the manure hits the land: economic and ... - PubAg - USDA

THE MANURE HITS THE LAND: ECONOMIC AND ENVIRONMENTAL IMPLICATIONS WHEN LAND APPLICATION OF NUTRIENTS IS CONSTRAINED JONATHAN D. KAPLAN, ROBERT C. JOHANSSON, AND MARK PETERS The discharge of manure nutrients into area waters from confined animal feeding operations is considered a leading contributor to U.S. water quality impairments. An option to mitigate these impairments is to constrain land application of manure. When these constraints are particularly binding, due to minimal acceptance of manure as a substitute for commercial fertilizer, potentially large and unanticipated changes in returns to agricultural production and water quality may occur. Moreover, some of the cost of meeting the constraints is passed on to consumers through higher prices and to a portion of rural economies through lower production rates and labor expenditures. Key words: animal agriculture, water quality, sector analysis.

Agricultural production in the United States annually discharges large amounts of nitrogen and phosphorus into area ground and surface waters. These nutrients from crop and animal production are found in 50% of impaired lakes and 20% of impaired rivers in the United States (Environmental Protection Agency 2002a). Current production of manure nutrients on animal feeding operations (AFOs) often exceeds the nutrient requirements of the surrounding cropland available for manure spreading, thereby increasing the potential nutrient discharge into surrounding waters (Kellogg et al.; Gollehon et al.). Growing public concern over manure and subsequent nutrient-related pollution is evident in recent state and federal legislation enacted to regulate livestock and poultry production (and in new government funding for livestock and poultry producers intended to assist in compliance with water quality regulations). Currently, thirty-five states have programs to control water pollution from AFOs; thirtyfour of these programs limited manureJonathan D. Kaplan is assistant professor, Department of Economics, California State University, Sacramento. Robert C. Johansson is economist, Economic Research Service, United States Department of Agriculture. Mark Peters is economist, Agricultural Marketing Service, United States Department of Agriculture. The authors thank Marcel Aillery, Marca Weinberg, and two anonymous referees for their valuable comments. The views expressed in this article are those of the authors and do not necessarily reflect the views of the Economic Research Service, the Agricultural Marketing Service or the United States Department of Agriculture.

nutrient-application rates prior to federal controls, and twenty-seven states require animal producers to develop manure-management plans (EPA 2002c). The primary federal environmental legislation affecting animal production is the Clean Water Act (CWA). The largest of the AFOs, or concentrated animal feeding operations (CAFOs), are required under the CWA to obtain a National Pollution Discharge Elimination System (NPDES) permit, which specifies how manure disposal is managed on each CAFO.1 The federal regulatory landscape changed for all AFOs in 1999 when the U.S. Department of Agriculture (USDA) and the U.S. Environmental Protection Agency (EPA) announced their Unified National Strategy for AFOs (USDA-EPA 1999). The Unified Strategy details various actions that the USDA and EPA would take under existing legal and regulatory authority to minimize water quality and public health impacts from improperly managed animal waste (Ch. 1, p. 9, USDA-ERS 2003a). The EPA recently revised the regulations for CAFOs to fulfill part of the goals of the Unified Strategy and to mitigate the actual and potential water quality impacts posed by animal production. These revisions changed the requirements for an NPDES permit and

1 EPA defines a CAFO as approximately 700 dairy cows, 1,000 beef cows, 4,000 swine, and 250,000 broilers (Gollehon et al.).

Amer. J. Agr. Econ. 86(3) (August 2004): 688–700 Copyright 2004 American Agricultural Economics Association

Kaplan, Johansson, and Peters

their associated Effluent Limit Guidelines for CAFOs by requiring permit holders to develop and implement nutrient management plans for manure nutrients. These nutrient management plans limit the land application of manure. The limits are generally agreed to be the agronomic nutrient demand on cropland—i.e., manure generated at CAFOs is applied to cropland and pastureland at a rate no greater than the rate at which the crops can assimilate the applied nutrients, thereby minimizing nutrient runoff from the fields (USDA-EPA 1999). This requirement is likely to constrain manure application rates based on cropland phosphorus demand rather than on nitrogen demand because animal waste generally contains more phosphorus than nitrogen relative to the nutrient demand for most crops. Our analysis focuses on the inclusion of this provision in the NPDES permitting process for the estimated 15,500 CAFOs covered by the CWA. Land application constraints on manure nutrient use may lower aggregate production of livestock and poultry commodities at the national level, resulting in increased prices. Changes in crop production will result from adjustments in the derived demand for cropland for spreading manure and the demand for crops as feed ration inputs. Moreover, the production changes at the national level will likely affect regional economies and their environments differently because economic and environmental conditions vary across regions. With such large-scale changes to consider, both a regional and a sector-wide assessment of economic adjustments are appropriate (Berck and Hoffman). An abundance of literature discusses the merits of judiciously applying manure nutrients to crops (see, e.g., Fleming, Babcock, and Wang; Lazarus and Koehler). Innes extends this literature by providing a comprehensive, albeit theoretical, treatment of the spatial and environmental issues of manure generation, management, and regulation. Recent empirical applications have examined the impact and effectiveness of restricting landbased applications of manure nutrients to achieve water quality goals at the farm level (Ribaudo, Gollehon, and Agapoff; Roe, Irwin, and Sharp) and at the national level (USDA-NRCS 2002b; EPA 2001; FAPRI). However, for different reasons these applications do not completely adhere to the comprehensive approach suggested by Innes. Our analysis of manure-nutrient application con-

