The World’s Largest Open Access Agricultural & Applied Economics Digital Library

This document is discoverable and free to researchers across the

globe due to the work of AgEcon Search.

Help ensure our sustainability.

Give to AgEcon Search

AgEcon Search
http://ageconsearch.umn.edu
[email protected]

Papers downloaded from AgEcon Search may be used for non-commercial purposes and personal study only.
No other use, including posting to another Internet site, is permitted without permission from the copyright
owner (not AgEcon Search), or as allowed under the provisions of Fair Use, U.S. Copyright Act, Title 17 U.S.C.
 Ind. Jn. of Agri. Econ.
Vol.67, No.3, July-Sept. 2012

Economic and Environmental Impact of Bt Cotton in India

K.R. Ashok, K. Uma, M. Prahadeeswaran and H. Jeyanthi*
I

COTTON IN INDIA

India is the largest cotton cultivating country in terms of acreage and second
largest in terms of production in the world. India accounts for 34 per cent of the
cotton area and 20 per cent of the cotton production in the world. But India ranks
only 46th in terms of productivity with a yield of about one tonne per hectare (FAO,
2011). The yield of cotton is one of the lowest among the leading cotton producing
countries in the world. The other major cotton producing countries in the world are
China, USA and Pakistan. In India there are about nine mega cotton growing states
with more than one lakh hectare area under cotton in 2008-09. Of the total cotton
area, two states, Maharashtra and Gujarat, alone account for nearly 58 per cent of the
cotton area in the country. In terms of production Gujarat alone accounts for nearly
32 per cent of the cotton production in the country though it has only 25 per cent of
the cotton area. It is because of high productivity of cotton in Gujarat. Gujarat ranks
third in terms of productivity in the country after Punjab and Haryana. On the other
hand, Maharashtra with 33 per cent of the area accounts only for 21 per cent of the
cotton production because of low productivity. The productivity of Maharashtra is
one of the lowest in the country. Tamil Nadu has a comparatively lower share in
production but ranks fourth in terms of productivity in India. Cotton is predominantly
grown as a rainfed crop in India. About 64 per cent of the cotton crop in the country
is grown without irrigation. The extent of irrigation varies widely across states, with
Punjab growing the entire cotton crop under irrigation and Maharashtra growing
nearly 95 per cent of the crop under rainfed condition. At the all India level only 36
per cent of the cotton crop is irrigated.

II

COMMERCIALIZATION OF AGRICULTURAL BIOTECHNOLOGY IN INDIA

Over the last fifteen years, modern agricultural biotechnology has been adopted
rapidly at the global level, including several developing countries. This trend has

*Professor (Agricultural Economics), Associate Professor (Agricultural and Rural Management), Assistant
Professor (Agricultural Economics), Tamil Nadu Agricultural University, Coimbatore-641003 and Administrative
Officer (R&D), Agriculture Insurance Company of India Ltd., respectively.
406 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

been most apparent for genetically modified (GM) crops. This is the fastest diffusion
of any new crop technology in the history of humankind (Qaim, 2005). The number
of countries growing biotech crops has increased steadily from 6 in 1996, the first
year of commercialisation, to 18 in 2003 and 29 in 2010. The growth rate between
1996 and 2010, the 15th anniversary of the commercialisation of biotech crops, was
an unprecedented 87-fold increase making it the fastest adopted crop technology in
recent history. In India 6.3 million small farmers benefited from planting 9.4 million
hectares of Bt cotton, equivalent to a high adoption rate of 86 per cent (Clive, 2010).
In India efforts to harness genetic engineering technology for boll worm resistance in
cotton began in the 1990s with the import of genetically modified cotton and
initiation of research programmes in national laboratories (APCoAB, 2006). The first
approval of commercial cultivation of Bt cotton in India was granted in 2002 to three
Bt cotton hybrids, MECH-12, MECH-162 and MECH-184, developed by the seed
company Mahyco following a series of trials approved by the Government of India.
In the first year of Bt cotton commercialisation in India, only three Bt hybrids had
been approved, which were grown on large areas in different regions. Since these
hybrids were not well adapted to all environments, the productivity advantage
associated with the Bt gene was partly offset by general germplasm disadvantages in
some locations (Qaim et al., 2006). The crop failure along with other concerns has
led to an intense scientific debate and public controversy. Much of this controversy
surrounds ethical arguments, or concerns for multinational control of the world’s
seeds, human health, or environmental risk (Herring 2007a, b).
Technological development and dissemination is not an isolated scientific
activity; rather it is embedded in and influenced by the social, economic, cultural and
political conditions. This holds true for modern biotechnology also. As the wider
global society struggles to come to terms with the benefits and (unknown) risks of the
biotechnology, better understanding of consumer interests and concerns is needed to
formulate and implement effective private and public policies (Onyango et al., 2004).
Factors influencing consumer preferences for bioengineered crops include their
perceptions of benefits and risks of bioengineered crops on human health and the
environment, their ethical stance toward genetic engineering, and their trust in
government regulations concerning risk assessment and management (OECD, 2000).
Hence a successful biotechnological programme needs to address the socio-economic
issues, biosafety aspects, intellectual property management, etc. to ensure that public
and policy makers are equipped to make informed choices. Analysing the socio-
economic impacts of this new technology is important in the context of India
emerging as one of the leading countries with large acreages under biotech crops.
Currently (in 2010) India with 9.4 million hectares under Bt cotton is the fourth
largest country in terms of acreage of biotech crops in the world after USA (66.8
mha), Brazil (25.4 mha) and Argentina (22.9 mha). Also out of the 15.4 million
biotech farmers in the world, 6.3 million are in India. When the study was conducted
in 2007-08 Bt cotton was the only GM crop approved for commercial cultivation in
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 407

India and all the biotech farmers in India were cotton farmers. In the above scenario
this study addressed issues like reduction in pesticide use, increase in yield and
income, profitability, gender dimension and environmental consequences of Bt cotton
technology in India.

