This is the accessible text file for GAO report number GAO-03-418 
entitled 'Agriculture Conservation: USDA Needs to Better Ensure 
Protection of Highly Erodible Cropland and Wetlands' which was released 
on May 21, 2003.

This text file was formatted by the U.S. General Accounting Office 
(GAO) to be accessible to users with visual impairments, as part of a 
longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov.

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately.

Report to the Ranking Democratic Member, Committee on Agriculture, 
Nutrition, and Forestry, U.S. Senate:

United States General Accounting Office:

GAO:

April 2003:

Agricultural Conservation:

USDA Needs to Better Ensure Protection of Highly Erodible Cropland and 
Wetlands:

GAO-03-418:

GAO Highlights:

Highlights of GAO-03-418, a report to Ranking Democratic Member, 
Committee on Agriculture, Nutrition, and Forestry, U.S. Senate 

Why GAO Did This Study:

Annually, over a billion tons of soil erodes from the nation’s 
cropland, and thousands of other acres, including wetlands, are 
converted to new cropland. Soil erosion reduces the land’s 
productivity and impairs water quality; drained wetlands reduce flood 
control. Under the 1985 Food Security Act, farmers risk losing federal 
farm payments if they do not apply conservation practices to reduce 
erosion or if they drain wetlands. Concerns about soil erosion and 
wetlands conversions continue, however, as do concerns about the U.S. 
Department of Agriculture’s (USDA) Natural Resources Conservation 
Service’s implementation of these provisions. GAO reviewed field 
offices’ and headquarters’ implementation and enforcement of the 1985 
act’s conservation compliance provisions.

What GAO Found:

USDA’s Natural Resources Conservation Service has not consistently 
implemented the 1985 Food Security Act’s conservation provisions. 
Inconsistent implementation increases the possibility that some 
farmers receive federal farm payments although their soil erodes at 
higher rates than allowed or they convert wetlands to cropland.  

According to GAO’s nationwide survey, almost half of the Conservation 
Service’s field offices do not implement the conservation provisions 
as required because they lack staff, management does not emphasize 
these provisions, or they are uncomfortable with their enforcement 
role. For example, field offices do not always find a farmer in 
violation for failing to implement an important practice, such as crop 
rotation, and do not always see whether a farmer has corrected the 
problem; they also do not always check for wetlands violations. 

The Conservation Service’s weak oversight of its field offices further 
impairs implementation of the provisions. In the process of selecting 
samples of cropland tracts to assess farmers’ compliance, the 
Conservation Service disproportionately emphasizes tracts with little 
potential for noncompliance, such as permanent rangelands. This 
selection process leads to inflated compliance rates. The Conservation 
Service also has no automated system to promptly inform its field 
offices of the tracts selected for compliance reviews or to enable the 
offices to efficiently report their review results. Therefore, the 
field offices cannot conduct timely reviews—during critical erosion 
periods—and provide headquarters with up-to-date information. 

Finally, the Farm Service Agency, the USDA agency responsible for 
withholding benefits for violations identified by the Conservation 
Service, often waives these noncompliance determinations without 
adequate justification. Without support from the Farm Service Agency, 
the Conservation Service’s field staff have less incentive to issue 
violations.

What GAO Recommends:

GAO recommends that USDA

* increase oversight of field offices’ compliance reviews to improve 
their accuracy and completeness,
* develop a more representative sample of tracts for review, 
* develop an automated system to manage the data needed for reviews, 
and 
* ensure that noncompliance waivers are supported.

USDA reviewed a draft of this report and concurred with the 
recommendations.  

Contents:

Letter:

Results in Brief:

Background:

Many NRCS Field Offices Are Not Implementing the Conservation 
Provisions As Required:

Weak Oversight Raises Doubts about NRCS's Assertion of a High Rate of 
Compliance:

Farm Service Agency Frequently Waives NRCS Noncompliance Decisions 
without Adequate Justification:

Conservation Provisions and Other Factors Have Significantly Reduced 
Soil Erosion and Wetlands Conversions, but Progress Has Slowed in 
Recent Years:

Conclusions:

Recommendations for Executive Action:

Agency Comments and Our Evaluation:

Appendix I: U.S. Department of Agriculture's Conservation Compliance 
Review Process:

Appendix II: Objectives, Scope, and Methodology:

Appendix III: Results of NRCS's Compliance Reviews, by State, 
Crop Years 2000 and 2001:

Appendix IV: USDA Benefits Denied Farmers Because of 
Conservation Compliance Violations, Crop Years 
1993-2001:

Appendix V: Results of Survey on the Implementation and Effectiveness 
of the Conservation Provisions:

Appendix VI: Comments from the U.S. Department of 
Agriculture's Farm Service Agency:

Appendix VII: GAO Contacts and Staff Acknowledgments:

Related GAO Products:

Tables:

Table 1: Results of NRCS's Compliance Reviews for Selected States for 
Combined Crop Years 2000-2001:

Table 2: USDA Benefits Denied Farmers by the Farm Service Agency for 
Violations of Conservation Provisions, Crop Years 1993-2001:

Table 3: Number of Field Offices Participating in Our Survey That 
Regularly Conduct Compliance Reviews and the Response Rate:

Table 4: Results of NRCS's Compliance Reviews Showing Violations and 
Waivers by State, Crop Year 2000:

Table 5: Results of NRCS's Compliance Reviews Showing Violations and 
Waivers by State, Crop Year 2001:

Table 6: USDA Benefits Denied Farmers by the Farm Service Agency for 
Violations of Conservation Provisions, Nationwide, Crop Years 1993-
2001:

Table 7: USDA Benefits Denied Farmers by the Farm Service Agency for 
Violations of Conservation Provisions by State, Crop Years 1993-2001:

Table 8: Reasons for Reinstating USDA Benefits by State, Crop Years 
1993-2001:

Figures:

Figure 1: Tons of Erosion Due to Water and Wind on Cropland, 1997:

Figure 2: Wetlands Acres on Cropland, 1992:

Figure 3: Percentage of NRCS Field Offices Indicating They Do Not 
Always Review Tracts the Year After Granting Compliance Waivers, 
Nationwide and Selected States:

Figure 4: Percentage of NRCS Field Offices Indicating They Do Not 
Always Issue a Violation When a Farmer Fails to Implement an Important 
Conservation Practice, Nationwide and Selected States:

Figure 5: Percentage of NRCS Field Offices Indicating Primary 
Hindrances in Carrying Out Conservation Compliance Provisions:

Abbreviations:

FSA: Farm Service Agency:

NRCS: Natural Resources Conservation Service:

USDA: U.S. Department of Agriculture:

United States General Accounting Office:

Washington, DC 20548:

April 21, 2003:

The Honorable Tom Harkin
Ranking Democratic Member
Committee on Agriculture,
 Nutrition, and Forestry
United States Senate:

Dear Senator Harkin:

Every year more than a billion tons of soil erodes from the nation's 
cropland while thousands of other acres, including wetlands, are 
converted into new cropland.[Footnote 1] Soil erosion gradually reduces 
the productivity of the land and impairs water quality by depositing 
sediment and other substances, such as pesticides and excess nutrients, 
into the nation's waters. When wetlands are drained, the ability to 
control floods and water quality can decrease, fish and wildlife 
habitat can be harmed, and recreational opportunities can be lost. To 
address these problems, the Food Security Act of 1985, as amended, (the 
1985 act) requires farmers who participate in federal farm programs to 
reduce erosion on highly erodible cropland and, with certain 
exceptions, prohibits the conversion of wetlands to croplands.

The 1985 act requires farmers to conserve highly erodible land and 
wetlands by linking their conservation activities with eligibility for 
federal farm program benefits. These benefits total over $20 billion 
annually from a number of commodity price support and loan programs. To 
be eligible, farmers must (1) have developed and implemented plans to 
apply approved conservation systems by 1995 to reduce erosion on highly 
erodible land they farmed in any year from 1981 through 1985 and (2) 
not have converted and farmed certain wetlands. Furthermore, farmers 
who plant on highly erodible land that they did not farm prior to the 
act's passage must apply a conservation system before planting (under 
the act's sodbuster provision). In general, farmers cannot plant on 
naturally occurring wetlands that were converted to cropland after the 
act's passage (under the act's swampbuster provision). The 1985 act's 
conservation provisions directed at controlling soil erosion cover 
about 104 million, or 28 percent, of the nation's 377 million acres of 
cropland in production.[Footnote 2]

The U.S. Department of Agriculture (USDA) is responsible for 
administering these conservation provisions, enforcing farmers' 
compliance, providing them with technical assistance, and assisting 
them with funding to implement conservation measures. Most of these 
activities fall to the Department's Natural Resources Conservation 
Service (NRCS), although another USDA agency, the Farm Service Agency, 
is responsible for withholding farm program benefits for noncompliance. 
To determine farmers' compliance, each year NRCS draws a random sample 
of cropland units, known as tracts, for compliance reviews. These 
tracts vary in size from a few acres to several thousand acres. 
Nationally, 4.5 million tracts are potentially subject to the act's 
conservation provisions.

To conduct a compliance review, NRCS field staff visit a tract to 
determine whether the farmer who owns it is applying approved 
conservation practices and whether these practices are effectively 
reducing soil erosion. The staff also determine whether that tract had 
any wetlands and if it did, whether the farmer drained them. When NRCS 
officials find noncompliance, they can either waive or recommend 
penalties. NRCS may grant a waiver if the violation occurred because of 
personal hardship or adverse weather, or if it was minor or technical. 
A waiver continues a farmer's eligibility for farm program benefits for 
12 months; the farmer is to take corrective measures during this time. 
If a waiver is not justified, NRCS staff are to find the farmer in 
noncompliance and notify the local Farm Service Agency office. This 
office then determines the amount of farm program benefits to be 
withheld. However, a farmer that NRCS finds in noncompliance can appeal 
this determination to the Farm Service Agency's field office. In 
response, that office may grant its own waiver if it believes the 
farmer acted in good faith--that is, the farmer did not intend to 
violate the conservation provisions.[Footnote 3] Appendix I provides 
further information on USDA's compliance review process.

You asked us to evaluate USDA's implementation of the conservation 
compliance provisions of the 1985 act. Specifically, you asked us to 
determine (1) how well NRCS's field offices are carrying out these 
provisions, (2) how effectively NRCS oversees its field offices' 
efforts to carry out these provisions, (3) how often the Farm Service 
Agency waives NRCS's noncompliance determinations, and (4) to what 
extent these conservation provisions have helped to reduce soil erosion 
and the loss of wetlands.

To conduct this work, we examined NRCS's national database on the 
results of compliance reviews for 1998 through 2001 to identify unusual 
patterns in compliance enforcement. Automated data for these reviews 
were not available for prior years. We also surveyed the official--
usually the district conservationist--responsible for compliance 
reviews in each of NRCS's approximately 2,500 field offices to obtain 
information on that official's understanding and implementation of the 
conservation provisions, as well as the official's views on the 
effectiveness of these provisions. In addition to responding to our 
survey questions, many of these officials also provided us with written 
comments. We received responses from almost 80 percent of the officials 
surveyed. We also conducted work in 20 NRCS field offices located in 19 
counties in 5 states (Arkansas, Colorado, Nebraska, North Dakota, and 
Texas) to review documentation of compliance decisions, including 
waivers, to determine the basis for these decisions. We selected these 
field offices on the basis of such criteria as the relative amount of 
land covered by the office that is subject to the 1985 act's 
conservation provisions; geographic dispersion; and apparent anomalies 
in USDA's data related to compliance checks, waivers, and penalties 
assessed. In addition, we examined the Farm Service Agency's database 
on violations and benefits withheld or waivers granted for crop years 
1993 through 2001 to determine trends in assessing penalties for 
noncompliance, and we spoke with Farm Service Agency field and 
headquarters staff regarding the reasons for waiving NRCS noncompliance 
determinations.[Footnote 4] Finally, regarding the environmental 
impacts of the conservation provisions, we reviewed the results of 
USDA's National Resources Inventory and other relevant studies and 
spoke with officials of various farm and conservation groups.

We conducted our review from April 2002 through February 2003 in 
accordance with generally accepted government auditing standards. 
Appendix II provides additional information on our scope and 
methodology. Appendix V summarizes the results of our survey. In 
addition, survey results stratified by state are included in a special 
publication entitled Agricultural Conservation: Survey Results on 
USDA's Implementation of Food Security Act Compliance Provisions (GAO-
03-492SP), which is available on the Internet at http://www.gao.gov/
cgi-bin/getrpt?gao-03-492SP.

Results in Brief:

Almost half of NRCS's field offices are not implementing one or more 
aspects of the conservation provisions of the 1985 act as required. 
Inconsistent implementation increases the likelihood that some farmers 
are still receiving federal farm payments even though they let soil 
erode at higher rates than allowed or convert wetlands to cropland. 
Specifically, according to our survey, field offices do not always 
follow all required procedures, such as (1) checking for wetlands 
violations during a compliance review (36 percent), (2) revisiting 
farms granted a waiver the previous year to determine whether the owner 
has taken measures to achieve compliance (16 percent), or (3) finding a 
farmer in violation for failing to implement an important conservation 
practice (19 percent). Our field office visits revealed a similar 
pattern. For example, in 14 of the 20 offices we visited, NRCS staff 
did not always conduct compliance reviews when cropland was most 
vulnerable to erosion, as agency guidelines require, such as when 
spring planting occurs; at this time, crop residue is at its lowest 
level and rains may be heavy. A number of factors--such as resource 
constraints, a de-emphasis on the conservation compliance provisions 
relative to other work, and a reluctance to assume the enforcement 
role--may be contributing to the implementation problems identified. 
These problems are compounded by a lack of training and unclear policy 
guidance concerning the implementation of the provisions. Finally, our 
analysis of NRCS's database on the results of compliance reviews 
underscores the variation in field offices' enforcement among the 
states: the number of waivers and violations issued as a percentage of 
total compliance reviews ranged from none to as much as 15 percent 
during crop years 2000 and 2001. Most reported violations occurred in a 
relatively few states.

