations. The raw sugar T-statistics are all larger as well. The fourth quarter beet sugar equation shows an insignificant relationship between the PPI and the same quarter refined beet sugar price. The fourth quarter PPI is significantly affected by third-quarter refined beet sugar spot prices, possibly reflecting the impact of contracting. All in all, estimation results suggest that the refined beet sugar spot price does not as accurately reflect market activity as does the No. 14 Contract raw sugar price. Although these results do not invalidate the use of beet sugar spot prices in policy contexts, they do suggest caution in interpreting beet sugar returns, especially in the fourth quarter when loans typically become due. Conclusion The price for beet sugar is imperfectly correlated with the raw sugar price. Econometric analysis shows no strong direct relationships between ending stocks-to-use ratios and the refined beet sugar prices. The best predictor of future quarter beet sugar pricing is typically the previous quarter's price. Other factors have differing roles depending on the quarter being analyzed. The price series itself is less substantive than the No. 14 New York Contract series for raw sugar. Econometric evidence shows that there is a weaker relationship between the refined beet sugar spot price and the beet sugar producer price index published by the Bureau of Labor Statistics than between the No. 14 Contract raw sugar price and the BLS raw sugar PPI. A credible hypothesis is that "higher-than-necessary" raw sugar prices to prevent forfeiture may be a reasonable policy goal given the statistically less reliable relationships between sugar stocks, end uses, and refined sugar prices. References United States General Accounting Office. Sugar Program: Changing the Method for Setting Import Quotas Could Reduce Cost to Users. GAO/RCED-99-209, Washington DC, July 1999. List of Tables World and U.S. Sugar and Corn Sweetener Prices 1. World refined sugar price, monthly, quarterly, and by calendar and fiscal year 2. World raw sugar price, monthly, quarterly, and by calendar and fiscal year 3. U.S. raw sugar price, duty fee paid, New York, monthly, quarterly, and by calendar and fiscal year 4. U.S. wholesale refined beet sugar price, Midwest markets, monthly, quarterly, and by calendar and fiscal year 5. U.S. retail refined sugar price, monthly, quarterly, and by calendar and fiscal year 6. U.S. wholesale list price for glucose syrup, Midwest markets, monthly, quarterly, and by calendar and fiscal year 7. U.S. wholesale list price for dextrose, Midwest markets, monthly, quarterly, and by calendar and fiscal year 8. U.S. spot price for HFCS-42, Midwest markets, monthly, quarterly, and by calendar and fiscal year 9. U.S. producer price index for HFCS and sugar, monthly 10. U.S. consumer price index for sugar and selected sweetener-containing products U.S. Sugar Supply and Use 11. U.S. sugarbeet crops: Area planted, acres harvested, yield per acre, and production, by State and region 12. U.S. sugarcane: Area, yield, production, output, recovery rate, and sugar yield per acre, crop years 13. U.S. beet and cane sugar production (including Puerto Rico), fiscal year and share of total 14. U.S. sugarbeet area, yield, and production, 1985-1999 15. U.S. production of beet sugar and cane sugar by State, monthly, quarterly, fiscal, calendar, and crop year 16. U.S. cane and beet sugar deliveries, monthly, quarterly, and by fiscal and calendar year 17. U.S. sugar deliveries for human consumption by type of user, quarterly and calendar year 18. U.S. sugar imports under tariff-rate quota (TRQ), by country, fiscal years 1996-99 19. U.S. sugar stocks held by primary distributors, by quarters 20. U.S. sugar (including Puerto Rico) supply and use, fiscal years 1991-2000 21. Monthly estimates of fiscal 1999 U.S. sugar supply and use 22. Monthly estimates of fiscal 2000 U.S. sugar supply and use Corn Sweetener Supply, Use, and Trade 23. U.S. wet-milled use of field corn, crop year 24. U.S. high fructose corn syrup (HFCS) deliveries (including Puerto Rico), quarterly, fiscal, and calendar year 25. U.S. high fructose corn syrup (HFCS) production (including Puerto Rico), quarterly, fiscal, and calendar year 26. U.S. high fructose corn syrup (HFCS) supply and use, calendar year 27. Net cost of corn starch to U.S. wet-millers, Midwest markets 28. U.S. corn sweetener exports to Mexico and Canada, fiscal years 1993-99 29. U.S. corn sweetener imports from Mexico and Canada, fiscal years 1993-99 30. U.S. HFCS trade with Mexico and Canada, monthly 1989-99 U.S. Imports of Sugar Syrups 31. U.S. total imports of sugar syrup, harmonized tariff code 1702.90.4000, monthly 1993-99 Special Article Tables A1. Sensitivity of FY quarterly No. 14 raw sugar prices to changes in interim and ending stocks-to-use ratios. A2. Sensitivity of FY quarterly refined beet sugar prices to changes in interim and ending stocks-to-use ratios. A3. Alternative specifications for estimating fourth quarter, FY, beet sugar prices A4. Relationships between sugar producer price indices and sugar prices used by USDA List of Figures Figure 1. U.S. Raw Sugar