estimating the effects of relaxing agricultural land use restrictions: wetland delineation in the...

Agricultural & Applied Economics Association

Estimating the Effects of Relaxing Agricultural Land Use Restrictions: Wetland Delineation inthe Swampbuster ProgramAuthor(s): Roger Claassen, Ralph E. Heimlich, Robert M. House and Keith D. WiebeSource: Review of Agricultural Economics, Vol. 20, No. 2 (Autumn - Winter, 1998), pp. 390-405Published by: Oxford University Press on behalf of Agricultural & Applied Economics AssociationStable URL: http://www.jstor.org/stable/1349997 .

Accessed: 25/06/2014 01:12

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Agricultural & Applied Economics Association and Oxford University Press are collaborating with JSTOR todigitize, preserve and extend access to Review of Agricultural Economics.

http://www.jstor.org

This content downloaded from 185.2.32.109 on Wed, 25 Jun 2014 01:12:17 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=oup

http://www.jstor.org/action/showPublisher?publisherCode=aaea

http://www.jstor.org/stable/1349997?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Review of Agricultural Economics--Volume 20, Number 2--Pages 390-405

Estimating the Effects of Relaxing Agricultural Land Use Restrictions: Wetland Delineation in the Swampbuster Program

Roger Claassen, Ralph E. Heimlich, Robert M. House, and Keith D. Wiebe

Wetland protection is an issue of ongoing debate. Although it is widely agreed that wetland loss to agriculture has been declining in recent decades, the role of policy remains contentious. We analyze the effect of changes in wetland delineation rules that were proposed but rejected by Congress during the 1996 farm bill debate. Our research combines detailed, site-specific information on wetlands with a broader model of the agricultural economy. Using site-specific data, we analyze the potential agricultural profitability of a representative sample of actual wetlands. We estimate wetland acreage that would have been exempted from swampbuster and Section 404 of the Clean Water Act under the proposed delineation changes, the acreage of exempted wetland that could be profitably converted to crop production, and the associated commodity price, crop acreage, and farm income effects. We find that up to 82.7 million wetland acres would be exempted under the proposed delineation changes, of which as many as 12 million acres could be profitably converted to crop production. This conversion would have a dampening effect on commodity price and farm income. We conclude that (a) accurately estimating the effect of resource policy depends critically on detailed information on resource quality and (b) commodity price and farm income effects imply that all agriculture producers- not only those who could expand cropland acreage through wetland drainage-have a stake in wetland policy.

During the past several decades, U.S. wetland policy has shifted from promot- ing exploitation (i.e., drainage and conversion to other uses) to encouraging

conservation and restoration of wetlands (Dahl and Allord; Heimlich, Carey, and

M Roger Claassen, Ralph E. Heimlich, Robert M. House, and Keith D. Wiebe are economists with the Resource Economics Division, Economic Research Service, US. Department of Agriculture.



Relaxing Agricultural Land Use Restrictions 391

Brazee). Policy actions that are evidence of this shift include the regulation of wetland dredge and fill under Section 404 of Clean Water Act (CWA) of 1977 and enactment of swampbuster as part of the Food Security Act (FSA) of 1985. This paper examines the potential economic and environmental effects of recent legis- lative proposals to relax these provisions.

Wetland protection programs are currently assessed by their contribution to the goal of "no net loss" of wetland areal (Conservation Foundation). To date, no net loss has been interpreted to mean that wetland loss to agriculture, urbanization, and so on must be offset through restoration or creation of similar wetlands. Whether no net loss has in fact been achieved is a matter of current debate (Heim- lich, Wiebe, and Claassen; Tolman). However, it is clear that the net rate of wetland conversion has been declining in recent decades and that much of this decline can be attributed to a decline in the gross rate of wetland conversion to other uses, especially agriculture.

Between 1954 and 1974, the average gross rate of wetland conversion to agriculture was 593,000 acres per year, roughly 80% of total gross wetland conversion of 730,000 acres per year (Frayer et al.). Between 1974 and 1984, total gross conversion of wetland to other uses fell to about 446,000 acres per year, and conversions for agriculture fell to roughly 50% of the total, about 235,000 acres per year (Dahl and Johnson). In the decade between 1982 and 1992, gross wetland loss dropped to 120,000 acres per year, only 20% of which were agricultural (Heimlich and Melanson).

The reasons for this decline are also a matter of debate. Contributing factors might include declining profitability of agricultural conversion, the swampbuster provisions of the 1985 and 1990 farm bills, elimination of tax incentives for wetland conversion in the Tax Reform Act of 1986, and continued implementation of reg- ulatory programs under Section 404 of the CWA. Some analysts believe that policy, along with economic factors, has played a critical role in slowing agricultural conversion (Heimlich and Melanson). Other analysts acknowledge (at least implicitly) that swampbuster has reduced the return to wetland conversion for agriculture but argue that conversion would be generally unprofitable even without swampbuster (Kramer and Shabman; Tolman).

A key question in the debate over no net loss and over wetland policy in general is whether progress toward no net loss, including reductions in the gross rate of wetland conversion, can be sustained if the swampbuster provision and Section 404 of the CWA are eliminated or relaxed (Heimlich et al.). A number of proposals to change wetland delineation criteria in the swampbuster program and Section 404 of the CWA (i.e., the extent of wetlands subject to sanction and/or regulation) were considered during the 104th Congress (Heimlich, Wiebe, and Claassen). One pro- posal, the so-called twenty-one-day exemption, was included in the House-passed legislation reauthorizing the CWA (U.S. House of Representatives) and was proposed in the context of the 1995-96 farm bill debate to make swampbuster consistent with that legislation. The twenty-one-day exemption would delineate wetlands subject to swampbuster as areas that are typically inundated (ponded or flooded) for at least twenty-one consecutive days during the growing season.2 Under the current swampbuster provision, wetland delineation requires the soil to be inundated for fifteen days during the growing season, except for prairie pothole, playa, or pocosin wetlands, which must be inundated for seven days (National Research Council).