III

METHODOLOGY

A. Sampling Procedure
The study was conducted through survey method. The data for the study was
collected from farmers through an interview schedule. The farmers were selected
through a multistage stratified sampling procedure to select a representative sample at
each stage. In the first stage, four states were selected considering the area,
production, productivity and the spread of Bt technology. Accordingly, the States of
Maharashtra, Gujarat, Andhra Pradesh and Tamil Nadu were selected for the study.
In the next stage, the major cotton producing districts/regions were identified
following the same criteria as used for selecting the States. For the purpose, districts
wise area under cotton for each State was collected from the State Department of
Agriculture/State Agricultural Universities/Department of Economics and Statistics.
From each district, two major cotton growing taluks/blocks, and from each
taluk/block three to five villages were identified considering the area under cotton
and spread of Bt technology in consultation with the Agricultural Development
Officers/Staff of State Agricultural Universities in the respective States. In all, 120
farmers were selected at random from each State comprising 80 Bt cotton growers
and 40 non Bt cotton growers, making an aggregate sample size of 480 farmers from
four States. The survey was conducted during 2007-08. State wise sample distribution
is given below.
(i) Gujarat
Gujarat is the largest cotton producer in India, and stands second in terms of area
under cotton. Surendranagar district alone cover 21 per cent of the total cotton area
in the State, while Bhavnagar (11.17 per cent), Rajkot (10.99 per cent), Vadodra
(8.86 per cent), Ahmedabad (8.47 per cent), Amreli (7.37 per cent) and Bharuch (7.36
per cent) are the other major cotton growing districts. Surendranagar and Rajkot
districts were selected based on the procedure discussed in the methodology and 120
Bt and non-Bt farmers were contacted at the village level as detailed in Table 1.
(ii) Maharashtra

Maharashtra is the second largest producer of cotton though it ranks first in terms
of area with 3.15 million ha under cotton. The productivity of cotton in Maharashtra
408 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

is one of the lowest among the major cotton producing States in the country. In
Maharashtra, nearly 38 per cent of cotton is cultivated in Amravati division followed
by Nashik and Latur divisions. From the Amravati region Amravati and Yawatmal
districts are the major cotton growing districts.

TABLE 1. VILLAGE WISE DISTRIBUTION OF SAMPLE FARMERS IN GUJARAT

District Block Villages No. of farmers

(1) (2) (3) (4)
Surendranagar Halvad Mayurnagar 10
(60) Juna Devalia 10
Nava Devalia 10
Vadhvan Moolchand 10
Bhadresh 10
Bakarthali 10
Rajkot (60) Doraji Torania 10
Naniparabadi 10
Motiparabadi 10
Upleta Motipaneta 10
Hariyasan 10
Kharachiya 10
Total 120

The selected blocks and villages and number of farmers contacted from each
village are given in Table 2. In each village, ten respondents were contacted
comprising both Bt and non-Bt cotton growers.

TABLE 2. VILLAGE WISE DISTRIBUTION OF FARMERS IN AMRAVATI AND YAWATMAL DISTRICTS

District Block Village No. of farmers

(1) (2) (3) (4)
Amravathi Amravathy Aasrigav poorna 10
Nanori 10
Javla Shapur 10
Chandoor Bazar New Akola 10
Sukhali 10
Nandorai 10
Yawatmal Ner Watfali 10
Manikwada 10
Dhanaj 10
Yawatmal Vagapur 10
Jambavadi 10
Laasina 10
Total 120

(iii) Andhra Pradesh

Andhra Pradesh is the third largest cotton growing state in India with 1.40 million
hectares under cotton. In Andhra Pradesh, the districts, viz., Adilabad, Guntur and
Warangal are the major cotton growing districts and each district covers an area of 15
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 409

per cent of the total cotton area in the state. The geographical and climatic features of
Adilabad district are similar to Yawatmal district of Maharashtra. Yawatmal district
was already covered under the study to represent Maharashtra and hence Adilabad
district was not selected. Guntur and Warangal districts were selected from Andhra
Pradesh and the sampling details are given in Table 3.

TABLE 3. VILLAGE WISE DISTRIBUTION OF FARMERS IN GUNTUR AND WARANGAL DISTRICTS

District Block Village No. of farmers

(1) (2) (3) (4)
Guntur Dachapalli Srinivasapuram 10
Dachepalli Village 10
Thadikonda Nidumukkula 10
Guntur rural Tonnalagadda 10
Warangal Bachennapet Itkyapalli 10
Nakkavarigudam 10
Thamidapalla 10
Chandrapuram 10
Devaruppala Devaruppala 10
Sitaramapuram 10
Kadavendi 10
Nirmala 10
Total 120

(iv) Tamil Nadu

Tamil Nadu was purposively selected as it is one of the earlier adopters of Bt

technology and has large area under Bt cotton. In Tamil Nadu, nearly 15 per cent of
the area under cotton is in Virudhunagar district, while Salem (9.04 per cent),
Madurai (8.38 per cent), Perambalur (7.59 per cent), Dharmapuri (6.84 per cent),
Coimbatore (6.53 per cent) and Thoothukudi (6.39 per cent) are the other major
cotton growing districts. Due to non-availability of adequate Bt cotton growers in the
southern districts (Virudhunagar and Madurai) of Tamil Nadu, Salem and Perambalur
districts were purposively selected where Bt cotton is widely cultivated. In this area

TABLE 4. VILLAGE WISE DISTRIBUTION OF FARMERS IN TAMIL NADU

District Block Village No. of farmers
(1) (2) (3) (4)
Salem Kolathur Ellaraimathikadu 10
Kannamoochi 10
Settiyur 10
Mecheri Aandikavundanur 10
Virudhasampatti 10
Mallikundham 10
Perambalur Kunnam Andhur 10
Chitthali 10
Anukur 10
Veppanthattai Nergunam 10
Kudikadu 10
Nergunam 10
Sub-total 120
410 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

Bt cotton cultivation is widespread and it was difficult to find the non-Bt cotton
growers in the same village where Bt cotton was grown. Hence in order to get Bt and
non-Bt samples in the same agro-climatic zone, six blocks were selected in total to
meet the adequate number of non-Bt cotton sample farmers. The details of blocks
and villages selected from the above two districts are given in Table 4.

(B) Analytical Methods

(i) Impact of Bt Technology

The impact of Bt technology on key variables like yield, pesticide use,

profitability etc. were estimated through the following Analysis-of-Variance (AOV)
model (Gujarati, 1988).

Y = α + βCBt + u
where,

Y = Variable for which impact is estimated

CBt = 1 for adoption of Bt cotton
= 0 for non- adoption of Bt cotton.

This model estimates whether Bt adoption makes any significant difference in the
yield, pesticide use, profitability, etc., assuming that all other variables are held
constant. For example take the impact of Bt on yield. The intercept α gives mean
yield of non-adopters and the slope β tells by how much the mean yield of adopters,
differs from the mean yield of non-adopters. α+β gives the mean yield of the Bt
adopter. The test of the null hypothesis that there is no difference between the yields
of Bt adopters and non-adopters (H0: β=0) can be made by running the regression in
the usual manner and finding out whether the estimated β̂ is statistically significant.

(ii) Yield Advantage in Bt Cotton: Technology Effect or Input Effect?

