INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1481

www.rsisinternational.org

Technical Efficiency and Production Constraints in Cotton Cultivation: An

Empirical Analysis of Hybrid and Local Varieties in Kushtia, Bangladesh

Nabila Hossain

, Dr. Rokeya Begum

, Tasmia Mahmuda Chowdhury

, Md. Masudul Hassan

Lecturer, Department of Agricultural Economics, Sher e-Bangla Agricultural University, Dhaka,

Bangladesh

Professor, Department of Agricultural Economics, Sher e-Bangla Agricultural University, Dhaka,

Bangladesh

Assistant Professor, Department of Development and Poverty Studies, Sher-e-Bangla Agricultural

University, Dhaka, Bangladesh

Lecturer, Department of Agricultural Finance and Management, Sher-e-Bangla Agricultural

University, Dhaka, Bangladesh

*Corresponding Author

DOI: https://dx.doi.org/10.47772/IJRISS.2025.91100117

Received: 13 November 2025; Accepted: 22 November 2025; Published: 02 December 2025

ABSTRACT

This study investigates the technical efficiency and production constraints of hybrid and local cotton varieties

in Kushtia district, Bangladesh, using primary survey data from 60 farmers (30 hybrid and 30 local). A Cobb–

Douglas stochastic frontier production function was applied to estimate technical efficiency and identify

determinants of inefficiency. Results indicate that hybrid cotton farmers achieved a mean technical efficiency

of 92 percent, compared with 81 percent for local farmers, implying potential output gains of 8 and 19 percent,

respectively, under existing input levels. For hybrid cotton, human labor, TSP, gypsum, and irrigation exerted

significant positive effects on production, while urea showed a negative impact. Farm size, training, and

farming experience were found to significantly reduce inefficiency. Local cotton farmers exhibited similar

patterns, with human labor, TSP, and irrigation positively influencing yield. The analysis further identified key

production constraints, including long cultivation duration (91.67 percent of farmers), low output prices (90

percent), insect infestations (83.33 percent), and adverse climatic conditions (75 percent). These findings

suggest that targeted interventions—such as farmer training, improved resource management, expansion of

farm size, and strategies to address systemic production constraints—can enhance cotton productivity and

efficiency. Strengthening these areas could reduce Bangladesh’s heavy reliance on cotton imports and support

the long-term sustainability of its textile sector.

Keywords: Cotton production, technical efficiency, stochastic frontier analysis, production constraints,

Kushtia district, Bangladesh.

INTRODUCTION

Cotton (Gossypium hirsutum) occupies a pivotal position in Bangladesh's agricultural economy as the second

most important cash crop after jute (Uddin and Mortuza, 2015). The textile and clothing industries, which are

heavily dependent on cotton fibers, constitute the largest manufacturing subsector in Bangladesh's economy,

contributing significantly to employment generation and export earnings (Dristy et al., 2024). Despite

Bangladesh's prominent global position as the fifth-largest raw cotton consumer and second-largest apparel

producer, the country faces a critical supply‒demand imbalance, producing only 2–3% of its cotton

requirements domestically while importing approximately 97% from countries including Uzbekistan, India,

Pakistan, and Turkmenistan (Uddin and Mortuza, 2015; Nadiruzzaman et al., 2019).

Bangladesh's annual cotton demand has experienced substantial growth, reaching 8 million bales in 2018,

while domestic production stagnates at approximately 150,000 bales (Nadiruzzaman et al., 2021). This striking

disparity between supply and demand imposes a significant economic burden on the country, costing 12–15

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1482

www.rsisinternational.org

thousand crore Taka annually in import expenditures. In response to this challenge, the Cotton Development

Board (CDB) has established an ambitious target to produce 2.5 lakh bales by 2021, which would satisfy

nearly 5–7% of the annual domestic demand and contribute to reducing import dependency.

