United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens GA 30613
Research and Development
EPA-600/S5-83-001 Sept. 1983
oEPA Project Summary
Effect of Agricultural Land Use
Practices on Stream Water
Quality: Economic Analysis
John A. Miranowski, Michael J. Monson, James S. Shortle, and Lee D. Zinser
Since nonpoint source pollution from
agricultural lands is to be controlled by
best management practices (BMPs), a
good understanding of the response of
farmers to incentive and regulatory
policies is needed if there is to be
effective implementation. This report
provides an economic analysis of agri-
cultural production activities and farmer
response to water quality controls using
the Four Mile Creek Watershed in Iowa
as a case study.
To address some of the limitations of
the existing information base on the
efficiency of BMPs, the investigation
included four components. First a
watershed survey provided a profile of
the farmers and the necessary data for
the empirical analysis. Second, a BMP
cost-effectiveness evaluation frame-
work was developed and applied. Third,
factors that are important in explaining
the adoption of BMPs, such as perceived
risk, were identified. Fourth, the impact
of increasing relative energy prices on
crop management practices and on the
effectiveness of nonpoint source pollu-
tion control policies was assessed.
The analysis showed that secondary
tillage practices are important in de-
signing cost-effective control strategies.
Knowledge of perceived risk associated
with practices is critical in determining
farmer adoption of BMPs and response
to control policies. Also, data on per-
ceived opportunity costs of certain
practices provided additional under-
standing of farmers' reluctance to a-
dopt those practices. A BMP evaluation
framework that accommodated the
scaling up from fields to watersheds
was ultimately developed. This analysis
showed that the marginal cost of soil
loss control increases significantly as
more stringent limits are adopted.
Also, it was determined that higher
energy prices are likely to enhance the
effectiveness of soil loss control policies.
This Project Summary was developed
by EPA's Environmental Research Lab-
oratory, Athens, GA, to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).
Introduction
Society's concern for improving water
quality is reflected in the Federal Water
Pollution Control Act Amendments (FWPCA)
of 1972 (PL 92-500) and in the Rural
Clean Water Program (RCWP) of 1977
(PL 95-217). Section 208 of FWPCA-
1972 requires states to develop areawide
plans for controlling nonpoint sources of
water pollution (including agricultural
sources) and the RCWP has provided a
framework and funding mechanism for
implementing the program. Since the
control or prevention of nonpoint source
pollution from agricultural land is directed
at site-specific management (implying that
compliance monitoring will be conducted
on land rather than water), it is essential
that economically feasible land use practices
or Best Management Practices (BMPs) be
selected. The Four Mile Creek evaluation
project is an attempt to develop the neces-
sary information base.
Previous nonpoint source (NPS) pollu-
tion control evaluation projects, based on
linear programming models designed to
maximize profits or minimize costs of
control, have utilized secondary data sources
-------
in developing coefficients for the model
and have assumed constant input prices.
Little attention has been given to the
actual information or data used by farmers
in forming their perceptions of profitability.
If farmers' yield expectations for new crop
management practices differ from those
employed in the model coefficients, the
actual responses of farm operations to
public incentives and regulations may be
significantly understated or overstated.
Likewise, if the secondary data sources
used m establishing the costs and returns
coefficients do not represent the actual
production environment, further bias will
be introduced into the programming model
results.
Past evaluations based solely on linear
programming models also suffer from the
assumption that only profitability matters
to farm operators. Other factors, such as
risk, however, may play an important role
in farm management decisions. If other
objectives enter into the farm operator's
decision function and if these factors are
ignored, the linear programming results
will not be accurate and may not be reliable
for planning purposes.
In any modeling effort, the investigator
must choose a set of input and output
prices that reflect farmer expectations or
professional judgment. Yet, relative price
patterns, especially for inputs, typically
change over time and affect the relative
profitability of different crop management
practices. For example, reduced tillage
practices may become more profitable
relative to moldboard plowings as energy
prices rise over time. Such changes in
relative prices, and in turn profitability,
may induce adjustments in tillage practices
that have favorable environmental impacts.
It is important not only to identify the
environmental consequences of relative
price changes but to recognize these ad-
justments when designing incentives and
regulations to induce the adoption of
BMPs.
In an effort to address some of the
limitations of the existing information base
on the efficiency of BMPs, this investigation
includes four major components. First
the watershed survey provides a profile of
the farmers and the necessary data for the
empirical analysis. Second, a BMP cost-
effectiveness evaluation framework is de-
veloped and an application is illustrated.