The yield advantage in Bt may be due to the Bt technology and/or due to

additional input use for Bt cultivation. Sources of the yield difference between Bt and
non-Bt cotton was estimated by decomposing the productivity change into technology
effect and input effect following the methodology of Bisaliah (1977).
Figure 1 depicts the production technology of conventional hybrids and Bt
hybrids. If the existing input use is X0, the yield from conventional hybrid is Y0 and
yield from Bt hybrid is C. Y0C is the yield gain due to technology. If the input use is
increased to X1, then the yield increases to Y1 for Bt hybrids. Hence CY1 is yield gain
due to additional inputs. The production technology is represented as:
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 411

Yield

Y1 Bt Hybrids
b
C c
Existing
Hybrids
Y0 a
Y0 C = Yield gain due to Technology

C Y1 = Yield gain due to additional

inputs

X0 X1 Input

Figure 1. Decomposition of the Impact of Technology

The Production function for non-Bt cotton
ln Y0 = ln α0+β0i lnX0i+u0 …. (1)
The Production function for Bt Cotton
ln Y1 = ln α1+β1i lnX1i+u1 ....(2)
Taking the difference of Eqn (1) and (2) and rearranging the terms, we get
ln [Y1/Y0] = [ln (α1/ α0) + (β1i- β0i) ln X0i] + β1i ln ( X1i/X0i)

Δ in yield = Δ in yield due to Bt technology + Δ in yield due to additional inputs

In the present study the following Cobb-Douglas production functions, for Bt and
non-Bt cotton cultivation were fitted:

LnYB = Lnb0 + b1LnCHEMB + b2SEEDB + b3LnFYMB + b4LnNB + b5LnPB

+b6LnKB + b7LnLABB + b8LnIRRIB + uB …. (3)
LnYN = Lnb9 + b10LnCHEMN + b11SEEDN + b12LnFYMN + b13LnNN + b14LnPN
+b15LnKN + b16LnLABN + b17LnIRRIN + uN …. (4)
Where,
CHEM = Cost of chemicals per acre
SEED = Seed rate in grams per acre
FYM = Manures in kg per acre
N = Total nitrogen in kg per acre
412 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

P = Total phosphorus in kg per acre

K = Total Potash in kg per acre
LAB = Total labour in man-days per acre
IRRI = Dummy for irrigated environment (I=1 otherwise 0)
bi = parameters to be estimated (i=1 to 8 for Bt and i=9 to 17 for Non-
Bt)
u = random error
Subscript B = Bt cotton
Subscript N = Non-Bt cotton

Taking differences between (3) and (4) and adding some terms and subtracting
the same terms and by rearranging terms, the equation becomes;

LnYB/YN ={Ln(b0/b9)}+{(b1-b10)LnCHEMN+(b2-b11)LnSEEDN+(b3-b12) LnFYMN

+(b4-b13) LnNN(b5-b14) LnPN +(b6-b15)LnKN+(b7-b16) LnLABN +(b8-
b17) LnIRRIN} + {b9)LnCHEMB/CHEMN) + b10Ln(SEEDB/SEEDN)+
b12Ln(FYMB/FYMN) + b13Ln(NB/NN)+ b14Ln(PB/PN)+ b15Ln(KB/KN)
+ b16Ln(LABB/LABN)+ b17Ln(IRRIB/IRRIN)}+{uB-uN} ….(5)

LHS of the Equation 5 denotes the difference in per acre productivity of Bt and
non-Bt cotton cultivation. The right handside (RHS) of the equation decomposes the
difference in productivity into change due to technology and change due to input use.
Equation 5 has three major terms within {}. The first term of RHS refers to the gap
attributable to neutral technological change, the second term refers to the gap
attributable to non-neutral technological change, and the third term refers to change
due to input use.

(iii) Economic Surplus and Distributional Impact of Bt Technology in India:

The Economic Surplus Model

Economic Surplus approach is widely used for estimating total benefits of a new
technology and to evaluate its distributional impact. An economic approach to
evaluating R&D begins with the basic, commodity market model of research benefits
depicted in Figure 1. S0 represents the supply function before the technical change,
and D0 represents the demand function. The initial price and quantity are P0 and Q0.
Suppose research generates yield increasing or input saving technologies. These
effects can be expressed as a per unit reduction in production costs, K, that are
modeled as a parallel shift down in the supply function to S1. This research-induced
supply shift leads to an increase in production and consumption to Q1(ΔQ=Q1-Q0),
and the market price falls to P1 (byΔP=P0-P1). Consumers are better-off because the
new technology enables them to consume more of the commodity at a lower price
(Wood et al., 2001).
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 413

Price
S0

a S1
P0

P1 e b

I0 D
d c
Change in Consumer surplus
I1 (ΔCS = area P0 ab P1)

Change in Producer surplus

(ΔPS = area P1 b I1 - area P0 a I0)

0 Q0 Q1 Quantity

Source: Alston et al., (1988), p.209.

Figure 2. The Economic Surplus Model

Although they receive a lower price per unit, the producers who adopt the new
technology are better-off, too, because their unit costs have fallen by an amount, K
per unit that is more than the fall in price. The change in consumer surplus which is
the measure of the consumer benefit is equal to area P0abP1. The change in producer
surplus which is the measure of the producer gain is equal to area P1 b I1 - area P0 a I0
in Figure 2. The total benefit/the economic surplus is obtained as the sum of producer
and consumer benefits. As an approximation, the cost-saving per unit multiplied by
the initial quantity, K.Q0, is often used. Thus, the size of the market, as indexed by
the initial quantity Q0, as well as the size of the research-induced savings in per unit
cost of production. K, are critical factors in estimating the economic benefits from
R&D. The economic surplus model was estimated through DREAM [Dynamic
Research Evaluation for Management] package, developed by International Food
Policy Research Institute (IFPRI).

(iv) Environmental Impact of Bt Technology

One of the important characteristics of Bt technology is that it reduces pesticide

use in cotton. There are different approaches to quantify the environmental impact of
pesticide use. Marginal productivity estimates provide an indirect measure of the cost
in terms of foregone agricultural output of reducing pesticide use to protect human
414 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

health and environment (Campbell 1976). Higley and Wintersteen (1992) used a
contingent valuation (CV) approach to assess the value to farmers of avoiding
environmental risks caused by pesticides. They considered effects of pesticides on
surface water ground water, aquatic organisms, birds, mammals, beneficial insects
and humans. Mullen et al., (1997) also used CV analysis to evaluate the impacts of
pesticides. Kovach et al., (1992) developed an Environmental Impact Quotient (EIQ)
to measure the environmental impact of different pesticides. Environmental Impact
Quotient (EIQ) value of individual pesticide refers to the average of its effect on farm
worker, consumer and ecology. Later Oskam and Vijftigschild (1999) modified the
EIQ field rating method to study the environmental impact of pesticides in
agricultural use. In this study Environmental Impact Quotient (EIQ) values were
adopted from Kovach et al., (1992).
EIQ = {C [(DT*5) +(DT*P)]
+ [(C*((S+P)/2)* SY)+(L)]
+ [(F*R)+(D*((S+P)/2)*3)+ (Z*P*3) + (B*P*5)]}/3
Where,
DT = dermal toxicity,
C = chronic toxicity,
SY = systemicity,
F = fish toxicity,
L = leaching potential,
R = surface loss potential,
D = bird toxicity,
S = soil half-life,
Z = bee toxicity,
B = beneficial arthropod toxicity,
P = plant surface half life.
To account for different formulations of the same active ingredient and different
use patterns, a simple equation called EIQ Field Use Rating was developed.
EIQ Field Use Rating = EIQ x per cent of active ingredient x Rate
With this method, comparisons of environmental impact between pesticides and
different pest management programmes can be made. Oskam and Vijftigchild (1999)
developed an indicator to measure the risk of pesticides to the environment and other
impact, RM using EIQ values as:

K
∑ (use of a.i. k × EIQ k ) for crop j
k =1
RM =
area of crop j
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 415

Where, RM = risk measure, a.i = active ingredient of pesticides, EIQ= Environmental

Impact Quotient, k = 1,….n, insecticides and j = 1,…m crops.