Cotton cultivation in Bangladesh currently covers approximately 42,000 hectares distributed across 35

districts, with Kushtia emerging as a major production hub (Nadiruzzaman et al., 2021). Two principal types of

cotton are cultivated: American upland cotton (Gossypium hirsutum) on plains and hill cotton (Gossypium

arboreum) on Chittagong Hill Tracts. The CDB has developed and released 14 upland cotton varieties (CB-1 to

CB-14), with CB-12, CB-13, and CB-14 being high-yielding varieties exhibiting favorable fiber characteristics

suitable for the textile industry.

Technical efficiency, which is conceptually defined as the ability to produce maximum feasible output from a

given set of inputs and available technology (Farrell, 1957; Koopmans, 1951), represents a crucial determinant

of agricultural productivity enhancement. Improving technical efficiency enables farmers to increase output

without requiring additional resource investments, thereby enhancing profitability and competitiveness

(Debreu, 1951). In resource-constrained agricultural systems characteristic of developing countries,

understanding efficiency levels and their determinants becomes paramount for designing effective policy

interventions (Biswas et al., 2022; Majumder et al., 2020).

While several studies have examined cotton profitability and production economics in Bangladesh (Biswas,

1992; Rahman et al., 2013; Rahman et al., 2018), comprehensive analyses of technical efficiency employing

stochastic frontier methods remain conspicuously limited. The literature gap is particularly pronounced

regarding comparative efficiency analysis between hybrid and local cotton varieties, which differ substantially

in their input responsiveness, management requirements, and yield potential. Understanding variety-specific

efficiency levels, identifying factors contributing to technical inefficiency, and documenting production

constraints faced by farmers are essential for policymakers and agricultural extension services to design

targeted interventions capable of increasing domestic cotton production and reducing import dependency.

Moreover, contemporary challenges, including climate variability, pest pressure, input price volatility, and

labor availability constraints, necessitate evidence-based strategies for sustainable cotton production

intensification (Nadiruzzaman et al., 2019; Dristy et al., 2024). The imperative to enhance domestic cotton

production while maintaining economic viability and environmental sustainability underscores the importance

of efficiency-oriented research.

Against this backdrop, the present study aims to (1) estimate the technical efficiency of hybrid and local cotton

production via stochastic frontier analysis; (2) identify and quantify factors affecting technical inefficiency in

both production systems; and (3) document and prioritize major production constraints faced by cotton farmers

in the Kushtia district. The findings are expected to provide actionable intelligence for agricultural

policymakers, extension services, and cotton development programs seeking to increase domestic cotton

production efficiency and competitiveness.

LITERATURE REVIEW

Theoretical Framework Of Technical Efficiency

The conceptual foundation of productive efficiency was established by Farrell (1957), who decomposed

economic efficiency into technical and allocative components. Technical efficiency refers to a firm's ability to

obtain maximal output from a given set of inputs, whereas allocative efficiency reflects the ability to use inputs

in optimal proportions given their respective prices and production technology (Farrell, 1957; Koopmans,

1951; Debreu, 1951). Farrell's seminal work provided the theoretical basis for empirical efficiency

measurement, which has since evolved through parametric and nonparametric approaches.

The stochastic frontier production function approach, independently introduced by Aigner et al. (1977) and

Meeusen and van den Broeck (1977), revolutionizes efficiency measurement by explicitly distinguishing

between random noise (beyond farmers' control) and technical inefficiency (within farmers' control). This

methodological innovation addresses a fundamental limitation of deterministic frontier approaches by

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1483

www.rsisinternational.org

accommodating measurement errors, weather shocks, and other stochastic factors affecting agricultural

production (Jondrow et al., 1982). The stochastic frontier approach has consequently become the predominant

method for measuring the technical efficiency of agricultural systems worldwide.

Technical efficiency in cotton production

Global cotton production efficiency has received considerable research attention because of the economic

importance and input-intensive cultivation requirements of cotton. Odedokun (2014) analyzed cotton

production efficiency in Zamfara State, Nigeria, employing stochastic frontier analysis to reveal a mean

technical efficiency of 0.67, indicating potential output increases of 23% through improved management

practices without additional input investments. The study identified farm size, education, and extension contact

as significant efficiency determinants, emphasizing the role of human capital in enhancing agricultural

productivity.