NRCS does not effectively oversee its field offices' implementation of 
the conservation provisions, among other things, calling into doubt its 
claim that 98 percent of the tracts reviewed are in compliance with the 
act's conservation provisions. First, NRCS's process for selecting 
tracts for compliance reviews disproportionately emphasizes tracts with 
little or no potential for noncompliance, such as permanent rangelands. 
Such tracts account for about 20 percent of the tracts selected 
annually. Second, NRCS does not have an automated system for promptly 
informing its field offices of the tracts selected for compliance 
reviews and for enabling the field offices to efficiently report the 
results of these reviews. As a consequence, in many cases the field 
offices do not have the information on the tracts to be reviewed until 
after the critical erosion control period has passed; the reviews 
should have been done during this period. Furthermore, without such a 
system, NRCS lacks accurate, up-to-date information for oversight to 
evaluate the field offices' implementation of the conservation 
compliance provisions. Third, NRCS does not consistently collect and 
analyze the results of the field offices' compliance reviews to 
identify unusual enforcement patterns across regions and states and 
over time. For example, until our review NRCS was not aware that 30 of 
the 50 "good faith" waivers granted nationally in 2000 occurred in just 
one state. Similarly, NRCS has not questioned the wide variation in 
other nationwide data on noncompliance determinations and waivers 
granted, which also suggests inconsistencies across the states and 
field offices in how the conservation provisions are being implemented. 
These inconsistencies are borne out by our survey results. Finally, 
USDA's Office of Inspector General has recently reported that 
improvements in NRCS's implementation of the conservation provisions 
are needed to strengthen the agency's ability to provide accurate and 
reliable assessments of farmers' compliance. Importantly, these 
improvements include the need for NRCS to refrain from issuing waivers 
that are not warranted.

In response to farmers' appeals, the Farm Service Agency waived NRCS's 
noncompliance determinations in 4,948 of 8,118 cases in which farmers 
were cited with violations, or about 61 percent, from 1993 through 
2001. Furthermore, because of these waivers, the Farm Service Agency 
reinstated about $40.4 million of the $59.6 million that was to be 
withheld for noncompliance determinations. These appeals were 
considered and ruled upon by local Farm Service Agency county 
committees. Because committees generally consist of farmers elected by 
other farmers in the county, some NRCS staff and conservation groups 
believe that the committee members are predisposed to approve farmers' 
appeals so as not to penalize a neighbor's eligibility for farm program 
benefits. In this regard, about one-third of NRCS's field offices 
indicated that the Farm Service Agency did not adequately justify its 
waiver decisions. Our field office visits generally reinforced this 
assertion. In the five offices we visited that had found farmers in 
violation of the conservation provisions, NRCS staff indicated that the 
Farm Service Agency's waivers were not adequately justified. 
Furthermore, the minutes of the Farm Service Agency's county committee 
meetings and other relevant records did not clearly describe the basis 
for waiving NRCS's noncompliance determinations in these cases. Without 
support from the Farm Service Agency, NRCS field office staff said that 
they have less incentive to find farmers out of compliance when 
warranted.

According to USDA, conservation groups, and farm organizations, in the 
17 years since the 1985 act, its conservation provisions have 
contributed to substantial reductions in soil erosion and wetlands 
conversions. For example, according to USDA data, soil erosion on land 
subject to these provisions declined by about 35 percent from 1982 
through 1997. Wetlands conversions for agricultural uses declined even 
more sharply, from 235,000 acres per year before 1985 to 27,000 acres 
per year from 1992 through 1997. However, because other factors have 
also influenced farmers' behavior, quantifying the impact of the 
conservation provisions is difficult. These other factors include 
economic incentives for agricultural producers to use new farming 
techniques and equipment that are more conserving of land and water 
resources. Despite the improvements made, concerns remain about 
continued high rates of soil erosion and wetlands losses in some 
regions. Specifically, although annual soil erosion on all cropland has 
declined to 5 tons per acre, annual soil erosion on about 27 percent of 
the land subject to the conservation compliance provisions still 
averages 24 tons per acre. Furthermore, USDA and other agricultural 
experts indicate that reductions in soil erosion have leveled off in 
recent years and that in some areas of the country soil erosion has 
even increased. In this regard, over 80 percent of NRCS field offices 
we surveyed reported that further reductions in soil erosion are 
feasible. Finally, the conservation provisions may be only marginally 
effective in protecting seasonal wetlands because USDA generally 
identifies these wetlands during the summer months when these lands are 
less likely to be saturated or exhibit other wetlands characteristics.

In light of the problems we have noted with NRCS's implementation of 
the 1985 act's conservation provisions, as well as continuing concerns 
related to soil erosion and wetlands conversion, we are making 
recommendations to USDA to improve the quality of NRCS's compliance 
reviews and the Farm Service Agency's documentation of its decisions 
regarding farmers' appeals of noncompliance determinations. In 
commenting on a draft of this report, NRCS and the Farm Service Agency 
concurred with the recommendations. The agencies also generally agreed 
with the report's findings, although NRCS stated that the report 
focuses too much on problems with the agency's implementation of the 
conservation compliance provisions and not enough on the provisions' 
positive accomplishments in reducing soil erosion and wetlands 
conversions. NRCS provided oral comments; the Farm Service Agency 
provided written comments, which are presented in appendix VI. The 
agencies also provided technical comments, which we have incorporated 
as appropriate.

Background:

Legislative Requirements:

The Food Security Act of 1985 introduced three conservation provisions 
to address environmental problems associated with highly erodible land 
and wetlands.[Footnote 5] Under the act, farmers must apply 
conservation systems to these lands or risk losing benefits.[Footnote 
6] First, under the "conservation compliance" provision, farmers must 
apply conservation systems to lands cropped in any year from 1981 
through 1985 to substantially reduce soil erosion. Second, the 
"sodbuster" provision applies to highly erodible land not farmed prior 
to the act's passage. For these lands, farmers must apply a 
conservation system before planting and must control soil erosion to a 
higher level than required under conservation compliance. Third, under 
the "swampbuster" provision, farmers are generally prohibited from 
converting wetlands to cropland. For the purpose of this report, we use 
the term "conservation compliance" to include all three conservation 
provisions of the 1985 act.

The Food, Agriculture, Conservation, and Trade Act of 1990 modified the 
conservation compliance provisions, giving USDA discretion to determine 
that a farmer, although in violation, acted in good faith--that is, 
without intending to violate the provisions.[Footnote 7] In such cases, 
USDA may reduce the farmer's benefits but the farmer would remain 
eligible to participate in federal farm programs if the farmer corrects 
the violation.[Footnote 8] In addition, the act revised the swampbuster 
provision to allow a farmer to retain eligibility for farm program 
benefits if the farmer mitigates a violation by restoring a wetland 
converted prior to the 1985 act.

The Federal Agriculture Improvement and Reform Act of 1996 mandated a 
variety of changes to help farmers comply with the provisions.[Footnote 
9] Among other things, the act allowed flexibility in developing and 
implementing conservation systems, and it allowed farmers to self-
certify compliance with their conservation systems.[Footnote 10] The 
act also required USDA field staff who provide technical assistance to 
a farmer and observe a potential compliance deficiency on the farmer's 
tract to, within 45 days, provide the farmer with specific information 
on how to correct the deficiency. If the farmer agrees to correct the 
deficiency and signs an approved conservation plan, the farmer is given 
a waiver. However, if the farmer does not implement corrective action 
within 12 months after the waiver, USDA will schedule the tract for a 
compliance review. In addition, the act provided farmers with more 
flexibility to offset wetlands losses through mitigation, including the 
enhancement of an existing wetland or the creation of a new wetland. At 
the same time, the act made easier the "good faith" provisions that the 
1990 act had previously added to the Food Security Act of 1985. 
Finally, the 1996 act removed crop insurance from the list of benefits 
that can be denied to farmers who violate the conservation 
provisions.[Footnote 11]

The 2002 Farm Security and Rural Investment Act did not change the 
conservation compliance provisions.[Footnote 12] However, the act 
provides that the Secretary of Agriculture's authority to make 
noncompliance determinations may not be delegated to any private person 
or entity.

Implementation of the Conservation Compliance Provisions:

NRCS monitors farmers' implementation of the conservation compliance 
provisions largely through compliance reviews.[Footnote 13] In addition 
to the random sample of tracts that NRCS draws annually for these 
reviews, field offices select other tracts based on referrals from 
other agencies, farmers who receive farm loans, whistleblower 
complaints, potential violations observed by NRCS employees when 
providing technical assistance, and tracts that maintained eligibility 
due to prior year waivers. In conducting these reviews, NRCS staff 
visit a land tract to determine if the relevant farmer is following the 
conservation system, including specific conservation practices, 
developed and approved for that tract. As discussed, the 1985 act 
requires farmers to develop these systems in order to remain eligible 
for farm program benefits.

In general, conservation systems are designed to be economically viable 
for a farmer while achieving substantial reductions in soil erosion. 
These systems are composed of one or more conservation practices. Some 
commonly used conservation practices include:

* conservation crop rotation--planting low-residue crops such as 
soybeans in one year, followed by a high residue crop, such as corn in 
the following year on the same field, in order to generate an average 
layer of residue from year to year (used on 81 percent of highly 
erodible cropland);

* conservation tillage--allowing the crop residue to stay on top of the 
field, rather than being plowed under when planting begins (used on 33 
percent of highly erodible cropland);

* terraces--creating an embankment or ridge (a terrace) at a right 
angle to sloping land in order to allow water to soak into the soil 
rather than to move down the slope, taking the soil with it (used on 13 
percent of highly erodible cropland); and:

* grassed waterways--creating a broad and shallow depression, usually 
below a terraced area, that is planted with grasses to mitigate erosion 
by slowing the flow of runoff, holding a bank, and filtering out soil 
particles (used on 9 percent of highly erodible cropland).
:

Regional Erosion Concerns:

The adoption of a particular conservation practice varies with climate, 
topography, soils, predominant crops, and preexisting production 
practices. For example, local environmental conditions in eastern 
Nebraska, and western Texas require different conservation practices. 
Eastern Nebraska primarily produces corn and soybeans and has a higher 
average rainfall and a more varied topography than western Texas. Thus, 
to control soil erosion from water, farmers in eastern Nebraska use a 
larger number of conservation practices--most frequently conservation 
crop rotation, conservation tillage, terraces, and grassed waterways. 
In western Texas, wheat and cotton are the predominant crops. In this 
area, where soil erosion from wind is the primary concern, most 
conservation practices consist of either applying conservation tillage 
or creating ridges on the field (roughening the surface) to prevent the 
soil from blowing away. Figure 1 shows areas of the country with 
cropland that has a high propensity for soil erosion due to water and 
wind.

Figure 1: Tons of Erosion Due to Water and Wind on Cropland, 1997:

[See PDF for image]

Note: Map includes land enrolled in the Conservation Reserve Program.

[End of figure]

According to the U.S. Environmental Protection Agency, soil erosion is 
a leading cause of water pollution. Soil deposits in streams, rivers, 
drainageways, and lakes degrade water quality by increasing turbidity 
and transporting attached nutrients, pesticides, pathogens, and toxic 
substances. In addition, soil erosion due to wind contributes to 
particulate matter in the air, which can cause respiratory illness and 
property damage.

Wetlands Conversion Concerns:

Of the estimated 220 million acres of marshes, bogs, swamps, and other 
wetlands in the contiguous United States during colonial times, over 
half have disappeared, and some remaining wetlands have been degraded. 
This decrease is due, primarily, to agricultural activities and 
development; wetlands were once regarded as unimportant areas to be 
filled or drained for these purposes. Pressure to use wetlands for such 
purposes continues, but in recent times, wetlands have become valued 
for a variety of ecological functions that they perform, including:

* providing vital habitat for wildlife and waterfowl, including about 
half of the threatened and endangered species;

* providing spawning grounds for commercially and recreationally 
valuable fish and shellfish;

* providing flood control by slowing down and absorbing excess water 
during storms;

* maintaining water quality by filtering out pollutants before they 
enter streams, lakes, and oceans; and:

* protecting coastal and upland areas from erosion. 

Recognizing the value of wetlands, in 1989, the administration set a 
national goal to protect against additional loss. Specifically, the 
first Bush administration established the national goal of "no net 
loss" of wetlands. Subsequently, the Clinton administration expanded 
the goal to achieve a net increase of 100,000 acres of wetlands per 
year by 2005.

In January 2001, the U.S. Supreme Court ruled that the U.S. Army Corps 
of Engineers (Corps) has no authority under the Clean Water Act to 
regulate certain isolated, intrastate, nonnavigable waters, including 
some wetlands.[Footnote 14] However, even if a wetland is no longer 
within federal jurisdiction under the Clean Water Act, it may still be 
protected under other federal or state laws. For example, in a January 
2003 Advance Notice of Proposed Rulemaking, the Environmental 
Protection Agency and the Corps stated that the federal government 
remains committed to wetlands protection through the Food Security 
Act's swampbuster requirements, among other programs.[Footnote 15] In 
this regard, NRCS officials indicated to us that they do not anticipate 
any change in how they implement the swampbuster provisions; in part, 
these provisions are directed at the protection of isolated, intrastate 
wetlands that occur on cropland, including "prairie potholes" in the 
upper Midwest.[Footnote 16]

Figure 2 shows areas of the country with wetlands on cropland, 
including permanent, seasonal, and prior-converted wetlands (cropped 
wetlands drained or filled prior to the 1985 act's conservation 
compliance provisions).