Prices Figure 2. U.S. Sugarbeet Acreage by Region Figure 3. U.S. Beet and Cane Sugar Production Figure 4. U.S. Wholesale Refined Beet Sugar Prices Figure 5. U.S. TRQ Sugar Imports Figure 6. U.S. Sugar Consumption Special Article Figures Figure A-1. Raw Sugar Price and Minimum Raw Sugar Price To Avoid Forfeiture Figure A-2. Refined Beet Sugar Price and Minimum Beet Sugar Price To Avoid Forfeiture Figure A-3. U.S. Raw Sugar Price Forecast, and Minimum Raw Sugar Price To Avoid Forfeiture, Fourth Quarter, FY Figure A-4. U.S. Beet Sugar Price Forecast, and Minimum Beet Price To Avoid Forfeiture, Fourth Quarter, FY Figure A-5. Alternative U.S. Beet Sugar Price Forecast, and Minimum Beet Price To Avoid Forfeiture, Fourth Quarter, FY Autofax Access to Sugar-Related Data (ERS) From your fax machine, call (202) 694-5700 and listen to voice prompts to have the following documents automatically downloaded to your fax machine. You may request up to three documents in one phone call. Directory Identification Numbers and Titles: Sugar and Sweeteners: 12626 World Agricultural Supply and Demand Estimates for Sugar 12627 3 pages (1) Wholesale refined beet sugar prices, Midwest market, 1960-present, monthly, quarterly, annual, fiscal (2) U.S. raw sugar price, duty-fee paid, New York, 1960-present, monthly, quarterly, annual, fiscal (3) U.S. retail refined sugar prices, 1975-present, monthly, quarterly, annual 12628 5 pages (1) Wholesale list prices for HFCS-42, Midwest market, 1975-1995, monthly, quarterly, annual, fiscal (2) Wholesale list prices for HFCS-55, Midwest market, 1981-1995, monthly, quarterly, annual, fiscal (3) Wholesale list prices for glucose corn syrup, Midwest market, 1975-present, monthly, quarterly, annual, fiscal (4) Wholesale list prices for dextrose, Midwest market, 1975-present, monthly, quarterly, annual, fiscal (5) U.S. producer price index for HFCS and sugar, monthly, annual, 1986-present 12629 2 pages (1) World raw sugar prices, 1960-present, monthly, quarterly, annual, fiscal (2) World refined sugar prices, 1980-present, monthly, quarterly, annual, fiscal 12630 2 pages (1) U.S. and world sugar prices, 1990-present, monthly, quarterly, annual, fiscal 12631 3 pages (1) Net cost of corn starch to U.S. wet-millers, Midwest markets, 1985-present 12632 3 pages (1) U.S. sugar (including Puerto Rico) supply and use, fiscal years, 1980/81-present 12633 10 pages (1) The Beet Sugar Industry of Minnesota and North Dakota: Current Situation and Prospects, by Ron Lord, September 1994 Sugar and Sweetener Situation and Outlook Report 12634 Text Boxes, Graphs, and Tables, September 1997 S&O Report, 10 pages "HFCS Trade Dispute With Mexico," by Jacqueline Salsgiver; "Origin of the U.S. Sugar Import Tariff-Rate Quota Shares," by Nydia Suarez; and "Changing Structure of the U.S. Refined Sugar Market," by Ron Lord and Robert Barry, September 1997 Sugar and Sweetener Situation and Outlook Report For more information on specially crops, contact the following subject matter specialists: Sugar & Sweeteners: Fannye Jolly, (202) 694-5249 Vegetables: Gary Lucier (202) 694-5253 Tobacco: Tom Capehart (202) 694-5311 Fruit & Tree Nuts: Susan Pollack (202) 694-5251 & Agnes Perez (202) 694-5255 Industrial Uses and Alternative Agriculture: Lewrene Glaser (202) 694-5246 Internet Access to Sugar-Related Data Home Pages Main Data Directory: http://usda.mannlib.cornell.edu/cgi-usda/agency.cgi?ers U.S. Department of Agriculture (USDA): http://www.usda.gov Economic Research Service (ERS): http://www.econ.ag.gov Office of the Chief Economist, USDA: http://www.usda.gov/agency/oce/ World Agricultural Outlook Board (WAOB): http://www.usda.gov/agency/oce/waob/waob.htm National Agricultural Statistics Service (NASS): http://www.usda.gov/nass/ Foreign Agricultural Service (FAS): http://www.fas.usda.gov/ Reports ERS Sugar & Sweetener Situation and Outlook Reports (including text of reports): http://usda.mannlib.cornell.edu/reports/erssor/specialty/sss-bb ERS Sugar Yearbook Data (May 1999): http://usda.mannlib.cornell.edu/data-sets/specialty/89019/ Sugar Statistical Compendium (1991): http://usda.mannlib.cornell.edu/data-sets/specialty/91006/ U.S. Corn Sweetener Statistical Compendium (1993): http://usda.mannlib.cornell.edu/data-sets/specialty/94002/ Farm Sector Cost of Production: Sugarbeets: Analysis, 1996-97: http://www.econ.ag.gov/briefing/fbe/car/beets2.htm Data, 1981-97: http://www.econ.ag.gov/briefing/fbe/car/beets3.htm Sugarcane: Analysis, 1995-96: http://www.econ.ag.gov/briefing/fbe/car/cane2.htm Data, 1981-96: http://www.econ.ag.gov/briefing/fbe/car/cane3.htm World Agriculture Supply and Demand Estimates Report (WASDE): http://www.usda.gov/oce/waob/wasde/wasde.htm February 1999 Outlook Forum: http://www.usda.gov/agency/oce/waob/outlook99/99speeches.htm Sugar: U.S. Sugar Re-export Programs, Foreign Agricultural Service (FAS): http://www.fas.usda.gov/htp/sugar/sugarpg.html Foreign Agricultural Service Report from Foreign Countries (includes sugar reports): http://www.fas.usda.gov/scriptsw/AttacheRep/attache_frm.idc Sweetener Market Data, Farm Service Agency (FSA): http://www.fsa.usda.gov/ao/epas/dsa/sugar/coversu.htm END_OF_FILE