In the context of genetically modified cotton adoption, Bennett et al. (2004) documented substantial financial

benefits for smallholder Bt cotton growers in South Africa, attributable to increased yields and reduced

insecticide costs. Their analysis revealed that Bt cotton technology generated efficiency gains through reduced

pesticide application and associated labor savings. Similarly, Loganathan et al. (2009) examined the

productivity and profitability impacts of Bt cotton cultivation in Tamil Nadu, India, and reported that

genetically modified varieties significantly enhanced the technical efficiency through superior pest resistance

and yield stability. Parameswaran and Cayalvizhi (2020) further demonstrated that fertilizer management, Bt

technology, and insecticide application contributed 60%, 23%, and 17%, respectively, to Bt cotton yield

increases in India from 2000--2014, underscoring the multifaceted nature of efficiency enhancement.

Cotton Production Research in Bangladesh

Research on cotton production in Bangladesh has evolved from descriptive analyses to more sophisticated

economic evaluations. Biswas (1992) conducted pioneering work identifying major problems in cotton

cultivation in Jessore District, documenting constraints including the nonavailability of quality seeds,

excessive input costs, inadequate technical knowledge among farmers, and limited access to credit facilities.

These foundational insights established the baseline understanding of production constraints affecting

Bangladeshi cotton farmers.

Rahman et al. (2013) undertook a comprehensive profitability analysis of major crops in Bangladesh, including

cotton, finding that cotton cultivation remained economically viable under proper management despite facing

numerous production challenges. The benefit‒cost ratio analysis revealed positive returns to cotton investment,

although profitability varied substantially across different production systems and agroecological zones.

Rahman et al. (2018) subsequently examined the productivity and profitability of improved versus existing

cropping patterns in the Kushtia region, highlighting the potential contribution of cotton to income

diversification and agricultural intensification.

More recently, Nadiruzzaman et al. (2019) conducted value chain analysis of climate-resilient cotton

production in Bangladesh, emphasizing the critical importance of improved varieties and adaptive management

practices for sustaining production under increasingly variable climatic conditions. Their work highlighted

systemic constraints extending beyond farm-level production, including postharvest handling, marketing

infrastructure deficiencies, and price volatility challenges. Dristy et al. (2024) further examined sustainable

practices for cotton production from economic and environmental perspectives, suggesting integrated

approaches that balance productivity enhancement with ecological sustainability concerns.

Despite these valuable contributions, systematic technical efficiency analysis employing stochastic frontier

methods remains limited in Bangladesh's cotton production literature. Biswas et al. (2022) applied stochastic

frontier analysis to maize cultivation in Bangladesh, demonstrating the approach's utility for identifying

efficiency gaps and improvement opportunities, but comparable analyses for cotton production are lacking.

Input Productivity and Resource Use Efficiency

Extensive research on input productivity in cotton production has identified human labor, fertilizers, and

irrigation as primary determinants of yield variation. Ramamoorthy (1990) examined the economics of cotton

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1484

www.rsisinternational.org

production in India, highlighting labor intensity as a distinguishing characteristic affecting profitability and

competitiveness. Reddy et al. (1997) conducted a comparative economic analysis of cotton cultivation in the

Guntur district, Andhra Pradesh, and reported significant yield responsiveness to phosphatic fertilizers and

irrigation applications.

The Cobb‒Douglas production function has been widely employed to quantify input‒output relationships in

cotton production systems. These studies consistently demonstrate that balanced fertilizer application, adequate

irrigation, and timely labor availability constitute critical determinants of cotton productivity (Ramamoorthy,

1990; Reddy et al., 1997). However, research has also revealed instances of overapplication of fertilizer,

particularly nitrogen, leading to diminishing returns and environmental externalities (Parameswaran and

Cayalvizhi, 2020).