Figure 2: Wetlands Acres on Cropland, 1992:

[See PDF for image]

[End of figure]

Many NRCS Field Offices Are Not Implementing the Conservation 
Provisions As Required:

According to our survey, almost half of NRCS field offices do not 
follow all required procedures in implementing the conservation 
compliance provisions, including, for example, checking for wetlands 
violations during a compliance review and finding a farmer in violation 
when the farmer fails to implement an important conservation practice. 
The inconsistent implementation of the conservation provisions 
increases the likelihood that some farmers are still receiving federal 
farm payments even though they let soil erode at higher rates than 
allowed or convert wetlands to cropland. Our field office visits 
revealed similar problems. Furthermore, the field offices may not be 
consistently enforcing the provisions, according to our analysis of 
NRCS's database on the results of compliance reviews: the number of 
waivers and violations issued as a percentage of total compliance 
reviews varied widely from state to state. Problems in the field 
offices' implementation of the conservation compliance provisions occur 
for a number of reasons, such as the lack of periodic training on how 
to conduct these reviews.

Many NRCS Field Offices Do Not Follow All Required Steps in Assessing 
Compliance:

Our survey results indicate that 48 percent, or 903, of the field 
offices, are not implementing one or more provisions for conducting 
compliance reviews included in NRCS's National Food Security Act Manual 
or other related guidance, as shown below:

* Nationwide, more than one-third, or 670, of the field offices, on 
average, do not check for wetlands violations when conducting 
compliance reviews. The lack of attention to potential wetlands 
violations varied by state, ranging from 15 percent to 63 percent of 
the field offices in each state; 18 states exceeded the national 
average of 36 percent.

* Nationwide, 16 percent, or 250, of the field offices do not always 
review tracts during the year after granting a compliance waiver to 
determine whether the farmer had taken measures to achieve compliance. 
The extent to which field offices do not follow this procedure varied 
considerably from state to state.

Figure 3 shows that the range varied from 4 to 39 percent for the 
selected states.[Footnote 17]

Figure 3: Percentage of NRCS Field Offices Indicating They Do Not 
Always Review Tracts the Year After Granting Compliance Waivers, 
Nationwide and Selected States:

[See PDF for image]

Note: GAO's survey results.

[End of figure]

* Nationwide, about 19 percent, or 324, of the NRCS field offices do 
not always find a farmer in violation when the farmer fails to 
implement an important conservation practice, as required by NRCS 
guidance. Figure 4 shows that from 4 percent to 38 percent of the field 
offices in selected states failed to cite farmers for a major 
violation. 

Figure 4: Percentage of NRCS Field Offices Indicating They Do Not 
Always Issue a Violation When a Farmer Fails to Implement an Important 
Conservation Practice, Nationwide and Selected States:

[See PDF for image]

Note: GAO's survey results.

[End of figure]

Our field office visits revealed similar problems as shown below:

* In 14 of the 20 offices, staff did not always conduct compliance 
reviews during critical soil erosion periods as required by NRCS's 
guidance; during these periods, the soil is most susceptible to water 
or wind erosion. Critical periods may include, for example, April, May, 
and June--when planting occurs, crop residue is at its lowest level, 
and rainfalls may be heavy. For example, in one office in Texas, none 
of the 25 compliance reviews done during the 4-year period we examined 
were conducted during the spring--the critical water erosion period. In 
another office in Texas, none of the 11 compliance reviews done in 2001 
that we reviewed were conducted during the critical wind erosion 
period--January through April. According to staff in the 14 offices, 
NRCS headquarters and state office delays in providing the lists of 
randomly selected tracts for compliance reviews generally prohibited 
field office staff from conducting the reviews within critical erosion 
periods.[Footnote 18]

* In five Nebraska field offices, staff improperly granted waivers in 
28 of the 60 minor or technical waiver cases we reviewed. According to 
NRCS guidance, staff may grant minor or technical waivers for 
conservation deficiencies if these deficiencies have little impact on 
erosion control. However, these 28 waivers were granted to farmers who 
had failed to implement a major soil-conserving practice, such as 
maintaining terraces or sufficient crop residue, thereby potentially 
allowing severe water-related soil erosion to occur.[Footnote 19]

* In one Texas field office, NRCS staff did not properly conduct a 
compliance review on a 9,878-acre tract in 2000. After the owner sold 
166 acres of the tract for a commercial cattle-feeding operation in 
1999, the Farm Service Agency assigned new tract numbers to both the 
166 acres and the remaining 9,712 acres of the original tract in order 
to ensure proper accounting for farm benefits. Nevertheless, NRCS 
requires that a compliance review be conducted on all land included 
under the original tract number. However, staff in the field office 
reviewed the 166-acre tract only, not the other 9,712 acres, and yet 
reported the original tract as being in compliance.

* Staff in one Colorado field office--responsible for conducting 
compliance reviews on about 40 tracts from 1998 through 2001--could 
find no evidence that these reviews had been done after a thorough 
search of their physical and electronic records. In contrast, these 
officials were able to produce documentation for reviews conducted in 
the years prior to 1998. Officials in this office indicated that it is 
doubtful that the reviews for 1998 through 2001 were done. 
Nevertheless, this office had reported that all the tracts were in 
compliance during these years.

We also identified other types of situations in which field staff 
missed opportunities to identify and correct noncompliance. First, 
according to NRCS guidance, the agency's field staff are required to 
report potential instances of noncompliance identified when they visit 
a farm to provide conservation technical assistance. In these cases, 
the guidance requires that the staff issue a 12-month waiver to allow 
the farmer time to take corrective measures, while continuing the 
farmer's eligibility for farm program benefits.[Footnote 20] However, 
fewer than 40 percent of NRCS field offices reported that supervisors 
either generally or strongly encouraged them to identify tracts in 
noncompliance when providing technical assistance. During crop years 
2000 and 2001, of the approximately 2.1 million technical assistance 
visits NRCS made, it identified deficiencies and issued waivers in only 
22 instances. According to NRCS headquarters and field staff, the 
agency's field staff are reluctant to identify deficiencies and issue 
waivers because they believe doing so would deter farmers from seeking 
technical assistance in the future; others cited NRCS guidance as 
unclear on when and how to issue a waiver for a deficiency discovered 
during technical assistance visits.

Second, NRCS field offices do not always include a sample of tracts 
related to farmers who participate in the Farm Service Agency's Farm 
Loan Program in their annual compliance reviews, as NRCS guidance 
requires.[Footnote 21] According to NRCS, in addition to the 
headquarters list of tracts selected for compliance reviews that 
includes farmers who receive farm program benefits and produce crops on 
highly erodible land, the field offices are to conduct compliance 
reviews on a 5 percent sample of the loan program participants who are 
producing crops on highly erodible land. This 5 percent sample is taken 
to ensure oversight over farmers who participate in the loan program 
but do not otherwise receive farm program benefits. However, we found 
that in half of the 20 NRCS offices we visited, NRCS and Farm Service 
Agency field staff did not ensure that they included this sample of 
borrowers in each of the years we examined.

Finally, NRCS field staff do not always maintain documentation 
supporting their decisions and do not always correctly report the 
results of their compliance reviews as required by the National Food 
Security Act Manual. For example, in 7 of the 20 offices we visited, 
the compliance review case file contained a worksheet documenting the 
decision but no evidence to show when the review was conducted, whether 
crop residue measurements were taken, or what on-site conditions were 
observed.

Significant Variation in the Number of Waivers and Violations May 
Indicate Inconsistent Enforcement among States:

During crop years 2000 and 2001, 5 percent of all compliance reviews 
resulted in waivers or violations, according to NRCS's database on the 
results of compliance reviews.[Footnote 22] However, as table 1 shows, 
this percentage varied significantly from state to state. For example, 
four states--Iowa, Nebraska, North Dakota, and Oklahoma--experienced 
significantly more waivers and violations as a percentage of reviews 
conducted than the national average, while 10 states--Alabama, 
Arkansas, Colorado, Georgia, Maryland, Michigan, North Carolina, South 
Dakota, Texas, and Virginia--experienced fewer. This variation suggests 
that NRCS's field offices are not consistently enforcing the 
conservation compliance provisions. Similarly, of the 1,810 waivers and 
violations issued during crop years 2000 and 2001, more than 80 percent 
occurred in only 10 states--Illinois, Iowa, Minnesota, Nebraska, New 
Mexico, North Dakota, Oklahoma, Pennsylvania, Vermont, and Wisconsin; 
these 10 states represent only 36 percent of all reviews, suggesting 
again a lack of enforcement consistency across states.

Table 1: Results of NRCS's Compliance Reviews for Selected States for 
Combined Crop Years 2000-2001:

State: Alabama; Tracts reviewed for compliance: 559; Tracts with NRCS 
waivers: 1; Tracts with violations: 6; Total waivers and violations: 7; 
Waivers and violations as a percentage of tracts reviewed: 1.3.

State: Arkansas; Tracts reviewed for compliance: 588; Tracts with NRCS 
waivers: 2; Tracts with violations: 5; Total waivers and violations: 7; 
Waivers and violations as a percentage of tracts reviewed: 1.2.

State: Colorado; Tracts reviewed for compliance: 883; Tracts with NRCS 
waivers: 15; Tracts with violations: 4; Total waivers and violations: 
19; Waivers and violations as a percentage of tracts reviewed: 2.2.

State: Georgia; Tracts reviewed for compliance: 691; Tracts with NRCS 
waivers: 7; Tracts with violations: 2; Total waivers and violations: 9; 
Waivers and violations as a percentage of tracts reviewed: 1.3.

State: Iowa; Tracts reviewed for compliance: 2,942; Tracts with NRCS 
waivers: 283; Tracts with violations: 130; Total waivers and 
violations: 413; Waivers and violations as a percentage of tracts 
reviewed: 14.0.

State: Maryland; Tracts reviewed for compliance: 235; Tracts with NRCS 
waivers: 1; Tracts with violations: 0; Total waivers and violations: 1; 
Waivers and violations as a percentage of tracts reviewed: 0.4.

State: Michigan; Tracts reviewed for compliance: 409; Tracts with NRCS 
waivers: 1; Tracts with violations: 2; Total waivers and violations: 3; 
Waivers and violations as a percentage of tracts reviewed: 0.7.

State: Mississippi; Tracts reviewed for compliance: 849; Tracts with 
NRCS waivers: 33; Tracts with violations: 0; Total waivers and 
violations: 33; Waivers and violations as a percentage of tracts 
reviewed: 3.9.

State: Nebraska; Tracts reviewed for compliance: 1,907; Tracts with 
NRCS waivers: 158; Tracts with violations: 115; Total waivers and 
violations: 273; Waivers and violations as a percentage of tracts 
reviewed: 14.3.

State: North Carolina; Tracts reviewed for compliance: 1,229; Tracts 
with NRCS waivers: 14; Tracts with violations: 5; Total waivers and 
violations: 19; Waivers and violations as a percentage of tracts 
reviewed: 1.5.

State: North Dakota; Tracts reviewed for compliance: 1,659; Tracts with 
NRCS waivers: 11; Tracts with violations: 246; Total waivers and 
violations: 257; Waivers and violations as a percentage of tracts 
reviewed: 15.5.

State: Oklahoma; Tracts reviewed for compliance: 706; Tracts with NRCS 
waivers: 74; Tracts with violations: 2; Total waivers and violations: 
76; Waivers and violations as a percentage of tracts reviewed: 10.8.

State: Pennsylvania; Tracts reviewed for compliance: 374; Tracts with 
NRCS waivers: 14; Tracts with violations: 7; Total waivers and 
violations: 21; Waivers and violations as a percentage of tracts 
reviewed: 5.6.

State: South Dakota; Tracts reviewed for compliance: 906; Tracts with 
NRCS waivers: 2; Tracts with violations: 2; Total waivers and 
violations: 4; Waivers and violations as a percentage of tracts 
reviewed: 0.4.

State: Texas; Tracts reviewed for compliance: 1,923; Tracts with NRCS 
waivers: 17; Tracts with violations: 3; Total waivers and violations: 
20; Waivers and violations as a percentage of tracts reviewed: 1.0.

State: Virginia; Tracts reviewed for compliance: 376; Tracts with NRCS 
waivers: 5; Tracts with violations: 1; Total waivers and violations: 6; 
Waivers and violations as a percentage of tracts reviewed: 1.6.

State: Washington; Tracts reviewed for compliance: 400; Tracts with 
NRCS waivers: 5; Tracts with violations: 14; Total waivers and 
violations: 19; Waivers and violations as a percentage of tracts 
reviewed: 4.8.

State: Wisconsin; Tracts reviewed for compliance: 1,460; Tracts with 
NRCS waivers: 59; Tracts with violations: 30; Total waivers and 
violations: 89; Waivers and violations as a percentage of tracts 
reviewed: 6.1.

State: Remaining 32 states; Tracts reviewed for compliance: 16,871; 
Tracts with NRCS waivers: 306; Tracts with violations: 228; Total 
waivers and violations: 534; Waivers and violations as a percentage of 
tracts reviewed: 3.2.

State: Total; Tracts reviewed for compliance: 34,967[A]; Tracts with 
NRCS waivers: 1,008[B]; Tracts with violations: 802[C]; Total waivers 
and violations: 1,810; Waivers and violations as a percentage of tracts 
reviewed: 5.2.

Source: NRCS.

Note: GAO's analysis of NRCS's data.

[A] Total compliance reviews include tracts of 13,025 and 13,544 in 
2000 and 2001, respectively, that were randomly selected by NRCS 
headquarters. The total also includes tracts added by NRCS field 
offices based on referrals from other agencies (e.g., tracts owned by 
employees of other USDA agencies), whistleblower complaints, tracts 
owned by Farm Loan Program participants, and tracts that maintained 
eligibility for farm benefits because of prior year waivers. These 
additional tracts numbered 4,234 and 4,164 in 2000 and 2001, 
respectively.