IV

RESULTS AND DISCUSSION

The study analyses the impact of Bt technology on key variables in the

production of Bt cotton like on yield, pesticide application, seed cost, labour and
profitability across the states. Discussion across States facilitates comparison of key
variables and help in understanding how the production environment in each state is
different and accordingly how the impact of Bt cultivation varies across the states in
terms of the key variables. Another way of presenting the results could be discussion
of all the key variables for each state (state wise), however we follow the former.

Impact of Bt Technology on Yield

One of the important advantages of Bt technology on cotton is higher yield. This

may be due to the prevention of yield loss due to pest attack. For example, the
analyses of several years of Indian field trial data demonstrated that Bt technology
can significantly reduce pesticide applications and increase effective yields under
experimental conditions [Qaim and Zilberman 2003; Qaim 2003; ICAR 2002; Naik
2001]. According to Clive (2008) yield increased by 31 per cent as per conservative
estimates for small farmers.
The impact of Bt technology on the yield of cotton in this study was estimated
through the Analysis-of-Variance (AOV) model. The intercept of the model gives the
mean yield of non-Bt. and the slope coefficient shows the difference in yield between
Bt and non-Bt. The difference in mean yield was statistically significant in all states
as shown in Table 5.

TABLE 5. IMPACT OF BT TECHNOLOGY ON YIELD

Yield (kg/ha)
States Intercept (Yield of Non Bt) Yield difference
(1) (2) (3)
Gujarat 1685.59*** 873.95***
(14.446) (6.115)
Maharashtra 828.50*** 338.08***
(14.022) (4.672)
Andhra Pradesh 1902.67*** 664.58***
(24.456) (6.975)
Tamil Nadu 1739.50*** 649.23***
(25.69) (7.828)
All India 1539.06*** 631.46***
(25.997) (8.709)
*** Significant at 1 per cent level.
416 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

The estimated mean yield of non-Bt was 1903 kg per ha in Andhra Pradesh, 1740
in Tamil Nadu, 1686 in Gujarat and 829 in Maharashtra. The slope coefficient shows
the difference in the yield of Bt and non-Bt. The mean difference in yield ranged
from nearly 338 kg per ha in Maharashtra to 874 kg per ha in Gujarat. At all India
level, mean difference in yield was 631 kg per ha which means the technology has
increased the productivity of cotton by 631 kg per ha.
The mean yield of Bt was estimated as the sum of slope coefficient and intercept
of the AOV model and the estimated mean yield of Bt and non-Bt is presented in
Table 6. The yield of Bt cotton was the highest in Andhra Pradesh with 2567 kg per
ha, followed by 2560 kg per ha in Gujarat, 2389 kg per ha in Tamil Nadu and 1167
kg per ha in Maharashtra. At all India average yield of Bt was 2171 kg per ha.

TABLE 6. YIELD OF BT AND NON-BT COTTON

T test
States N Mean* (kg/ha) T P
(1) (2) (3) (4) (5) (6)
Gujarat Bt 80 2559.54 6.115 0.00
Non-Bt 40 1685.59
Maharashtra Bt 80 1166.58 4.672 0.00
Non-Bt 40 828.50
Andhra Pradesh Bt 80 2567.26 6.975 0.00
Non-Bt 40 1902.67
Tamil Nadu Bt 80 2388.74 7.828 0.00
Non-Bt 40 1739.50
All India Bt 320 2170.53 8.709 0.00
Non-Bt 160 1539.06
* Differences in mean yields of Bt and non Bt are significant at one per cent level.

Impact of Bt technology on Pesticide Application

The Impact of Bt technology on pesticide application in cotton was estimated

through the Analysis-of-Variance (AOV) model. The model shows whether Bt
adoption makes any significant difference in the quantity of pesticide applied. The
intercept of the model gives the mean expenditure on pesticides in non-Bt. The
estimated mean expenditure on pesticides in non-Bt was Rs. 4707 per ha in Andhra
Pradesh, 4262 in Gujarat, 2713 in Tamil Nadu and 2391 in Maharashtra. The slope
coefficient of Bt shows the difference in the expenditure on pesticides in Bt and non-
Bt. The expenditure on pesticides in Bt was less than non Bt and statistically
significant in all states as shown in Table 7. The reduction in expenditure on
pesticides ranged from Rs. 1718 per ha in Andhra Pradesh, Rs. 1713 per ha in
Gujarat, Rs.1442 per ha in Tamil Nadu and Rs. 1057 per ha in Maharashtra.
The expenditure on pesticides for Bt was estimated as the sum of slope and
intercept of the AOV model and the estimated expenditure on pesticides is presented
in Table 8. The expenditure on pesticides in Bt cotton was the highest in Andhra
Pradesh with Rs. 2989 per ha, followed by Rs. 2548 per ha in Gujarat, Rs. 1334 per
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 417

ha in Maharashtra and Rs. 1271 per ha in Tamil Nadu. At all India average
expenditure on pesticides in Bt was Rs. 2036 per ha. The Differences in average
amount spent on pesticides between Bt and non Bt are significant at one per cent
level in all the cases.

TABLE 7. IMPACT OF BT TECHNOLOGY ON EXPENDITURE ON PESTICIDE APPLICATION

Expenditure on plant protection chemicals (Rs./ha)

States
Intercept (Expenditure – Non-Bt) Difference in expenditure
(1)
(2) (3)
Gujarat 4262.323*** -1713.82***
(9.388) (-3.082)
Maharashtra 2391.014*** -1057.041***
(15.128) (-5.461)
Andhra Pradesh 4707.203*** -1718.05***
(24.824) (-7.398)
Tamil Nadu 2713.036*** -1441.73***
(22.45) (-9.74)
All India 3518.394*** -1482.662***
(23.847) (-8.205)
*** Significant at 1 per cent level.