The Manure Hits the Land

689

straints (henceforth, nutrient constraints) departs from and builds upon the recent empirical literature by focusing on regional and sector economic adjustments, as suggested by Innes, and Berck and Hoffman. We adopt a regional optimization approach, which solves endogenously for price and production adjustments across the U.S. agricultural sectors when nutrient constraints are imposed on the largest of the AFOs. This illustrates the explicit impacts of nutrient constraints on livestock and poultry producers, crop producers, consumers, rural economies, and the environment. Given our policy assumptions and scope of analysis, our results are most closely comparable to the Food and Agricultural Policy Research Institute (FAPRI) (for sector impacts) and EPA (for environmental impacts) studies, which employ representative farm models of livestock production across sectors and regions. That said, there are several underlying differences between these approaches and our analysis, which may explain why there are differences in our findings. In our analysis, we focus on the costs of meeting nutrient standards for all manure generated under different assumptions of land availability; whereas the FAPRI and EPA analyses include additional costs facing livestock CAFOs under the new EPA regulation but do not consider alternative land availability. Also, the FAPRI analysis accounts for potential farm exit (not entry) behavior in response to adverse economic impacts, whereas our regional treatment allows the number of animals produced to vary, which implicitly captures exit (reduction in production) and entry (increase in production) influences. In addition, we incorporate the interaction between the cropping and animal sectors through land availability for nutrient spreading and through animal feed prices, which the other studies do not consider. We also build on these other approaches by examining potential impacts on rural economies through changes in labor expenditures, an indicator of regional welfare. Given these similarities and differences, our empirical application suggests that nutrient constraints result in reduced livestock production and increased prices for their products, similar to the FAPRI findings. In both analyses, the livestock sector experiences similar relative declines in production (approximately 1% to 2%). However, we find the greatest price and production changes are in the poultry sector (approaching 6% and nearly 8%,

690

August 2004

respectively).2 Turning to a comparison of results on changes in water quality, our analysis suggests that the reduction in CAFO generated nutrients discharged to the edge-of-thefield may exceed 18%, which is comparable to the EPA finding of 24% (p. 7,239, Federal Register). Although the percentages appear to be similar, in absolute terms our estimates differ because the EPA does not account for excess manure sent off farm—a substantial amount of manure in many cases. Furthermore, we find that under some scenarios the potential for environmental degradation increases in some regions, contrary to both the EPA findings and the policy objective that motivated the imposition of nutrient constraints. This difference can be explained by our model’s ability to capture changes in crop and animal production in response to both the policy and the secondary price effects. The next section describes the empirical analysis used to evaluate the economic and environmental implications of these constraints. We develop a sector model that illustrates the potential changes in production and prices, given the imposition of land application constraints for manure nutrients. The third section presents the results of the policy simulations and discusses the potential changes in the livestock and poultry sectors, crop sectors, rural economies, and the environment. The article closes with a summary of findings and potential implications for regulating the environmental impacts of animal production. Empirical Analysis Nutrient constraints essentially force animal production and crop production within a geographic area to be in balance, likely reducing the quantity of manure nutrients that reach U.S. ground and surface waters. We assume that when an operation meets nutrient constraints, the manure generated from that operation is applied to cropland at no greater than the agronomic demand for the limiting nutrient. This could increase the cost of production for affected AFOs as they seek available cropland for manure spreading, incur higher hauling costs, and invest in associated nutrient

Amer. J. Agr. Econ.

management services. Furthermore, if a region has more manure nutrients than can be assimilated by available cropland, then it is out of balance. Of the several changes that can occur to allow a region to achieve a balanced state, we consider endogenous changes in cropping and animal composition, technologies, and production levels. With a sector model in mind, we see that the imposition of these nutrient constraints and thus manure management costs on CAFOs are transmitted throughout the entire agricultural sector, across animal and crop sectors, and across regions, because animal and crop production is intricately linked through input and output markets. For instance, crop producers supply feed grains to the animal sectors. If animal production declines due to the increased cost of production, the demand for feed grains will shift downward, affecting the feed grain markets. Cross-sector shifts in demand and supply will continue throughout the agricultural economy as the markets adjust to changes in relative prices for complement and substitute goods. Ultimately, the prevailing demand and supply elasticities throughout the input and output markets will determine the regional and sector-wide impacts of imposing nutrient constraints.3 To gauge the responsiveness of the U.S. agricultural sector to nutrient constraints, we focus attention on the availability of agricultural cropland for spreading manure, which in turn depends on crop producers’ willingness to substitute manure for commercial fertilizer.4 In the following analysis, we consider the case where only CAFOs meet nutrient constraints. These facilities represent 4.47% of the total AFOs in the United States. However, the quantity of manure generated by CAFOs exceeds 200 million tons, more than 46% of the United States’ total from confined operations. Regional differences are also notable (i.e., table 1). The percentage of animal operations categorized as CAFOs in the Southeast and Pacific regions is significantly higher than in other regions. Additionally, CAFOs generate more than 60% of the manure from all confined operations in the Northern Plains, Appalachia, Mountain, and Pacific regions. 3

2

The FAPRI analysis does not examine the poultry sector. The EPA analysis considers the poultry sector and estimates minimal poultry sector price changes but does not impose nutrient constraints to off-farm manure applications. This serves to relax the nutrient constraints for those CAFO facilities, which do not have extensive land holdings.

Elasticities can be provided on request. New technological innovations that might allow animal producers alternative means to curtail manure nutrient generation are not considered. Examples might include supplements to livestock and poultry feed and alternative manure storage and treatment options that would serve to diminish the nutrient content of animal manure. 4



691

Table 1. Operations with Confined Livestock and Manure Distribution USDA Farm Production Region

Total AFO

% CAFO

Total AFO

% CAFO

CAFO Manure Concentrationa (Tons/Acre)

31,350 52,498 71,252 26,087 22,776 12,635 12,252 10,500 7,780 7,654 254,784

1.59 1.64 3.18 4.77 7.46 10.79 7.48 7.00 8.43 14.85 4.47

39 59 73 65 66 23 19 46 33 40 462

15.42 25.10 39.55 64.01 62.29 43.31 39.04 38.22 69.31 60.55 46.36

0.42 0.39 0.29 0.57 2.25 1.33 0.42 0.56 0.80 2.43 0.64

Operations

Northeast Lake Corn Belt Northern Plains Appalachia Southeast Delta Southern Plains Mountain Pacific Totals

Manure (Million Tons)

a Tons of manure and acres of cropland are measured at the regional level.

