August 1990 AGRICULTUR,AL ..- TRADE Government support Calculations Under the : U.S.-Canada Free ‘. Trade Agreement GAO; RCED-90-210 GAO United States General Accounting Office Washington, D.C. 20548 Resources, Community, and Economic Development Division B-240062 August 81990 The Honorable Kent Conrad United States Senate Dear Senator Conrad: Your letter of October 26, 1989, expressed concerns about the method used to calculate levels of government support in the U.S.-Canada Free Trade Agreement. The agreement uses a formula, which calculates the amount of support governments provide their agricultural sectors, to determine the conditions that are required for Canada to open its mar- kets to U.S. grain and grain products. You raised several questions regarding the accuracy of the 1989 government support calculations for U.S. wheat, which we address in this report. We will address your other question-regarding the appropriateness of the agreement’s method of comparing government support for agriculture-in a subsequent report. The 1989 government support totals were generally calculated accu- Results in Brief rately, based on our review of 93 percent of the total amount of govem- ment support calculated. However, we noted several opportunities to adjust state budget expenditure data that would show more precisely how much support state governments are providing their agricultural sectors. I State budget expenditures represent a relatively small portion-about 2 percent-of government transfers provided to U.S. wheat producers. Consequently, even very large adjustments made to state budget expenditures would result in only a small change in the total govern- ment support percentage. In addition, some adjustments would increase and others would decrease the amount of state budget expenditures. Thus, the magnitude of these changes would probably be even less because these adjustments would tend to offset each other. However, when the two countries’ government support calculations are close, as they were in 1990, these adjustments could affect whether Canada IS required to lift its grain import restrictions. ‘Annex 705.4 of the US-Canada Free Trade Agreement refers to this category as State Budget 1hit- hys. However, the formula measures agricultural expenditures by state governments. Conrqwnt~y. throughout the report, we refer to these data as state budget expenditures. Page 1 GAO/RCED-W210 Agrkulturd Trade - B-240062 Background trading partners-entered into a Free Trade Agreement. The countries negotiated the agreement, which became effective January 1. 1989, Lvith the goal of eventually eliminating trade barriers to improve market access for each country’s goods and services. As part of the agreement, both countries agreed to eliminate virtually all tariffs between the coun- tries within ten years. The agricultural chapter of the agreement includes a provision on market access for grain and grain products. Canada protects its agricul- tural sector by requiring import permits for some U.S. grain products. Canada uses these permit requirements to protect its domestic pro- ducers from import competition. This provision sets conditions for Canada to eliminate its import permit requirements for wheat, barley, and oats and associated products when the United States’ level of gov- ernment support for these commodities becomes equal to or less than Canada’s Because the level of IJS. government support for oats was lower than Canada’s in 1989, Canada no longer requires import permits for U.S. oats and oat products. To measure the level of government support, the agreement includes a Calculating the formula that measures the amount of producer revenue that is derived Agreement’s Levels of from government policies. These measures of government support are Government Support calculated for each crop year, but for the purposes of determining kvhen import restrictions are eliminated, officials use an average of the two most recent crop years. The U.S.-Canadian Free Trade Agreement repre- sents the first time that such a measure of government support had been used in a trade agreement. The level of government support is expressed as a ratio. The numerator expresses the total of both direct and indirect government support for each commodity (wheat, oats, and barley) in one crop year. Payments to divert land from production are an example of direct government pay- ments to producers. Inspection and research programs, which help farmers, but are not directly paid to them, are examples of indirect sup- port. The denominator expresses the total value of production for each of these commodities, plus direct government payments. For example, the calculations of government support for crop >.tbar1987 showed that the U.S. government provided wheat producers $:3 1Z!j bil- lion in direct payments and $2.178 billion in indirect support i 5icbtl;~pp. I.) Added together, the total of $5.307 billion represents go~~t~rnmc~nt Page 2 GAO/RCEDS@21O .Agncultural Track B-240062 transfers to producers, which serves as the measure’s numerator. To derive the measure’s denominator, known as the adjusted producer value, analysts added the value of wheat production for 19872 ($5.497 billion) to the amount of direct payments ($3.1‘29 billion) to obtain $8.626 billion. The ratio of the numerator and the denominator show that in crop year 1987, U.S. wheat producers derived 61.5 percent