392 Review of Agricultural Economics

Both the swampbuster program and CWA Section 404 were continued without change in delineation criteria (U.S. Congress). However, the wetland delineation debate highlighted the need for a methodology to assess the effect of such proposals in terms of the prospects for achieving and maintaining no net loss of wetland resources and associated economic consequences. Previous research on agricultural wetland conversion has largely been confined to site-specific simulation models (Heimlich and Langner, Kramer and Shabman, U.S. Department of the Interior). These models generally contain significant detail on local resource conditions (e.g., productivity) and farm structure (e.g., the size and crop mix for farms). As such, they provide conclusions regarding economic incentives to wetland conversion for a generalized farm on a specific site. One exception is an econometric estimation of the role of federal flood control in facilitating wetland conversion in the Mississippi delta (Stavins and Jaffee). None of these models provides the nationwide or multiregional analysis necessary to assess the broader economic or environmental consequences of the proposed delineation change in the swampbuster program or other changes in land use policy.

Our research fills this void by combining detailed, site-specific data on wetlands and data on local economic conditions with a broader model of the agricultural economy. We analyze data on wetland hydrology and potential agricultural productivity for nearly 50,000 wetland sample points that are aggregated to make regional and national estimates of wetland area excluded from swampbuster under the twenty-one-day exemption that could be profitably drained for crop production. The site-specific nature of the data allows us to draw regional and national conclusions on the basis of the potential agricultural productivity of a representative sample of actual wetlands rather than using county average productivity or other assumptions that might obscure important variations in resource quality. The national scope of our study allows us to (a) quantify potential wetland losses and assess policy proposals in terms of consequences for achieving and maintaining no net loss and (b) estimate potential equilibrium adjustments crop acreage, commodity prices, farm income, and the regional distribution of farm income.

The organization of the article is as follows. We present our methodology in the next section, followed by results and conclusions. Estimates presented include wetland acreage exempted from swampbuster and CWA Section 404 under the twenty- one-day exemption, the acreage of exempted wetland that could be profitably converted to crop production, and long-run changes in commodity prices, crop acreage, and farm income. We also discuss prospects for achieving and maintaining no net loss and some potential environmental consequences of wetland conversion under the twenty-one-day exemption.

Methods Economic and environmental effects are estimated using a three-step process.

First, we estimate total wetland acreage that would be exempted from swampbuster and Section 404 under the twenty-one-day rule. Next, we estimated the acreage of exempted wetland that could be profitably farmed at expected (baseline) crop prices and production and conversion costs. This step provides a measure of the potential for wetland conversion, assuming no feedback effect on prices and costs from increased production due to wetland conversion. Finally, potentially convert-




ible wetland acreage is used to augment land supply in a national/ interregional model of U.S. agriculture to obtain estimates of long-run adjustment in commodity prices, crop acreage, and farm income.

Step 1: Estimating Exempted Acreage Exempted wetland is difficult to estimate precisely because of the configuration

of the data. Data on land use and condition (e.g., hydrology) is drawn from the National Resources Inventory (NRI) and Soil Interpretive Record (SIR) databases. NRI and SIR point data files are collected and maintained by the Natural Resources Conservation Service (NRCS) of the USDA and contain detailed data on land use and condition for more than 800,000 points nationwide. Whereas the twenty-one- day rule exempts land that is inundated for less than twenty-one consecutive days during the growing season, SIR classify soils according to the total annual days of inundation, not necessarily consecutive days or days during the growing season. For SIR, duration is considered short for soils that are typically inundated for less than seven days annually, long for total annual inundation of seven to thirty days, and very long for inundation of more than thirty days.

Acreage exempted from swampbuster under the twenty-one-day criterion is estimated as acreage (a) not currently exempted from swampbuster and (b) typically inundated for a total of up to thirty days annually. The estimate might overstate the exemption by including some land that is inundated for twenty-one or more consecutive days during the growing season. However, some land inundated for a total of twenty-one or more days annually might be inundated for less than twenty- one consecutive days during the growing season, reducing the magnitude of the potential bias.

Step 2: Estimating Wetland Conversion at Baseline Prices and Costs Potential wetland conversion under baseline conditions is estimated at nearly

50,000 wetland sites for which data are available from the NRI database. These individual decisions are aggregated to a regional level (using NRI acreage expansion factors) for use in subsequent economic modeling. Wetlands are considered potentially convertible when the net present value (NPV) of expected returns to crop production after conversion exceeds total costs of conversion (the NPV of return to land in its wetland condition plus drainage and clearing costs) by a threshold level. If 0k is a zero-one indicator, then

Ok = 1 iff max(NPVkj) - (NPVk + CC~) > max[r(NPVi + CCki), m]

and regional acreage (regions are defined below) is estimated as

AR = B kAk ke R

where NPVk is the estimated net present value of returns after drainage to crop j (corn, sorghum, barley, oats, wheat, rice, cotton, or soybeans) at site k, NPVk is the opportunity cost of land in its wetland condition use (forestry, pasture, or crop production) at site k, CCk is the cost of draining and clearing land for crop production at site k, AR is the estimated acreage of potentially convertible wetland in region R, Ak is the NRI acreage expansion factor for site k, r is the landowner's




discount rate, and m is the minimum return per acre needed to justify the investment in drainage and clearing for crop production.

Expected commodity prices are prices projected for 2001 in the current USDA long-term agricultural baseline (USDA 1997c). Price projections for 2001 were chosen because they approximate the mean of current baseline price projections over the next ten years. Net present values are defined over a finite time horizon that varies by wetland condition land use. For forested wetland, the time horizon is the length of a single forestry rotation. Other sites are assigned a ten- to twenty-year time horizon, depending on the drainage technology used. The minimum return requirement effectively excludes sites at which conversion might not be undertaken because of low overall returns. A full description of data used for step 2 of the analysis is presented in the appendix.

Potentially convertible wetland is estimated for two scenarios. In the low-conversion scenario, we consider only that land that NRI indicates has some probability of conversion to crop production in the foreseeable future. We assume a discount rate (r) of 6%3 and a minimum return requirement (m) of $100 per acre. NRCS field technicians who collect NRI data assess the probability that noncropland sites (including wetlands) will be converted to crop production on the basis of potential agricultural returns, the cost of developing wetlands for crop production, and whether similar land had been converted to crop production in the past three years (USDA 1991). Because NRI estimates of conversion potential are based in part on economic considerations that can vary over time, we also estimated a high-conversion scenario, assuming a discount rate of 6% and a minimum return of $500 per acre for land for which NRI indicates no conversion potential and a 6% discount rate and a $100 minimum return otherwise. In both cases, wetland sites that are projected to be enrolled in the Conservation Reserve Program (CRP) are excluded from consideration.4

Our approach to estimating potential wetland conversion assumes that landowners will drain wetland when it is profitable to do so. An alternative approach would be to econometrically estimate a model relating the probability of wetland conversion to economic and resource conditions. However, in the present context a useful econometric estimation would require data on wetland conversion under economic and policy conditions similar to what would be created under the FAIR Act with the twenty-one-day exemption in place. No such data are available. We note that Parks, Kramer, and Heimlich compare these two approaches in estimating the cost of enrolling acreage in a wetland reserve program. They found that, for a given level of payment, the econometric model predicted fewer enrolled acres. One issue not addressed by Parks et al. that might influence observed producer behavior is subsequent equilibrium adjustments in commodity and factor markets. We take up estimation of these adjustments in step 3.