Determinants of Technical Inefficiency

Empirical studies have identified diverse socioeconomic and farm-specific factors influencing the technical

inefficiency of agricultural production. Farm size has been consistently associated with efficiency levels,

although the relationship direction varies across contexts. Larger farms often exhibit greater efficiency due to

economies of scale, better resource access, and enhanced management capacity (Odedokun, 2014). However,

some studies report superior efficiency among smaller farms attributable to intensive management and family

labor utilization.

Human capital variables, particularly education, training, and farming experience, have demonstrated

significant effects on technical efficiency. Educated farmers typically exhibit greater receptivity to

technological innovations and improved management practices (Biswas et al., 2022; Majumder et al., 2020).

Agricultural training programs enhance farmers' technical knowledge and decision-making capabilities,

directly contributing to efficiency improvement. Similarly, farming experience facilitates learning-by-doing

processes, enabling farmers to optimize resource allocation over time.

Farmer age has shown mixed effects on efficiency across different studies. While younger farmers may

demonstrate greater innovation adoption propensity, older farmers often possess accumulated tacit knowledge

valuable for production management. Extension service access and market proximity have also emerged as

significant determinants of efficiency, affecting farmers' information access and input‒output market

integration (Nadiruzzaman et al., 2019).

Production Constraints in Cotton Cultivation

Cotton production faces multifaceted constraints ranging from biophysical challenges to market and

institutional limitations. Pest and disease pressure, particularly from lepidopteran pests and boll rot infections,

represents a persistent challenge, causing substantial yield losses and escalating production costs (Bennett et

al., 2004; Loganathan et al., 2009). Climate variability, including erratic rainfall patterns, drought stress, and

extreme temperature events, increasingly threatens cotton production stability (Nadiruzzaman et al., 2019;

Dristy et al., 2024).

Market-related constraints, including price volatility, limited storage infrastructure, and weak farmer

bargaining power, undermine economic viability and discourage production expansion (Uddin and Mortuza,

2015). Input market imperfections, characterized by quality adulteration, supply chain inefficiencies, and price

fluctuations, further complicate production management. Institutional constraints, including inadequate

extension services, limited credit access, and insufficient research-extension-farmer linkages, impede

technology adoption and efficiency improvement (Biswas, 1992; Nadiruzzaman et al., 2019).

Recent studies emphasize the importance of integrated approaches addressing production constraints

holistically rather than in isolation. Garbole (2025) demonstrated in Ethiopian agriculture that sustainable

intensification requires simultaneous attention to agrobiodiversity conservation, input optimization, and farmer

knowledge enhancement. Similarly, contemporary cotton research advocates for systems-oriented interventions

that combine improved varieties, precision nutrient management, integrated pest management, climate risk

mitigation, and value chain development (Dristy et al., 2024).

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1485

www.rsisinternational.org

Research Gap and Study Design

Despite substantial global literature on cotton production efficiency, significant gaps persist in understanding

the determinants of technical efficiency in Bangladesh's cotton sector. Existing studies have focused primarily

on profitability analysis and general production constraints, with limited application of stochastic frontier

methods for rigorous efficiency measurement. The comparative efficiency analysis between hybrid and local

cotton varieties remains unexplored, despite varieties' potentially divergent input responsiveness and

management requirements.

Furthermore, previous constraint assessments have been largely descriptive, lacking systematic prioritization

on the basis of farmer perceptions and impact severity. The interaction between technical efficiency levels and

specific production constraints requires further investigation to inform targeted intervention design.

Additionally, variety-specific inefficiency determinants need to be identified to enable differentiated extension

strategies appropriate for hybrid and local cotton production systems.

The present study addresses these gaps by providing a comprehensive stochastic frontier analysis of the

technical efficiency of both hybrid and local cotton varieties, identifying variety-specific inefficiency

determinants, and systematically documenting and prioritizing production constraints. The findings contribute

empirical evidence supporting policy formulation and program design aimed at enhancing domestic cotton

production efficiency and reducing import dependency, thereby advancing Bangladesh's agricultural

development and textile industry competitiveness.