of the value of their production from government support. Equivalent calcula- tions for the same crop year showed that Canadian wheat producers derived 46.7 percent of their income from government support. Table 1 identifies the United States’ and Canada’s level of government support for wheat (in terms of 2-year averages) since they have been calculated under the agreement. As a result of both years’ calculations, U.S. wheat producers are still subject to Canada’s import restrictions. Table 1: Levels of Government Support for U.S. and Canadian Wheat Level of Government Support Country 198P 1 990b United States 61.62 45 80 Canada 46.28 44 83 aAverage of levels of government support for crop years 1986 and 1987. ‘Average of levels of government support for crop years 1987 and 1988 The U.S. government support calculations for wheat are comprised of 13 categories of direct and indirect support. (See app. II.) In total, these categories are comprised of 25 elements; the largest element by far is deficiency payments to producers.” Other significant elements include Commodity Credit Corporation (CCC)loan forfeiture benefits, Export Enhancement Program benefits, and CCC commodity loans. The Free Trade Agreement specifically identifies the sources of data and the for- mulas used to derive each country’s level of government support. kakulated by multiplying 57.36 million metric tons of wheat by the average producer pnce 1It $96.84 per ton. .%eeappendix II for a listing of each element. Page 3 GAO/lZCED9@210 Agricultural Trade B-240062 U.S. Level of tion of the 1989 U.S. government support totals for wheat. We noted, Government Support however, several opportunities to adjust state budget expenditure data for Wheat Generally that would show more precisely how much state governments are sup- porting their agricultural sectors. Calculated Correctly State budget expenditures represent a relatively small portion of the government support numerator. In crop year 1987, for example, state budget expenditures totaled $106 million, approximately 2 percent of the $5.3 billion in government transfers provided to U.S. wheat pro- ducers. Consequently, even very large adjustments made to state budget expenditures would result in only a small change in the total govern- ment support percentage. In addition, we identified adjustments that would both increase and decrease the amount of reported state budget expenditures. Thus, the magnitude of the change would probably be even less because these adjustments would tend to offset each other. The agreement’s formula for state budget expenditures measures the amount of agricultural expenditures by state governments, by crop, excluding any federal funds provided for those purposes. The basis for this calculation is the amount of state government agricultural expendi- tures reported by the U.S. Census Bureau. The Census Bureau data for state agriculture expenditures, however, are not collected for the pur- pose of calculating the level of government support. Consequently, Census Bureau officials suggested making certain adjustments to the Bureau’s state budget expenditure data that would provide a more pre- cise total of the amount that states are spending to support their agricul- tural sectors. These adjustments involve the reporting of (1) user charges and check- off fees, (2) interest income, (3) interest expense, (4) state contributions to retirement, and (5) state expenditures for experimental stations and the extension service. Both the Census Bureau and USDA officials agreed that these adjustments were merited. However, they both noted that in some cases determining the magnitude of these adjustments could be difficult and costly. User Charges and Check- States collect user charges and check-off fees for various agricult~lral related purposes. For example, state agricultural promotion boards are Off Fees usually funded by check-off fees, which are financed as a tax on the first buyer’s purchase of an agricultural product. User charges U-P often Page 4 GAO/WED-9@210 Agricultural Trade 8240062 collected by states on the sale of agricultural products. These funds are used for state inspection or grading services. The agreement specifically excludes expenditures based on user contri- butions from being included in the government support calculation because these funds are generated by the agricultural community for the agricultural community. The Census Bureau’s reports, however, only identify a portion of revenues generated from user contributions-those contributions generated primarily by state fairs and shows and the sale of products from agriculture experiment station farms-and do not identify revenues generated from check-off fees. All revenues generated from user contributions need to be identified to make appropriate adjustments to the Census Bureau state budget expenditure totals. USDA and Census Bureau officials agreed that expenditures funded by check-off fees should also be excluded from these totals because, like user charges, these revenues are collected from the agricultural commu- nity for use by the agricultural community. Without adjustments to these figures, the Census Bureau reports would overstate, for the purposes of the Free Trade Agreement, the amount of state budget expenditures. At our recommendation, the Census Bureau provided USDA with adjustments to the 1990 government support calcu- lation