AGRICULTURAL POLICY BRIEF

Transcription

1 Center for Agricultural Policy and Trade Studies North Dakota State University AGRICULTURAL POLICY BRIEF No. 27 May 13 Potential Changes in Federal Crop Insurance Premiums in North Dakota Under a Whole Farm Federal Crop Insurance Program Introduction Richard D. Taylor Won W. Koo* The main topic during the recent discussions of the 13 Farm Bill has been the federal budget deficit. Most recent spending bills have been reviewed closely to determine areas where spending can be removed from the legislation. Farm bill spending on the commodity title has been limited recently to direct payments, since commodity prices have been well above target prices. However both the premium subsidies and actuarial soundness of the federal crop insurance program has come under review as a possible source of further budget reductions. During the previous four years, the loss ratios for federal crop insurance for the ed States were.58 in 9,.56 in 1,.9 in 11 and 1.56 in 12. This indicates that only once during the past four years the crop insurance losses were greater than the premiums. The loss ratio for North Dakota during those same years was.65,.67, 1.55, and.25. Nationwide, the federal crop insurance premium subsidy amounted to about $26 per acre or 38% of the premium. In North Dakota the premium subsidy was about $32 per acre or 38%. In 12 and 11 the total federal crop insurance premium subsidy was $7. billion and $7.5 billion, respectively. In North Dakota the subsidy has been $618 million in 12 and $679 million in 11. To reduce crop insurance premiums, it has been proposed that a whole farm crop insurance option be made available to producers which might lower premiums. The SURE program in the 8 Farm Bill was an example of a whole farm insurance program. The entire crop production of an operation would be determined before any benefits were made. This option would insure all crop production under one package, which in theory would reduce indemnity losses and therefore insurance premiums. Under statistical theory, the variation of data, in this case commodity yields, are reduced as the data are aggregated. For example, the standard deviation of wheat yields in Stutsman County in North Dakota is 7.65 bushels per acre. The standard deviation of wheat yields for North Dakota is 6.56 bushels per acre and the standard deviation for wheat yields in the ed States is 3.95 bushels per acre. Another example is soybean yield, the standard deviation in Cass County is 6.69 bushels per acre, 5.1 bushels per acre in North Dakota and 4.19 bushels per acre in the * Research Scientist, and Professor and Director, respectively, in the Center for Agricultural Policy and Trade Studies in Fargo, North Dakota. Center for Agricultural Policy and Trade Studies North Dakota State University Fargo, North Dakota, 5815 (71) Fax: (71)

2 2 Center for Agricultural Policy and Trade Studies ed States. Variation in data is lower as the data set becomes more aggregated and yields under a whole farm crop insurance are more aggregated than data for a single unit. The objective of this study is to estimate the differences of crop returns under a whole farm crop insurance program compared to a unit crop insurance program. From those differences, crop insurance premium differences could be estimated and determine the amount of savings that producers could experience under a whole farm option. Method Since variation, i.e., standard deviation, within a data set become smaller as the data becomes more aggregated, potential crop losses due to yield variations should be smaller under a whole farm insurance program than under a unit farm insurance program. The North Dakota Representative Farm model, operational at the Center for Agricultural Policy and Trade Studies at North Dakota State University will be used to estimate federal crop insurance payments to is used to generate observations, using the standard deviation and mean of commodity yields, to represent producer s crop production variation. The difficulty is that standard deviations for individual producers or cropping units are not readily available. Proxies were developed using data from the last six years of the North Dakota Farm and Ranch Business Management Education Programs. DIVIDE BURKE RENVILLE BOTTINEAU ROLETTE TOWNER CAVALIER PEMBINA WILLIAMS MOUNTRAIL WARD 2. PIERCE RAMSEY WALSH MCHENRY MCKENZIE BENSON NELSON GRAND FORKS 4. DUNN MCLEAN MERCER SHERIDAN WELLS EDDY FOSTER GRIGGS STEELE TRAILL GOLDEN VALLEY BILLINGS STARK OLIVER MORTON BURLEIGH KIDDER 3. STUTSMAN BARNES 1. CASS SLOPE HETTINGER GRANT LOGAN LA MOURE RANSOM BOWMAN ADAMS EMMONS RICHLAND SIOUX MCINTOSH DICKEY SARGENT Region 1. Red River Valley (RRV) Region 2. North Central (NC) Region 3. South Central (SC) Region 4. Western (West) Figure 1. North Dakota Farm and Ranch Business Management Regions The North Dakota Representative Farm model divides the state into four regions. They are the Red River Valley (RRV), North Central (NC), South Central (SC), and Western (West) (Figure 1). The farms in each region are representative of the average, high, and low-profit farms; and small, medium, and large-size farms enrolled in the North Dakota Farm and Ranch Business Management Education Program. For this study the medium size farm was used for the simulation.