METHODOLOGY

Study Area and Sampling

The study was conducted in Kushtia district, which is located in the southwestern region of Bangladesh (23°42'

to 24°12' N latitude and 88°42' to 89°22' E longitude). Three upazilas—Bheramara, Daulatpur, and Mirpur—

were purposively selected on the basis of high cotton production concentration, homogeneous soil and climatic

conditions, and accessibility.

A purposive random sampling technique was employed. From a sampling frame of 250 cotton farmers

purposive random sampling technique was employed. From a sampling frame of 250 cotton farmers provided

by the Department of Agricultural Extension (DAE), 100 were identified as small farmers (0.05--2.49 acres)

with a minimum of three years of cotton cultivation experience. Sixty farmers were randomly selected: 30

cultivating hybrid varieties (CB-10, CB-11, CB-12) and 30 cultivating local varieties.

Table 1: Sample distribution

SL. No.

Variety

Number of Respondents

Hybrid

Local

Total

Source: Field Survey

Data collection

Primary data were collected from March 1 to April 15, 2020, through face‒to‒face interviews via a pretested

structured questionnaire. The survey schedule covered the following:

 General and sociodemographic information

 Farm holding status

 Cotton production details (inputs and outputs)

 Farmers' perceptions and constraints

Multiple visits were conducted to ensure data reliability, particularly since farmers did not maintain formal

records. Inconsistencies were verified through neighboring farmers.

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1486

www.rsisinternational.org

ANALYTICAL FRAMEWORK

Stochastic Frontier Production Function

The Cobb‒Douglas stochastic frontier production function was specified as follows:

ln Y = ln α + β₁ln X₁ + β₂ln X₂ + β₃ln X₃ + β₄ln X₄ + β₅ln X₅ + β₆ln X₆ + β₇ln X₇ + (Vi - Ui)

where:

 Y = Gross return per hectare (Tk/ha)

 ln α = Intercept

 X₁ = Cost of human labor (Tk/ha)

 X₂ = Cost of urea (Tk/ha)

 X₃ = Cost of the TSP (Tk/ha)

 X₄ = cost of MoP (Tk/ha)

 X₅ = Cost of gypsum (Tk/ha)

 X₆ = cost of irrigation (Tk/ha)

 X₇ = Cost of insecticide (Tk/ha)

 β₁...β₇ = Coefficients to be estimated

 Vi = random error term, N(0, σᵥ²)

 Ui = Technical inefficiency effect, |N(0, σᵤ²)|

Technical Inefficiency Model

The effects of technical inefficiency were modeled as follows:

Ui = δ₀ + δ₁Z₁ + δ₂Z₂ + δ₃Z₃ + δ₄Z₄ + δ₅Z₅ + δ₆Z₆ + Wi

where:

 Z₁ = Farm size (hectares)

 Z₂ = Respondent age (years)

 Z₃ = Respondent education (years)

 Z₄ = training (1 if received, 0 otherwise)

 Z₅ = Cotton farming experience (years)

 Z₆ = Market distance (km)

 Wi = Random variable with positive half-normal distribution

The model was estimated simultaneously via STATA software with maximum likelihood estimation.

RESULTS AND DISCUSSION

Sociodemographic profile

Table 2: Average Family Size and Distribution by Sex

Particulars

Bheramara

Upazila

Daulatpur

Upazila

Mirpur

Upazila

All

Farmers

National

Average

Number

Percent

Number

Percent

Number

Percent

Number

Percent

Family Size

Male

3.56

62.24

3.15

53.94

2.97

53.23

3.23

56.47

4.06

Female

2.16

37.76

2.69

46.06

2.61

46.77

2.49

43.53

Total

5.72

100

5.84

100

5.58

100

5.72

100

Source: Field Survey

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1487

www.rsisinternational.org

The average family size of the cotton farmers was 5.72 persons, which was slightly above the national average

of 4.06. Male members constituted 56.47% of the household members. The primary age group was 20--40

years (52%), indicating active working-age farmers. In terms of educational attainment, 41.7% had secondary

or higher secondary education, whereas 8.3% had graduation-level education, as described in Table 2.