TABLE 8. PESTICIDE EXPENDITURE ON BT AND NON-BT COTTON

T test
States N Mean (Rs./ha) T P
(1) (2) (3) (4) (5) (6)
Gujarat Bt 80 2548.49 -3.082 0.00
Non-Bt 40 4262.33
Maharashtra Bt 80 1333.97 -5.461 0.00
Non-Bt 40 2391.01
Andhra Pradesh Bt 80 2989.16 -7.398 0.00
Non-Bt 40 4707.20
Tamil Nadu Bt 80 1271.31 -9.73961 0.00
Non-Bt 40 2713.04
All India Bt 320 2035.73 -8.205 0.00
Non-Bt 160 3518.39

Earlier studies about the agronomic and economic impacts of Bt cotton

cultivation demonstrated that the adopting farmers benefit from income increase
through reduced pest control costs and higher effective yields (Morse et al., (2004);
Qaim and Zilberman 2003, Qaim and Traxler, 2005). The benefits will vary
according to varying pest infestation levels in different years and locations. In China,
for instance, infestation levels of lepidopteron pests are the highest in the northern
and eastern parts of the country, and hence the benefits of Bt cotton were most
pronounced there. This was reflected in much higher adoption rates, as compared to
western China (Pray et al., 2002). In South Africa, 90 per cent of the smallholder Bt
cotton producers achieved significant reduction in pesticide use (Ismael et al., 2001).
In the Maharashtra State of India Narayanamoorthy and Kalamkar (2006) found that
418 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

farmers cultivating Bt cotton had marginally higher expenditure of pesticides. They

argued that the farmers spray the same quantity of pesticides as before because of the
fear of bollworm attack. However Clive (2008) reported that insecticide application
by small farmers in India decreased by 39 per cent on an average by conservative
estimates.

Impact of Bt technology on Seed Cost

The GM cotton in India was commercialised in India by the multinational

company Monsanto through co-operation with local seed companies. In GM cotton
the Bt technology is embedded in seed and hence the cost of seed is higher in Bt
cotton. The impact of Bt technology on seed cost in cotton was estimated through the
Analysis-of-Variance (AOV) model. The model shows difference in the cost of seed
purchased by Bt and non Bt farmers and the results were presented in Table 9.

TABLE 9. IMPACT OF BT TECHNOLOGY ON SEED COST

Seed cost (Rs./ha)

States Intercept (Seed cost in Non-Bt) Difference in seed cost
(1) (2) (3)
Gujarat 413.59*** 1696.43***
(6.715) (22.49)
Maharashtra 1574.93*** 889.05***
(13.325) (6.141)
Andhra Pradesh 1621.25*** 1119.84***
(15.515) (8.750)
Tamil Nadu 1695.04*** 618.03***
(24.95) (7.430)
Average 1630.40***1 776.63***
(28.628) (11.630)
Note: *** Significant at 1 per cent level; 1=excluding Gujarat.

The intercept of the model gives the mean expenditure on seed by the non- Bt
farmers. The estimated mean expenditure on seed for non-Bt farmers was around Rs.
1600 per ha except in Gujarat, where the expenditure on seed was very low compared
to other States. The slope coefficient shows the difference in the expenditure on seed
cost in Bt and non-Bt. The expenditure on seed cost in Bt was more than non-Bt and
statistically significant in all states. The increased expenditure on seed ranged from
Rs. 618 per ha in Tamil Nadu to Rs. 1696 per ha in Gujarat. The seed costs of Bt and
non-Bt cotton are compared in Table 10.
The average Bt seed cost in India was Rs. 2407 per ha. The seed cost was highest
in Andhra Pradesh at Rs. 2741 per ha followed by Maharashtra, (Rs. 2464/ha) in
Rs.2313 in Tamil Nadu and Rs. 2110 in Gujarat. The average non-Bt seed cost in
India was Rs. 1630 per ha.
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 419

TABLE 10. SEED COST IN BT AND NON-BT COTTON

T test
States Bt_Non-Bt N Mean* (Rs./ha) T P
(1) (2) (3) (4) (5) (6)
Gujarat Bt 80 2110.02 22.490 0.00
Non-Bt 40 413.59
Maharashtra Bt 80 2463.98 6.141 0.00
Non-Bt 40 1574.93
Andhra Pradesh Bt 80 2741.08 8.750 0.00
Non-Bt 40 1621.25
Tamil Nadu Bt 80 2313.07 7.43 0.00
Non-Bt 40 1695.04
Average Bt 320 2407.04 11.630 0.00
Non-Bt 160 1630.411
*Mean difference significant at 1 per cent level; 1=excluding Gujarat.

Impact of Bt Technology on Cost of Cultivation

One of the areas of concern on Bt cultivation was that the adoption of Bt

technology would result in increased cost of cultivation as a result of high seed cost
and due to increased use of other inputs.
The estimates of cost of cultivation of non-Bt and the difference in cost between
Bt and non-Bt cotton is given in Table. 11. The cost was higher in Bt cotton
cultivation and the difference was statistically significant in all the States. The
difference in cost was Rs. 5071 at all India level and among the States the difference
in cost was the highest in Tamil Nadu and the lowest in Maharashtra.

TABLE 11. IMPACT OF BT TECHNOLOGY ON COST OF CULTIVATION

Cost of Cultivation (Rs./ha)
Intercept Difference in
States (Cost of Cultivation in Non-Bt) Cost of Cultivation
(1) (2) (3)
Gujarat 22088.90*** 5810.67***
(13.492) (2.898)
Maharashtra 13652.15*** 1830.82***
(23.62) (2.587)
Andhra Pradesh 29855.04*** 3251.93***
(42.442) (3.775)
Tamil Nadu 28046.82*** 9388.65**
(7.146) (1.953)
All India 23410.73*** 5070.52***
(18.900) (3.342)
*** and ** Significant at 1 and 5 per cent level, respectively.

A comparison of the cost of cultivation between Bt and non-Bt cotton is given in

Table 12. At all India level the cost of non-Bt was Rs. 23,411 per ha and cost of
cultivation of Bt was Rs. 28481. There was wide variation in the cost of cultivation of
Bt ranging from Rs. 15483 per ha in Maharashtra to Rs. 37435 per ha in Tamil Nadu.
In Maharashtra the cost of cultivation was comparatively low for both Bt and non-Bt
cotton.
420 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

TABLE 12. COST OF CULTIVATION OF BT AND NON-BT COTTON

T test
States N Mean (Rs./ha) T P
(1) (2) (3) (4) (5) (6)
Gujarat Bt 80 27899.57 2.898 0.00
Non-Bt 40 22088.90
Maharashtra Bt 80 15482.98 2.587 0.00
Non-Bt 40 13652.15
Andhra Pradesh Bt 80 33106.98 3.775 0.00
Non-Bt 40 29855.04
Tamil Nadu Bt 80 37435.47 1.953 0.03
Non-Bt 40 28046.82
All India Bt 320 28481.25 3.342 0.00
Non-Bt 160 23410.73
Differences in cost of cultivation are significant at 1 per cent level except Tamil Nadu (5 per cent level).

Impact of Bt Technology on Profitability

The higher yield and reduced pesticide expenditure offset the higher cost of
cultivation in Bt cotton resulting in higher profitability. The estimated mean
profitability of non-Bt was Rs. 9516 per ha in Gujarat, Rs. 3502 per ha in
Maharashtra, Rs. 7997 per ha in Andhra Pradesh and Rs. 4111 per ha in Tamil Nadu.
The difference in mean profitability was statistically significant in all states as shown
in Table 13.