Source: 1997 U.S. Census of Agriculture (USDA-NASS 1997). Northeast = CT, DE, MA, MD, ME, NH, NJ, NY, PN, RI, and VT; Lake = MI, MN, and WI; Corn Belt = IA, IL, IN, MO, and OH; Northern Plains = KS, ND, NE, and SD; Appalachia = KY, NC, TN, VA, and WV; Southeast = AL, FL, GA, and SC; DELTA = AR, LA, and MS; Southern Plains = OK and TX; Mountain = AZ, CO, ID, MT, NM, NV, UT, and WY; Pacific = CA, OR, and WA.

Table 1 illustrates the regions where meeting nutrient constraints might be more difficult than others. Appalachia, Southeast, and Pacific regions have greater manure production per acre of cropland than do other regions. This indicates that possible changes in economic performance throughout these regions could be more severe than in other regions. That said, we might also see greater environmental improvement in these regions. To evaluate the implications of meeting nutrient constraints, we simulate a constrained partial equilibrium, regional optimization model, which seeks to maximize profits from livestock, poultry, and cropping enterprises, max (P j − V Cr j )xactr j (1a) xactr j ,xactri j

− T Cr − FCr ) + (Pi − V Cri )xactri − AVCr , i

subject to nutrient constraints (1b) ( jr × man nutrjf × xactrj ) j

≤

(substitute × Ag nutrif

regime) in region r; Pj and VCj are equilibrium prices and variable costs for livestock and poultry products; and Pi and VCi are equilibrium prices and variable costs for crops. We also include the fixed costs (FC) essential for meeting a nutrient constraint, transportation costs (TC) associated with manure spreading, and additional variable costs (AVC) for soil testing and fertilizer savings.5 Transportation costs for each region are a function of the distance traveled and the quantity and type of manure transported, (2a) T Cr = (Tonr j × xactrj ) j

× (Spreadrj + Disr × Haulrj ), where Ton is the amount of manure produced per animal species, Spread and Haul represent spreading and hauling charges for each animal species and region, and Dis is the average distance greater than a mile traveled by a CAFO to spread manure. Manure storage, handling, and spreading technologies will vary by farm and by region. For example, pump-sprayfield technologies require pumping the manure only one direction from a lagoon, and will have lower hauling costs compared to hauling slurry

i

× (xactri ), ∀r, f. Here xactrj represents regional production of livestock and poultry species j in region r; xactri represents regional acres planted under cropping enterprise i (crop rotation and tillage

5 These fixed costs include per operation manure testing and management plan development of $200 and $400, respectively, and soil testing cost estimates ranging from $0.08 to $0.40 per acre per year across regions based on soil type (USDA-NRCS 2002b). Commercial fertilizer savings when manure is substituted for commercial fertilizer amount to $0.185 per pound for nitrogen and $0.30 per pound for phosphorus (Ribaudo, Gollehon, and Agapoff).

692

August 2004

manure, requiring a return trip. Poultry litter on the other hand, while dry and relatively inexpensive to haul, involves relatively high handling costs. We choose the commercial spreading and hauling charges to reflect the predominant manure storage and handling practices for each animal species in each region (USDA-NRCS 2002b; Borton et al.; Pease, Pelletier, and Kenyon; Fleming, Babcock, and Wang). We do not account for the fixed costs of specific technology types, such as the cost to purchase hauling equipment or to extend irrigation infrastructure. To determine the regional distance per affected AFO, we regionalize the Fleming, Babcock, and Wang methodology for calculating manure-spreading acreage Acr (2b) Disr = − 1, (1 − r ) × T Or × 640 where Acr is the total acres available for spreading manure (Acr is divided by 640 to convert acres to square mileage), which is a function of the nutrient constraints and the endogenous crop acreage choices (xactri ). ∈ (0, 1] describes the regional share of production from CAFOs, and TO is the total number of AFOs in that region. approaches one as the number of CAFOs within a region increases, effectively centralizing the location of affected operations toward the middle of the region, thereby allowing the manurespreading acreage to be the greater distance needed to spread manure from a few highly concentrated operations. We assume that transportation of manure only occurs within a region because the farm production regions are already large (incorporating from two to eleven states). The nutrient constraint (equation (1b)) explicitly requires the regional agricultural sector to maintain nutrient balance through altering animal and crop production. That is to say, within region r, the sum of each manure nutrient generated (man nutrjf ) by CAFOs must be less than or equal to a fixed proportion (substitute) of the sum of the agronomic nutrient demand (Ag nutrif ) for each cropping activity i, where f indexes nitrogen and phosphorus, respectively.6 The available de6 Estimates of available manure nutrients by animal type are the net losses attributable to prevailing storage and handling technology (Kellogg et al.). Agronomic demand is calculated using crop uptake values for nitrogen and phosphorus, accounting for losses due to denitrification, subsurface flow, runoff, and leaching.

Amer. J. Agr. Econ.

mand for manure nutrients may be enhanced by increasing the manure substitution rate or by increasing the agronomic nutrient demand. For instance, the intensive cropping in the Southeast of bermudagrass, which has a relatively high agronomic nutrient demand compared to the conventional crops contained in the simulation model (Darst et al.) would allow manure to be spread over less acreage. rj represents the CAFO portion of available manure generation for each region and species. Note that man nutrjf and Ag nutrif are endogenously determined given optimal levels of animal and crop production, but the fixedproportion, substitute, is not. We exogenously define the willingness to substitute manure nutrients (substitute) as a fixed proportion of a region’s agronomic nutrient demand that is met by manure nutrients because the simulation model used in the analysis does not consider the decisions of individual operators, and is therefore unable to endogenize substitution rates. Such substitution rates are difficult to determine from existing data and will most likely vary by crop and by region. For example, 58% of farms in the Lake States and Northeast surveyed in 1996 report using manure as a fertilizer on their corn fields, compared to 20% in the Corn Belt (Foreman). However, manure use is not as prevalent in soybean production: 25% of farms in the Lake States and Northeast surveyed in 1997 reported using manure as a fertilizer on their soybean fields, compared to 11% in the Corn Belt (Foreman and Livezey). Across all regions and farm sizes an estimated 17% of corn producers and 8% of soybean producers supplement commercial fertilizer with manure as part of their crop fertilization regime (USDA-ERS 2003b). These percentages are not manure substitution rates per se, but do indicate that manure is widely used in crop production. It is unclear to what extent substitution rates might change as CAFOs meet nutrient constraints. Some crop producers may be reluctant to accept manure as a substitute for inorganic fertilizers given that manure nutrients are not packaged as uniformly as commercial fertilizers, may contain pathogens, and are generally more difficult to handle (Risse et al.). Nevertheless, it is not unrealistic to assume that the use of manure nutrients as a substitute for commercial fertilizer may increase, especially in regions facing binding nutrient constraints. Therefore, we allow the substitution rate to vary between 20% and 40% (i.e., substitute ∈