[B] "Waivers" refers to NRCS variances and exemptions. An NRCS waiver 
does not change the fact that the agency has made a noncompliance 
determination. The farmer receiving the waiver is still considered to 
have committed a violation that must be corrected, unless the waiver 
was given for severe or unusual conditions related to weather, disease 
or pests.

[C] Total tracts with violations largely reflect NRCS's preliminary 
noncompliance determinations, as of the date these data were compiled. 
Many of these determinations were subsequently reversed by NRCS--
through granting variances or exemptions--on appeal from the affected 
farmers. Because NRCS's data on tracts with violations reflect a number 
of preliminary determinations that were subsequently reversed by NRCS, 
the total number of violations reported for crop years 2000-2001 in 
this table is greater than the total number shown for these years in 
Table 2. This latter table, based on Farm Service Agency data, shows 
the actual number of tracts with violations referred by NRCS to the 
Farm Service Agency for action.

[End of table]

The Soil and Water Conservation Society and the Sustainable Agriculture 
Coalition--two conservation groups--maintain that the wide state-to-
state differences may indicate inconsistent application or differing 
interpretations of conservation compliance procedures. We also hold 
this view. These groups also noted that some of these differences may 
be explained by the differences in topography, local weather 
conditions, and farmers' ability to comply with the conservation 
provisions. More detailed information on the results of NRCS's 
compliance reviews is contained in appendix III.

Several Factors, Including Lack of Training, Contribute to Problems in 
Implementing Conservation Compliance:

Our survey and field office visits identified key reasons for the 
problems in implementing the conservation compliance provisions. As 
figure 5 shows, on the basis of our survey results, field offices 
reported lack of staff, reversal of noncompliance decisions, and 
unwillingness to assume an enforcement role as the primary hindrances 
in carrying out the provisions.

Figure 5: Percentage of NRCS Field Offices Indicating Primary 
Hindrances in Carrying Out Conservation Compliance Provisions:

[See PDF for image]

Notes: GAO's survey results.

[End of figure]

"Other" includes lack of NRCS guidance; lack of appropriate 
information, such as maps; compliance reviews that are not a priority 
with supervisor; sample tracts that are received at inconvenient times; 
and, external influences.

"Reversal of noncompliance decisions" includes decisions overturned by 
NRCS or USDA's National Appeals Division and waivers issued by the Farm 
Service Agency. However, Farm Service Agency officials noted that the 
issuance of a "good faith" waiver by their agency does not, technically 
speaking, represent a reversal of the noncompliance determination. They 
explained that although the farmer involved remains eligible for farm 
program benefits, the farmer has committed a violation and must 
undertake corrective measures within 12 months or risk losing these 
benefits at that time. Nevertheless, for the purpose of our work, we 
made a distinction: a compliance review either results in a violation, 
leading to the loss of a farmer's eligibility for farm program 
benefits, or a waiver, allowing a farmer to continue his/her 
eligibility.

In addition, 244 field staff elaborated on one of these hindrances--the 
undesirability of the enforcement role--in their written comments to 
our survey. For example, some wrote that it is difficult to provide 
assistance to farmers most of the year in the small communities where 
the field staff live and work and then have to cite some of the same 
farmers for noncompliance, which may result in the loss of their farm 
program benefits.

In this regard, our past work has noted this cultural conflict in NRCS 
between its regulatory role under the 1985 act and its traditional role 
of advising and helping farmers.[Footnote 23] Specifically, for the 
past 70 years, NRCS's role, including that of its predecessor 
organization, the Soil Conservation Service, has largely been to work 
cooperatively with farmers to provide technical assistance and foster 
voluntary conservation. With the addition of the 1985 conservation 
compliance provisions, NRCS is often in the conflicting position of 
acting as advisor to and regulator of farmers. Our past evaluation and 
many of the studies we reviewed found that this internal conflict 
contributes to the reluctance of the agency's field office staff, with 
whom most contacts with farmers take place, to cite farmers with 
violations in their conservation plans because such violations could 
cause farmers to lose their farm program benefits.

In addition to the primary hindrances noted by the survey respondents, 
36 percent of respondents reported that since the mid-1990s, the 
agency's management has de-emphasized the conservation compliance 
provisions. Instead, NRCS has shifted its emphasis to providing 
technical assistance and to enrolling farmers in incentive-based 
conservation programs that provide cost-share and other financial 
assistance. For example, in January 2003, NRCS headquarters officials 
indicated that over the past 20 years, the number of conservation 
programs that the agency is responsible for implementing has doubled 
from 6 to 12. Furthermore, the funding for many of these programs has 
increased markedly in recent years, with a corresponding increase in 
workload for the agency's staff. However, according to these officials, 
NRCS staff level has declined by about 25 percent in the last 20 
years.[Footnote 24]

These problems, which may be difficult to address, are exacerbated by a 
lack of training on the conservation compliance provisions and a need 
for further clarification of NRCS's written guidance. Periodic training 
has generally not been available since the mid-1990s, which has led to 
errors in assessing and reporting on compliance. For example, in 11 of 
the 20 field offices we visited, staff had not received periodic 
training on how to conduct these reviews and, as a result, these field 
staff did not always correctly report the results of their compliance 
reviews.

Similarly, nearly 80 percent of the 114 survey respondents who provided 
written comments on training issues noted the lack of recent training 
on highly erodible land and wetlands conservation provisions, including 
conducting compliance reviews. For example, one respondent commented 
that he received excellent training in the conservation compliance 
provisions when they were first implemented in the early 1990s but that 
many of the younger employees have not received such training. Another 
respondent, who had transferred from one state to another noted 
differences in how the compliance reviews were conducted and suggested 
that staff needed training to ensure that reviews are conducted 
uniformly. In discussing this issue with NRCS headquarters officials in 
January 2003, they acknowledged that periodic training has not been 
provided in recent years and agreed that this lack of training is a 
problem.

Moreover, more than 50 percent of survey respondents reported that 
NRCS's National Food Security Act Manual (the guidance manual) needs 
clarification in key areas of conservation compliance--highly erodible 
cropland, sodbuster, swampbuster, compliance reviews, and appeals of 
noncompliance decisions. USDA's Office of Inspector General has also 
reported on the need for clarification in NRCS's written guidance for 
implementing the conservation compliance provisions. For example, in a 
September 2002 report, the Inspector General noted that weaknesses in 
guidance manual procedures had reduced the effectiveness of NRCS 
administration of the highly erodible land provisions and the agency's 
ability to accurately evaluate producers' compliance with these 
provisions.[Footnote 25] The weaknesses noted include incorrect 
procedural cross-references, inconsistent guidelines for applying the 
provisions, and inconsistent instructions for executing the provisions. 
For example, the Inspector General noted that the guidance manual does 
not provide specific guidance on the action required if a farmer and 
field office personnel disagree over the conservation practices to be 
included in the farmer's conservation plan. In reviewing these types of 
cases in Kansas, the Inspector General found that farmers were 
generally granted a special problem waiver in lieu of a noncompliance 
determination, although the guidance manual provides that this type of 
waiver is authorized only when a farmer is actively applying an 
approved conservation plan. In another case, the Inspector General 
found that although the guidance manual allows "properly trained 
personnel" to make visual estimates of crop residue during on-site 
compliance reviews, the manual does not specify the training necessary 
or who will determine whether the person doing a review is qualified to 
make the estimate.

The Inspector General also noted a need for NRCS to better coordinate 
with the Farm Service Agency to ensure consistency among their 
respective policies and procedures concerning the compliance 
provisions. For example, the Inspector General found that although the 
NRCS guidance manual states that the Farm Service Agency rules 
applicable to the agency's Farm Loan Program require annual compliance 
reviews for 5 percent of borrowers producing commodity crops on highly 
erodible cropland, the manual does not explain how this sample should 
be drawn. Similarly, relevant Farm Service Agency handbooks do not 
provide this guidance either. As a consequence, the Inspector General 
found that sample selection methods were inconsistent across the field 
offices examined, leading to inefficiencies such as the inclusion of 
farmers whose loans are not affected by the compliance provisions. The 
Inspector General also found that farmers are not subject to a 
mandatory compliance review at the time a loan is requested, yet once a 
loan is closed, it is unlikely the loan will be called because of a 
conservation compliance violation.

In light of these findings, the Inspector General recommended that NRCS 
undertake certain actions to clarify its written guidance and to better 
coordinate its guidance with guidance issued by the Farm Service 
Agency. In a letter signed by its Acting Deputy Chief for Programs, 
NRCS accepted these recommendations and promised corrective actions by 
August 2003.

Weak Oversight Raises Doubts about NRCS's Assertion of a High Rate of 
Compliance:

NRCS does not effectively oversee its field offices' implementation of 
the conservation provisions, among other things, calling into doubt its 
claim that 98 percent of the tracts reviewed are in compliance with the 
act's conservation provisions. First, NRCS's process for selecting 
tracts for compliance reviews disproportionately emphasizes tracts with 
little or no potential for noncompliance, such as permanent rangelands. 
Such tracts account for about 20 percent of the tracts selected 
annually. Second, NRCS does not have an automated system for promptly 
informing its field offices of the tracts selected for compliance 
reviews and for enabling the field offices to efficiently report the 
results of these reviews. As a consequence, in many cases, the field 
offices do not have the information on the tracts to be reviewed until 
after the critical erosion control period has passed; the reviews 
should be done during these periods. Third, NRCS does not consistently 
collect and analyze the results of the field offices' compliance 
reviews to identify unusual enforcement patterns across regions and 
states and over time. For example, NRCS was not aware, until our 
review, that 30 of the 50 "good faith" waivers granted nationally in 
2000 occurred in just one state. Finally, USDA's Office of Inspector 
General has noted that improvements in NRCS's implementation of the 
conservation provisions are needed to strengthen the agency's ability 
to provide accurate and reliable assessments of farmers' compliance. 
Importantly, these improvements include the need for NRCS to refrain 
from issuing waivers that are not warranted.

NRCS's Compliance Sample Includes Tracts with Little Potential for 
Noncompliance:

NRCS's process for selecting land tracts for compliance reviews gives a 
disproportionate emphasis to tracts that have little potential for 
noncompliance, potentially inflating the farmers' compliance rate 
reported by the agency. To conduct the compliance reviews, NRCS 
randomly selects about 13,000 tracts of land from a Farm Service Agency 
database containing more than 4.5 million tracts of land owned or 
leased by farmers receiving USDA program benefits. Of the 4.5 million 
tracts, 1.7 million are designated as highly erodible land, and the 
remaining 2.8 million are designated as potential wetlands.

Of the 13,000 sample tracts, about 60 percent are selected for highly 
erodible land. The remaining 40 percent are selected from tracts that 
have the potential to contain wetlands. This latter group of tracts is 
separated into two groups--hydric and nonhydric. The hydric group 
includes tracts located in counties where more than 20 percent of the 
soil is classified as hydric--that is, the land is flooded long enough 
during a growing season to support plants that can grow in water or in 
soil too waterlogged for most plants to survive. The remaining tracts 
are placed in the nonhydric group. In general, tracts placed in this 
group have a very low potential to be subject to the conservation 
compliance provisions concerning wetlands. For example, many of these 
tracts are located in arid parts of the country and include permanent 
rangelands. Nevertheless, NRCS draws a relatively large sample from the 
nonhydric group--amounting to about 20 percent of the 13,000 sample 
tracts overall--even though the applicability of the conservation 
compliance provisions to these tracts is very unlikely.

The inclusion of nonhydric tracts for which the conservation compliance 
provisions have little applicability tends to inflate the farmer 
compliance rate reported by NRCS. In 2001, this rate was reported as 98 
percent compliance. For example, 73 field offices providing written 
comments to our survey reported that many tracts selected for 
compliance reviews were not subject to the conservation provisions 
because the tracts did not contain wetlands or highly erodible land or 
because the tracts contained rangeland, timber, or permanent cover 
grass. The results of our field office work tended to reinforce this 
conclusion. For example, we found that 36 tracts selected for the 
wetlands compliance review contained no wetlands. Moreover, some of 
these tracts consisted of permanent rangeland, which is not subject to 
the conservation provisions. Nevertheless, the field offices involved 
reported these tracts as being in compliance, even though the 
conservation provisions were not applicable to them. A number of NRCS 
officials told us that it would be more appropriate for the agency to 
reduce the number of nonhydric tracts reviewed in favor of increasing 
the number of highly erodible land and hydric tracts reviewed.

Another potential issue concerning the sample of tracts selected by 
NRCS for review concerns the size of that sample. For example, two 
groups with whom we spoke, the Wildlife Management Institute and the 
Sustainable Agriculture Coalition, assert that the sample size is too 
small to serve as a deterrent to farmers who may be violating the 
conservation compliance provisions. They note that this sample size has 
dropped from 42,000 in 1997 to about 13,000 annually beginning in 1998; 
the latter number represents about one-quarter of 1 percent of the 4.5 
million tracts potentially subject to the conservation provisions 
nationwide.[Footnote 26] However, these groups have not questioned the 
statistical validity of the sample drawn for projecting to the universe 
of all farmers associated with these 4.5 million tracts. In response, 
NRCS officials told us that they reduced the sample size because of 
resource constraints; higher-priority work related to its other 
programs; and the absence of demonstrated, widespread noncompliance in 
past reviews. In addition, they maintain that this lower number of 
reviews is adequate and statistically valid for projecting nationally. 
However, at the same time, they note that the sample is not large 
enough to project on a state-by-state basis.