Table 3: Agricultural Work and Income Sources

Sector

Average Annual Income (Tk)

Crops

60,897.87

Poultry

34,989.80

Livestock

36,800.00

Fisheries

25,678.00

Mean Agricultural Income

158,365.67

Non-Agricultural Income

105,171.40

Total Average Annual Income

263,537.07

Average Annual Expenditure

220,989.60

Average Annual Savings

42,547.00

Source: Field Survey

Table 3 shows that agricultural activities, particularly crop production, constituted the primary income source.

The average annual household income was Tk. 263,537.07, with savings of Tk. 42,547 annually.

Table 4: Agricultural training and organizational membership

Indicator

Bheramara

Daulatpur

Mirpur

All Farmers

No.

Received Training

86.7

Organization Member

71.7

Source: Field Survey

Table 4 shows that 86.7% of the farmers received cotton cultivation training, primarily from the Cotton

Development Board (CDB) and Bangladesh Agricultural Development Corporation (BADC). This high degree

of training participation suggests good extension service coverage in the study area.

Stochastic Frontier Production Function Results

Hybrid Cotton Production

Table 5: ML estimates for the Cobb‒Douglas stochastic frontier production function - hybrid cotton

Variables

Parameter

Coefficients

T-ratio

Stochastic Frontier:

Constant (X₀)

β₀

6.38**

2.05

Human Labor (X₁)

β₁

0.6774**

2.21

Urea (X₂)

β₂

-0.3483***

-3.40

TSP (X₃)

β₃

0.7819***

4.10

MoP (X₄)

β₄

-0.1930

-0.37

Gypsum (X₅)

β₅

0.2310***

3.78

Irrigation (X₆)

β₆

0.0970***

3.71

Insecticide (X₇)

β₇

-0.05232

-0.11

Inefficiency Model:

Constant

δ₀

-5.46*

-2.44

Farm Size (Z₁)

δ₁

-0.0673**

-2.25

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1488

www.rsisinternational.org

Age (Z₂)

δ₂

0.1170**

1.89

Education (Z₃)

δ₃

-0.0944

-0.42

Training (Z₄)

δ₄

-0.1523**

-2.50

Experience (Z₅)

δ₅

-0.3242*

-1.70

Market Distance (Z₆)

δ₆

0.8291

1.05

Log-likelihood Function

42.27

*Note: **, ** and * indicate significance at the 1%, 5% and 10% levels, respectively. Source: Field survey

Table 5 shows that the stochastic frontier production function results for hybrid cotton indicate that several

inputs significantly influence productivity. Human labor had a positive and significant effect (β₁ = 0.6774,

p<0.05), suggesting that increased labor use increases output, with a 1 percent rise in labor cost contributing to

a 0.68 percent increase in yield. Among the inputs, TSP fertilizer had the largest positive coefficient (β₃ =

0.7819, p<0.01), underscoring the importance of phosphorus in increasing cotton output, whereas gypsum (β₅

= 0.2310, p<0.01) also contributed positively, highlighting the benefits of calcium and sulfur for crop growth.