(0.2, 0.3, 0.4)) to reflect a feasible range of possible rates. We simulate this constrained optimization problem using the U.S. Regional Agricultural Sector Mathematical Programming Model (USMP), a comparative-static, spatial and market equilibrium model that incorporates agricultural commodity, supply, demand, environmental impacts, and policy measures (House et al.). This model has been applied to climate change mitigation (Peters et al.), water quality policy (Ribaudo et al.), wetlands policy (Claassen et al.), and sustainable agriculture (Faeth). The model includes forty-five geographic subregions based on the intersection of the ten USDA Farm Production Regions and the twenty-five USDA Land Resource Regions. Twenty-three inputs are included, as are production and consumption of 44 agricultural commodities and processed products.7 The model differentiates more than 5000 crop production enterprises at the subregion level according to cropping rotations, tillage practices, and fertilizer rates. The more than ninety livestock and poultry production enterprises are delineated at the region level by species, given less regional and subregional heterogeneity in livestock and poultry production relative to crop production. USMP calibrates production levels and enterprises to regularly updated production practice surveys using a positive math programming approach (Howitt), the USDA multi-year baseline (USDA World Agricultural Outlook Board 2001), and the National Resources Inventory (USDA Soil Conservation Service 1994). Regionally specific extensive (animal and crop production levels) and intensive (crop rotations, tillage, and fertilizer practices) management practices are endogenously determined. Substitution among the cropping activities is achieved in the model using nested constant elasticity of transformation functions. The transformation elasticities are specified so that model supply response at the national level is consistent with domestic supply response in the USDA’s Food and Agriculture Policy Simulator (Westcott, Young, and Price) and with trade response in the USDA 7 USMP accounts for production of the major crop (corn, soybeans, sorghum, oats, barley, wheat, cotton, rice, hay, and silage) and confined livestock (beef, dairy, swine, and poultry) categories comprising approximately 75% of agronomic production and more than 95% of confined livestock production (USDA-NASS 1997). We do not consider potential applications of manure to rangeland, vegetable, horticulture, sugar, peanut, or silviculture operations.


693

Economic Research Service/Penn State Model (Abler). In addition, the nonlinear supply response functions are constructed to reflect declining marginal rates of transformation between crop rotations and between tillage activities. This implies that changes in land allocated to various production enterprises will not occur in a bang–bang fashion, but will smoothly adjust to changes in relative returns across production enterprises. Optimal crop and animal production decisions simulated in USMP are linked to edge-of-field environmental variables using the Environmental Policy Integrated Climate Model (EPIC) (Mitchell et al.). For each crop production activity, EPIC simulates erosion (sheet, rill, and wind), nutrient, and pesticide cycling as a function of crop management (rotation, tillage, and fertilizer rates), given historic weather, hydrology, soil temperature, and typography data. Our estimates of field-level discharge represent mean values for a sixty-seven-year time horizon. The movement of nutrients, pesticides, and sediment across the landscape is then calibrated to USGS estimates of regional pollutant loads (Smith, Schwartz, and Alexander). Estimates of CAFO and AFO spreading practices on swine operations taken from Ribaudo, Gollehon, and Agapoff allow us to account for prior land application of manure in the simulations. On average, CAFOs currently spread manure on the 155 nearest acres to their operation and the smaller AFOs currently spread manure on the ninety nearest acres. While these numbers do not reflect the wide variety of animal operations across the United States, they are reasonable for initial estimates of the environmental effects of excess manure utilization at the Farm Production Region scale. Because many livestock facilities have little or no land on which to dispose of manure, the above levels provide a lower bound on our estimated benefits to meeting nutrient constraints. Given the acres currently receiving manure nutrients, we calculate the quantity of manure nutrients in excess of the crop requirements on those acres. These excess nutrients are available for potential leaching into ground waters and/or transport across the landscape into surface waters. Results By simulating sector-wide optimization, we obtain results portraying a potential range

694

August 2004

Amer. J. Agr. Econ.

Table 2. Percent Change in Prices and Quantities by Manure Substitution Rate Prices Commodity Eggs (dozen) Broilers (carcass lbs.) Turkey (carcass lbs.) Manufactured milk (cwt) Butter (lbs.) American cheese (lbs.) Fed beef (cwt) Pork (cwt) Corn (bu) Sorghum (bu) Barley (bu) Oats (bu) Wheat (bu) Rice (cwt) Soybeans (bu) Cotton (bale)

Quantities

40%

30%

20%

40%

30%

20%

−0.04 0.15 0.20 0.60 1.05 0.50 0.42 0.01 −0.15 −0.12 −0.07 −0.83 −0.02 0.00 −0.03 0.00

1.65 0.89 1.37 1.10 1.91 0.91 0.45 0.00 −0.38 −0.16 −0.12 −0.81 −0.03 −0.14 −0.17 −0.06

6.17 3.13 5.17 2.33 4.04 1.92 0.55 0.73 −1.27 −0.50 −0.28 −3.77 −0.15 −0.34 −0.88 −0.19