Step 3: Estimating Economic Effects Commodity price, crop acreage, and farm income effects of the twenty-one-day

exemption are derived as comparative static impacts of augmenting land supply in forty-five land resource regions in the U.S. Regional Agricultural Sector Model (USMP).5 An agriculture sector spatial equilibrium model as described in McCarl and Spreen, USMP incorporates agricultural commodity supply, use, and policy




measures (House). USMP has been applied to project the effects on U.S. national and regional agriculture of changes in export levels and variability (Miller et. al.), trade agreements (Burfisher, House, and Langley), imports (Spinelli et al.), input taxes (Peters, McDovells, and House), irrigation policy (Homer et al.), ethanol production (House et al.), wetlands policy (Heimlich et al.), sustainable agriculture policy (Faeth), and various other policy and program scenarios.

USMP models the production of ten crops: corn, sorghum, oats, barley, wheat, rice, cotton, soybeans, hay, and silage. Sixteen primary livestock production enter- prises are included, the principal being dairy, swine, beef cattle, and poultry. Co- efficients in crop and livestock enterprise budgets were developed from USDA NRI, Cropping Practices Survey (CPS), and Farm Costs and Returns Survey (FCRS) data. CPS and FCRS data are collected and analyzed by the Economic Research Service and National Agricultural Statistical Service of the USDA. Several dozen processed and retail products are included in the model structure, the principal being dairy products, pork, fed and nonfed beef, poultry, soy meal and oil, livestock feeds, and corn milling products. Acreage, commodity supply/use, conservation reserve program acreage, prices, production practices, and so on are validated exactly to USDA baseline projections for 2001 (USDA 1997c) and corresponding geographic information. For example, USMP's base U.S. corn acreage planted in 2001 equals the USDA baseline projection and corn acreage in each model region/practice stratum is determined by share information from NRI and CPS regional data. On the demand side, domestic use, exports, ending stocks, and price levels for crop and livestock commodities and most processed or retail products are endogenously determined within the model structure with domestic consumption, commercial stock, export, and other demand functions specified with elasticities from the FAP- SIM econometric simulation model (Green and Price).

We use regional acreage of potentially convertible wetland to shift land supply curves in the USMP model. We assume that converted wetlands might then be cropped at the regional marginal cost of production that existed prior to the wetlands conversion. Comparative static adjustments to the wetland delineation acreage shifts explain how the sector changes (through both aggregate indicators such as U.S. farm income and detailed indicators such as acreage in corn-bean rotation in the central Corn Belt) between the base period and several years later, when the change has worked itself out and the sector returns to equilibrium. USMP acreage planted/ commodity supply response uses a positive mathematical programming formulation (Howitt) with U.S. aggregate commodity supply response calibrated to supply response elasticities from the FAPSIM model. Responses in individual region, tillage practice, rotation, and other strata follow nested adjustment functions that are part of the PMP calibration and sum up to aggregate response. (No bounds or flexibility constraints are used.)

Results

Wetland Acreage Exempted We estimate that 82.7 million acres would be exempted from swampbuster under

the twenty-one-day exemption, about 90% of remaining FSA wetlands. Enactment of the twenty-one-day exemption would have been the functional equivalent of




Table 1. Wetland acres exempted and potentially drained at baseline prices under twenty-one-day exemption

Acreage Profitable to Convert

Farm Production Exempted Acreage at Baseline Prices (Millions)a

Region (Millions) Low Conversion High Conversion

Northeast 12.3 0.5 0.9 Lake States 19.5 0.6 1.4 Corn Belt 3.9 0.4 0.5 Northern Plains 6.3 0.8 0.8 Appalachia 6.0 0.7 1.9 Southeast 17.6 0.9 3.8 Delta States 7.9 1.4 2.3 Southern Plains 3.6 0.2 0.4 Mountain States 3.3 * * Pacific Coast 2.4 * * U.S. total 82.7 5.6 12.0

a Asterisks indicate fewer than 50,000 acres.

ending the swampbuster program and regulation of agricultural wetland conversion under CWA Section 404. Table 1 provides exempted wetland acreage by USDA farm production region. Exempted wetlands are concentrated in the Lake States (24%), the Southeast (21%), the Northeast (15%), the Delta States (9%), and the Northern Plains (8%).6

Potential Wetland Conversion Potentially convertible wetland and long-run equilibrium acreage changes, by

farm production region, are reported for low- and high-conversion scenarios in table 1. For the low-conversion scenario, wetland conversion or improved drainage for crop production would be profitable at baseline prices on an estimated 5.6 million acres (6.1% of wetlands subject to swampbuster): 868,000 acres of cropped wetland, 921,000 acres of pasture or rangeland, 3.4 million acres of forested wetland, and 436,000 acres of land classified as swamp, marsh, and so on.7 For the high-conversion scenario, we estimate that conversion is profitable on 12 million acres (13.1% of wetlands subject to swampbuster): 868,000 acres of cropped wetland, 1.2 million acres of pasture or rangeland, 8.8 million acres of forested wetland, and 1.1 million acres of land classified as swamp, marsh, and so on.

Regionally, the largest increases in potential wetland conversion in the high- conversion scenario over the low-conversion scenario are for forested wetlands in the Appalachian and Southeast farm production regions. There is little or no change in wetland acreage that is profitable to convert in the Northern Plains, Mountain States, or Pacific Coast States. In the Southeast, for the high-conversion scenario, 3.8 million acres of wetland are estimated to be potentially profitable in crop production-a huge pool of land when compared to a total cropland base of roughly 18 million acres (Daugherty). In the Appalachian region, the high-conversion scenario estimate of 1.9 million acres of potentially convertible wetland is a somewhat




smaller, but still large, proportion of the roughly 29 million acres of existing cropland in that region.