3 North Dakota State University 3 Several assumptions were made before the standard deviations could be estimated. First, small farms, the smallest 3%, tended to be located in one location while large farms, largest 3%, tended to be in multiple locations. Second, production practices and management abilities were similar between the two groups. Finally, yield variations were due to location and not to management. Several of these assumptions are questionable but they needed to be made because the necessary data were not available from other sources. This study was interested in variations of return rather than averages. Once the standard deviations of total crop revenue for both the small 3% and the large 3% were determined, the North Dakota Representative Farm model is run under the two scenarios; unit crop insurance and whole farm crop insurance, to estimate the differences in potential crop insurance payments. Table 1. Acreage Size Range and Standard Deviation for the Small and Large Farms Used to Estimated Standard Deviations Size Range of the Size Range of the Small 3% Large 3% Small Large Acres Acres Standard deviation as % of per acre total crop return Red River Valley ,79 5, North Central ,522 7, South Central ,167 13, West ,64 9, Table 1 shows the standard deviation of the small and large farms. The standard deviations were reported as percent of per acre total crop return. For example, if a farmer in the North Central region had an average total crop return of $278 per acre the standard deviation for that farmer would be $ per acre for the small or unit farm and $13.14 per acre for the large or whole farm. Those standard deviations were used a proxies for the standard deviations of the unit (single location) and whole (multiple locations) farm scenarios. It is predicted that the crop insurance indemnities for the unit coverage is higher than for the whole farm coverage because the standard deviations are greater. The larger standard deviation will results in larger variations in crop returns and those variations are the reason for crop insurance was used to generate 1 observations for each scenario. Those observations were used to calculate potential crop insurance payments based of a 75% coverage level which is average coverage level for the state. Results Table 2 shows the dollar value of the standard deviations for each region of the state, potential average crop insurance indemnity and frequency for those losses. From the table it is clear that the standard deviations of unit coverage are larger in every region than the whole farm coverage. The average insurance payment for the Red River Valley region is $15,494 for unit coverage and Center for Agricultural Policy and Trade Studies North Dakota State University Fargo, North Dakota, 5815 (71) Fax: (71)

4 4 Center for Agricultural Policy and Trade Studies $6,516 for whole farm coverage. The average insurance payment in the North Central region is $21,734 compared to $11,651 for the whole farm coverage. In the South Central region, unit coverage averaged $18,47 per year compared to $1,327 for whole farm coverage. Insurance payments in the West region average $27,188 for the unit coverage compared to $14,456 for the whole farm coverage. In addition to larger crop insurance payments for the unit coverage, insurance payments were made more frequent. In the Red River Valley payments were made 17.6% of the time under the unit scenario compared to 13.4% under the whole farm scenario. The frequency of payments was also higher for the unit coverage in all other regions. Table 2. Standard Deviations, Federal Crop Insurance Payments, and Frequency of Payments Whole Whole Difference Whole Red River Valley North Central South Standard Deviation ($) Crop Insurance Payments ($) Frequency of Payments (%) 141,932 14,939 15,494 6,516 8, ,125 8,916 21,734 11,651 1, ,123 96,566 18,47 1,327 7, Central West 139,772 14,31 27,188 14,456 12, State 86,288 66,545,616 1,737 1, Figure 2 shows the distribution of total crop revenue for North Dakota representative farm in the four regions of the state under the two scenarios. The area of interest for this study is the difference between the curves at the left side of each figure. For example, in the Red River Valley, the area between $4 thousand and $64 thousand is the area where federal crop insurance payments would be made. Since the distribution line for the unit coverage is higher than for the whole farm coverage, larger crop insurance payments would be made for the unit farms. The same is true in other regions. It can be seen that the larger standard deviations also impact the right side of the figures. However the means are similar under both scenarios. Table 3 shows the calculations of the per acre differences in average crop insurance payments between unit insurance coverage and whole farm insurance coverage. In the Red River Valley, the difference is $6.43 per acre while the difference is $5.34, $6., and $7.91 per acre in the North Central, South Central and the West regions, respectively. Reductions of per acre crop insurance payments indicate the possible premium reduction of the whole farm insurance coverage program. Average crop insurance premiums for North Dakota in 12 were $41.5 per acre for $ per acre insurance coverage with a $26.79 premium subsidy. There would be a $6.-$7. reduction in premiums costs under the whole farm coverage scenario in North Dakota. Total savings in insurance premium is $149 million per year in the state. However, it would also lower insurance coverage.