0.00 −0.37 −0.28 −0.18 −0.30 −0.18 −1.70 −0.02 −0.15 −0.31 −0.21 −0.43 0.04 −0.03 −0.04 0.00

−0.09 −2.23 −1.97 −0.34 −0.54 −0.33 −1.83 −0.01 −0.27 −0.16 0.02 −0.48 0.09 1.36 0.06 0.07

−0.35 −7.87 −7.47 −0.71 −1.14 −0.70 −2.20 −1.56 −0.64 −0.01 −0.37 −2.38 0.40 3.35 0.80 0.26

of national and regional changes in the U.S. agricultural sector following the adoption of nutrient constraints by CAFOs. In general, production decreases for livestock and poultry commodities and increases for crop commodities, while their respective prices move in the opposite direction. At the regional level, large decreases in animal production are noted in the Southeast, Appalachia, and Pacific regions as expected. Conversely, increases in production are predicted for the Northeast, Lake, Corn Belt, and Delta regions. However, due to price changes, net returns increase in many sectors and regions throughout the United States. Prices and Quantities The magnitude of production decreases and price increases decline across the livestock and poultry sectors as manure substitution increases (table 2).8 The largest quantity and price changes occur in the poultry sector. When substitution rates are at or near current levels (i.e., 20%), broiler production decreases by 7.8%, and turkey production decreases by 7.5%. The corresponding prices increase by approximately 3.1% for broilers and 5.2% for turkeys. Although egg production declines by 0.35%, the price increases by 6.2%, perhaps due to a low elasticity of demand for eggs. Price

8 These changes are in reference to the USDA baseline projections for the year 2010 (USDA World Agricultural Outlook Board 2001).

changes for dairy products are also noticeable under the 20% substitution rate where butter and milk prices increase by as much as 4% and 2.3%, respectively. These price increases suggest that consumers of these commodities will bear some of the cost associated with the imposition of nutrient constraints. The quantity and price changes in the crop sector are not as general as are those for the livestock and poultry sectors. This is in part due to the dual role of cropland as a sink for manure nutrients and as a source of feed grains for livestock and poultry operations. This sink role creates an incentive to plant crops that require relatively high quantities of phosphorus (assuming that phosphorus is the limiting nutrient). For example, corn production falls, as does its price. This occurs because the derived demand for corn as a feed ingredient shifts inward with a decline in animal production more than the derived demand for corn acreage as a means of disposal shifts outward when nutrient constraints are imposed. This may be explained by the decrease in poultry production arising from an upward shift in the poultry supply function given nutrient management costs, a major user of corn in its feed, and by the fact that corn production occurs in regions with relatively low poultry concentrations. In contrast, production of hay and wheat increases while their prices fall. This may be occurring because hay and wheat are relatively high consumers of (i.e., sinks for) phosphorus, and the nutrient constraints force an outward shift in the demand for these activities, which outweighs



the inward shift in derived demand for these activities as livestock and poultry feed. Regional Response At the regional scale we see how different the impacts of nutrient constraints will be on the number of animal units produced, the number of cropland acres planted, and regional agricultural labor expenditures (table 3). At a 40% substitution rate, only minor changes occur because most regions have sufficient cropland for spreading manure. When the manure substitution rate is 20%, livestock and poultry production falls by more than 20% in the Appalachia, Southeast, and Pacific regions. Conversely, livestock and poultry production increases in the Northeast (6.2%), Lake (5.9%), Corn Belt (7.1%), Delta (7.4%), and Southern Plains (1.6%) regions. Although crop acreage changes less than 1% across most scenarios, it increases in the Appalachia, Southeast, and Pacific regions (i.e., those regions with the most binding constraints on manure nutrient production, reflecting the high cost of meeting nutrient constraints). These acreage increases provide additional sinks for manure nutrients and potentially lessen any livestock and poultry production reductions that otherwise might have been required to meet the nutrient constraints. Other regions, with a general abundance of land for spreading manure, do not experience significant changes in cropland acreage. These changes in regional production patterns trickle down through rural agricultural economies, creating a relationship between nu-

695

trient constraints and agricultural labor expenditures. Under all substitution rate scenarios the amount paid to agricultural labor falls nationally, indicating that rural agricultural economies may face decreasing purchasing power. A closer look at the regional heterogeneity reveals that most of this decrease occurs in the Northern Plains, Appalachia, and Pacific regions. The remaining regions increase expenditures on agricultural labor, particularly when manure substitution rates are 20% with the greatest increases in the Southern Plains and Mountain regions. Sector Responses Table 4 shows the changes in net returns to livestock, poultry, and crop sectors at the national and regional levels. Essentially, when net returns to the livestock and poultry sectors increase due to prevailing elasticities and the stringency of the nutrient constraints, net returns to the crop sectors decrease, and vice versa. While we can generalize these effects at the national level, the sectors experience a range of changes across regions depending on the extent to which crop producers are willing to substitute manure nutrients for commercial fertilizer. Nationally, secondary price effects have an overall positive effect on the livestock and poultry sectors under the 20% manure substitution rate scenario. In addition, as substitution rates increase, the livestock and poultry sectors see increasing losses because the cost of meeting nutrient constraints is not offset by a fully compensating increase in prices. When the manure substitution rate is at or near

Table 3. Regional Change in Animal Units, Crop Acreage, and Labor Expenditure by Manure Substitution Rate Animal Units (%) Region Northeast Lake Corn Belt Northern plains Appalachia Southeast Delta Southern plains Mountain Pacific United States

Crop Acreage (%)

40%

30%

20%

40%

30%

20%

0.45 −0.31 −0.61 −3.03 0.50 0.65 −0.22 0.95 −3.50 −0.95 −1.06

1.35 0.33 −0.18 −2.73 0.91 −7.39 0.67 1.36 −3.11 −8.22 −1.13

6.24 5.85 7.11 −0.37 −20.53 −26.52 7.40 1.62 −1.55 −22.72 −2.23

−0.10 −0.13 −0.05 −0.08 −0.14 −0.08 −0.07 −0.03 −0.07 −0.05 −0.08

−0.17 −0.24 −0.12 −0.14 −0.28 4.65 −0.16 −0.05 −0.13 2.66 0.04

−0.48 −0.75 −0.41 −0.42 6.17 12.66 −0.49 0.29 −0.27 6.05 0.46

Labor Expenditure ($Million) 40% 3.0 −0.3 −7.1 −19.1 1.0 1.9 −0.2 5.5 3.9 −11.8 −23.1

30% 6.4 2.3 −4.8 −16.4 3.2 5.9 0.0 15.8 11.7 −41.6 −17.5

20% 16.4 13.8 16.7 −4.2 −46.3 10.8 1.8 20.9 28.0 −109.0 −51.0

696

August 2004

Amer. J. Agr. Econ.