Third, factors that are important in explain-
ing the adoption of BMPs are identified.
Finally, the impact of increasing relative
energy prices on crop management prac-
tices and on effectiveness of NPS pollution
policies is assessed.
Profile of Four Mile Creek
Watershed
To establish a primary data base for the
analysis, a survey of farmers in Four Mile
Creek, Iowa, was undertaken. A five-
section questionnaire designed to inven-
tory machinery, crop management practices,
livestock management practices, risk atti-
tudes, and background information was
administered. The objectives of the survey
were to determine (1) existing manage-
ment practices, (2) how farmers would
adjust their production practices given
incentives, and (3) coefficients for BMP
evaluation and programming models.
A few findings of the survey, crucial to
the economic analysis conducted, should
be highlighted. First, as more erosive
primary tillage practices were reduced,
secondary tillage practices increased in
the watershed. Thus, the potential soil
loss reduction attributed to the reduction
of primary tillage operations may be some-
what overstated when farmers shift to
conservation tillage. Second, watershed
farmers expected the highest yields from
fall tillage activities, followed by spring
tillage, and then no-till. In addition to
having the lowest expected yield, no-till
had the highest risk rating assigned by
farmers. Knowledge of perceived yield
impacts of different tillage systems as well
as perceived risk associated with practices
is critical in determining farmer adoption
of BMPs and response to NPS pollution
control policies. Third, the data suggest
that Four Mile Creek farmers tend to be
risk neutral to risk averse in their behavioral
preferences. Different formulations of risk
questions were employed and relatively
consistent responses were received. Final-
ly, data on perceived opportunity costs of
terracing, including both time and yield
reduction costs, provided further under-
standing of farmers' reluctance to terrace
cropland.
BMP Cost-Effectiveness
Evaluation Framework
To evaluate the cost-effectiveness of
alternative BMPs, an activity analysis
framework was developed. The framework
accounts for the various costs associated
with different management practices and
levels of residual control. The impacts of
scaling-up from fields to farms, farms to
subwatersheds, and subwatersheds to
watersheds can be evaluated.
The preliminary results indicated that
contouring is the most cost-effective BMP
of the subset considered. Yet, contouring
is only effective as an erosion management
practice over a relatively narrow range of
residual control. In other words, contouring
can reduce gross soil loss in the Four Mile
Creek Watershed from 1 7 to 9 tons per
acre per year but no further. The same soil
loss reduction with terracing or pasturing
is far less cost-effective. Also, as supported
by other studies, the marginal cost of
additional increments of soil loss (residual)
control increases significantly regardless
of the BMP imposed.
To compare and contrast the imposition
of BMPs at different levels of aggregation,
the Four Mile Creek Watershed was divided
into five subwatersheds based on dif-
ferences in soil type and erosion potential.
Supporting the argument frequently made
for targeting soil conservation funding, the
average cost of reducing soil loss is signifi-
cantly higher in subwatersheds with less
serious erosion hazards. Put another way,
specific BMPs are more cost-effective in
areas with greater erosion hazards, other
things equal. Further investigation of the
implications of BMP targeting is needed.
Adoption of Conservation Tillage
Practices
Given the emphasis that EPA has placec
on management practices for the improve
ment of water quality, it is important tc
understand why farmers do or do no
adopt particular BMPs. Specific attentior
was given to the conservation tillage prac-
tice adoption decision. To isolate the
factors that influence the adoption of con-
servation tillage practices (i.e., a BMP), £
linear probability model of tillage choice
was developed. The probability that i
farmer will choose a tillage practice othe
than moldboard plowing is related tc
characteristics of the farm operator, in-
cluding experience, tenure, education, anc
risk perceptions, and to characteristics o
the farm, including field topography, crop
rotation, and farm size.
The adoption model was fitted to 23£
observations of primary tillage choice f rorr
the Four Mile Creek Watershed survey
The results should be useful in the desigr
of more effective soil loss control policy
First, farmers with more education anc
experience had a higher probability o
adopting conservation tillage practices
These results point to the importance o
considering the human capital character
istics of farmers when designing education
incentive, and targeting programs to induce
soil conservation practices.
Second, the structural characteristics o
the farm operation produced some surpris
ing results. Farm size proved to be <
negative factor in the adoption decision
i.e., the probability of adoption was in
-------
versely related to the size of farm operation.