Sensitivity Analysis of Potential Wetland Conversion We tested the sensitivity of the model to changes in expected agricultural com-

modity prices and discount rates. To test price sensitivity, we estimated potential conversion in the low-conversion scenario using actual prices for 1992 and prices projected for baseline year 1998 and compared these estimates to results described above (using prices projected for baseline year 2001). Prices for 1992 are generally 20% to 25% lower than baseline projections for 2001, except soybeans (15% lower) and rice (45% lower), whereas 1998 price projections fall between 1992 prices and 2001 price projections for all eight commodities. Discount rate sensitivity is tested by estimating potential conversion in the low-conversion scenario for discount rates of 4%, 8% and 10% in addition to the initial rate of 6%.

Estimates of potential wetland conversion are quite sensitive to changes in expected commodity prices. Potentially convertible wetland is estimated to be 3.4 million acres using price projections for 1998 and only 1.3 million acres using actual 1992 prices compared with 5.6 million acres using price projections for 2001 (as reported above). The sensitivity is not especially surprising given that an expansion of the land base into wetland area would be expected only during a period in which agriculture is quite profitable. Moreover, the difference between 1.3 million acres and 5.6 million acres appears large because it is the sensitivity of the marginal response of cropland acreage to price changes. When viewed in terms of the change in the overall cropland base, the difference is relatively small. Draining 5.6 million acres of wetland is a 0.87% increase in cropland, whereas draining 1.3 million acres is a 0.40% increase.

Potential wetland conversion is somewhat less sensitive to a change in the real interest rate. Using projected commodity prices for baseline year 2001, estimates range from 6.6 million potentially convertible wetland acres using a real discount rate of 4% to 4.3 million acres with a 10% rate of discount. One reason the model is less sensitive to interest rates than to agricultural commodity prices is that the net present values of pasture and forestry also adjust. A decline in the discount rate increases the net present value of all three land uses.

Long-Run Economic Effects Long-run economic effects8 are reported as changes from the prices, crop acre-

ages, and farm income anticipated by the USDA baseline. As new land is brought into crop production, commodity prices decline. The net amount of additional land used for crop production is a market equilibrium result of acreage adjustment to the point at which marginal returns from production equal marginal costs of production. Commodity price results depend on both the economics of production (supply functions) and demand across different markets (feed, food, ending stocks, and export use).

Table 2 provides baseline prices and absolute and percentage changes in price due to wetland conversion. All eight commodity prices are reduced in both scenarios. In the low-conversion scenario, percentage reductions are lowest for wheat (-0.7%) and barley (-0.8%) and largest for rice (-5.3%), soybeans (-3.1%), and




Table 2. Long-run price effects of twenty-one-day exemption

Price Change Baseline PrBaselinea Low Conversion High Conversion Pricea

Crop Unit ($/Unit) ($/Unit) Percentage ($/Unit) Percentage Corn bu 2.80 -0.07 -2.5 -0.19 -6.8 Sorghum bu 2.50 -0.07 -2.8 -0.18 -7.2 Barley bu 2.60 -0.02 -0.8 -0.05 -1.9 Oats bu 1.70 -0.04 -2.4 -0.17 -10.0 Wheat bu 4.30 -0.03 -0.7 -0.06 -0.9 Rice cwt 10.31 -0.55 -5.3 -0.96 -9.3 Soybeans bu 6.45 -0.20 -3.1 -0.48 -7.4 Cottonb lb - - -2.9 - -6.5

a Baseline prices for 2001 are from USDA (1997b). b The USDA is prohibited from publishing cotton price projections.

cotton (-2.9%). These results are not surprising because potentially convertible wetlands are concentrated in the South, where rice, soybeans, and cotton are major crops, and because wheat and barley are grown in regions with comparatively few potentially convertible wetlands.

In terms of overall cropland acreage, the low-conversion scenario would result in a 2.1 million acre increase in cropland acreage, 0.6% higher than the baseline acreage of 328.3 million acres. In the high-conversion scenario, total crop acreage would rise by 4.5 million acres from baseline, a 1.4% increase. In both scenarios, the long-run acreage increase is about 38% of the potentially convertible wetland acreage provided to the USMP model.

However, gross conversion of wetland to crop production might not be limited to the long-run increase in crop acreage. Wetlands might be initially converted and then removed from production as prices fall, or other marginal land that had been in production might be removed from production as prices fall. At long-run equilibrium (lower) prices, little of the wetland acreage estimated to be profitable to convert becomes unprofitable, suggesting that converted wetlands are likely to remain in production while other marginal land is pushed out. For the low-conversion scenario, 5.1 million wetland acres are still profitable at long-run equilibrium prices, 91% of the 5.6 million acres profitable at baseline prices. For the high-conversion scenario, 9.4 million acres remain profitable at long-run equilibrium prices, 78% of the 12 million acres profitable at baseline prices. Even if converted wetlands were removed from production, there is little reason to believe that they would be effectively restored to wetland condition.

Nationally, long-run aggregate net farm income drops by more than $1.6 billion in the low-conversion scenario and by about $3.0 billion in the high-conversion scenario, reductions of 2.4% and 4.6%, respectively. Note that deficiency payment, supply control, export promotion, and other features of pre-FAIR farm legislation that served to mitigate the magnitude of income declines are no longer authorized.

Regional changes in crop acreage and farm income are detailed in table 3 and table 4, respectively. In both scenarios, aggregate farm income also declines in most




Table 3. Long-run crop acreage change under twenty-one-day exemption

Farm Production Baseline Acreage Cropland Acreage Change (Millions) Region (Millions) Low Conversiona High Conversion

Northeast 14.9 0.4 0.6 Lake States 39.0 0.1 0.3 Corn Belt 92.1 -0.3 -1.2 Northern Plains 72.3 0.0 -0.6 Appalachia 18.0 0.5 1.5 Southeast 7.1 0.8 3.1 Delta States 15.3 0.9 1.5 Southern Plains 35.4 -0.2 -0.4 Mountain States 25.2 * -0.1 Pacific Coast 9.1 * -0.1 U.S. total 328.3 2.1 4.5

a Asterisks indicate fewer than 50,000 acres.

farm production regions, indicating that increased returns on converted acres (the output effect) is outweighed in most regions by reduced returns due to price declines. However, the Southeast, Delta, and Appalachian regions enjoy small increases in aggregate net farm income. These regions have large amounts of potentially convertible wetland but relatively small existing cropland bases on which to suffer losses due to the price effect. The largest aggregate reduction in income is suffered in the Corn Belt, where few unconverted wetlands remain and the existing cropland base is large and highly productive. Substantial declines in farm income also occur in the Northern Plains, Southern Plains, and Lake States.