Table 4. Regional Change in Net Returns to Livestock and Poultry Sector and Crop Sector by Manure Substitution Rate Livestock and Poultry Sector ($Million) Region Northeast Lake Corn Belt Northern plains Appalachia Southeast Delta Southern plains Mountain Pacific United States

40% 18.4 −41.2 −104.2 −306.3 −68.0 −44.8 −36.2 −2.0 −70.1 −175.6 −830.1

30% 120.8 50.1 4.7 −245.0 25.1 −221.9 45.2 88.8 −10.4 −497.2 −639.8

current levels we see that net returns increase for the livestock and poultry sector in all regions except in the Northern Plains, Southeast, and Pacific. The lack of suitable cropland for spreading manure in these regions implies that the cost effect will be too large for the corresponding national price increase to compensate, thus resulting in declining returns. As the substitution rate increases and the national price falls for livestock and poultry products, we see the Mountain region experience net returns below pre-constraint levels. If the substitution rate were 40%, then the bulk of the manure management cost would fall on CAFOs, given little to no change in the national price level, and net returns would fall below pre-constraint levels for all regions. For the crop sector, the changes in production, prices, and net returns are smaller in magnitude than changes in the livestock and poultry sector. When the substitution rate is 20%, the overall crop sector sees decreases in net returns associated with a large decrease in animal production and thus a decrease in the demand for feed grains such as corn and soybeans. This is most noticeable in the Corn Belt where net returns fall by as much as $350 million. With greater substitution rates, we see that the losses to the crop sector decline. This can be explained by the corresponding growth in livestock and poultry production, and thus feed grain demand as the substitution rate increases. These results suggest that overall the decrease in demand for feed grains outweighs the increase in demand for crop acreage as a sink for manure nutrients.

20% 427.9 369.5 526.4 −1.0 105.5 −743.3 310.2 267.9 132.3 −1110.9 284.8

Crop Sector ($Million) 40% 1.9 1.8 −9.0 0.8 13.4 8.6 −0.1 0.2 0.3 0.9 18.8

30% −1.9 −11.1 −73.7 −19.1 6.1 16.0 −1.9 −1.2 −0.9 7.0 −80.8

20% −17.2 −65.2 −350.5 −97.0 −21.0 16.6 −17.6 −5.9 −5.7 12.0 −551.6

Environmental Impacts The use of USMP and EPIC allows us to examine the environmental implications resulting from our policy scenarios. In particular, we estimate the quantity of phosphorus discharged into surface water and the quantity of nitrogen discharged into surface and ground waters (table 5). We find that under all nutrient constraint scenarios the nitrogen and phosphorus discharged to surface waters falls by approximately 400 and 100 million pounds, respectively. These declines correspond to significant overall reductions in nitrogen (10%) and phosphorus discharge (29%). However, it appears as though the quantity of nitrogen leached to groundwater increases at low substitution rates due to induced cropping and livestock production changes. Phosphorus reductions across the regions range from 9% in the Northeast region when substitution remains at or near current levels to 70% in the Pacific region at the higher levels of substitution. The greatest reductions in pounds of phosphorous occur in Appalachia and the Corn Belt regions. Reduction in nitrogen discharged to surface waters ranges from 2% (40 million pounds) in the Corn Belt to 40% (80 million pounds) in the Pacific region. Although the majority of regions exhibit declines in nitrogen leaching across all scenarios, the amount of nitrogen leached to groundwater increases nationally by 2.6%, or 45 million pounds under the 20% substitution rate scenario. Increased nitrogen leaching is particularly evident in the Appalachia (5.8%), Southeast (10.5%), and Pacific (32%)



697

Table 5. Reduction of Phosphorus and Nitrogen Discharged (Million Pounds) by Manure Substitution Rate Phosphorus to Surface Water Region Northeast Lake Corn Belt Northern plains Appalachia Southeast Delta Southern plains Mountain Pacific United States

40% 4.1 8.9 23.1 10.9 21.1 11.1 4.4 6.6 11.7 11.0 112.8

30% 3.5 8.5 23.1 10.9 20.2 11.8 4.0 6.3 11.6 11.6 111.5

Nitrogen to Surface Water 20%

1.6 5.1 19.5 10.3 20.8 14.1 2.6 5.6 11.2 12.9 103.7

40% 20.7 33.5 38.6 33.5 55.8 63.4 16.2 17.8 43.2 81.2 403.9

regions when substitution rates are at or near current levels. As substitution rates increase nitrogen leaching falls in all regions, indicative of less severe changes in regional agricultural production. One factor leading to the somewhat unanticipated results for nitrogen leaching is when the nutrient constraint becomes more binding, an incentive exists to increase cropland acres and to grow crops that consume relatively more nutrients and that leach more nitrogen on those acres. Because the phosphorus constraint binds sooner than the nitrogen constraint, cropland producers will have to supplement the new crop composition and acreage planted with additional commercial nitrogen fertilizer, which in effect serves to undermine the reductions in manure production (at least vis-a-vis ` nitrogen leaching). Nevertheless, the nitrogen prevented from reaching surface waters is of a greater magnitude than the relatively small increases in nitrogen leaching in all regions and scenarios. Summary of National-Level Analysis Efforts are underway at the local, state, and federal level to reduce the discharge of manure nutrients into U.S. ground and surface waters. Recently the EPA promulgated its final rule for CAFOs (EPA 2002b). These regulations, slated for implementation in 2006, will require CAFOs to meet nutrient constraints. Many possible outcomes of this policy exist, affecting the location and production decisions of U.S.

