|* Presentation slides are available at the bottom of the page.|
The 2010 USDA Agricultural Resource Management Survey (ARMS) Corn data will be used to examine the price paid for manure as a function of type of manure (i.e. species), form of manure, distance, size of farm, location, yield goal and whether the application rates of manure were influenced by Federal, State or local policies. Based on economic theory and the few empirical studies on manure use, it is hypothesized that swine manure will command a lower price than manure from cattle or poultry operations, all else equal. Liquid manure, due to dilution and volatilization of nutrients, will have a negative effect on price received. Due to transportation costs, which are included in the ARMS manure cost question, distance is hypothesized to have a positive effect on price. Farms in areas with high nutrient demand, such as the Corn Belt, are hypothesized to pay higher prices for manure while those in areas with excess manure nutrients, such as the Chesapeake Bay area, will pay lower prices or even be compensated to accept manure. Similarly, if policy affects application of manure, it is an indication that there are problems of excess manure in the area so prices are expected to be lower. Higher yield goals are expected to be positively associated with the price paid for manure since nutrient requirements will be higher.
Increasing fertilizer prices may affect demand for manure nutrients. If manure is viewed as a valuable resource it is more likely to be well-managed. Examining the factors affecting the price paid for manure can indicate to what extent it is viewed as a resource but also indicate what might be done to increase the real or perceived value of manure for crop production.
The USDA Agricultural Resource Management Survey (ARMS) is actually a set of surveys. Some are for the operation as a whole and some are for specific crops. The Corn Production Practices and Costs Report was conducted for the 2010 crop year by the National Agricultural Statistics Service. A specific, randomly selected field is identified and interviewers ask a variety of questions regarding management practices, yields, etc. on that field. The total size of the dataset was 2654, but this research focuses on the 919 farmers who indicated that manure was applied to the field in question. Summary statistics were calculated for this subset of the data.
The relationship between the price paid for the manure and various explanatory variables was examined using OLS regression as a first step. For the regression analysis, only farmers who indicated the manure came from off the farm were included, which reduced the number of observations to 206, 169 of which had no missing data. The cases of biosolids, sheep, and other were deleted since the numbers were too low once the farmers who applied their own manure were dropped. The dependent variable was calculated and represents the total cost of manure purchased for the field. This combined answers for the total cost for the field plus the product from answers to the amount per unit times number of units (e.g. $ per gallon times number of gallons) since farmers had a choice of how to answer the question.
A few summary statistics on the full dataset are provided for comparison purposes. For the operation as a whole, an average of 345 acres of corn was planted in 2010. The size of the sample field was 48.5 acres. The value of corn production as a percentage of the total farm value of production was 38.4%. Only 0.34 % of farms indicated that this field was certified organic. For the initial subset of the data, 190 acres of corn was planted, the size of the sample field was 30.3 acres, the value of corn production was 18.4%, and 0.33% of farms indicated the field was certified organic. Therefore, the use of manure on the field does not seem to be associated with a higher rate of organic production. The farms that use manure were smaller than farms in the dataset as a whole and less specialized in corn production. As we see below, the majority of farmers who applied manure produced it themselves which also indicates these farms are more diversified.
The data for the subset of farmers who applied manure can be used to generally characterize manure use on farmers’ corn fields. The mean number of acres that manure was applied to in the field was 25.4, less than the average field size. The average expected corn production on the field was 5505 bu. The distance from the source of the manure to the field was on average 3.29 miles with a range from 0 to 320 miles. Of these farmers, 16% indicated that federal, state or local regulations had affected the manure application rate. The manure was applied by a custom applicator in 17% of cases. The mean custom application cost for purchased manure was $297 per field. Manure was tested in 22% of cases.