Irrigation (β₆ = 0.0970, p<0.01) had a positive, although smaller, effect, reflecting the significance of an

adequate water supply. In contrast, urea had a negative and highly significant coefficient (β₂ = -0.3483,

p<0.01), indicating problems of overapplication or inefficient nitrogen use, which reduced output. Other

inputs, such as MoP and insecticides, were not statistically significant, suggesting limited or inconsistent

impacts on hybrid cotton yields. The inefficiency effects model identified several socioeconomic and

management factors influencing farm-level performance. Farm size (δ₁ = -0.0673, p<0.05) significantly

reduced inefficiency, confirming that larger farms benefit from economies of scale and better resource

management. Training (δ₄ = -0.1523, p<0.05) also had a negative and significant effect on inefficiency,

suggesting that farmers who received training were more efficient, reflecting the effectiveness of extension

services. Similarly, experience (δ₅ = -0.3242, p<0.10) lowered inefficiency, indicating that experienced farmers

manage production more effectively, which is consistent with learning-by-doing. Age (δ₂ = 0.1170, p<0.05),

however, had a positive effect on inefficiency, suggesting that older farmers were less efficient, possibly due to

reluctance to adopt modern practices. Education did not have a significant effect, implying that formal

schooling alone may not directly improve cotton production efficiency. Overall, these findings highlight that

while labor, fertilizers, and irrigation remain critical to hybrid cotton productivity, efficiency is also shaped by

farm size, training, and farmer experience, underscoring the importance of targeted extension support and

balanced input management.

Local Cotton Production

Table 6: ML estimates for the Cobb‒Douglas stochastic frontier production function - local cotton

Variables

Parameter

Coefficients

T-ratio

Stochastic Frontier:

Constant (X₀)

β₀

2.93*

1.78

Human Labor (X₁)

β₁

0.0868**

2.22

Urea (X₂)

β₂

-0.1764**

-2.20

TSP (X₃)

β₃

0.2544**

2.30

MoP (X₄)

β₄

-0.1061***

-3.18

Gypsum (X₅)

β₅

-0.2202

-0.70

Irrigation (X₆)

β₆

0.2175*

1.71

Insecticide (X₇)

β₇

-0.2533

-0.58

Inefficiency Model:

Constant

δ₀

-12.39*

-1.69

Farm Size (Z₁)

δ₁

-0.0154***

-3.47

Age (Z₂)

δ₂

0.5766

0.59

Education (Z₃)

δ₃

0.3070***

2.81

Training (Z₄)

δ₄

0.0601**

2.21

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1489

www.rsisinternational.org

Experience (Z₅)

δ₅

0.1035

0.66

Market Distance (Z₆)

δ₆

0.6807*

1.63

Log-likelihood Function

20.57

*Note: **, ** and * indicate significance at the 1%, 5% and 10% levels, respectively. Source: Field survey

Table 6 shows that the results of the stochastic frontier production function for local cotton production

highlight several important relationships between input use and output efficiency. Human labor had a positive

and significant effect (β₁ = 0.0868, p<0.05), although the coefficient was smaller than that of hybrid cotton,

indicating relatively lower labor productivity. Triple superphosphate (TSP) also had a significant positive

influence (β₃ = 0.2544, p<0.05), whereas irrigation (β₆ = 0.2175, p<0.10) contributed positively, suggesting

that water availability plays a critical role in local cotton yields. In contrast, urea (β₂ = -0.1764, p<0.05) and

Muriate of Potash (MoP) (β₄ = -0.1061, p<0.01) had negative effects, implying that excessive or imbalanced

fertilizer application may reduce productivity. The effects of gypsum and insecticides were not significant,

reflecting the limited impact on yield variation in this context.

The inefficiency effects model revealed that farm size (δ₁ = -0.0154, p<0.01) significantly reduced inefficiency,

confirming that larger farms are more efficient in utilizing resources. Interestingly, education (δ₃ = 0.3070,

p<0.01) and training (δ₄ = 0.0601, p<0.05) were associated with greater inefficiency in local cotton production,

which contrasts with the hybrid cotton results. This may suggest that training programs are less adapted to

local varieties or that farmers are reluctant to apply modern techniques to traditional crops. Other factors, such

as age, experience, and market distance, were not statistically significant, although market distance showed a

positive tendency, indicating that remoteness may contribute to inefficiency. Overall, the model suggests that

improving the input balance and designing tailored extension programs could substantially increase the

efficiency of local cotton farming.