that resulted in a reduction of about $550 million in the Bureau’s reported state budget expenditures. These adjustments represented close to 20 percent of the unadjusted totals for state budget expendi- tures and reduced the calculated amount of government support for wheat by $24.3 million in crop year 1988 and $20.8 million in crop year 1987.” Interest Income Agricultural organizations funded by user charges and check-off fees earn interest on some of their funds, primarily because they often receive total funding at the beginning of a fiscal year and earn interest until the money is spent. Just like offsets needed for revenues generated from user charges and check-off fees, revenues generated from interest earned on these funds also need to be offset. Bureau officials told us that the amount of these funds can be identified in most states. Offset- ting state budget expenditures for interest earned on these revenues would decrease the totals for state budget expenditures for agriculture. ‘Wheat represents about 4 percent of the total value of U.S. agricultural production T?wreftIrr I he $660 million ac&stment for all U.S. agricultural production translates into a much smaller dJu>rment for wheat. Page 5 GAO/RCED-B&210 Agricultural Trade 5240062 Interest Expense States often issue public debt to fund certain projects, some of which may be related to agriculture. The Census Bureau’s reports of state budget expenditures for agriculture do not include the interest on the debt for these expenditures. Census Bureau officials and a i3DA repre- sentative agreed that the portion of interest on this debt that is dedi- cated to agricultural projects should be added to the state budget expenditures. However, reporting this expense would be very difficult because public debt is used for many types of projects, only some related to agriculture. Identifying the agriculture-related portion of this debt to add to state budget expenditures would be technically difficult, they said. State Contributions to State agricultural budget expenditures do not report state contributions to retirement, unemployment compensation, and workman’s compensa- Retirement tion funds for state agriculture department employees, extension agents. and experimental station employees. These costs equal about 7 to 10 percent of these employees’ salaries, according to Census Bureau offi- cials. For Census Bureau reporting purposes, these costs are listed as accruals and recorded as expenses under categories other than agricul- ture when they are realized. Bureau officials said that these costs can be readily identified in most states and would increase total state expendi- tures for agriculture. Extension Service and The Census Bureau’s state budget expenditure reports are supposed to include state expenditures for the Extension Service and LSD;\ esperi- Experimental Stations mental stations. A small amount of these expenditures, however. are not included in the Census Bureau’s report of state budget expenditures for agriculture. These amounts are generally undercounted because very often colleges and universities in their reports to the Department of Education do not accurately identify their agriculture-related expenses. The Census Bureau, which is aware of the problem, does not expect to solve this data collection problem until it begins collecting the data Itself beginning in fiscal year 1991. Including the accurate total for these state costs would add to the Bureau’s reported total for state agriculr l1r-t’ budget expenditures. - ~__ For the most part, U.S. levels of government support are bemy (.;III II- Conclusion lated correctly, although we noted several opportunities to adJu4 -rate budget expenditure data showing how much support state goi t*r-nrr~nts are providing their agricultural sectors. These adjustments to ! t1t8 Page 6 GAO/RCED!W210 Agnrul~ural Trade B-240062 Census Bureau reports-both adding and subtracting from current totals-would be needed to provide more precise totals. Calculating some of these adjustments, however, could be costly because doing so would require additional analysis and data collection. Making all of these adjustments also may not be cost beneficial because state budget expenditures represent only about 2 percent of the entire level of government support total; that is, adjustments that affected state budget expenditure totals by even as much as $100 million would only affect the final government support percentage by less than 2 per- centage points. The magnitude of these changes would probably be even less because these adjustments would tend to offset each other. Never- theless, for the 1990 government support calculations-in which the U.S. and Canada wheat results were different by less than 1 percentage point-an adjustment of that magnitude, if it was in the United States’ favor, would have been sufficient to cause Canada to remove its require- ment for wheat import permits. Given the potential importance of even small changes in the U.S. govern- Recommendationto ment support calculations to the lifting of Canadian import restrictions the Secretary of for grain, USDA and the Cens& Bureau should jointly identify a cost- beneficial method for adjusting the Census Bureau’s report of state Agriculture and the budget expenditures for agriculture that would show more precisely Secretary of how much state governments are supporting their agricultural sectors. Commerce -.