Transcript Pest control

Pesticide Economics
David Zilberman
EEP 101/Econ 125:
Spring 2003
7/21/2015
General Overview
• There are three major classes of pesticides:
– Insecticides
– Fungicides
– Herbicides
• Pesticides are useful in controlling agricultural pests.
The adverse human health effects of pesticides tend
to depend on the similarity between human biology
and the target pest:
– That is, a “monkeycide” would be worse for human health
than a “raticide.”
– Fungicides are generally worse for human consumption than
insecticides.
7/21/2015
Pesticides: Damage Control Agents
• Pests include:
– Big animals (elephants, coyotes)
– Small creatures (mice, birds)
– Insects
– Viruses
– Weed
• Control types--Chemical
– Agronomic: fences,hoes, tractors, traps
– Biological: cats, dogs, predators of pests
– Seed varieties including genetically modified crops:
7/21/2015
pest resistant, pesticides tolerant
Changes in Pesticide Use in the U.S.
7/21/2015
A Brief History of Pesticide Use
• Herbicides: From 1965 to 1980, growth in the relative price of labor
increased the use of herbicide as a factor of production. This occurred
because herbicide is a substitute for labor-intensive work. During the 1980s,
lower agricultural commodity prices and reduced crop acreage led to a
reduction in herbicide use.
• Insecticides: During the 1970’s, the creation of the EPA and an increase in
energy prices led to a reduction in insecticide use.
• Fungicide: Fungicide use has remained relatively stable over the past 30
years, although recent legislation banning the use of carcinogenic chemicals
in the Delaney Clause will soon outlaw many fungicides (and several
popular insecticides and herbicides).
7/21/2015
Modeling Pest-Control Choices
•
•
•
•
•
•
•
•
•
•
Y = OUTPUT Z = INPUT (fertilizer)
Q = g(Z) - potential output
X = pesticides-damage control agent
d(N) = fraction damaged, N = final pest population
N = h(X, M) M = initial pest population, pesticides
reduce population from M to N
Y = g(Z)*(1 - d(h(X))
Firms aim to maximize profits
P = output price, W = input price, V = pest-control
price
A = fixed application cost
Profit = Pg(Z)*(1 - d(h(X,M)) - Z*W - V*X - A
7/21/2015
Pest Population and Pest Control
•
•
•
•
•
•
•
At optimal solution
VMPZ = P(∂g/ ∂ Z) *(1 - d(n(X,M)) = W.
Value of marginal product of input = input price.
VMPX= -P g(Z)*∂d/∂N *∂n/∂X = V.
Value of marginal product of pest control = its price.
Larger initial population requires more application.
If initial population is sufficiently small and does not
cover fixed application cost, do not apply.
• Application is warranted if a population threshold has
exceeded. Apply only if M > threshold.
• Estimation population is costly.
7/21/2015
Effective Pest Management & Biology
• Predator-prey consideration
– Suppose two pests cause damage, and N1 &
N2 denote their populations.
– Pest 1 is the predator of 2 so
– N1 = n1(X1, M1) & N2 = n2 (N1, M2)
– The optimal rule for applying pest control 1 -X1
VMPX1= MCN2 + V
– The value of pesticides in controlling pest 1 =
marginal cost of larger population of pest 2 + pest
control cost.
• Control of pests that are also predators of other
pests should recognize their benefits and
7/21/2015
reduce application levels accordingly.
Preventive vs. Responsive Application
• Pest population arrival time and size are random.
• Preventive applications. Based on average
performance; may lead to overspraying. Standard
spraying based on a large population will occur when
pests do not arrive or population is small.
• Responsive application requires costly monitoring of
population; will save chemicals but require costly
monitoring and may lead to slow or incomplete
response to invasion.
• Integrated pest management. Relies on monitoring of
pest population and combines a mixture of strategies
that aim to minimize use of chemicals.
7/21/2015
The Economic Threshold
• When the pest damage rises above the social cost of a
pesticide application, it is welfare-maximizing to apply the
pesticide.. The economic threshold, , is the pest population
leveln0
at which it becomes optimal to apply the pesticide.
The economic threshold is determined by setting total pest
damage equal to the total cost of a single pesticide application
and solving
for
:
n0
Pg(Z)D(n0 )  w
• where the value of pest damage to output (at the threshold
level) is and w is the cost of applying pesticide.
7/21/2015
Unknown Pest Populations and Pest
Population Monitoring
• With preventive application, pesticides are
applied without an attempt to determine
potential pest populations. Instead, based on
experience or historical data, the farmer
makes educated guesses about the
probabilities of various pest population levels
occurring.