TABLE 13. IMPACT OF BT TECHNOLOGY ON PROFITABILITY

Profitability (Rs./ha)
Intercept Difference in
States (Profitability in Non-Bt) Profitability
(1) (2) (3)
Gujarat 9516.44*** 27945.76***
(3.308) (7.931)
Maharashtra 3502.03*** 5644.43***
(2.789) (3.671)
Andhra Pradesh 7997.32*** 10276.33***
(4.383) (4.599)
Tamil Nadu 4110.73 16651.80***
(0.864) (2.858)
All India 6281.63*** 15129.58***
(3.851) (7.573)
*** Significant at 1 per cent level.

The profitability for Bt was estimated as the sum of slope and intercept of the
AOV model and presented in Table 14. The profitability in Bt cotton was the
highest in Gujarat with Rs. 37462 per ha, followed by Rs. 20762 per ha in Tamil
Nadu, Rs. 18274 per ha in Andhra Pradesh and Rs. 9146 per ha in Maharashtra.
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 421

TABLE 14. PROFITABILITY OF BT AND NON-BT COTTON

T test
States N Mean (Rs./ha) T P
(1) (2) (3) (4) (5) (6)
Gujarat Bt 80 37462.19 7.931 0.00
Non-Bt 40 9516.44
Maharashtra Bt 80 9146.46 3.671 0.00
Non-Bt 40 3502.03
Andhra Pradesh Bt 80 18273.65 4.599 0.00
Non-Bt 40 7997.32
Tamil Nadu Bt 80 20762.53 2.858 0.00
Non-Bt 40 4110.73
All India Bt 320 21411.21 7.573 0.00
Non-Bt 160 6281.63
Differences in profitability are significant at one per cent level.

The average profitability of Bt at all India level was Rs. 21411 per ha.
Differences in profitability between Bt and non-Bt were significant at one per cent
level in all states. The profitability was relatively very low in Maharashtra compared
to other States.

Decomposition of Yield Increase into Technology Effect and Input Effect

A careful observation of these results justifies the necessity for decomposing the
yield increase into technology effect and yield effect. It is evident that the higher
profitability of Bt cotton was mainly due to higher yield and reduced expenditure on
pesticides. But it is also important to note the high cost of cultivation of Bt cotton
despite the reduced expenditure on pesticides which implies higher use of other
inputs in Bt cotton cultivation. This indicates the higher yield in Bt cotton is due to Bt
technology and higher input use. Hence the technology effect and input effect on
yield increase was decomposed and presented in Table 15.

TABLE 15. DECOMPOSITION OF SOURCES CAUSING THE YIELD DIFFERENCE

(per cent)
Details Gujarat Maharashtra Andhra Pradesh Tamil Nadu India
(1) (2) (3) (4) (5) (6)
Due to technology 23.79 26.22 27.44 14.96 17.40
Due to input use 19.05 8.42 4.48 15.01 14.78
Estimated yield difference 42.83 34.64 31.91 29.98 32.18
Observed yield difference 51.85 40.81 34.93 37.32 36.95
Estimated yield difference as per 82.60 84.88 91.35 80.33 87.09
cent of observed yield difference

The results show that the contribution of technology was higher than the
contribution of higher input use to yield increase. At the all India level, the
contribution of technology was 17 per cent of the higher yield and the contribution of
higher input use was 15 per cent. The contribution of technology to yield increase
422 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

varied from 27 per cent in Andhra Pradesh to 15 per cent in Tamil Nadu. In Gujarat,
technology contributed for 24 per cent and higher input use contributed for 19 per
cent of the higher yield. In Maharashtra the contribution by technology was 26 per
cent while higher input use accounted for only 8 per cent of the higher yield. In
Andhra Pradesh also contribution by the technology to yield increase was very high
compared to the contribution by the higher input use. In Tamil Nadu the contribution
by the technology and higher input use contributed 15 per cent each. The
decomposition model accounted for about 80 to 90 per cent of the observed yield
difference.

The Distributional Impact of Bt Technology: Economic Surplus Approach

The results in the foregoing sections indicate the gains from Bt technology. The
nationwide impacts of higher production of cotton influence the macro supply and
consequently the gains of producers and consumers. To understand this distributional
impact was analysed through economic surplus approach and the results were
presented in Table. 16.

TABLE 16. ESTIMATED ECONOMIC SURPLUS DUE TO BT COTTON

(Rs.)
Supply Elasticity*
1.2 0.31 0.13
(1) (2) (3) (4)
Producer surplus 147.29 254.53 292.36
(40.00) (72.07) (86.02)
Consumer surplus 220.93 98.63 47.51
(60.00) (27.93) (13.98)
Economic surplus 368.22 353.16 339.87
(100.00) (100.00) (100.00)
Note: Supply elasticities were taken from Poonyth et al. (2004), Shepherd (2006) and Sumner (2003).

The producer surplus and consumer surplus were estimated under three supply
elasticity scenarios of cotton. Under relatively inelastic supply assumption, 0.31 and
0.13, producers benefit more from the technology than the consumers and at elastic
supply consumers benefit more than the producers.

Environmental Consequences

Under the Environment Protection Act (1986), the Ministry of Environment and
Forests, has notified the Rules for the Manufacture, Use, Import, Export and Storage
of Hazardous Microorganisms/Genetically Engineered Organisms or Cells. Bt cotton
being a transgenic crop requires environmental clearance under Rule 7-10 of the 1989
“Rules for Manufacture, Use, Import, Export and storage of hazardous
microorganisms/Genetically Engineered Organisms or Cells” notified under the
Environment (Protection) Act, 1986. The bio-safety and environmental issues related
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 423

to this GM crop includes molecular characterisation of induced gene, biochemical

characterization of the expressed protein, estimation of the level of the expressed
proteins in cotton, proteins in cotton products, safety of the expressed proteins to non-
target organisms, environmental fate of the Bt protein, and agronomic, compositional
and food and feed safety evaluation of Bt cotton compared to non-Bt cotton seed. The
different aspects of environmental consequences based on the survey data is
presented below.

Incidence of Bollworm Attack in Bt and Non-Bt Cotton

Bollworm attack is the major cause of indiscriminate use of pesticides in cotton.