Table 4. Long-run farm income changes under twenty-one-day exemption

Farm Production Baseline Farm Farm Income Change (Million $) Region Incomea (Million $) Low Conversion High Conversion Northeast 3,956.3 -17.0 -21.4 Lake States 7,546.6 -204.7 -373.3 Corn Belt 19,190.2 -807.1 -1,918.8 Northern Plains 9,372.7 -362.9 -814.0 Appalachia 5,105.4 11.6 132.9 Southeast 2,539.1 146.2 696.0 Delta States 4,491.0 58.6 -27.6 Southern Plains 6,210.1 -224.2 -411.7 Mountain States 2,906.5 -73.2 -108.6 Pacific Coast 3,852.4 -100.3 -137.5 U.S. total 65,170.2 -1,572.9 -2,984.0

a Baseline farm income is from USDA (1997b).




Environmental Effects Even the most conservative wetland loss scenario outlined above (a long-run

loss of 2.1 million wetland acres in the low-conversion scenario) would be a serious blow to achieving and maintaining no net loss of wetlands. Converting 2.1 million acres, even over a period of years, would be a significant increase in gross conversion of wetland acreage from recent levels. Between 1982 and 1992, gross conversion of wetland for crop production was about 309,000 acres (Heimlich and Melan- son). Conversion of 2.1 million acres over a ten-year period would represent a sevenfold increase in the rate of wetland conversion for agriculture, although it would be less than one-half of the 593,000 acres converted annually, on average, between the mid 1950s and mid 1970s (Frayer et al.). This level of wetland conversion would also far exceed currently planned federally funded efforts to restore wetlands previously converted to agricultural production. The Wetlands Reserve Program (WRP) funds wetland restoration efforts but is capped at a maximum enrollment of 975,000 acres (U.S. Congress). Roughly 533,000 acres are already enrolled, leaving a total additional federally funded restoration effort of 442,000 acres (USDA 1997b).

The largest environmental effect of adopting the twenty-one-day exemptions would be on bottomland hardwood forests in the Delta, Appalachian, and South- east farm production regions. These wetlands store flood water, help maintain water quality, and provide winter habitat for waterfowl. In the lower Mississippi alluvial plain, about 80% of forested wetlands have already been lost, mostly to crop production (Dahl). Although the acreage of cropped wetland that would be converted is small, much of it is located in the prairie pothole region (eastern Montana, North Dakota, South Dakota, Minnesota, and Iowa), an important area for waterfowl breeding. In some years, prairie pothole wetlands produce up to one- half of U.S.-bred migratory waterfowl (Kantrud, Krapus, and Swanson; Stewart). About 50% of these wetlands have already been lost, mostly to crop production (Dahl).

Conclusion We have described a unique method of determining the potential environmental

and economic consequences of changes in resource quality-related land use restrictions. Our approach incorporates site-specific resource data in a national level analysis of the agricultural economy, providing broad-based results about both the economic and the environmental consequences of policy. Our method is demon- strated in the context of wetland delineation changes but could be extended to a wider range of land use policy questions. As we discussed in the description of step 2 of the modeling procedure, some evidence suggests that an econometric model might produce smaller estimates of potentially convertible wetland acreage; however, the data resources required for an appropriate economic estimation are not available. Moreover, our estimates incorporate equilibrium adjustments in commodity and factor markets that moderate the total change in cropland acreage.

The analysis suggests two conclusions. First, it is not possible to draw broad conclusions about the effect of resource policy without incorporating detailed information on resource quality into the analytic framework. Our research indicates that the potential agricultural productivity of unconverted wetlands varies significantly. The use of average productivity in previous representative farm models




masked the existence of wetlands with high potential productivity that could be profitably converted to crop production. Although conversion of wetlands for agriculture has slowed significantly in recent decades, a relaxation of swampbuster and CWA Section 404, coupled with favorable commodity prices, could prompt a return to substantially higher rates of wetland conversion. Although exact environmental effects (in terms of lost wetland functions and values) cannot be accurately quantified, achieving or maintaining no net loss could be jeopardized.

Second, local or regional models cannot provide information on commodity price and farm income effects that are critical to policy analysis. Our research shows that price and farm income effects can be significant and that farm income effects can vary widely between regions. In the wetlands example, potential declines in farm income of between 2.4% (in the low-conversion scenario) and 4.6% (in the high-conversion scenario) demonstrate that farmers and landowners who do not drain wetlands have a significant economic stake in wetland policy. Farmers and agricultural landowners who actually drain wetlands for crop production are likely to see their incomes rise. However, these individuals are likely to be a small minority of farmers and agricultural landowners. Other producers would suffer reduced incomes due to lower commodity prices. Although land use restrictions of any kind have never been popular among farmers or landowners, our analysis shows that reducing the coverage of swampbuster and CWA Section 404 by relaxing wetland delineation would be contrary to the economic interests of a large majority of farmers and landowners.

Disclaimer The views expressed here are those of the authors and not necessarily those of the USDA.

Endnotes ' Although it would be desirable to estimate both the costs and the benefits of wetland protection,

such an approach might not be currently practical. Many wetland benefits flow from nonmarketed services, making valuation difficult when compared to marketed goods. For example, wetlands help maintain water quality by filtering water of pollutants, reduce flooding and recharge groundwater by storing runoff, and serve as critical habitat for a wide variety of wildlife (Novitski, Smith, and Fretwell). Existing wetland valuation studies (see Leitch and Ekstrom, Leitch and Ludwig) report a wide variation in wetland values and are concentrated on coastal wetlands rather than wetlands potentially subject to conversion for agriculture. Some authors have expressed concern that the wide variation in wetland values might be due to flaws in estimation methods rather than real variations in physical attributes and locational characteristics that underlie wetland valuation (Anderson and Rockel, Shabman and Batie). Very little of the currently available evidence regarding the value of specific wetlands might be trans- ferable to other wetlands (Scodari), especially to inland wetlands with agricultural potential.