Even though larger scale operations gene-
rally would be expected to have an incentive
to acquire and adopt improved technology
because they can spread the costs over
more units, the smaller operations actually
exhibited a higher probability of adoption.
Only if adoption of conservation tillage
practices had negative scale effects or if it
were generally less economical would this
result be expected. Existing experimental
and survey data do not support these
contentions. Also contrary to the findings
of some previous studies as well as to a
popularly held belief, a more tenured
operator (i.e., a larger share of owner-
operated land) was not more likely to
adopt conservation tillage practices than a
less tenured operator. Economic theory
supports the hypothesis that tenure should
not matter if landowners are attempting to
maximize the long run net returns to their
investment
Third, potential erosivity or field topo-
graphy was also found to be a strong factor
in explaining the probability of adoption.
As was to be expected, farm operators on
hilly land had a higher probability of adopt-
ing conservation tillage practices. Addition-
ally, failure to account for potential erosion
problems may explain why previous studies
found a significant positive relationship
between tenure and conservation practice
adoption.
Finally, differences in risk perceptions
and expected yields were considered.
Differences in risk perceptions influenced
the probability of adoption but the statistical
significance of the result was somewhat
weak. The result suggests that improved
information directed toward developing
more accurate perceptions of the riskiness
of reduced tillage may be useful. Expected
yield differences were not significant in
explaining adoption, but this result may be
due to the small expected variation in yields
in the region.
How far the results from the Four Mile
Creek Watershed can be generalized is
open to question, but they do indicate
some potentially important factors that
should be considered in designing future
soil conservation programs.
Effects of Rising Energy
Prices on Nonpoint Source
Pollution and Its Control
Changing relative input prices may con-
found the implementation of NPS pollution
control options. Such price changes may
complement the desired policy impacts, or
alternatively, may compete with the policy
objectives. Rising relative energy prices
are of particular interest because of the
impact of energy prices on the choice of
tillage, rotation, and cultural practices.
Thus, the formation of NPS pollution con-
trol policies should be cognizant of the
impacts of increasing relative prices of
farm fuel and other energy related inputs.
To assess the impacts of rising energy
prices on soil loss and on the potential
effectiveness of IMPS pollution control
policies, a linear programming model for
the Iowa River Basin was used. Depart-
ment of Energy 1985 and 1990 price
projections for fuels were used to generate
model solutions to compare with the 1978
baseline case.
The overall policy implications were that
higher energy prices tended to bring about
shifts to practices that decrease both soil
erosion and the amount of fertilizer used
by farmers, even without encouragement
from NPS pollution policies. These reduc-
tions are caused by a projected shift from
continuous corn to corn-beans rotation
and by a partial shift from fall chisel to no-
till planting. Significant changes occurred
in both soil loss and net farm income by
1985, with 22 and 19 percent decreases,
respectively. Little additional adjustment
occurred between 1985 and 1990.
Combining the impacts of increasing
energy prices and per acre soil loss restric-
tions, soil loss restrictions are only effective
if the average annual is set at less than 10
tons per acre. Subsidies for soil loss
abatement combined with increasing energy
prices do provide ambiguous results. At
lower subsidy levels, increasing energy
prices will enhance the effectiveness of
the subsidy policy, but at higher subsidy
levels increasing energy prices may weaken
the impact of the abatement subsidy.
Generally, in the politically relevant range
of subsidies as well as over the likely range
of soil loss restrictions, higher energy
prices will likely enhance the effectiveness
of soil loss control policies.
John A. Miranowski, MichaelJ. Monson, James S. Shortle, andLeeD. Unset are
with Iowa State University, Ames, IA 50011.
Thomas E. Waddell is the EPA Project Officer (see below).
The complete report, entitled "Effect of Agricultural Land Use Practices on Stream
Water Quality: Economic Analysis," (Order No. PB 83-217 836; Cost: $14.50,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
Athens, GA 30613
<5U.S GOVERNMENT PRINTING OFFICE 1983-659-017/7183
-------
United States Center for Environmental Research Fees Paid
Environmental Protection Information Environmental
Agency Cincinnati OH 45268 Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
Pt> 0000329
U S ENVIR PROTECTION AGENCY
REGION 5 LIBRARY
230 S OEARbORN STREET
CHICAGO IL 60604
------- |