A number of categorical variables are also of interest. In the majority of cases (76.9%) farmers were applying their own manure. For manure that was sourced off the farm, 129 farmers purchased it, 79 received it for free, and 4 were compensated to accept it, showing that it is viewed as a valuable resource in most cases. The form of manure was primarily solid (635 farmers or 69%), followed by lagoon liquid (147) and slurry (136). For those who applied their own manure the percentage of solid manure was similar to all farmers applying manure (481/706 or 68%) and the lagoon form was only slightly higher (16.9% vs 16.0%). For those who applied manure from off the farm, there were slightly more cases of slurry than lagoon liquid (30 vs 28). The indicated units for manure application amounts are instructive; the majority (558) indicated tons, 232 indicated gallons, and the rest (129) indicated bushels. The number who indicated gallons is larger than the number who indicated the use of lagoon liquid so some slurry was also measured this way.
The most common animal source of the manure was dairy (453/919 or 49.3%), followed by beef (24.0%), poultry (16.4%), swine (7.6%), other (1.2%), equine (0.8%), sheep (0.3%), and there was one case of biosolids and zero cases of food waste. It should be noted that a separate set of questions related to compost, which was applied by about 2% of all the corn farmers surveyed.
The survey also indicated the ERS region for each farm. For farmers with corn who indicated that they used manure on the selected field, the majority (399 or 43.4%) were in the Northern Crescent (Lake States and Northeast). The next most common region was the Heartland (Corn Belt) with 38.7% of farmers, followed by Eastern Uplands (5.1%), Southern Seaboard (4.5%), Prairie Gateway and Northern Great Plains (each at 3.9%), and two cases each in the Basin and Range, and Fruitful Rim Regions.
For the regression analysis, only farmers who indicated the manure came from off the farm were included, which reduced the number of observations to 206, 169 of which had no missing data. The cases of biosolids, sheep, and other were deleted since the numbers were too low once the farmers who applied their own manure were dropped. The dependent variable was calculated and represents the total cost of manure purchased for the field. Farmers were given a choice as to how to answer the question, a total cost or answering two questions, number of units and price per unit. The dependent variable combined answers for the total cost for the field plus the product from answers to the amount per unit times number of units (e.g. $ per gallon times number of gallons).
The following results from the OLS regression are preliminary. The adjusted R2 for the model was 0.417. There was no effect of size of the corn operation overall as measured by total acres of corn planted. Total value of farm production would probably be a better measure of size but a fairly large number of farmers didn’t answer that question. Expected corn production on the field did not have a significant effect and the coefficient was negative. Whether the field was organic was excluded from the regression by SAS and may be partly due to the low number of observations. As one would expect, given the nature of the dependent variable, the area the manure was applied to had a positive and significant impact on the price paid. Distance between source and field was positive and significant (p=0.057). For the form of manure, the base was slurry and the price paid for lagoon liquid was higher (p=0.028), contrary to expectations. There was no price difference between solid manure and slurry. The value of beef, dairy and poultry manure was higher than that for swine manure, in line with expectations. There was no effect of government regulations. The value was significantly higher if the manure was custom applied. Contrary to expectations, there were no regional differences after controlling for the other factors.
Alternate specifications of the model, including alternative versions of the dependent variable, will be examined to check for robustness. A final paper will relate the findings to the small but growing literature in this area.
Laura McCann, Associate Professor, University of Missouri; McCannL@missouri.edu
Ali, Sarah, Laura McCann, Jessica Allspach. "Manure Transfers in the Midwest and Factors Affecting Adoption of Manure Testing". Journal of Agricultural and Applied Economics Vol. 4 (4), November 2012, pp. 533-548.
Nunez, Jennifer and Laura McCann. “Determinants of Manure Application by Crop Farmers” Journal of Soil and Water Conservation Vol. 63 (5) September/October 2008, pp. 312-321.
We acknowledge research funding by USDA-NIFA, Integrated Water Quality Grant Program, 110D.
The authors are solely responsible for the content of these proceedings. The technical information does not necessarily reflect the official position of the sponsoring agencies or institutions represented by planning committee members, and inclusion and distribution herein does not constitute an endorsement of views expressed by the same. Printed materials included herein are not refereed publications. Citations should appear as follows. EXAMPLE: Authors. 2013. Title of presentation. Waste to Worth: Spreading Science and Solutions. Denver, CO. April 1-5, 2013. URL of this page. Accessed on: today’s date.