NRCS Lacks an Automated System to Promptly Inform Its Field Offices of 
Tracts Selected for Compliance Reviews:

NRCS does not have a nationwide, automated system, such as a web-based 
system, for promptly informing its field offices of the tracts selected 
for compliance reviews and for enabling the field offices to 
efficiently report the results of these reviews. As a consequence, in 
many cases, the field offices do not have the information on the tracts 
to be reviewed until after the critical erosion control period during 
which the reviews should have been done. As a result, NRCS does not 
have a comprehensive picture of erosion during critical periods. 
Furthermore, without such a system, NRCS lacks accurate, up-to-date 
information for oversight to evaluate the field offices' implementation 
of the conservation compliance provisions.

According to NRCS officials, the agency had a nationwide, automated 
system in the mid-1990s that it used to promptly inform its field 
offices of tracts selected for review and to receive the results of 
these reviews. However, NRCS discontinued using this system in 1998 
because it was unsatisfactory for other agency operations, such as 
developing conservation plans for farmers or taking applications for 
USDA's various incentive-based conservation programs. In place of the 
original system, NRCS implemented a new system that more efficiently 
collects information related to these other operations.[Footnote 27] 
However, the new system does not provide a means to efficiently 
disseminate and collect information on compliance reviews.

At present, NRCS uses a cumbersome, multi-step process to disseminate 
and collect information on compliance reviews that does not allow for 
the efficient sharing of information. Specifically, in order to 
disseminate information on tracts selected for review, NRCS 
headquarters must first provide its state offices with the list of 
selected tracts. The state offices then sort and transmit this 
information to supervisory area field offices, which, in turn, sort and 
transmit the information to individual field offices.[Footnote 28] This 
process generally tends to delay the transmittal of information on 
tracts selected for reviews to the field offices that must do these 
reviews until after the critical erosion control periods in which the 
reviews should be done have passed, such as during the spring planting 
period when the residue (ground cover) from previous crops is low, 
rains may be heavy, and the soil is being disturbed for planting. For 
example, in 2000, NRCS's Texas state office received information from 
NRCS headquarters on the tracts selected for review on March 31, but 
did not transmit this information to area offices until April 26, 2000. 
The area office for west Texas did not transmit the selected tracts to 
its field offices until May 2, 2000. As a result of the delays at each 
step in this process, the west Texas field offices did not receive the 
list of tracts for compliance reviews until well after the critical 
erosion control period, which ended in early April.

The current process for disseminating and collecting information on 
compliance reviews also interferes with timely compliance checks that 
should be done in the fall, such as at the time when winter wheat is 
sown. For example, NRCS requires its field offices to complete 
compliance reviews no later than the early fall, which is before the 
critical erosion period in some areas of the country associated with 
crops planted in late fall. NRCS requires reporting by early fall in 
order to allow time for it to enter the information on the results of 
these reviews into its compliance review database before the end of the 
calendar year.[Footnote 29] As with the dissemination of information 
from headquarters to the field offices on the tracts to be reviewed, 
the roll up of the results of these compliance reviews must repeat the 
multi-step process in reverse: field office to supervisory area office; 
area office to state office; and state office to headquarters. In some 
cases, the results of the reviews are provided in electronic files 
attached to emails; in others, the field offices provide the results in 
paper documents that the receiving office must enter into an electronic 
file before it can be passed on to the next level.

USDA's Office of Inspector General has also reported on the need for 
more timely compliance reviews to strengthen the agency's ability to 
provide accurate and reliable assessments of farmers' compliance. In 
its September 2002 report, the Inspector General concluded that 
compliance reviews are not always performed during critical erosion 
control periods because of the untimely distribution of compliance 
review tract selection lists to NRCS state and field office personnel. 
For example, in Kansas, where the critical wind erosion control period 
is March 1 through April 15, the list of tracts selected for compliance 
reviews in 2000 was not received in one field office until April 19 of 
that year. Moreover, guidance from the NRCS's Kansas state office on 
how to download and use the 2000 compliance review list was not issued 
to its field offices until April 27. As a consequence, it was 
impossible for these field offices to perform the required compliance 
reviews during the critical erosion control period.

NRCS headquarters and field staff acknowledge the need for a 
nationwide, automated system, such as a web-based system, for promptly 
informing the agency's field offices of the tracts selected for 
compliance reviews and for enabling these offices to report the results 
of their reviews as they are completed. In general, these staff 
referred to the current process for disseminating and collecting this 
information as piecemeal and inefficient. Furthermore, in a letter 
responding to the Inspector General's September 2002 report, the 
agency's Acting Deputy Chief for Programs stated that NRCS would 
reengineer its compliance review process, including the development of 
new software, to provide for a more timely distribution of the status 
review lists. This official stated that this action would be taken by 
August 2003. However, in January 2003, NRCS headquarters officials 
indicated that uncertainties regarding the agency's appropriation for 
fiscal year 2003 would preclude NRCS from taking this action by August 
2003.[Footnote 30] In addition, these officials said that the 
reengineering of the agency's compliance review process must be weighed 
against other agency priorities for available funding. At present, NRCS 
officials said the timeframe for completing this reengineering is 
uncertain, although they indicated that the agency's plan is, at some 
point, to web-base the data entry for compliance reviews.[Footnote 31]

NRCS Does Not Consistently Collect and Analyze Monitoring Data:

NRCS has not established and maintained a consistent methodology for 
collecting and summarizing compliance review data so that it can (1) 
reliably compare farmers' compliance with conservation provisions from 
year to year and (2) assess its field offices' conduct of compliance 
reviews. According to Office of Management and Budget Circular A-123, 
agencies are required to implement management controls such as policies 
and procedures to reasonably ensure that programs achieve their 
intended results; laws and regulations are followed; and reliable and 
timely information is obtained, maintained, reported, and used for 
decision making. From 1993 through 1997, NRCS collected detailed 
information about the results of its compliance reviews to identify 
trends and anomalies in monitoring farmers' compliance with the 
conservation provisions. NRCS analyzed data at the state level and 
reported information such as the percentage of farmers in 
noncompliance, the types of waivers granted, and the soil erosion rates 
both before and after the application of conservation practices. 
However, in January 2003, NRCS headquarters officials said that 
beginning in 1998 the agency significantly reduced the information it 
gathers because staff reductions and an increasing workload associated 
with its other programs made the collection of this information 
burdensome. In addition, these officials noted that after the 
conservation systems required under the compliance provisions were in 
place by the mid-1990s, NRCS placed less emphasis on collecting data 
related to compliance reviews. For example, NRCS no longer collects 
soil erosion rates before and after conservation practices have been 
applied and has only periodically collected information on the types of 
waivers granted. Furthermore, although NRCS still collects the results 
of the field offices' compliance reviews, it no longer analyzes these 
results to determine consistency across regions and states and over 
time.

As a result of these changes, NRCS is no longer able to determine 
whether the conservation provisions are being consistently applied 
across states and over time. For example, until we brought it to their 
attention, NRCS headquarters staff were unaware that 30 of the 50 good 
faith waivers granted nationally in crop year 2000 occurred in just one 
state. In another case, NRCS was unaware that of the approximately 2.1 
million technical assistance visits that its staff made to farms during 
crop years 2000 and 2001, as discussed, these staff identified 
deficiencies and issued waivers in only 22 instances.

USDA's Office of Inspector General Has Also Noted the Need to Improve 
the Reliability of NRCS's Assessments of Farmers' Compliance:

In addition to improving the timeliness of compliance reviews, as 
discussed, USDA's Office of Inspector General has noted other areas in 
need of improvement regarding NRCS's implementation of the conservation 
provisions. For example, in reports issued in August and September 
2002, the Inspector General cited specific examples in Kansas where 
special problem waivers were approved for circumstances that did not 
appear to meet the established criteria.[Footnote 32] NRCS may grant 
this type of waiver to a farmer if a violation occurred because of 
personal hardship or adverse weather, or if the violation was minor, or 
if it was technical. The Inspector General also noted cases where 
potential compliance deficiencies, identified by NRCS field staff when 
providing conservation technical assistance to a farmer, were not 
subject to follow up status reviews.

Regarding the use of special problem waivers, the Inspector General 
found cases where NRCS area and state office personnel authorized these 
waivers because of agency procedural errors even though the farmers 
involved were not actively applying approved conservation plans or 
systems, as required by the conservation provisions and the agency's 
guidance manual. For example, in one case, the Inspector General found 
that a farmer received a waiver for 2 consecutive years on the basis of 
a minor procedural issue although the district conservationist found 
the farmer to be in noncompliance with the sodbuster provision. The 
waiver was recommended by the area office on the basis of the local 
field office's failure to complete in-office paperwork related to the 
compliance reviews within prescribed time frames although this missed 
internal deadline had no bearing on the district conservationist's 
noncompliance determination. The Inspector General concluded that 
special problem waivers were being used inappropriately to prevent 
farmers from being found to be noncompliant with the conservation 
compliance provisions. As a result, according to the Inspector General, 
the farmers involved were allowed to continue to receive USDA farm 
program benefits even though they were violating the conservation 
provisions.

Concerning compliance deficiencies noted during the on-farm provision 
of conservation technical assistance, the Inspector General found cases 
in one field office where, although the NRCS staff had seen violations 
during the provision of this assistance, these staff did not include 
the farmers involved on a list of producers who would be determined to 
be noncompliant if all noted deficiencies were not corrected. In 
general, this occurred in cases where a farmer requested, but did not 
receive, cost-share assistance for planned conservation practices under 
other USDA programs. Specifically, the Inspector General found that 
this office's philosophy was that the farmer should not be penalized, 
despite the existence of compliance violations, for voluntary efforts 
to apply conservation measures. However, NRCS's national guidance 
manual is clear in these cases: (1) NRCS should inform the farmer of 
actions or practices needed when potential compliance deficiencies are 
noted while providing routine technical assistance. (2) The farmer is 
then required to agree to correct the deficiency, sign a conservation 
plan within 45 days, and implement the necessary conservation system 
within 1 year to remain compliant and eligible for farm program 
benefits. (3) NRCS should conduct a compliance review after a year to 
determine if the farmer took the necessary actions. Again, because of 
the problems noted, the Inspector General concluded that farmers who 
were potentially noncompliant with the conservation compliance 
provisions remained eligible for USDA farm program benefits.

In response to the Inspector General's findings, NRCS agreed to take 
corrective actions. For example, the NRCS Kansas state office indicated 
that it will review all special problem waiver requests on an ongoing 
basis and approve waivers only for those situations that meet the 
criteria for special problems established in national guidance. The 
state office also concurred that potential compliance deficiencies 
observed while providing routine technical assistance are subject to a 
follow-up compliance review, regardless of the presence or absence of 
cost-share assistance; the state office promised a clarifying directive 
to all of its field offices by January 1, 2003, if it found that this 
problem was occurring in other offices.

Farm Service Agency Frequently Waives NRCS Noncompliance Decisions 
without Adequate Justification:

The Farm Service Agency frequently waives NRCS's noncompliance 
determinations but does not always adequately support its decisions. 
From 1993 through 2001, the Farm Service Agency waived NRCS's 
noncompliance determinations in 4,948 of 8,118 cases in which farmers 
were cited with violations, or about 61 percent, in response to 
farmers' appeals. Of the 4,948 appeals leading to waivers, 3,966, or 80 
percent, were considered by the Farm Service Agency's local county 
committees, which found that the farmers had acted in good faith--that 
is, they did not intend to violate the conservation 
provisions.[Footnote 33] Regarding the role of the county committees, 
41 NRCS field offices providing written comments in response to our 
survey noted that because the committee members are fellow farmers, 
they are predisposed to approve farmers' appeals so as not to penalize 
a neighbor's eligibility for farm program benefits. In addition, about 
one-third of our NRCS survey respondents indicated that the Farm 
Service Agency does not adequately justify its waiver decisions. In 
Illinois, Indiana, and Iowa, half of the survey respondents shared this 
view. Without support from the Farm Service Agency, some NRCS field 
office staff said that they have less incentive to conduct compliance 
reviews and issue violations when warranted. In addition, in discussing 
this issue with NRCS headquarters officials in January 2003, they 
expressed surprise when informed of the frequency with which the Farm 
Service Agency waives NRCS's noncompliance determinations.

Our field office visits reinforced the assertion that the Farm Service 
Agency does not adequately justify its waiver decisions. In the five 
offices we visited that had found farmers in violation of the 
conservation provisions, NRCS officials indicated that the Farm Service 
Agency's waivers were not adequately justified. Furthermore, our review 
of the minutes of the Farm Service Agency's county committee meetings 
and other relevant records revealed that these documents did not 
clearly explain the basis for waiving NRCS's noncompliance 
determinations in 34 of the 48 waivers we examined. For example, in 
2001, in one office in Nebraska, we found that the county committee 
waived 8 NRCS noncompliance determinations for a single farmer, even 
though NRCS had already waived 16 violations for this farmer from 1999 
through 2001. However, the committee's minutes and other documentation 
did not clearly state the reason for these waivers.

Because of the waivers granted by its county committees, the Farm 
Service Agency reinstates most benefits that farmers would otherwise be 
ineligible to receive. As shown in table 2, for crop years 1993 through 
2001, of the $59.6 million in benefits that were to be denied because 
of compliance violations, about $40.4 million was reinstated after the 
Farm Service Agency considered farmers' appeals and made its final 
ruling. The table also shows that the benefits actually denied as a 
percentage of benefits to be denied has generally declined over time. 
More detailed information on the benefits denied farmers for 
conservation compliance violations is contained in appendix IV.

Table 2: USDA Benefits Denied Farmers by the Farm Service Agency for 
Violations of Conservation Provisions, Crop Years 1993-2001:



Crop year: 1993; Tracts 
reviewed for compliance[A]: 53,878; Tracts with 
violations: 2,085; Farmers with violations: 
2,860; Benefits to be denied before appeals: 
$17,211; Benefits reinstated by Farm Service 
Agency[B]: $10,416; Benefits reinstated by 
others[C]: $2,067; Benefits denied: $4,483; 
Percentage of benefits denied: 26.0.