2 For currently used U.S. Army Corps of Engineers and Natural Resources Conservation Service (USDA) wetland delineation manuals, the growing season is defined as the portion of the year when the soil temperature 19.7 inches below the soil surface is higher than biological zero (410F) (National Research Council).

3 The 6% real interest rate is roughly equal to the real rates of interest on long-term farm (real estate) loans in recent years from the Farm Credit System (FCS). Since 1989, the average annual real rate of interest on FCS real estate loans has varied between 5.23% and 6.73% (the real rate of interest is the nominal rate as reported in USDA 1997a) less the change in the gross domestic product implicit price deflator. Moreover, current income return to agricultural assets has consistently been less than 6%. Between 1960 and 1991, current income return to agricultural assets averaged 4.1%, calculated as ratio of current returns to value of farm assets, in constant dollars, from ERS data (USDA 1993). Total return (current income plus real capital gains) averaged 4.6% over the same period. Previous studies of investment in land use conversion have recognized these historically low rates of return and used discount




factors of 4% (USDA 1988) and 5% (Kramer and Shabman). Hardie recognizes 4% as a long-term return to agricultural assets and uses discount rates ranging from 3% to 10%.

4 NRI points that are most likely to be enrolled in a 36.4-million-acre CRP, given potential economic and environmental benefits, projected by Tim Osborn, Economic Research Service, USDA.

5 Regions are specified as the intersection of the ten USDA farm production regions and the twenty- six USDA land resource regions (USDA 1981).

6 USDA farm production regions: Appalachian (Kentucky, North Carolina, Tennessee, Virginia, and West Virginia); Corn Belt (Illinois, Indiana, Iowa, Missouri, and Ohio); Delta (Arkansas, Louisiana, and Mississippi); Lake States (Michigan, Minnesota, and Wisconsin); Mountain (Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Utah, and Wyoming); Northeast (Connecticut, Delaware, Maine, Mary- land, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, and Vermont); Northern Plains (Kansas, Nebraska, North Dakota, and South Dakota); Pacific (California, Oregon, and Washington); Southeast (Alabama, Florida, Georgia, and South Carolina); and Southern Plains (Oklahoma and Texas).

7 Improving drainage on the 868,000 acres of cropped wetland is the equivalent of adding an additional 377,000 acres of new cropland. Conversion of 299,000 acres of noncropped wetland for crop production for which double cropping is profitable is the equivalent of adding two acres to cropland supply for each drained acre, a total of 540,000 acres. Thus, the 5.61 million actual acres are an acreage equivalent of 5.36 million acres for the purpose of augmenting land supply in USMP.

8 Sensitivity analysis performed in step 2 of the analysis is not carried forward to the long-run results. However, it is clear that changes in prices and interest rates will affect long-run economic effects. In particular, significantly lower prices (as in the 1992 price scenario), which result in far lower potential conversion, will also result in minimal price and farm income effects.

References Anderson, R, and M. Rockel. "Economic Valuation of Wetlands." Discussion Paper No. 65, American

Petroleum Institute, Washington DC, 1991. Burfisher, M.E, R.M. House, and S.V Langley. "Effects of a Free Trade Agreement on U.S. and

Southern Agriculture." S. J. Agr. Econ. 24(July 1992):61-78. Conservation Foundation. Protecting America's Wetlands: An Action Agenda. The Final Report of the

National Wetlands Policy Forum. Washington DC: Conservation Foundation, 1988. Dahl, T.E. Wetlands Losses in the United States, 1780's to 1980's. Washington DC: Department of the

Interior, U.S. Fish and Wildlife Service, 1990. Dahl T.E, and G.J. Allord. "History of Wetlands in the Conterminous United States, Mid-1970's to

Mid-1980's. National Summary on Wetland Resources. J.D. Fretwell, J.S. Williams, and P.J. Redman, eds., pp. 79-86. Washington DC: U.S. Department of the Interior, U.S. Geological Survey, 1996.

Dahl, T.E, and C.E. Johnson. Status and Trends of Wetlands in the Conterminous United States, Mid-1970's to Mid-1980's. Washington DC: U.S. Department of the Interior, Fish and Wildlife Service, 1991.

Daugherty, A.B. Major Uses of Land in the United States, 1992. Agr. Washington DC: USDA, Economic Research Service, Agr. Econ. Rep. No. 723, 1995.

Faeth, P. "Growing Green: Enhancing the Economic and Environmental Performance of U.S. Agri- culture." World Resources Institute, Washington DC, April 1995.

Frayer, W.E, T.J. Monahan, D.C. Bowden, and FA. Graybill. Status and Trends of Wetlands in the Conterminous United States, 1950's to 1970's. Fort Collins CO: Colorado State University, Department of Forest and Wood Sciences, 1983.

Green, R.C, and J.M. Price. "Estimates of Short-Run Price Elasticities for Major U.S. Field Crop and Livestock Commodities Using FAPSIM." Unpublished, Economic Research Service, USDA, December 1987.

Hardie, I.W. "Comparative Rents for Farmland and Timberland in a Subregion of the South." S. J. Agr. Econ. 16(December 1984):45-53.

Haynes, R.W. An Analysis of the Timber Situation in the United States: 1989-2040. Fort Collins CO: USDA, Forest Service, Rocky Mountain Forest and Range Experiment Station, Gen. Tech. Rep. RM-199, 1990.

Heimlich, R.E, M.B. Carey, and R.J. Brazee. "Beyond Swampbuster: A Permanent Wetland Reserve." J. Soil and Water Conserv 44(September-October 1989):445-50.

Heimlich, R.E, and L.L. Langner. Swampbusting: Wetland Conversion and Farm Programs. Washington DC: USDA, Economic Research Service, Agr. Econ. Rep. No. 551, 1986.

Heimlich, R.E, and J. Melanson. "Wetlands Lost, Wetlands Gained." Nat. Wetlands Newsletter 173(1995):1-5.

Heimlich, R.E, K.D. Wiebe, and R. Claassen. "Sustaining Our Wetland Gains." Nat. Wetlands News- letter 19, 4(1997):5-9.




Heimlich, R.E, K.D. Wiebe, R. Claassen, and R.M. House. "Recent Evolution of Environmental Policy: Lessons from Wetlands." J. Soil and Water Consern 52, 3(1997):157-61.