Bt technology is specifically targeted against bollworm incidence in cotton. Farmers’
experience with bollworm incidence in different states is given in Table 17.
In Tamil Nadu, there was no attack of American bollworm and spotted boll worm
among the Bt farmers and 95 per cent of the farmers reported no attack of pink boll
worm. In non-Bt there was moderate incidence of American bollworm in 68 per cent
cases and severe incidence in 10 per cent cases. There was moderate incidence of
pink bollworm and spotted bollworm in 38 per cent and 28 per cent of the farms
respectively. In Maharashtra, there was no attack of American bollworm in 35 per
cent of the Bt samples, mild attack in 51 per cent and moderate attack in 14 percent
of the cases. In the case of pink bollworm 50 per cent reported no attack, 25 per cent
reported mild attack, 20 per cent moderate and 5 per cent reported severe attack. In
the case of spotted bollworm, 54 per cent of the farmers reported moderate attack. In
Andhra Pradesh in Bt there was no attack of American boll worm in 88 per cent of
the farms and there was mild attack in rest of the 11 per cent farms. In the case of
pink worm there was no attack in 38 per cent of the farms, mild attack in 45 per cent
and moderate attack in 18 per cent of the farms. There was no attack of spotted boll
worm in 89 per cent of the respondents. In non-Bt 95 per cent of the cases reported
moderate attack of American bollworm and pink boll worm. In Gujarat 84 per cent of
the farmers reported no attack of American bollworm in Bt cotton, while 6 per cent
reported mild attack and 10 per cent reported moderate attack. There was no
incidence of pink bollworm in 96 per cent of the cases, but 35 per cent reported
moderate attack of spotted boll worm. In non-Bt also in 70 per cent of the cases there
was no attack of American boll worm while there was moderate attack of pink boll
worm and spotted boll worm in 55 and 65 per cent of the cases respectively. In Tamil
Nadu there was no attack of American boll worm in Bt crop while in non Bt there
was no attack in 20 per cent cases. But in non-Bt there was moderate incidence in 68
per cent cases. Generally there was low incidence of boll worm compared to the past
which indicates that farmers grow resistant non-Bt hybrids and in areas where boll
worm incidence is generally low. There was no attack of pink worm in 95 per cent of
cases in Bt and in 40 per cent of cases in non-Bt. In Tamil Nadu there was no attack
of spotted boll worm in Bt while 28 per cent reported moderate attack in non-Bt.
424 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

TABLE 17. SEVERITY OF BOLLWORM ATTACK IN BT AND NON BT COTTON

Gujarat Maharashtra Andra Pradesh Tamil Nadu
Non- Non- Non- Non-
Pests Severity Bt Bt Bt Bt Bt Bt Bt Bt
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
American No attack 83.8 70.0 35.0 22.5 88.8 - 100.0 20.0
BW Mild attack and no 6.3 2.5 51.3 45.0 11.3 5.0 - 2.5
pp measures
needed
Moderate 10.0 20.0 13.8 20.0 - 95.0 - 67.5
Severe - 7.5 - 12.5 - - - 10.0
Pink BW No attack 96.3 42.5 50.0 30.0 37.5 - 95.0 40.0
Mild attack and no 1.3 2.5 25.0 37.5 45.0 5.0 1.3 2.5
pp measures
needed
Moderate 2.5 55.0 20.0 20.0 17.5 95.0 3.8 37.5
Severe - - 5.0 12.5 - - - 20.0
Spotted BW No attack 57.5 32.5 22.5 20.0 88.8 - 100.0 67.5
Mild attack and no 7.5 2.5 23.8 32.5 11.3 5.0 - 5.0
pp measures
needed
Moderate 35.0 62.5 53.8 30.0 95.0 - 27.5
Severe - 2.5 - 17.5 - - - -

Cultivation of Refuge Crops

Fields with Bt cotton are required to cultivate non-Bt cotton refuge to help
control resistance. The refuge crop supply non-resistant insects to mate with possible
resistant insects in Bt cotton to produce non-resistant insects. Thus refuge crop is
meant to prevent the emergence of resistant pests which may require even higher
dose of chemicals. Details of refuge cotton cultivation are given in Table 18.

TABLE 18. DETAILS ON REFUGE CROP CULTIVATION

(Per cent)
Details Tamil Nadu Maharashtra Andhra Pradesh Gujarat
(1) (2) (3) (4) (5)
Awareness of refuge seed supplied 80 98 100 86
Use of refuge seeds 6 96 28 29
Awareness of susceptibility of refuge crop 100 53 36 43
Spraying of refuge crops 40 73 13 20

The Bt seed companies provide a small packet of refuge seeds along with Bt
seeds to cultivate around the Bt fields. Most of the farmers in all the four States were
aware of this refuge seeds. But only less than 30 per cent of the farmers used the
refuge seeds for the purpose for which it was given. In Maharashtra nearly all the
farmers used the refugee seeds as normal seeds and also sprayed the refuge crop to
save from the boll worms, thus defeating the very purpose for which it was supplied.
Awareness of the susceptibility of refuge crop was also very low except in Tamil
Nadu. The reasons for not using the refuge seeds were analysed and presented in the
Figure 3.
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 425

29%
Fear of pest attack & more
Maintenance needed
28%
Small holding size & no space to
grow refugee crop

Attract more pest & also increase

16%
Infestation of pest in Bt

Not aware of importance of growing

13%
refugee crop

Yield reduce in that rows 7%

Other reasons 7%

Figure 3. Reasons for Not Using Refugee Crop

The most important reason for not using refuge seeds was that farmers feared
severe pest attack and more expenditure for pest control and reduced yield in refuge
crop. In 28 per cent of the cases farmers found it difficult to comply with this
measure because of small land holdings. The other important reasons were fear of
pest attack in Bt cotton from refuge cotton, lack of awareness about the importance of
growing refuge crop and fear of yield reduction.

Environment Impact Quotient (EIQ) of Cultivating Bt and Non-Bt Cotton

Environment Impact Quotient (EIQ) for Bt cotton was lower in all the states. The
decrease in EIQ value was 15 per cent in Gujarat, 19 per cent in Andhra Pradesh, 50
per cent in Maharashtra and 68 per cent in Tamil Nadu. The variation in reduction in
the EIQ could be due to the difference in pesticides used in different states and due to
the difference in the incidence of sucking pests in different states. The pest control
measure differs based on the level of sucking pest incidence.

TABLE 19. EIQ FIELD FOR BT AND NON-BT COTTON

(rating/ha)
Percent change
State Bt cotton Non-Bt Difference over non-Bt
(1) (2) (3) (4) (5)
Gujarat 43.23 50.64 -7.41 -14.63
Maharashtra 34.50 68.72 -34.22 -49.80
Andhra Pradesh 58.33 71.74 -13.41 -18.69
Tamil Nadu 26.58 82.31 -55.73 -67.71
426 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

Macro Trends: Before and After Bt Cotton

The macro trends in the area, production and yield along with a high level of
adoption rate (86 per cent in 2010) brings out the level of acceptance of technology in
the country. The area of cotton in the major producing states, which were nearly
stagnant or declining for nearly a decade before the introduction of Bt in 2002-03,
showed significant increase in most of the States. This is primarily due to the jump in
productivity of cotton after the introduction of Bt cotton as evident from the Figure 4.
This macro data confirms the results of this study on yield advantage. But there is
decline in productivity in recent years. Decline in yield in one or two years could be
due to climatic factors or other constraints. Nevertheless these data throws some light
on the sustainability of the production potential of Bt in the long run and more
rigorous studies are essential before making any conclusions.