Crop year: 1994; Tracts 
reviewed for compliance[A]: 49,314; Tracts with 
violations: 1,639; Farmers with violations: 
2,483; Benefits to be denied before appeals: 
14,845; Benefits reinstated by Farm Service 
Agency[B]: 9,415; Benefits reinstated by 
others[C]: 1,803; Benefits denied: 3,625; 
Percentage of benefits denied: 24.4.

Crop year: 1995; Tracts 
reviewed for compliance[A]: 44,983; Tracts with 
violations: 633; Farmers with violations: 940; 
Benefits to be denied before appeals: 2,838; 
Benefits reinstated by Farm Service Agency[B]: 
1,639; Benefits reinstated by others[C]: 224; 
Benefits denied: 975; 
Percentage of benefits denied: 34.4.

Crop year: 1996; Tracts 
reviewed for compliance[A]: 49,986; Tracts with 
violations: 498; Farmers with violations: 632; 
Benefits to be denied before appeals: 2,302; 
Benefits reinstated by Farm Service Agency[B]: 
967; Benefits reinstated by others[C]: 405; 
Benefits denied: 930; 
Percentage of benefits denied: 40.4.

Crop year: 1997; Tracts 
reviewed for compliance[A]: 49,636; Tracts with 
violations: 183; Farmers with violations: 277; 
Benefits to be denied before appeals: 2,305; 
Benefits reinstated by Farm Service Agency[B]: 
1,622; Benefits reinstated by others[C]: 279; 
Benefits denied: 403; 
Percentage of benefits denied: 17.5.

Crop year: 1998; Tracts 
reviewed for compliance[A]: 15,385; Tracts with 
violations: 205; Farmers with violations: 268; 
Benefits to be denied before appeals: 3,895; 
Benefits reinstated by Farm Service Agency[B]: 
1,988; Benefits reinstated by others[C]: 1,175; 
Benefits denied: 731; 
Percentage of benefits denied: 18.8.

Crop year: 1999; Tracts 
reviewed for compliance[A]: 14,136; Tracts with 
violations: 180; Farmers with violations: 245; 
Benefits to be denied before appeals: 4,959; 
Benefits reinstated by Farm Service Agency[B]: 
4,241; Benefits reinstated by others[C]: 355; 
Benefits denied: 362; 
Percentage of benefits denied: 7.3.

Crop year: 2000; Tracts 
reviewed for compliance[A]: 17,259; Tracts with 
violations: 153; Farmers with violations: 197; 
Benefits to be denied before appeals: 4,870; 
Benefits reinstated by Farm Service Agency[B]: 
4,168; Benefits reinstated by others[C]: 39; 
Benefits denied: 634; 
Percentage of benefits denied: 13.0.

Crop year: 2001; Tracts 
reviewed for compliance[A]: 17,708; Tracts with 
violations: 118; Farmers with violations: 170; 
Benefits to be denied before appeals: 6,385; 
Benefits reinstated by Farm Service Agency[B]: 
5,941; Benefits reinstated by others[C]: 155; 
Benefits denied: 289; 
Percentage of benefits denied: 4.5.

Crop year: Total; Tracts 
reviewed for compliance[A]: 312,285; Tracts with 
violations: 5,694; Farmers with violations: 
8,072; Benefits to be denied before appeals: 
$59,610; Benefits reinstated by Farm Service 
Agency[B]: $40,397; Benefits reinstated by 
others[C]: $6,502; Benefits denied: $12,432; 
Percentage of benefits denied: 20.9.

Source: GAO.

Notes: GAO's analysis of the Farm Service Agency's data.

Benefits denied include price and income support payments, conservation 
payments, disaster payments, guaranteed loans, and crop insurance 
(through 1996). Benefits denied do not include $361,441 that could not 
be directly associated with specific tracts.

Data in table do not include benefits denied participants in the 
peanut-marketing quota program or tobacco-marketing quota program, 
which are reported in pounds. The benefits denied total 2.1 million 
pounds for peanuts and 3.6 million pounds for tobacco.

The decline in the number of tracts and farmers with violations over 
time is attributable, in part, to the reduction in the number of 
compliance reviews being performed each year. Other factors, as noted 
in the results of our survey and field office visits, as well as in 
reports issued by USDA's Inspector General, likely include a misuse of 
waivers, decreasing management emphasis on conservation compliance 
relative to NRCS's other responsibilities, and a continuing reluctance 
to assume the enforcement role called for by the compliance provisions.

[A] Compliance reviews are conducted by NRCS.

[B] Includes $33.9 million reinstated because of good faith waivers and 
$6.5 million reinstated because of tenant and landlord waivers.

[C] Includes benefits reinstated by NRCS State Conservationists, USDA's 
National Appeals Division, and judicial courts after considering 
related farmers' appeals.

[End of table]

In discussing the waiver issue with Farm Service Agency officials in 
January 2003, they noted that the Federal Agriculture Improvement and 
Reform Act of 1996 eliminated graduated payment reductions as a penalty 
for conservation compliance violations, except for sodbuster. Thus, the 
agency's county committees currently have few options when considering 
a farmer's appeal of a noncompliance determination. According to these 
officials, the committees are faced with an "all or nothing" decision: 
either the committee must grant a good faith waiver, continuing the 
farmer's eligibility for benefits and giving the farmer 12 months to 
get back into compliance, or deny the appeal, making the farmer 
ineligible for farm program benefits. These officials added that the 
decision to grant a good faith waiver is, by its nature, subjective, 
not technical; despite the violation, the committee must decide whether 
the farmer acted in good faith and without intent to violate the 
conservation provisions. However, as discussed, one-third of our NRCS 
survey respondents indicated that the Farm Service Agency's waivers are 
not adequately justified, and some field staff commented that the 
granting of these waivers acts as a disincentive to them to make future 
noncompliance determinations when warranted.

Conservation Provisions and Other Factors Have Significantly Reduced 
Soil Erosion and Wetlands Conversions, but Progress Has Slowed in 
Recent Years:

According to USDA, conservation groups, and farm organizations, in the 
17 years since the 1985 act, its conservation provisions have 
contributed to substantial reductions in soil erosion and wetlands 
conversions. However, other factors, such as economic incentives for 
farmers to use new farming techniques and equipment that are more 
conserving of land and water resources, have also contributed. In 
addition, reductions in soil erosion and wetlands conversions have 
leveled off in recent years, and in some areas of the country, soil 
erosion has even increased.

According to NRCS's National Resources Inventory, substantial 
reductions in soil erosion occurred during the 1980s and 1990s as the 
conservation provisions were being implemented.[Footnote 34] The 
nation's soil erosion on all cropland--both highly erodible and 
nonhighly erodible cropland--fell from 3.1 billion tons, or about 7 
tons per acre, in 1982 to 1.9 billion tons, or about 5 tons per acre, 
in 1997, the most recent year for which data are available.[Footnote 
35] The soil erosion rate on highly erodible cropland--land that is 
subject to the conservation provisions--declined by 35 percent from 
1982 through 1997. In 1982, the average annual soil erosion rate 
attributable to water on these lands was about 8 tons per acre, but by 
1997, the rate was about 5 tons per acre.

Less soil erosion helps maintain soil productivity.[Footnote 36] 
According to NRCS, from 1982 through 1997, the percentage of cropland 
on which long-term soil productivity is being depleted declined from 67 
percent to 55 percent. Similarly, for the same period, the percentage 
of highly erodible cropland on which long-term soil productivity is 
being depleted declined from 45 percent to 33 percent. Furthermore, 
reducing soil erosion on cropland has benefits for the general public 
that may be substantial, such as improving water and air quality. For 
example, according to a 2001 USDA study, the societal benefits of 
reducing erosion through conservation compliance exceed $1.4 billion 
per year.[Footnote 37]

However, the proportion of this soil erosion reduction that can be 
attributed to conservation compliance provisions is difficult to assess 
because other factors have affected farmers' decisions. For example, 
some farmers adopted erosion-reducing conservation tillage practices 
over this period because these practices can reduce their crop 
production costs, resulting in increased profits. During this time, new 
machinery and technology allowed farmers to plant their crops with less 
tillage, thereby saving time and money, while also keeping soil-
conserving crop residue on the field. In addition, conservation 
programs, such as USDA's Environmental Quality Incentives Program, 
provided farmers with financial assistance to encourage them to adopt 
conservation practices. Finally, federal, state, and local laws 
addressing other environmental concerns might have also encouraged the 
adoption of conservation practices. Thus, even in the absence of the 
conservation compliance provisions, some farmers would have employed 
conservation practices, and some of this soil erosion reduction would 
have been achieved.[Footnote 38]

Nevertheless, while conservation progress has been substantial, USDA 
considers soil erosion a continuing problem and believes that progress 
in reducing soil erosion has slowed in recent years. In this regard, 
soil erosion on about 27 percent, or 28 million acres, of the cropland 
subject to the 1985 act's provisions is still much higher than on 
cropland generally--an average of 24 tons per acre annually compared 
with the national average of about 5 tons per acre. In addition, about 
50 million of the nearly 273 million acres of the nation's cropland 
that is not subject to conservation compliance is experiencing a high 
rate of erosion and loss in long-term soil productivity, according to 
USDA. Regarding the slowing in progress noted, the Conservation 
Technology Information Center, a nonprofit conservation organization 
that reports biennially on farmers' use of conservation tillage 
practices, found in 2002 that this usage continued a slight decline 
that began in 1998 after increasing during the period 1990 through 
1997. The center also reported that farmers slightly increased their 
use of intensive tillage practices--which result in higher soil erosion 
levels than when conservation tillage is used.

NRCS field offices believe that further reductions in soil erosion are 
possible. Over 80 percent of our survey respondents reported that 
further declines in soil erosion are economically feasible for farmers 
to achieve. In addition, in counties where high levels of soil erosion 
are permitted by alternative conservation systems, 74 percent of the 
field offices reported that further reductions in erosion are 
feasible.[Footnote 39]

With respect to wetlands, wetlands conversions due to agriculture fell 
sharply from an average of 235,000 acres per year before the 1985 
provisions (from 1974 through 1983) to an average of 27,000 acres per 
year after the provisions (from 1992 through 1997), according to USDA. 
However, as with provisions to reduce soil erosion, factors other than 
the wetlands conservation provisions may be responsible, in part, for 
the reductions. According to a USDA study, about half of the original, 
naturally occurring wetlands in the continental United States had been 
drained by 1985 and many of the remaining wetlands might not have been 
converted because these wetlands were not economically feasible to 
convert.[Footnote 40] Furthermore, according to the American Farm 
Bureau Federation's Senior Director of Government Relations, the 
positive effects of the wetlands conservation provisions cannot be 
determined without knowing how other federal, state, and local 
regulations affect wetlands conversions. However, a 2001 USDA report 
cites the large decline in wetlands conversions and credits the 
conservation compliance provisions with discouraging the conversion of 
as much as 3.3 million acres of wetlands.[Footnote 41] In addition, the 
Director of the Wildlife Management Institute stated that the wetlands 
provisions have been very effective in protecting permanent and 
semipermanent wetlands. However, this official said that the provisions 
are only marginally effective in protecting temporary and seasonal 
wetlands because USDA generally identifies these wetlands during the 
summer months when the wetlands are often smaller or completely dry and 
less likely to exhibit other wetlands characteristics. Officials in the 
Department of the Interior's Fish and Wildlife Service also noted that 
the wetlands provisions are less effective for temporary or seasonal 
wetlands for this reason.[Footnote 42]

Conclusions:

The compliance review process serves as NRCS's principal tool for 
monitoring farmers' adherence to the 1985 act's conservation compliance 
provisions. In field offices and in headquarters, however, NRCS's use 
of this tool has fallen short. Improper implementation of the 
conservation provisions increases the likelihood that some farmers are 
still receiving federal farm payments even though they let soil erode 
at higher rates than allowed or convert wetlands to cropland.

As the results of our survey and field office visits indicate, NRCS's 
field offices often do not implement one or more of the key 
conservation compliance provisions designed to control erosion or 
prevent wetlands conversion. Provisions that are often neglected 
include checking for wetlands violations, revisiting farms granted 
compliance waivers the previous year, and citing farmers with 
violations for failing to implement important conservation measures. 
Because of these implementation problems, NRCS cannot be assured that 
its field offices' reports of farmers' compliance with the conservation 
provisions are accurate. A number of conditions contribute to these 
implementation problems, such as resource constraints, the lack of 
management emphasis, and a reluctance to assume an enforcement role. 
Even under these conditions, however, implementation could be improved 
if field office staff received clearer guidance and training so that 
they better understood their roles and responsibilities in implementing 
the compliance provisions, as well as the importance of these 
provisions.

Moreover, flaws in NRCS's oversight monitoring make questionable USDA's 
claim that 98 percent of the nation's cropland tracts subject to the 
conservation provisions are in compliance. First, NRCS's sample of 
tracts selected for compliance reviews reduces confidence in this 
claim. Twenty percent of the sample reviewed includes tracts that are 
not subject to the conservation provisions, such as permanent 
rangelands. Nevertheless, NRCS reports these tracts in compliance. 
Second, the current system for providing the field with information, 
and receiving information from it, does not enable field offices to 
visit tracts when the land is most vulnerable to erosion and to observe 
the effectiveness of farmers' compliance efforts. Third, NRCS is 
collecting less information about the results of the compliance reviews 
than it has in the past, making it difficult to compare farmers' 
compliance with the conservation provisions from year to year. Finally, 
although the information yielded by the compliance reviews may not be 
fully credible, it does suggest inconsistent enforcement. For example, 
10 states issued most of the waivers and violations in crop years 2000 
and 2001. However, NRCS has not used this information to investigate 
the enforcement issues raised.