Homer, G, S.A. Hatchett, R.M. House, and R.E. Howitt. "Impacts of San Joaquin Valley Drainage- Related Policies on State and National Agricultural Production." National Impact of Drainage-Related Poli- cies. University of California and San Joaquin Valley Drainage Program, 1990.

House, R.M. "USMP Regional Agricultural Model." NED working paper, Economic Research Service, USDA, Washington DC, June 1987.

House, R, M. Peters, H. Baumes, and W.T. Disney. Ethanol and Agriculture: Effect of Increased Production on Crop and Livestock Sectors. Washington DC: USDA, Economic Research Service, Agr. Info. Bull. 667, May 1993.

Howitt, R.E. "Positive Mathematical Programming." Amer. J. Agr. Econ. 77, 2(1995):329-42. Kantrud, H.A, G.L. Krapu, and G.A. Swanson. Prairie Basin Wetlands of the Dakotas: A Community

Profile. Washington DC: U.S. Department of the Interior, Fish and Wildlife Service, Biological Rep. No. 85(7.28), 1989.

Kramer, R.A. and L. Shabman. "The Effects of Agricultural and Tax Policy Reform on the Economic Return to Wetland Drainage in the Mississippi Delta Region." Land Econ. 69, 3(1993):249-62.

Leitch, J.A, and B.L. Ekstrom. Wetland Economics and Assessment: An annotated Bibliography. New York: Garland Publishing Co., 1989.

Leitch, J.A, and H.R. Ludwig, Jr. Wetland Economics 1989-1993, a Selected, Annotated Bibliography. Westport CT: Greenwood Press, 1995.

McCarl, B, and T. Spreen. "Price Endogenous Mathematical Programming as a Tool for Sector Anal- ysis." Amer. J. Agr. Econ. 62, 1(1980):86-102.

McClure, J.P, and H.A. Knight. Empirical Yields of Timber and Forest Biomass in the Southeast. Asheville NC: USDA, Forest Service, Southeastern Forest Research Station, Res. paper SE-245, 1984.

Miller, T, J. Sharples, R. House, and C. Moore. Increasing World Grain Market Fluctuations: Implications for UlS. Agriculture. Washington DC: USDA, Economic Research Service, Agr. Econ. Rep. No. 541, October 1985.

National Research Council. Wetlands: Characteristics and Boundaries. Washington DC: National Re- search Council, 1995.

Norris, Frank, ed. Timber Mart South Yearbook. Highlands NC: Timber Marts, Inc., 1986 (plus annual updates).

Novitski, R.P, R.D. Smith, and J.D. Fretwell. "Wetland Functions, Values, and Assessment." National Summary on Wetland Resources. J.D. Fretwell, J.S. Williams, and P.J. Redman, eds., pp. 79-86. Washington DC: U.S. Department of the Interior, U.S. Geological Survey, 1996.

Parks, PJ. "Explaining 'Irrational' Land Use: Risk Aversion and Marginal Agricultural Land." J. Environ. Econ. and Manage. 28(January 1995):34-27.

Parks, P.J, R.A. Kramer, and R.E. Heimlich. "Simulating Cost-Effective Wetlands Reserves: A Com- parison of Positive and Normative Approaches." Natur. Resour. Modeling 9(Winter 1995):81-95.

Pavelis, G.A. "Conservation Program Investment in the United States: Guidelines for RCA Studies." Unpublished, Economic Research Service, USDA, March, 1994.

Peters, M.E, EH. McDowell, and R.M. House. "Environmental and Economic Effects of Taxing Nitrogen Fertilizer." Paper presented at the annual meeting of the American Agricultural Economics Association, Toronto, July 1997.

Pierce, EJ, W.E. Larson, R.H. Dowdy, and W.A.E Graham. "Productivity of Soils: Assessing Long- Term Changes Due to Erosion." J. Soil and Water Conserz. 38(January-February 1983):39-44.

Scodari, EE Measuring the Benefits of Federal Wetland Programs. Washington DC: Environmental Law Institute, 1997.

Shabman, L.A, and S.S. Batie. "Socioeconomic Functions and Values of Wetlands: A State-of-the-Art Review." U.S. Army Corps of Engineers Tech. Rep. Y-85, Waterways Exp. Sta., Vicksburg MS, 1985.

Spinelli, F, W.T. Disney, J. Blackwell, and H. Metcalf "U.S. Economic Impact of Uncooked Beef Imports from Argentina." Paper presented at the annual meeting of the American Agricultural Econom- ics Association, San Antonio TX, July 1996.

Stavins, R.N. and A.B. Jaffee. "Unintended Impacts of Public Investments on Private Decisions: The Depletion of Forested Wetlands." Amer. Econ. Rev 80(1990):337-52.

Stewart, R.E, Jr. "Wetlands as Bird Habitat." National Water Summary on Wetland Resources. J.D. Fre- twell, J.S. Williams, and P.J. Redman, eds; pp. 49-56. Washington DC: U.S. Department of the Interior, U.S. Geological Survey, 1996.

Tolman, J. "How We Achieved No Net Loss." Natl. Wetlands Newsletter 19, 4(1997):1-22. U.S. Congress. Federal Agricultural Improvement and Reform Act of 1996. Public Law 104-127.

Washington DC: April 4, 1996. U.S. Department of Agriculture, Economic Research Service. Agricultural Income and Finance: Situa-

tion and Outlook. AIS-64, Washington DC, February 1997a. -. Economic Indicators of the Farm Sector, National Financial Summary, 1991. ECIFS 11-1, Washington

DC, January 1993.




-. "Wetland Reserve Programs." AREI Updates 1997. No. 6, November 1977b. U.S. Department of Agriculture, Forest Service. The South's Fourth Forest: Alternatives for the Future.