• With reactive application, a fixed monitoring
cost is paid to determine the pest population level,
and then the optimal X is chosen for the specific
7/21/2015
pest level.
Resistance
• Occurs when efficacy of pesticides declines as
use of chemical increases over time.
• Since pests move across farms, it is a common
problem. Individual producers ignore future
resistance cost associated with pesticides use.
• Policy intervention
• (Ideally) Incentives (tax or subsidy) to reflect
the social cost of resistance.
– Use regulation to limit the use of materials to
“worthwhile“ situations.
– Research to identify alternatives.
7/21/2015
Pesticide Resistance management
• Short run pesticide control problems in a given
season will be inefficient if long term resistance
effects are not considered. Therefore, the
calculation of optimal dosage of pesticide should
take into account:
–resistance buildup (pesticide effectiveness is an
"exhaustible resource" and should be modeled as
such)
–use of alternative chemicals or alternative pest
control methods (such as the use of alternative
cropping methods, crop rotation, natural diseases
and predator-prey relationships).
7/21/2015
One Person’s Pest Is Another
Person’s Game
• The definition of pests is relative:
–Elephants damage farms but can be a source of
eco-tourism income.
–Feral pigs cause damage to field crops, but
many will pay to hunt them.
• Pest management strategies should take
advantage of activities that will take
commercially utilize pests and reduce the
cost of pest control.
• The beneficiaries of “green” pest control
7/21/2015
methods should pay to support them.
Regional Cooperation in Pest
Control Activities
• Pests do not recognize property rights.
– Pest control districts e.g., mosquito control districts are
introduced to overcome externality problem in pesticides
application,.
– Their activities include monitoring activities, coordinating
crop management, and pesticide spraying
. Current pesticide policy separates pesticide
economics from health considerations.
New policy is triggered solely by health
considerations: when a chemical is found to be
carcinogenic or damaging to the environment, it is
banned, or “canceled”.
7/21/2015
Economic Impacts of Pesticide
Cancellation
• To estimate overall short-term impacts, the impacts on
yield per acre and cost per acre are evaluated using
one of the following methods:
– Delphi method: The delphi method uses "Guesstimates by
experts", which are easy to obtain but are arbitrary and
sometimes baseless. (Named after the famous Òracle at
Delphi in ancient Greece.)
– Experimental studies: These studies are based on data from
agronomical experiments, but experimental plots often do
not reflect real farming situations.
– Econometric studies: Statistical methods based (ideally) on
data gathered from real farming operations. However, these
studies are often not feasible because of data limitations and
7/21/2015
the difficulty of isolating the specific effects of pesticides.
Cost-Budgeting Method
y ij=
= output per acre of crop i at region j with pesticides
Pij»
= price crop i, region j
•
Aij
= acreage crop i, region j.
•
y ij
= yield reduction per acre because of cancellation
•
c ij
=cost reduction per acre because of cancellation
•
•
Under a partial crop budget, impacts on social welfare are estimated
as:
•
I
J
  P Y
ij
ij
 Cij Aij
i1 j 1
• or,
a pesticide cancellation causes losses in revenue from lower yields per
7/21/2015
acre and increased costs per acre, which is multiplied the total acreage in all
The cost budgeting approach has
several limitations.
1. It ignores the effect of a change in output on
output price. This tends to overestimate
producer loss and underestimate consumer
loss.
2. It ignores feedback effects from related
markets.
• In general, this method does not consider the
interaction of supply and demand, and does not
attempt to find the new market equilibrium after
the application of a pesticide ban.
7/21/2015
Impact assessment should recognize
price effects
• Banning pesticides will reduce supply and
increase output prices.
• If demand is inelastic producers may gain.
Consumers-especially poor ones, will lose.
• Environmentalists may gain
demand
Supply without pesticides
Supply with
pesticides
7/21/2015
Simulated impacts of Proposition 128 on five crops in
California.
7/21/2015
Impacts of pesticide bans on major
commodities.
Impact
Wheat Barley Rice
Corn
Cot ton Soybean SorghumPeanut
(% change)
Yield
Production
Price
Export
-25
-29
-57
-32
-39
-37
-20
-70
-9
-12
-39
-18
-30
-26
+4
-17
6
23
83
38
34
100
13
146
-15
-22
-64
-26
-46
-50
-35
-8
Source: Knutson, Taylor, Penson, Smit h (10).