Lastly, the Farm Service Agency waived NRCS's noncompliance 
determinations about 61 percent of the time during crop years 1993 
through 2001. We found little documentation in the files to support 
these waivers. The frequency and questionableness of these waivers 
undermines NRCS's enforcement efforts. Without support from the Farm 
Service Agency, NRCS's field office staff have less incentive to issue 
violations when warranted.

Recommendations for Executive Action:

To improve USDA's implementation of the conservation compliance 
provisions of the Food Security Act of 1985, as amended, and to better 
protect the highly erodible croplands and wetlands covered by those 
provisions, we recommend that the Secretary of Agriculture direct the 
Chief of NRCS to:

* increase oversight of field offices' conduct of compliance reviews to 
improve the accuracy and completeness of the reviews;

* periodically provide training for field office staff on how 
compliance reviews should be conducted;

* develop a more representative sample of tracts selected for 
compliance reviews that excludes land that is not subject to the 
compliance provisions;

* establish and maintain a consistent methodology for collecting, 
analyzing, and summarizing data to identify patterns and trends in 
enforcement across regions and states and over time; and:

* develop a nationwide, automated system, such as a web-based system, 
for efficiently managing information needed to conduct compliance 
reviews and report results.

In addition, we recommend that the Secretary of Agriculture direct the 
Administrator of the Farm Service Agency to ensure that decisions by 
the Farm Service Agency's field offices to waive NRCS's findings of 
noncompliance are justified and documented.

Agency Comments and Our Evaluation:

We provided USDA with a draft of this report for its review and 
comment. We received oral comments from NRCS officials, including the 
Deputy Chief for Strategic Planning and Accountability and the Director 
for Operations Management and Oversight. We also received written 
comments from the Farm Service Agency.

NRCS officials concurred with our recommendations and indicated that 
they have begun steps to implement them. These officials also said that 
they generally agreed with the report's findings but found the tone of 
the report to be overly negative. Specifically, these officials said 
that the report focuses too much on problems with the agency's 
implementation of the conservation compliance provisions and not enough 
on the provisions' positive accomplishments in reducing soil erosion 
and wetlands conversions.

We do not believe that the report is overly critical. NRCS's written 
guidance sets an expectation that its field staff consistently follow a 
set of procedures to determine farmers' compliance with the 
conservation provisions. On the basis of our survey responses from over 
2,000 NRCS field offices, we found that nearly half of these offices do 
not consistently follow one or more of these procedures when conducting 
compliance reviews. Moreover, our field office visits revealed a 
similar pattern, thus reinforcing the survey results. Regarding the 
provisions' positive accomplishments, the report discusses the 
substantial reductions in soil erosion and wetlands conversions 
attributed to the provisions by USDA and others. The report also 
presents data on the extent of these accomplishments.

NRCS officials also objected to our characterization of NRCS's role in 
implementing the conservation compliance provisions as having an 
"enforcement" component. These officials said that NRCS is not an 
enforcement agency. They explained that NRCS makes technical 
determinations of farmers' compliance or noncompliance with the 
conservation provisions and that the decision to withhold farm program 
benefits for noncompliance rests with the Farm Service Agency. While we 
understand NRCS's sensitivity to this issue, we nevertheless believe it 
is accurate to describe its role in the provisions' implementation as 
including an enforcement component.

The conservation compliance provisions require that farmers who receive 
federal assistance meet standards for environmental quality. By setting 
such standards for agricultural activity, the provisions represent a 
departure from USDA's traditional role of implementing soil and water 
conservation programs that are voluntary and incentive based. NRCS is 
the lead agency for administering the conservation compliance 
provisions. Its responsibilities include the performance of compliance 
reviews to verify farmers' implementation of the conservation systems 
required by the provisions. These reviews are an integral step in the 
enforcement of the provisions. A determination of noncompliance, 
potentially leading to a farmer's loss of eligibility for farm program 
benefits, rests with NRCS. This responsibility exists and is intrinsic 
to enforcement even if another agency must take action, on the basis of 
NRCS's finding and recommendation, to withhold this eligibility.

The Farm Service Agency also agreed with the report's recommendations. 
In addition, the agency generally agreed with the report's findings, 
but it downplayed the significance of our finding that the agency's 
county committees waived 61 percent of the NRCS noncompliance 
determinations made during 1993 through 2001. For example, the agency 
said that its waivers do not negate or overrule an NRCS determination 
that a violation occurred. The agency noted that although a "good 
faith" waiver granted by a county committee allows a violating farmer 
to continue receiving program benefits, the farmer must still take 
corrective action within 1 year, incurring any associated costs. In 
addition, the Farm Service Agency noted that the issuance of good faith 
waivers has aided the restoration of converted wetlands and the 
implementation of conservation systems on highly erodible land tracts 
that were brought back into compliance. According to the agency, it 
could be argued that many of these tracts would not have been brought 
back into compliance if eligibility for benefits had not been 
reinstated under the good faith waiver. The agency added that the 
overall purpose of the conservation provisions is not to deny benefits, 
but rather to achieve conservation compliance.

We agree with the Farm Service Agency's assessment that the purpose of 
the conservation provisions is not to deny farmers benefits. However, 
the Farm Service Agency's written guidance requires that county office 
committees grant good faith waivers judiciously and only when supported 
by conclusive evidence indicating that the farmer did not intend to 
violate the provisions. In addition, NRCS officials were concerned at 
the extent to which the Farm Service Agency waives NRCS's noncompliance 
determinations. These officials expressed the view that many of these 
waivers are not justified and that the high number of waivers tends to 
undermine NRCS's implementation of the conservation compliance 
provisions, thus giving its field staff less incentive to issue 
violations when warranted.

NRCS and the Farm Service Agency also provided technical corrections, 
which we have incorporated into the report as appropriate. The Farm 
Service Agency's written comments are presented in appendix VI.

As agreed with your office, unless you publicly announce its contents 
earlier, we plan no further distribution of this report until 30 days 
from the date of this letter. We will then send copies to interested 
congressional committees; the Secretary of Agriculture; the Director, 
Office of Management and Budget; and other interested parties. We will 
make copies available to others on request. In addition, the report 
will be available at no charge on GAO's Web site at http://www.gao.gov.

If you have any questions about this report, please contact me at (202) 
512-3841. Key contributors to this report are listed in appendix VII.

Sincerely yours,

Lawrence J. Dyckman
Director, Natural Resources and Environment:

Signed by Lawrence J. Dyckman:

[End of section]

Appendix I: U.S. Department of Agriculture's Conservation Compliance 
Review Process:

[See PDF for image]

Note: GAO's analysis of U.S. Department of Agriculture's (USDA) 
compliance review process.

[A] In addition to the random sample of tracts identified, NRCS field 
offices select additional tracts for review on the basis of referrals 
from other agencies (e.g., tracts owned by employees of other USDA 
agencies), tracts associated with farmers who receive farm loans, 
whistleblower complaints, tracts with potential violations observed by 
NRCS employees when providing technical assistance; and tracts 
associated with farmers granted a waiver the prior year.

[B] We use the term "waiver" in this case to refer to variances and 
exemptions given by NRCS. A variance continues a farmer's eligibility 
for federal farm program benefits when the farmer is unable to apply a 
conservation system or practice because of severe or unusual conditions 
such as weather, disease, or pests; because the farmer experienced an 
extreme personal hardship or unusual occurrence, such as illness or 
death; or because the deficiency is minor and technical in nature. An 
exemption maintains a farmer's eligibility for benefits when a 
violation is identified while NRCS staff are providing on-site 
technical assistance. The 1-year period to correct the deficiency does 
not apply to the severe or unusual conditions waiver.

[C] Farmer has 30 days to request a field review for reconsideration or 
mediation of NRCS's initial noncompliance decision. If the farmer does 
not exercise this option within the 30-day period, NRCS's noncompliance 
determination becomes final and NRCS refers the matter to FSA for 
further action.

[D] The benefits to be withheld from a farmer may include price and 
income support payments, conservation payments, disaster payments, and 
access to guaranteed loans.

[E] Farmer may appeal FSA's ineligibility determination first to the 
local FSA county committee and then, if necessary, to USDA's National 
Appeals Division. If the farmer disagrees with the National Appeals 
Division's decision, the farmer may file suit in federal district 
court.

[F] We use the term "waiver" in this instance to refer to variances and 
exemptions given by the local FSA county committees. A variance 
continues a farmer's eligibility when the farmer is unable to implement 
a conservation system because doing so would cause undue economic 
hardship. An exemption maintains the farmer's eligibility when the 
farmer acted in good faith and without intent to violate the 
conservation provisions or when a landlord prevents a tenant farmer 
from implementing an approved conservation system. Similarly, in cases 
in which the tenant farmer's violation is not attributable to actions 
of the landlord, the landlord may receive an exemption that continues 
the landlord's eligibility for benefits regarding other tracts. The 1-
year period to correct the deficiency does not apply to the tenant and 
landlord waivers.

[End of section]

Appendix II: Objectives, Scope, and Methodology:

At the request of the Ranking Democratic Member of the Senate Committee 
on Agriculture, Nutrition, and Forestry, we reviewed USDA's 
implementation of the conservation provisions of the Food Security Act 
of 1985. Specifically, we agreed to determine (1) how well NRCS's field 
offices are carrying out these provisions; (2) how effectively NRCS 
oversees its field offices' efforts to carry out these provisions; (3) 
how often FSA waives NRCS's noncompliance determinations; and (4) the 
extent to which these conservation provisions have helped to reduce 
soil erosion and the loss of wetlands.

To determine how well NRCS's field offices are carrying out the 
conservation provisions, and how effectively NRCS's headquarters 
oversees its field offices' efforts to carry out these provisions, we 
examined NRCS's national database on the results of compliance reviews 
for crop years 1998 through 2001 to identify unusual patterns in 
compliance enforcement. Automated data for these reviews were not 
available for years prior to 1998. We also examined the guidance that 
NRCS's field offices use to monitor farmers' compliance with the 
conservation provisions, including relevant laws; the Code of Federal 
Regulations, title 7, part 12; and agency guidance, including NRCS's 
National Food Security Act Manual and related state amendments and 
bulletins.

We also surveyed staff--usually the district conservationist--
responsible for compliance reviews in each of NRCS's 2,549 field 
offices that conducted compliance reviews during the period 1998 
through 2001 to obtain information on that official's understanding and 
implementation of the conservation provisions, as well as the 
official's views on the effectiveness of these provisions. To obtain 
the views of each field office official, we developed an electronic 
questionnaire that was posted on GAO's home page on the Internet. In 
developing the questionnaire, we met with officials in NRCS's 
headquarters to gain a thorough understanding of highly erodible land 
and wetlands conservation issues. We also shared a draft copy of the 
questionnaire with these officials who provided us comments including 
technical corrections. We then pretested the questionnaire with two 
NRCS district conservationists in Texas, and one in Maryland and 
Virginia. During these visits, we asked the officials to fill out the 
survey over the Internet. After completing the survey, we interviewed 
the respondents to ensure that (1) the questions were clear and 
unambiguous, (2) the terms we used were precise, (3) the questionnaire 
did not place an undue burden on the agency officials completing it, 
and (4) the questionnaire was independent and unbiased. On the basis of 
the feedback from the pretests, we modified the questions as 
appropriate.

Information about accessing the questionnaire was provided via E-mail 
for those NRCS staff selected to participate in the survey. The survey 
was activated, and staff informed of its availability on September 10, 
2002; it was available until October 2, 2002. To ensure security and 
data integrity, we provided each NRCS field official with a password 
that allowed him or her to access and complete a questionnaire for the 
local office. No one else could access that questionnaire or edit its 
data. We also provided these officials with a pledge of confidentiality 
to ensure their candor in completing the survey.

We received responses from 2,015, or 79 percent, of the officials 
surveyed. Table 3 shows the number of field offices that participated 
in our survey and each state's response rate. The results of our survey 
are summarized in appendix V.[Footnote 43] For survey results 
stratified by state, see a special publication entitled Agricultural 
Conservation: Survey Results on USDA's Implementation of Food Security 
Act Compliance Provisions (GAO-03-492SP), which is available on the 
Internet at http://www.gao.gov/cgi-bin/getrpt?gao-03-492SP.

Table 3: Number of Field Offices Participating in Our Survey That 
Regularly Conduct Compliance Reviews and the Response Rate:

State: Alabama; Field offices surveyed: 62; Field offices responding: 
50; Percent response rate: 80.6.

State: Alaska; Field offices surveyed: 4; Field offices responding: 4; 
Percent response rate: 100.0.

State: Arizona; Field offices surveyed: 14; Field offices responding: 
12; Percent response rate: 85.7.

State: Arkansas; Field offices surveyed: 63; Field offices responding: 
45; Percent response rate: 71.4.

State: California; Field offices surveyed: 49; Field offices 
responding: 33; Percent response rate: 67.3.

State: Colorado; Field offices surveyed: 55; Field offices responding: 
44; Percent response rate: 80.0.

State: Connecticut; Field offices surveyed: 1; Field offices 
responding: 1; Percent response rate: 100.0.

State: Delaware; Field offices surveyed: 1; Field offices responding: 
1; Percent response rate: 100.0.

State: Florida; Field offices surveyed: 42; Field offices responding: 
31; Percent response rate: 73.8.

State: Georgia; Field offices surveyed: 84; Field offices responding: 
61; Percent response rate: 72.6.

State: Hawaii; Field offices surveyed: 5; Field offices responding: 5; 
Percent response rate: 100.0.

State: Idaho; Field offices surveyed: 39; Field offices responding: 28; 
Percent response rate: 71.8.

State: Illinois; Field offices surveyed: 97; Field offices responding: 
77; Percent response rate: 79.4.