Forest Resource Rep. No. 24, Washington DC, 1988. U.S. Department of Agriculture, National Agricultural Statistical Service, Agricultural Statistics

Board. Agricultural Prices: Annual Summary. Washington DC, various issues. -. Crop Production. Washington DC, various issues. U.S. Department of Agriculture, Natural Resources Conservation Service. Summary of the Wetlands

Reserve Program and Emergency Wetlands Reserve Program. Washington DC, August 22, 1996. U.S. Department of Agriculture, Soil Conservation Service. Instructions for Collecting 1992 National

Resources Inventory Sample Data. Washington DC, 1991. -. Land Resource Regions and Major Land Resource Areas of the United States. Agriculture Handbook

296, Washington DC, December 1981. -. Second Resources Conservation Act Appraisal. Washington DC, 1987. U.S. Department of Agriculture, World Agricultural Outlook Board. Agricultural Baseline Projections

to 2005, Reflecting the 1996 Farm Act. Staff Rep. WAOB-97-1, Washington DC, February 1997c. U.S. Department of the Interior, Fish and Wildlife Service. The Impact of Federal Programs on Wetlands,

Volume 1. Washington DC, 1988. U.S. House of Representatives, Committee on Transportation. Clean Water Act Amendments of 1995.

House Report 104-112, Washington DC, 3 May 1995.

Data Appendix This appendix details data used to screen NRI wetland sites for profitability in crop production.

Data used to calculate crop returns are generally consistent with data used to calibrate the USMP model, although the screening model uses more extensive spatial disaggregation where possible. Also, the screen utilizes estimates of return to pasture and bottomland hardwood forestry that USMP does not provide.

Crop returns Eight commonly grown crops are considered: barley, corn, cotton, oats, rice, sorghum, soybeans, and

wheat, including summer fallow rotations and double cropping, where appropriate. County-level crop prices are devised as follows. Prices from the posted county price database for market year 1994 (Murray, personal communication) are divided by U.S. average prices from Agricultural Prices: Annual Summary, 1994 (USDA) to obtain a relative price for each county. Relative prices are multiplied by USDA baseline projections (USDA 1997b) for national average prices to obtain prices used in the simulation. Site-specific crop yields are devised by multiplying county average crop yields for 1991-95, obtained from Crop Production (USDA), by an index of relative productivity calculated from the productivity index (PI) developed by Pierce et al. and calculated from the SIR database. The relative productivity index for a particular crop is the ratio of site-specific PI to average PI for sites within the county where NRI cropping history shows production of that crop.

Crop production costs are estimated by the Economic Research Service, USDA, from the FCRS data. Estimates of variable and fixed costs of production by crop and by state are used. The most recent state data, from 1989, is updated to 1995 by multiplying state costs by the ratio of 1995 to 1989 production costs at a regional level. We assume that the purchase cost of land is sunk and will not enter into the conversion decision and that unpaid operator labor costs will not increase because of wetland drainage.

Pasture returns State average pasture rental rates for 1994 are adjusted to site-specific conditions using the PI and

to USDA baseline economic conditions using estimates of the percentage change in pasture rental rates from a 1% change in beef prices and costs. These estimates are obtained from regional regression models of pasture rents on beef prices and costs:

R, = H- H x,E

which can be written as

In Rs = C dsln as + W fln Xi + In E

where RPs is the state-average pasture rental rate for state s, ds is a dummy variable for state s, the X is a beef price and cow-calf production cost index, the Ina and 0 are parameters to be estimated, and lnE is an error term. Regions are north-central states (Corn Belt and lake state farm production regions), the




Table A. Estimated percentage change in pasture rental rates

North South Plains West

Beef price index 0.92 0.50 1.11 -0.31 (8.51) (6.09) (6.64) (-0.44)

Cow-calf costs -1.42 -1.16 -2.09 -0.82 (-22.72) (-22.34) (-13.12) (0.72)

Adjusted R2 0.85 0.74 0.69 0.89

Note: t-ratios are in parentheses.

South (the Southeast, Delta, and Appalachian farm production regions), Plains states (Northern and Southern Plains), and the west (Mountain and Pacific Coast States).

Parameter estimates (which are also the elasticities in this specification) are reported in table A. In the West, parameter estimates for the beef price index and cow-calf cost variables are not significantly different from zero, perhaps because of the large amounts of federal land on which grazing fees are established by nonmarket procedures. For western states, average pasture rental rates are assumed to be constant at 1994 levels. Estimated pasture rental rates average $13.06 per acre but range up to $66 per acre. Beef price data are from various issues of Agricultural Prices: Annual Summary (USDA). Cow- calf production costs were estimated by the Economic Research Service, USDA, from Farm Costs and Returns Survey data.

Forest returns Returns to bottomland hardwood rotations are calculated for thirteen southern states. In the North,

where rotation lengths are considerably longer than in the South, forestry is assumed to be a residual land use; that is, wetland that cannot be profitably drained for other uses is retained in forested wetland, although some management might eventually be undertaken to encourage desirable species (Luppold, personal communication). In western states, data limitations preclude estimation of forest opportunity costs.

Compared with crop production and pasture, calculation of forestry returns involved considerable uncertainty. In part, this uncertainty derives from data limitations. Yield data, for example, are available only on a regional level (although regional yields are available for various productivity classes and are assigned to sites on the basis of site productivity). Uncertainty is also a function of unobserved factors that might influence landowner decisions involving forestry land use. Because of the long-term nature of the forestry investment, the role of price expectations, degree of risk aversion, and cash flow are potentially more important than for agricultural land use. Parks shows that each of these factors can reduce the net present value of benefits from forestry land use relative to agriculture.

Bottomland hardwood yields (oak-cypress-gum stands) are from McClure and Knight. Yields for high-medium-, and low-productivity sites are matched to NRI sites using the site index (SI) from the SIR database. Soils with an SI less than 60 are classified as low, SI 60 to 78 as medium, and SI above 78 as high. Hardwood sawtimber and pulpwood prices are from Norris. Regeneration costs are from the USDA (1988). Base prices are a three-year average and are adjusted up by 1.5% per year to match USDA Forest Service projections (Haynes).

Returns are calculated for a single rotation of thirty to forty years, depending on expected prices and site productivity. Rotation length is chosen on the basis of maximum NPV. Timber is assumed to be harvested in a clear-cut operation yielding both sawtimber and pulpwood. Returns are estimated for a 6% discount rate. The average NPV of expected returns to bottomland hardwood forestry is $137 per acre but ranges up to $442 per acre, depending on site productivity and stumpage prices.

Drainage and clearing costs The costs of drainage, annual drainage maintenance, and land clearing were also estimated by local

technical experts (USDA 1987) by major land resource area and state. These costs were adjusted to 1995 levels using the index of purchases of nonresidential farm structures (Pavelis).



estimating the effects of relaxing agricultural land use restrictions: wetland delineation in the...

Documents