Ban will increase prices reduce export may benefit
producers but will hurt consumers. Bad for livestock,
7/21/2015
good for filed crops
Pesticide Regulation and
Agricultural Policy The assessment of the
cost of banning should
take into account
subsidies-when food is
subsidized, the
marginal cost of food
lost is smaller than the
price-since the price is
inflated by the
subsidy.Sometimes
ban is improving
welfare as it will reduce
oversupply
7/21/2015
Health Risks of Pesticides
• Food safety—mortality or morbidity resulting
from chemical residues—include:
– Acute impacts—poisoning, allergic responses.
Poisoning when packaging materials used for food
consumption
– Chronic impacts—cancer.
– Much uncertainty about the food safety effect.
• Worker safety—damage to mixer applicator and
farmers may be high, especially if caution is not
taken.
– May be control by protective clothing reentry
regulations etc.
7/21/2015
Environmental side effects of pesticides
– Damages to fish, birds,beneficial species,beneficial
insects.
– Some pesticides are possible or definite endocrine
disrupters (block the action of male hormones)
– Some categories of pesticides are more damaging
to the environment than others. DDT was banned
for its environmental effects
– Impact depends on mode of operation and
persistence.
– Biological control also may have significant effects
on the environment ( the mongoose in Hawaii)
7/21/2015
Controlling Pesticide Externalities
• Registration requirements. Before a product is
introduced, it must pass a battery of test to identify
obviously risky products (carcinogens).
• Incentives. Taxes and subsidies to pay for damages.
• Limits on total use. Tradable permits to users.
• Ban. Complete or partial bans on chemicals?.
• Restrictions on applications. Limits on when, where,
and how chemicals are applied (e.g., not near schools,
when it is windy, or aerially spraying).
• Direct control. Protective clothing, food treatment
requirements, and reentry regulation to sprayed fields.
• Education and information. Notification regarding
spraying
activities and possible exposure risks.
7/21/2015
•
Possible Pesticide Use Levels
Price
Social optimum=point A
Monopoly =Price C Quantity B
Competitive outcome=point E
Monopoly
Price
Demand
MB of production
E
B
MEC=Marginal
Externality Cost
MC +MRC+MEC
A
7/21/2015
MB=Marginal Benefit
MRC=Marginal
Resistance Cost
C
Marginal
revenue
MC=Marginal Cost
MC
Quantity
Possible Use Levels of Pesticides
• If a manufacturer is a monopoly (has a patent), there
may be under-use of pesticides if a monopoly price
is greater than marginal externality and
resistance costs.
• Social optimum occurs if marginal benefits of
pesticides in production equals sum of marginal
externality, resistance, and production cost.
• Without intervention, the most use occurs where
marginal benefits equal marginal cost of production.
7/21/2015
Pesticides in Developing Countries
• Under-application in some situations.
– Many developing countries are in the humid tropic with major
pest problems, but not many have pest control tools, since
most pesticides have been developed for problems in
developed countries and temperate zones.
– Adaptation of pest control solutions is costly, and ability to
pay for companies’ investment are limited.
– Pesticide application equipment is costly, and peasants
frequently face credit constraints.
• Techniques such as bio-control (mealybug in cassava) and
GMOs are especially useful (and easy to apply and spread).
– Safety rules may not be followed, and there are cases of
overappliation.
7/21/2015
Overapplication in Others
• There is lack of enforcement of environmental
regulation, resulting in overuse and exposure.
• Pesticide patents may not be registered or
recognized, and cheap old generic ones are
used.
• Pesticides may be subsidized in some countries
(China).
• Cheap materials may be combined with cheap
application equipment, and unregulated setup
will
lead
to
environmental
dangers.
7/21/2015
The Good and Bad Sides of Pesticide Use
• Average pest losses in Indian cotton are 50-60%.
Insect pests losses In the United States and China are
12% &15%, respectively..
• Yield-increasing pesticides may prevent deforestation
and acreage of farming.
• The low productivity effect of pesticides in rice in the
Philippines and Indonesia, combined with worker
safety effects, suggests much overuse there.
• Banning chemicals in most cases is suboptimal. The
problem is not chemicals but how they are being used.
7/21/2015
From Chemicals to GMO
• Pesticide regulations have triggered introduction of
new chemicals and GMOs.
• Bt cotton has reduced pesticide applications in US
and china by 50-60%,but yield effects are between
0-5%. In India yield effects are +50%.
• The high pest pressure in developed countries and
lack of pesticides suggest high yield-increasing
potential for GMOs.
• Effort in adaptation and development of appropriate
genetic materials and access to Intellectual Property
Rights are needed.