30% 20.9 34.1 39.8 34.1 56.2 62.1 16.4 16.6 43.3 79.5 402.9

Nitrogen to Ground Water 20%

21.4 36.7 44.8 37.3 43.5 53.2 12.7 13.2 43.5 77.3 383.5

40% 0.3 1.8 0.6 0.5 1.5 0.5 0.4 0.2 0.1 0.2 6.0

30% 0.4 3.4 0.7 0.6 2.2 −6.8 0.8 0.2 0.1 −7.4 −5.8

20% 1.1 10.1 1.2 1.1 −23.3 −19.2 2.0 −0.3 0.1 −17.6 −44.9

agricultural producers. We consider potential economic and environmental implications for a case when only the largest animal facilities, CAFOs, meet nutrient constraints. These operations account for nearly 50% of confined livestock and poultry production. Because the impact of nutrient constraints is dependent upon the acres available for spreading manure nutrients, we have varied the percentage of crop producers that are willing to substitute manure nutrients for commercial fertilizer. Furthermore, because we allow the production decisions of both livestock and crop producers to respond to secondary price effects, we add to the body of literature attempting to quantify the costs and benefits of such nationwide agrienvironmental policies. However, this analysis cannot reveal how individual operations would benefit or suffer due to the application constraints. If only the largest AFOs meet nutrient constraints, the costs of compliance would potentially fall most heavily on the CAFOs and the benefits from secondary price effects will accrue to the smaller AFOs. We see that when crop producers’ substitution rates for manure nutrients remain at or near current levels, the secondary price effects are sufficient to compensate most livestock and poultry sectors for the costs of meeting nutrient standards. However, at higher acceptance rates the costs of transporting manure, manure testing, soil testing, and developing a manure management plan outweigh compensating price effects and foregone commercial fertilizer purchases, resulting in reduced net returns for the

698

August 2004

animal and crop sectors. At the most, these losses could approach $830 million (1.6% of baseline returns), when 40% of agronomic nutrient requirements are met with manure nitrogen and phosphorus. We note that the costs we have included in our analysis (namely transportation costs, manure testing costs, soil testing costs, and nutrient management plan development) do not include all the costs livestock and poultry producers face as they adjust to meet nutrient standards. Additional costs might include relocation costs, and investments in new manure storage and handling infrastructure. In addition to the costs and benefits borne by agricultural producers in meeting nutrient constraints, consumers could face higher food prices under the various policy scenarios. In particular, prices for poultry and dairy products could increase when substitution rates remain low. Due to reductions in livestock production under all scenarios, labor expenditures incurred in animal production are likely to decrease between $17.5 million and $51 million. As with the sector performances, these impacts on the labor market are heterogeneous across regions and scenarios. Motivating many of the agri-environmental policies aimed at reducing discharge of manure nutrients into U.S. surface and ground water has been a realization of the adverse ecological effects attributed to these pollutants. We illustrate that however well intentioned policies are, potentially undesirable secondary effects could arise from shifting market equilibria. Studies that have looked at the potential benefits to restricting land applications of manure nutrients have until now ignored secondary price effects on production choices and subsequent environmental ramifications. We show that under modest assumptions (i.e., when crop producers’ willingness to substitute manure nutrients for commercial fertilizer remains at or near current levels) nitrogen leached to ground water may increase, especially in the Appalachia, Southeast, Southern Plains, and Pacific regions. However, as intended, in all regions phosphorus discharged to surface waters and net nitrogen discharged (surface plus ground water) to the environment falls significantly across all scenarios. In sum, the potential for unanticipated impacts in certain areas of the United States exist due to heterogeneous economic and environmental conditions. In some regions these impacts may be positive, such as in-

Amer. J. Agr. Econ.

creasing net returns to livestock and poultry producers; in others they may be undesirable, such as increased nitrogen leaching to groundwater. Such variable effects illustrate the need to perform regional and sector-wide analyses when evaluating such far-reaching policies. [Received March 2003; accepted September 2003.] References Abler, D. “The ERS/Penn State Trade Modeling Project.” Available at http://coldfusion. aers.psu.edu/wto/, 2001. Berck, P., and S. Hoffman. “Assessing the Employment Impacts of Environmental and Natural Resource Policy.” Environmental and Resource Economics 22(2002):133–56. Borton, L.R., C.A. Rotz, H.L. Person, T.M. Harrigan, and W.G. Bickert. “Simulation to Evaluate Dairy Manure Systems.” Applied Engineering in Agriculture 11(1995):301–10. Claassen, R., R.E. Heimlich, R.M. House, and K.D. Wiebe. “Estimating the Effects of Relaxing Agricultural Land Use Restrictions: Wetland Delineation in the Swampbuster Program.” Review of Agricultural Economics 20(1998):390–405. Darst, B.C., J.N. Pratt, H.D. Pennington, and W.B. Gass. “Nutrient Demand High in Bermudagrass.” Fluid Journal 15(1996):22–25. Faeth, P. “Growing Green: Enhancing the Economic and Environmental Performance of U.S. Agriculture.” World Resources Institute, Washington DC, 1995. Federal Register. “National Pollutant Discharge Elimination System Permit Regulation and Effluent Limitation Guidelines and Standards for Concentrated Animal Feeding Operations (CAFOs); Final Rule.” National Archives and Records Administration, Friday, 12 February 2003. Fleming, R., B. Babcock, and E. Wang. “Resource or Waste? The Economics of Swine Manure Storage and Management.” Review of Agricultural Economics 20(1998):96–113. Food and Agricultural Policy Research Institute (FAPRI). “FAPRI’s Analysis of the EPA’s Proposed CAFO Regulations.” University of Missouri, FAPRI-UMC Report No. 06-01, Columbia, MO, July 2001. Foreman, L. “Characteristics and Production Costs of U.S. Corn Farms.” Economic Research