State: Indiana; Field offices surveyed: 79; Field offices responding: 
63; Percent response rate: 79.7.

State: Iowa; Field offices surveyed: 108; Field offices responding: 81; 
Percent response rate: 75.0.

State: Kansas; Field offices surveyed: 109; Field offices responding: 
94; Percent response rate: 86.2.

State: Kentucky; Field offices surveyed: 91; Field offices responding: 
73; Percent response rate: 80.2.

State: Louisiana; Field offices surveyed: 49; Field offices responding: 
45; Percent response rate: 91.8.

State: Maine; Field offices surveyed: 12; Field offices responding: 11; 
Percent response rate: 91.7.

State: Maryland; Field offices surveyed: 22; Field offices responding: 
15; Percent response rate: 68.2.

State: Massachusetts; Field offices surveyed: 7; Field offices 
responding: 6; Percent response rate: 85.7.

State: Michigan; Field offices surveyed: 60; Field offices responding: 
48; Percent response rate: 80.0.

State: Minnesota; Field offices surveyed: 83; Field offices responding: 
69; Percent response rate: 83.1.

State: Mississippi; Field offices surveyed: 80; Field offices 
responding: 66; Percent response rate: 82.5.

State: Missouri; Field offices surveyed: 104; Field offices responding: 
80; Percent response rate: 76.9.

State: Montana; Field offices surveyed: 61; Field offices responding: 
45; Percent response rate: 73.8.

State: Nebraska; Field offices surveyed: 82; Field offices responding: 
73; Percent response rate: 89.0.

State: Nevada; Field offices surveyed: 7; Field offices responding: 5; 
Percent response rate: 71.4.

State: New Hampshire; Field offices surveyed: 6; Field offices 
responding: 5; Percent response rate: 83.3.

State: New Jersey; Field offices surveyed: 7; Field offices responding: 
4; Percent response rate: 57.1.

State: New Mexico; Field offices surveyed: 35; Field offices 
responding: 24; Percent response rate: 68.6.

State: New York; Field offices surveyed: 47; Field offices responding: 
39; Percent response rate: 83.0.

State: North Carolina; Field offices surveyed: 86; Field offices 
responding: 72; Percent response rate: 83.7.

State: North Dakota; Field offices surveyed: 56; Field offices 
responding: 50; Percent response rate: 89.3.

State: Ohio; Field offices surveyed: 76; Field offices responding: 61; 
Percent response rate: 80.3.

State: Oklahoma; Field offices surveyed: 79; Field offices responding: 
62; Percent response rate: 78.5.

State: Oregon; Field offices surveyed: 27; Field offices responding: 
21; Percent response rate: 77.8.

State: Pennsylvania; Field offices surveyed: 46; Field offices 
responding: 40; Percent response rate: 87.0.

State: Rhode Island; Field offices surveyed: 2; Field offices 
responding: 1; Percent response rate: 50.0.

State: South Carolina; Field offices surveyed: 42; Field offices 
responding: 29; Percent response rate: 69.0.

State: South Dakota; Field offices surveyed: 65; Field offices 
responding: 54; Percent response rate: 83.1.

State: Tennessee; Field offices surveyed: 71; Field offices responding: 
55; Percent response rate: 77.5.

State: Texas; Field offices surveyed: 201; Field offices responding: 
157; Percent response rate: 78.1.

State: Utah; Field offices surveyed: 21; Field offices responding: 15; 
Percent response rate: 71.4.

State: Vermont; Field offices surveyed: 8; Field offices responding: 6; 
Percent response rate: 75.0.

State: Virginia; Field offices surveyed: 54; Field offices responding: 
41; Percent response rate: 75.9.

State: Washington; Field offices surveyed: 34; Field offices 
responding: 26; Percent response rate: 76.5.

State: West Virginia; Field offices surveyed: 32; Field offices 
responding: 23; Percent response rate: 71.9.

State: Wisconsin; Field offices surveyed: 61; Field offices responding: 
52; Percent response rate: 85.2.

State: Wyoming; Field offices surveyed: 23; Field offices responding: 
12; Percent response rate: 52.2.

State: Total; Field offices surveyed: 2,549; Field offices responding: 
2,015; Percent response rate: 79.1.

Source: GAO.

[End of table]

We also visited 20 NRCS field offices located in 19 counties in 5 
states (Arkansas, Colorado, Nebraska, North Dakota, and Texas) to 
discuss implementation of the compliance provisions with relevant staff 
and to review documentation of compliance decisions, including waivers, 
in order to determine the basis for these decisions. We selected these 
field offices on the basis of such criteria as the amount of highly 
erodible cropland covered by the office that is subject to the 
conservation provisions, geographic dispersion, and apparent anomalies 
in USDA data related to compliance reviews, waivers, and penalties 
assessed (such as an office that appears to issue an inordinately large 
number of waivers). Specifically, in these field offices, we spoke with 
the district conservationist and/or other staff and reviewed 
documentation on file, including Form AD-1026 (referrals for highly 
erodible land and/or wetlands determinations); aerial photography; 
conservation assistance notes; conservation plans; soil loss 
computations; status review results; and correspondence, including 
requests for waivers.

In addition, we reviewed relevant studies prepared by USDA's Office of 
Inspector General and the Congressional Research Service, as well as 
our own past reports. We also reviewed annual Food Security Act status 
(compliance) reviews prepared by NRCS or FSA, and interviewed officials 
from these agencies in the field and headquarters. Furthermore, we 
reviewed USDA's Fiscal Year 2003 Annual Performance Plan and Revised 
Plan for Fiscal Year 2002 and NRCS's Initial Performance Plan for 
Fiscal Year 2003 and Revised Plan for Fiscal Year 2002 to determine 
what performance goals and measures USDA has established for soil 
erosion reduction and wetlands conservation.

To determine how often FSA waives NRCS's noncompliance decisions, we 
examined FSA's database on violations and benefits withheld or waivers 
granted for crop years 1993 through 2001 to determine trends in 
assessing penalties for noncompliance. We also spoke with FSA field and 
headquarters staff regarding the reasons for waiving NRCS's 
noncompliance determinations and reviewed relevant documentation, 
including the minutes of county committee meetings, correspondence 
files, and appeal files. In addition, we reviewed FSA guidance, 
including its manual on highly erodible land conservation and wetlands 
conservation provisions.

Finally, to determine the extent to which the conservation provisions 
have helped to reduce soil erosion and loss of wetlands, we reviewed 
the results of USDA's National Resources Inventory and other related 
studies. In addition, we spoke with officials from USDA's Economic 
Research Service; the Department of the Interior's Fish and Wildlife 
Service; farm organizations, including the American Farm Bureau 
Federation and American Soybean Association; and conservation groups, 
including Ducks Unlimited, the Soil and Water Conservation Society, the 
Sustainable Agriculture Coalition, and the Wildlife Management 
Institute.

We conducted our review from April 2002 through February 2003 in 
accordance with generally accepted auditing standards. Although we did 
not independently assess the accuracy and reliability of the USDA data 
we used, we reviewed the data for reasonableness among regions and 
compared them with data in other USDA reports.

[End of section]

Appendix III: Results of NRCS's Compliance Reviews, by State, Crop 
Years 2000 and 2001:

Table 4: Results of NRCS's Compliance Reviews Showing Violations and 
Waivers by State, Crop Year 2000:

[See PDF for image]

Source: NRCS.

Notes: GAO's analysis of NRCS's data.

Violations occur when a farmer does not apply conservation practices on 
highly erodible land or the farmer's tract contains a potential 
wetlands violation. Waivers allow farmers to maintain eligibility for 
federal farm benefits when they are unable to apply a conservation 
practice because the farmer was prevented by severe or unusual 
conditions such as weather, disease, or pests; because the farmer acted 
in good faith, and without intent to violate the conservation 
provisions; because the farmer experienced an extreme personal hardship 
or unusual occurrence, such as illness or death; because failures were 
minor and technical in nature; because the farmer was identified in 
noncompliance by NRCS while providing on-site technical assistance; or 
because the farmer was conditionally applying practices until the 
approved conservation practices are completed. An on-site visit is not 
required when an exemption for a specific tract is approved.

[A] Tracts reviewed for compliance includes 13,025 tracts randomly 
selected by NRCS headquarters and another 4,234 tracts added by NRCS 
field offices on the basis of referrals from other agencies, 
whistleblower complaints, and tracts that maintained eligibility for 
farm benefits based on previous year waivers.

[B] Total tracts with violations largely reflects NRCS's preliminary 
noncompliance determinations, as of the date these data were compiled. 
Many of these determinations were subsequently reversed by NRCS--
through granting variances or exemptions--on appeal from the affected 
farmers. Tracts referred b;y NRCS to FSA are only those tracts for 
which NRCS sustains its preliminary noncompliance determination.

[End of table]

Table 5: Results of NRCS's Compliance Reviews Showing Violations and 
Waivers by State, Crop Year 2001:

[See PDF for image]

Source: NRCS.

Notes: GAO's analysis of NRCS's data.

Violations occur when a farmer does not apply conservation practices on 
highly erodible land or the farmer's tract contains a potential 
wetlands violation. Waivers allow farmers to maintain eligibility for 
federal farm benefits when they are unable to apply a conservation 
practice because the farmer was prevented by severe or unusual 
conditions such as weather, diseases, or pests; because the farmer 
acted in good faith, and without intent to violate the conservation 
provisions; because the farmer experienced an extreme personal hardship 
or unusual occurrence, such as illness or death; because failures were 
minor and technical in nature; because the farmer was identified as 
being in noncompliance by NRCS while providing on-site technical 
assistance; or because the farmer was conditionally applying practices 
until the approved conservation practices are completed. An on-site 
visit is not required when an exemption for a specific tract is 
approved.

[A] Tracts reviewed for compliance includes 13,544 tracts randomly 
selected by NRCS headquarters and another 4,164 tracts added by NRCS 
field offices on the basis of referrals from other agencies, 
whistleblower complaints, and tracts that maintained eligibility for 
farm benefits based on previous year waivers.

[B] Total tracts with violations largely reflects NRCS's preliminary 
noncompliance determinations, as of the date these data were compiled. 
Many of these determinations were subsequently reversed by NRCS--
through granting variances or exemptions--on appeal from the affected 
farmers. Tracts referred by NRCS to FSA are only those tracts for which 
NRCS sustains its preliminary noncompliance determination.

[End of table]

[End of section]

Appendix IV: USDA Benefits Denied Farmers Because of Conservation 
Compliance Violations, Crop Years 1993-2001:

Table 6: USDA Benefits Denied Farmers by the Farm Service Agency for 
Violations of Conservation Provisions, Nationwide, Crop Years 1993-
2001:

Dollars in thousands:

Crop Year: 1993; Highly erodible land provisions: Tracts with 
violations: 1,893; Highly erodible land provisions: Farmers with 
violations: 2,592; Highly erodible land provisions: Benefits 
to be denied before appeals: $12,748; Highly erodible 
land provisions: Benefits denied: $3,005; Wetlands provisions: Tracts 
with violations: 192; Wetlands provisions: Farmers with violations: 
268; Wetlands provisions: Benefits to be denied before appeals: 
$4,463; Wetlands provisions: Benefits denied: 
$1,478; Wetlands provisions: Total benefits 
denied: $4,483.

Crop Year: 1994; Highly erodible land provisions: Tracts with 
violations: 1,530; Highly erodible land provisions: Farmers with 
violations: 2,303; Highly erodible land provisions: Benefits 
to be denied before appeals: 10,692; Highly erodible 
land provisions: Benefits denied: 2,243; Wetlands provisions: Tracts 
with violations: 109; Wetlands provisions: Farmers with violations: 
180; Wetlands provisions: Benefits to be denied before appeals: 
4,153; Wetlands provisions: Benefits denied: 
1,382; Wetlands provisions: Total benefits 
denied: 3,625.

Crop Year: 1995; Highly erodible land provisions: Tracts with 
violations: 605; Highly erodible land provisions: Farmers with 
violations: 892; Highly erodible land provisions: Benefits 
to be 
denied before appeals: 2,674; Highly erodible 
land provisions: Benefits 
denied: 968; Wetlands provisions: Tracts with violations: 28; Wetlands 
provisions: Farmers with violations: 45; Wetlands 
provisions: Benefits to be denied before appeals: 0; 
Wetlands provisions: Benefits denied: 0; 
Wetlands provisions: Total benefits denied: 968.

Crop Year: 1996; Highly erodible land provisions: Tracts with 
violations: 402; 
Highly erodible land provisions: Farmers with violations: 
491; Highly erodible land provisions: Benefits 
to be 
denied before appeals: 1,266; Highly erodible 
land provisions: Benefits 
denied: 492; Wetlands provisions: Tracts with violations: 96; Wetlands 
provisions: Farmers with violations: 141; 
Wetlands provisions: Benefits to be denied before appeals: 1,036; 
Wetlands provisions: Benefits denied: 439; 
Wetlands provisions: Total benefits denied: 931.

Crop Year: 1997; Highly erodible land provisions: Tracts with 
violations: 150; 
Highly erodible land provisions: Farmers with violations: 
215; Highly erodible land provisions: Benefits 
to be 
denied before appeals: 1,391; Highly erodible 
land provisions: Benefits 
denied: 334; Wetlands provisions: Tracts with violations: 33; Wetlands 
provisions: Farmers with violations: 62; Wetlands 
provisions: Benefits to be denied before appeals: 913; 
Wetlands provisions: Benefits denied: 69; 
Wetlands provisions: Total benefits denied: 403.

Crop Year: 1998; Highly erodible land provisions: Tracts with 
violations: 167; 
Highly erodible land provisions: Farmers with violations: 
220; Highly erodible land provisions: Benefits 
to be 
de