• Possible externalities need to be inspected.
7/21/2015
A General Problem: Policies to
Control Environmental Risks
• The impacts of policies are uncertain, and the
environment is subject to stochastic forces.
• Methodologies to both model risk and analyze choices
under risk are crucial for effective policymaking.
• There are alternative approaches to risk. Economic
and decision theoretic models measure risk as
deviations from the norm or average. They
emphasize assessing the impact of such deviations on
behavior and their cost.
• Public health develops risk assessment techniques
that define risk explicitly as the probability of data
outcome.
7/21/2015
Properties of Risk Assessment Models
• Risk = probability that a member of a population
will die or get sick during a period of time.
• Risk-generating functions = relationship between
risk and processes that cause it.
• The knowledge needed for risk-generation
functions is interdisciplinary. It provides the base
for both estimation and policymaking.
• Risk assessment models can be used to assess
– Human health risk
– Environmental health risk (risk to fish)
– Food security
7/21/2015
Chemical Application Risk
Pollution control
policies
Risk
=
Barriers/filters
Contamination
Protective
clothing
Medical
treatment
Dose/
Transfer&
Exposure
*
*
* Response
fate
Risk of chemical residues can be reduced by
*Reducing application levels through taxes, direct
control,etc.
*Blocking movement of residue to and in bodies of water
(can be induced by incentives).
*Reducing
human exposure by filters,protective clothing.
7/21/2015
*Treatment in case of poisoning and injury.
Farm Worker Pesticides Risk
• Let r = represent individual health risk where
• r = f1(X,B1) f2(B2)
f3(B3)
initial exposure exposure dose/response
• X = pollution on site (i.e., the level of pesticide use)
• B1 = damage control activity at the site (i.e., protective
clothing; re-entry rules)
• B2 = averting behavior of individuals (i.e., washing
fruits and vegetables)
• B3 = the medical control of pollution dosage.
7/21/2015
Modeling Environmental Risk
• The modeling principles used to model human
health risk from pesticides also apply to
modeling risk to, say, birds.
• There are processes of contamination transfer
and fate exposure and dose response (transfer
and fate and contamination are most important in this context).
• These processes are controlled through
policies.
7/21/2015
Policy Optimization under Risk
• A reasonable policymaking principle• The objective is to maximize economic welfare
subject to the constraint
•
Probability (Risk < R) > 
–
–
R = target level of risk
. = safety level (measures the degree of social risk
aversion)
 might represent the degree of confidence we have
in our risk estimate.
• For example, policymakers may aim to maximize
economic surplus given that risk from pesticides
cannot
exceed 1 million with a 95% probability.
7/21/2015
Uncertainty and Assessment
• Use of higher degree of statistical reliability, 
leads to higher risk estimate. The risk of a chemical
may not increase .05 with  .95, but may exceed
it with  .995 .
• Is useful to use consistent reliability requirements for
all risk estimates to allow comparisons.
• It may be useful to identify a target group in the
population (say, top 95% in terms of vulnerability) and
compare how policies affect risk to this group.
7/21/2015
Sources of Variability
• Coefficients of risk-generation functions vary.
– We may not have a reliable number representing coefficients
of specific processes.
– The risk function may be r =  * b * g *X and the coefficients
may be stochastic.
• The causes of variability:
– Heterogeneity can be handled by more specific analysis.
– Randomness.
– Uncertainty (lack of knowledge) can be reduced by learning.
7/21/2015
Pesticide Registration
• Main form of policy is pretesting and registration.
• Objective is to eliminate “risky” pesticides and
minimize side effects.
• Once a product is discovered to be problematic, it may
be banned or its use restricted.
• Intensive testing is beneficial to corporations because
it increases entry costs ($50+ million to introduce a
new chemical) and assures their market power.
• It reduces availability of new products and results in
“orphan” diseases, especially used in specialty crops
and developing countries.
• Governments and donors may need to subsidize
introduction
of new products beneficial to society but
7/21/2015
not to private sectors.
Pesticide “Doctors”
• Productivity of pesticides can be enhanced, and
their environmental impact reduced if their
performance is monitored and decisions on what
and when to apply are optimized.
• One approach is to restrict diagnosis of pesticides
to certified pesticide consultants and applications
to certified applicators.
• Extension can train both types of professionals.
They can also be required to document pesticide
applications and may be liable for wrong choices.
• Optimal sharing of liability for mistakes is a
7/21/2015
challenge, but if done correctly can improve policy.
7/21/2015