Service, USDA, Statistical Bulletin No. 974, Washington DC, 2001. Foreman, L., and J. Livezey. “Characteristics and Production Costs of U.S. Soybean Farms.” Economic Research Service, USDA, Statistical Bulletin No. 974–4, Washington DC, 2002. Gollehon, N., M. Caswell, M. Ribaudo, R. Kellogg, C. Lander, and D. Letson. “Confined Animal Production and Manure Nutrients.” Economic Research Service, USDA, Agriculture Information Bulletin No. 771, Washington DC, 2001. Howitt, R.E. “Positive Mathematical Programming.” American Journal of Agricultural Economics 77(1995):329–42. House, R.M., H. McDowell, M. Peters, and R. Heimlich. “Agriculture Sector Resource and Environmental Policy Analysis: An Economic and Biophysical Approach.” Environmental Statistics: Analyzing Data for Environmental Policy. New York : John Wiley and Sons, 1999. Innes, R. “The Economics of Livestock Waste and Its Regulation.” American Journal of Agricultural Economics 82(2000):97–117. Kellogg, R.L., C.H. Lander, D.C. Moffitt, and N. Gollehon. “Manure Nutrients Relative to the Capacity of Cropland and Pastureland to Assimilate Nutrients: Spatial and Temporal Trends for the United States.” Natural Resource Conservation Service, Economic Research Service, USDA, Washington DC, 2000. Lazarus, W.F., and R.G. Koehler. “The Economics of Applying Nutrient-Dense Livestock Waste at Low Rates.” Review of Agricultural Economics 24(2002):141–59. Mitchell, G., R.H. Griggs, V. Benson, and J. Williams. “Environmental Policy Integrated Climate Model.” Available at http://www.brc. tamus.edu/epic/, 1998. Pease, J., B.A. Pelletier, and D. Kenyon. “Poultry Litter Transport Alternatives for Land Application in Virginia.” Selected paper presented at the Annual Southern Agricultural Economics Association meetings, Ft. Worth Texas, 2001. Peters, M., J. Lewandrowski, R. House, and H. McDowell. “Economic Impacts of Carbon Charges on U.S. Agriculture.” Climatic Change 50(2001):445–73. Ribaudo, M., N. Gollehon, and J. Agapoff. “Land Application of Manure by Animal Feeding Operations: Is More Land Needed?” Journal of Soil and Water Conservation 58(2003):30–38. Ribaudo, M., R. Heimlich, R. Claassen, and M. Peters. “Least-Cost Management of Non-


699

point Source Pollution: Source Reduction vs. Interception Strategies for Controlling Nitrogen Loss in the Mississippi Basin.” Ecological Economics 37(2001):183–97. Risse, L.M., M.L. Cabrerea, A.J. Franzluebbers, J.W. Gaskin, J.E. Gilley, R. Killorn, D.E. Radcliffe, W.E. Tollner, et al. “Land Application of Manure for Beneficial Reuse.” National Center for Manure and Animal Waste Management white paper, 2001. Roe, B., E.G. Irwin, and J.S. Sharp. “Pigs in Space: The Spatial Structure of Hog Production in Traditional and Nontraditional Production Regions.” American Journal of Agricultural Economics 84(2002): 259–78. Smith, R.A., G.E. Schwarz, and R.B. Alexander. “Regional Interpretation of Water-Quality Monitoring Data.” Water Resources Research 33(1997):2781–98. U.S. Department of Agriculture (USDA), Economic Research Service (ERS). Manure Management for Water Quality: Costs to Animal Feeding Operations of Applying Manure Nutrients to Land. Agricultural Economic Report No. AER824, 2003a. ——. Agricultural Resources and Environmental Indicators. Agricultural Handbook No. AH722, Washington DC, 2003b. U.S. Department of Agriculture, National Agricultural Statistical Service (NASS). Census of Agriculture —1997. Washington DC, Available at www.nass.usda.gov/census/, 1997. U.S. Department of Agriculture, National Resource Conservation Service (NRCS). Costs Associated with Development and Implementation of Comprehensive Nutrient Management Plans. Review Draft, USDA, NRCS, Washington DC, 2002b. U.S. Department of Agriculture, Soil Conservation Service. Summary Report: 1992 National Resources Inventory. Washington DC, 1994. U.S. Department of Agriculture, World Agricultural Outlook Board. USDA Agricultural Baseline Projections to 2010, Staff Report WAOB-20011. Office of the Chief Economist, Washington DC, 2001. U.S. Department of Agriculture-U.S. Environmental Protection Agency (USDA-EPA). Unified National Strategy for Animal Feeding Operations. Available at www.epa.gov/npdes/pubs/ finafost.pdf, 1999. U.S. Environmental Protection Agency (EPA). Environmental and Economic Benefit Analysis of Proposed Revisions to the National Pollutant Discharge Elimination System Regulation and

700

August 2004

the Effluent Guidelines for Animal Feeding Operation. Office of Water: EPA 821-R-01-002, Washington DC, 2001. ——. 2000 National Water Quality Inventory. Office of Water Report EPA-841-R-02-001, 2002a. ——. Concentrated Animal Feeding Operations (CAFO)—Final Rule. Available at http://cfpub. epa.gov/npdes/afo/cafofinalrule/npdes/afo/ cafofinalrule.cfm, 2002b.

Amer. J. Agr. Econ.

——. State Compendium: Programs and Regulatory Activities Related to Animal Feeding Operations. Available at www.epa.gov/owm/ stcpfin.pdf, 2002c. Westcott, P.C., C.E. Young, and J.M. Price. “The 2002 Farm Act: Provisions and Implications for Commodity Markets.” U.S. Dept. Agr, Agricultural Information Bulletin No. 778, Washington DC (November), 2002.