‘h m In developing our responses for this report, ye obtained information from the USDA and the Department of Commerce, Bureau of the Census. In agreement with the requester, we limited our analysis to the verifica- tion of 7 of the 25 elements comprising the 1989 U.S. government sup- port calculations for wheat. These elements together account for 93 percent of the total amount of support calculated.” Our verification of the U.S. government support calculations for wheat included deter- mining whether (1) the data sources were complete and appropriate. ( 2) the calculations were consistent with the agreement, and (3) the calcula- tions were made correctly. To determine the accuracy of the calcula- tions, we traced all numbers back to original sources. We also performed ‘Payments of the Commodity Credit Corporation, Acreage Reduction Program, CCC Loan Fwt*alt~ire Benefits, Price Enhancing Aspects of Government Programs, CCC Commodity Loans. State HrilWr Outlays, and Farm Credit Programs. Page 7 GAO/RCED9O-210 Agricultural Trade B-240062 tests on numbers derived from formulas to ensure that these were calcu- lated accurately. We did not verify the Canadian government support calculations because questions were not raised about their accuracy. We discussed our findings with officials from USDA and the Census Bureau and incorporated their comments in the report. Our work was conducted in accordance with generally accepted government auditing standards. As arranged with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 7 days after the date of this letter. At that time we will send copies of this report to the Secretary of Agriculture; the Secretary of Commerce; the Director, Office of Management and Budget; and other interested parties. Copies will be provided to others on request. This work was performed under the direction of John W. Harman. Director, Food and Agriculture Issues, who may be reached on (202) 275-5138. Major contributors to this report are listed in appendix III. Sincerely yours, c/ Assistant Comptroller General Page 8 GAO/RCED9@210 Agrirulrural Trade Page 9 GAO/RCED-90210.4gricultrrral Trade Contents Letter Appendix I Calculation of the U.S. Wheat Level of Government Support for Crop Year 1987 Appendix II 13 U.S. Government Support for Wheat, Crop Years 1986 and 1987 Appendix III 14 Major Contributors to This Report Table Table 1: Levels of Government Support for U.S. and 3 Canadian Wheat Abbreviations ccc Commodity Credit Corp. Page 10 GAO/RCEDW210 AgriculturalTrade . Page 11 GAO/RCEDBO-210 Agricnltud Trade Calculation of the U.S. Wheat Level of Govemment Support for Crop Year 1987 Dollars In billions Government transfers to producers Direct payments 93 129 Indirect payments 2 178 Total $5.307 Adjusted producer value Value of wheat productton $5 497 ~___-.. Direct payments 3 129 Total $8.626 Level of government support (5 30718626) = 61 5°C Page 12 GAO/RCEDW210 Agrirulrural Tradr Appendix II U.S. Government Support for Wheat, Crop Years 1986 and 1987 Dollars in millions (sections III-VII) 1986 1987 I. Level of production in million metric ton8 55.92 57.36 II. Producer price-dollars per ton 988.61 $95.84 Ill. Value of production SW44.35 t&496.83 IV. Direct payments $3.750.78 S3,129.09 V. AdJurted producer value s&795.13 W625.92 VI. Policy transfers to producers A. Direct payments $3,750.78 $3,129 09 1. Payments of the CCC 3,799.07 3,279x6 a. Deficiency payments 3d56.84 3,279.06 b. Disaster payments 126.96 0.00 c. Diversion payments 215.27 0.00 2. CCC storage payments 170.33 144.75 3. Conservation Reserve Program 11.95 102.47 4. Acreage Reduction Program -355.46 -479.54 5. Cerficate premiums and discounts 124.89 82.35 6. Other support 1,677.37 2,178.27 6. CCC loan forfeiture benefits 487.37 105.51 7. Price enhancement 457.80 1,433.Oo 8. Advance payments benefits 3.78 9.65 9. Crop Insurance 48.76 164 10. Government service programs 93.92 103.56 i. Federal grain inspection 0.62 0.35 ii. Research and extension 43.03 44.05 iii. Irrigation 3.65 4.16 iv. Inland waterways freight 16.57 1760 v. Conservation 22.09 27 34 vi. Rail freight 0.65 0.79 vii. Low interest loans for rail 1.42 162 viii. Cooperator export programs 0.89 1 07 ix. Marketing services 0.61 0.64 x. Plant disease and pest control 4.26 4.59 xi. Targeted export assistance 0.13 134 11. CCC commodity loans 376.48 30847 12. State budget outlays 106.00 106.00 13. Farm credit programs 103.27 11043 VII. Total government support $5,42&l 5 s5Jo7.37 VIII. Qovemment support (per-t) 61.72 61.53 Average for 1936 and 1987 (percent) 61.62 Note: Pursuant to Article 705 and Annex 705.4 of the United States-Canada Free Trade Agreemenr Source: USDA, adapted by GAO. Page la GAO/WED-BWlO Agrldtplrl Trade Appendix III Major Contributors to This Report Flora H. Milans, Associate Director Resources, Jeffrey E. Heil, Assistant Director Community, and Jeffrey Itell, Evaluator-in-Charge Jack M. Pivowar, Evaluator Economic Development Division, Washington, D.C. Page 14 GA0/RCED9%210 Agriirultural Tradr (150007) ____ [enited States (kneral Accounting Offiw -~- Washington, II.<‘. 2O.‘i-t8 Official Business Penalty for Private Use $300
Agricultural Trade: Government Support Calculations Under the U.S.-Canada Free Trade Agreement
Published by the Government Accountability Office on 1990-08-08.
Below is a raw (and likely hideous) rendition of the original report. (PDF)