5 North Dakota State University 5 RRV NC Whole Whole thousand $ thousand $ SC 5 56 thousand $ Whole We st thousald $ Figure 2. Distribution of total crop revenue under the unit and whole farm scenario in the four regions of the state Whole Table 3. Crop Acres, Change in Crop Insurance Payments Between Insurance and Whole Farm Insurance, and Difference in Per Acre Payments Insurance Crop Acres Difference in Insurance Payments $ Per Acre Difference $ Red River Valley 1,397 8, North Central 1,889 1, South Central 1,286 7,7 6. West 1,69 12, State 1,545 1, So far this study has been dealing with total crop revenue. What is important for farmers is net farm income. Tables 4 and 5 show the frequency of potential net farm income reduction under various levels of crop losses. For example under the unit standard deviations, a farmer who normally has a net income of $15, in the South Central region would have a 6.7% or 1 out of 15 chance of having a 75% loss in net farm income. That same farmer under the whole farm scenario would experience a 75% loss in net farm income once every 28 years. A farmer in the Red River Valley could expect a 5% loss in net farm income once every eighteen years. Most production losses are small, less than 5%, however the potential for large losses require serious strategies for dealing with those risks. Center for Agricultural Policy and Trade Studies North Dakota State University Fargo, North Dakota, 5815 (71) Fax: (71)

6 6 Center for Agricultural Policy and Trade Studies Table 4. Potential Reduction in Net Farm Income Under the Scenario Standard Deviations Without Crop Insurance Net Farm Income Loss percent RRV NC SC West 75% % % Table 5. Potential Reduction in Net Farm Income Under the Whole Farm Scenario Standard Deviations Without Crop Insurance Net Farm percent Income Loss RRV NC SC West 75% % % RRV NC Who le 3 Who le d ollars/a cre dollars/acre SC 1 14 d ollars/a cre Who le West dollars/acre Who le Figure 3. Distribution of Federal Crop Insurance Payments Under the Alternative Scenarios in the Four Regions of the State Figure 3 presents the distribution of crop insurance payments across the state. The vast majority of the payments are small, $-$3 per acre, however it shows that the unit crop insurance program provides more frequent large payments because of the larger standard deviations.

7 North Dakota State University 7 Summary Recent discussions in Congress have addressed the possibility of the reduction or removal of the premium subsidies for the Federal Crop Insurance program. With the current Federal budget deficit all programs and all features of those programs are being scrutinized for possible budget savings. Those subsidies for U.S. agriculture amount to over $7 billion per year, and about $65 million per year in North Dakota. It has been proposed that a whole farm crop insurance option would reduce federal crop insurance premiums compared to unit crop insurance pricing. Under this option the entire crop production of a farmer would be insured under one program no matter where the crops are located. The idea is that production in several locations combined into one package would lower insurance losses and then, premiums. This study was conducted to estimate insurance premium differences. To conduct the computer program along with the North Dakota Representative Farm model was used to simulate cropping histories under both plans to estimated potential differences in crop insurance uses the standard deviation and sample means of crop yields to simulate observations. The means are readily available but standard deviations are not. Statistical theory maintains that the variation (standard deviation) of a data set become smaller as that data is aggregated with similar data. For example, yield variations for a crop in the ed States are smaller than yield variation for that crop a state. Likewise, yield variations for a state are smaller than yield variation for a county within that state. Logically, yield variations for a single section of land within a county would be greater than for that county. However, that standard deviation is not available. Two proxies were developed using the past six years data from the North Dakota Farm and Ranch Business Management Education Program. The proxy for unit insurance (current) came from the standard deviation from the small farms in the program. It was assumed that they were typically in one or two locations while large farms were located in many areas. The standard deviations from the large farms were used as proxies for the whole farm insurance program. The standard deviation for the small (unit) and large (whole) were substantially different but the differences in crop insurance losses varied between $5.34 per acre in the North Central region and $7.91 per acre in the West. Those differences would indicate the potential reduction in crop insurance premiums from the unit price crop insurance to the whole farm price crop insurance. Center for Agricultural Policy and Trade Studies North Dakota State University Fargo, North Dakota, 5815 (71) Fax: (71)

8 8 Center for Agricultural Policy and Trade Studies References North Dakota Agricultural Statistics. Various issues. North Dakota Agricultural Statistics Service, Fargo. North Dakota Farm and Ranch Business Management Annual Reports 9, 1 and 11. North Dakota State Board for Vocational Education, Bismarck. Taylor, D. Richard, Won W. Koo and Andrew L. Swenson. 12 North Dakota Agricultural Outlook: Representative Farms, Agribusiness and Applied Economics Report No June 12. Center for Agricultural Policy and Trade Studies, Department of Agribusiness and Applied Economics, North Dakota State University Fargo, North Dakota USDA Agricultural Projections to 21. Long-term Projection Report OCE-112. February 12. ed States Department of Agriculture. Office of the Chief Economist. Washington DC. National Agricultural Statistics Service website. Quick Stats 2..