Figure 4. Productivity of Cotton in Major States

IV

CONCLUSION

Over the last ten years, modern agricultural biotechnology has been adopted
rapidly at the global level, including several developing countries. In India this trend
has been most apparent for Bt cotton in India in the last few years. But the
introduction of Bt cotton in the country in 2002 led to an intense scientific debate and
public controversy. The results of the study reveal that the farmers in the major cotton
growing states in India benefitted significantly from adopting Bt technology through
higher profitability mainly due to reduced pest control costs and higher yields, though
there was considerable variation in key variables like yield, cost, pesticide use, etc.,
Cotton is one of the highest pesticide consuming crops which impact the environment
through polluting land and water and poisoning humans, animals and insects. The
environmental impact quotient was significantly lower for Bt cotton because of
 ECONOMIC AND ENVIRONMENTAL IMPACT OF BT COTTON IN INDIA 427

reduced pesticide consumption. The long term environmental impacts and bio-safety
aspects of GM crops were not covered in this study.

REFERENCES

Alston, Julian M., George W. Norton and Philip G. Pardey (1988), “Science under Scarcity: Principles
and Practices for Agricultural Research Evaluation and Priority Setting”, CAB International
(CABI), Wallingford, U.K.
APCoAB (2006), Bt Cotton in India – A status Report, Asia-Pacific Consortium of Agricultural
Biotechnology, New Delhi, p.34
Bisaliah, S. (1977), “Decomposition Analysis of Output Change under New Production Technology in
Wheat Farming: Some Implications to Returns on Investment”, Indian Journal of Agricultural
Economics, Vol.32, No.3, July-September, pp. 193-201.
Campbell, H.E. (1976), “Estimating the Marginal Productivity of Agricultural Pesticides: The Case of
Tree Fruit Farms in the Okangan Valley, Canadian Journal of Agricultural Economics, Vol.24, pp.
23-30.
Clive, James (2008), Global Status of Commercialized Biotech/GM Crops: 2008, ISAAA Brief No. 39,
ISAAA: Ithaca, NY.
Clive, James (2010), Global Status of Commercialized Biotech/GM Crops: 2010. ISAAA Brief No. 42,
ISAAA: Ithaca, NY.
FAO (2011), FAO Statistics Accessed in November 2011.
Gujarati, D.N. (1988), Basic Econometrics, New York: McGraw-Hill Book Company, p. 432.
Herring, R.J. (2007a), “The Genomics Revolution and Development Studies: Science, Politics and
Poverty,” Journal of Development Studies, Vol. 43, No. 1, pp. 1-30.
Herring, R.J. (Ed.) (2007b), Transgenics and the Poor: Biotechnology and Development Studies,
Oxford, Routledge.
Higley, L.G. and W.K. Wintersteen (1992), “A Novel Approach to Environmental Risk Assessment of
Pesticides as a Basis for Incorporating Environmental Costs into Economic Injury Level”,
American Entomologist, Vol.38, pp.34-39.
ICAR (2002), All India Coordinated Cotton Improvement Project: Annual Report 2001-02, Central
Institute for Cotton Research, Indian Council for Agricultural Research, Coimbatore.
Ismael Y.R., R. Bennet and S. Morse (2001), Farm Level Impact of Bt Cotton in South Africa,
Biotechnology and Development Monitor, Vol. 48, pp.15-19.
Kovach, J., C. Petzoldt, J. Degni and J. Tette (1992), A Method to Measure the Environmental Impact of
Pesticides, New Yorks Food and Life Sciences Bulletin No.139, New York State Agriculture
Experiment Station, Cornell University.
Morse, S., R. Bennett, and Y. Ismael (2004), “Why Bt Cotton Pays for Small-Scale Producers in South
Africa”, Nature Biotechnology, Vol.22, pp.379– 380.
Mullen, J.D., G.W. Norton and D.W. Reaves (1997), “Economic Analysis of Environmental Benefits of
Integrated Pest Management”, Journal of Agricultural and Applied Economics, Vol.29, No.2, pp.
243 – 253.
Naik, G. (2001), “An Analysis of Socio-Economic Impact of Bt Technology on Indian Cotton Farmers”,
mimeo, Centre for Management in Agriculture, Indian Institute of Management, Ahmedabad.
Narayanamoorthy, A. and S.S. Kalamkar (2006), “Is Bt Cotton Cultivation Economically Viable for
Indian Farmers? An Empirical Analysis”, Economic and Political Weekly, June 30.
Onyango, B., G. Ramu, G. Hallman, Ho-Min Jang and V.S. Puduri (2004), Consumer Acceptance of
Genetically Modified Foods in Korea: Factor and Cluster Analysis, Food Policy Institute Working
Paper No. WP1104-015, Rutgers University.
Organization for Economic Co-operation and Development (2000), “Modern Biotechnology and
Agricultural Markets: A Discussion of Selected Issues”, Working Party on Agricultural Policies and
428 INDIAN JOURNAL OF AGRICULTURAL ECONOMICS

Markets, December 2000.

Oskam, A. and Rob Vijftigschild (1999), Towards Environmental Pressure Indicators for Pesticide
Impacts, Environmental Indicators and Agricultural Policy (Eds.) M. Brouwer and Crabtree, CAB
International.
Poonyth, Daneswar, Alexander Sarris, Ramesh Sharma and Shangnan Shui (2004) , FAO Commodity
and Trade Policy Research Working Paper-8.
Pray, C.E., J. Huang, R. Hu, and S. Rozelle (2002), Five Years of Bt Cotton in China – The Benefits
Continue, The Plant Journal, Vol.31, No.4.
Qaim, M. (2003), “Bt Cotton in India: Field Trial Results and Economic Projections”, World
Development 31, pp. 2115-2127.
Qaim, M. and D. Zilberman (2003), “Yield Effects of Genetically Modified Crops in Developing
Countries”, Science 299, pp. 900-902.
Qaim, M. (2005), “Agricultural Biotechnology Adoption in Developing Countries”, American Journal of
Agricultural Economics, Vol.87, No.5, pp. 1317–1324.
Qaim, M., A. Subramanian, G. Naik and D. Zilberman (2006), “Adoption of Bt Cotton and Impact
Variability: Insights from India”, Review of Agricultural Economics, in press.
Qaim, M. and G. Traxler (2005), “Roundup Ready Soybeans in Argentina: Farm Level and Aggregate
Welfare Effects”, Agricultural Economics, Vol.32, pp.73–86.
Shepherd, Ben. (2006), Estimating Price Elasticities of Supply for Cotton: A Structural Time-Series
Approach, FAO Commodity and Trade Policy Research Working Paper-21.
Sumner, Daniel (2003), A Quantitative Simulation Analysis of the Impacts of U.S. Cotton Subsidies on
Cotton Prices and Quantities, Mimeo.
Wood Stanley, Liangzhi, You and Wilfred Baitx (2001), DREAM Version 3, User Manual,
International Food Policy Research Institute, Washington, D.C., U.S.A.