Technical Efficiency Distribution

Table 7: Frequency distribution of technical efficiency

Efficiency (%)

Hybrid Variety

Local Variety

No. of Farms

Percentage

No. of Farms

Percentage

0-60

6.67

20.00

61-80

6.67

20.00

81-90

20.00

16.67

91-99

66.67

43.33

Total

100.00

Minimum

0.53

0.36

Maximum

0.99

0.98

Mean

0.92

0.81

0.12

0.18

Source: Field Survey

The analysis of technical efficiency revealed significant differences between hybrid and local cotton farmers in

the Kushtia district, as shown in Table 7. The mean technical efficiency for hybrid cotton was 92 percent, with

the majority of farmers (66.67 percent) operating at very high efficiency levels between 91 and 99 percent. The

efficiency scores ranged from 53 to 99 percent, suggesting that although most farmers are close to the

production frontier, there is still potential to increase output by approximately 8 percent through better input

use and improved management practices. In contrast, local cotton farmers recorded a lower mean technical

efficiency of 81 percent, with only 43.33 percent achieving efficiency levels of 91–99 percent. Their efficiency

range was much wider (36–98%), reflecting greater variability in performance. This implies that local variety

farmers could increase their output by approximately 19 percent if existing resources were managed more

efficiently. The higher mean efficiency and lower standard deviation observed in hybrid cotton (92 percent and

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1490

www.rsisinternational.org

0.12) than in local cotton (81 percent and 0.18) highlight the superior consistency and productivity of hybrid

cultivation while also underscoring the scope for improvement among local variety growers.

Production Constraints

Table 8: Cotton production problems and constraints

Type of Problems

No. of Farmers

Percentage

Rank

Long duration of cotton cultivation

91.67

Low price of cotton

90.00

Insect attack in cotton field

83.33

Adverse climate

75.00

Boll rot attack in cotton field

75.00

High price of hybrid cotton seed

66.67

Lack of operating capital

66.67

Natural calamities

50.00

Shortage of human labor

50.00

Lack of scientific knowledge of farming

46.67

Adulteration of fertilizer, insecticide, and pesticide

41.67

High price of fertilizers

36.67

Need high crop management

30.00

Lack of necessary advice from concerned authority

20.00

Source: Field Survey

Table 8 shows that the study identified a wide range of constraints affecting cotton production in Kushtia

district, with long cultivation durations emerging as the most critical problem, reported by 91.67% of the

farmers. Cotton’s six-month growing period restricts land availability for other crops and increases household

expenses and loan repayment pressure, making it less attractive than short-duration alternatives are. Low

market prices were highlighted by 90% of the respondents, as farmers are often forced to sell during harvest

when prices are depressed due to inadequate storage and weak bargaining power. Pest incidence was another

major challenge, with 83.33% of the farmers reporting severe insect infestations from bollworms and

Spodoptera species, which reduce yields and increase production costs, while ineffective insecticide use further

resulted in compound losses. Climatic adversities, particularly droughts, floods, and erratic rainfall, were cited

by 75% of the farmers, underscoring the role of climate variability as an uncontrollable risk factor. Similarly,

75% of the respondents experienced boll rot, a disease that damages bolls late in the season after significant

investment, necessitating costly and additional crop management. High prices of hybrid cotton seeds and other

inputs, along with a lack of operating capital, were reported by 66.67% of the farmers, limiting their ability to

adopt recommended input packages. Labor shortages during peak seasons (50 percent) and knowledge gaps

(46.67 percent) further constrained production efficiency, with many farmers unable to access timely technical

advice or training. The adulteration of fertilizers and pesticides, cited by 41.67 percent, also undermined

productivity, reflecting weaknesses in input market regulation. High fertilizer prices (36.67 percent), crop

management difficulties (30 percent), and inadequate extension services (20 percent) were also noted as

significant barriers. Overall, these findings suggest that cotton farmers face a combination of biological,

climatic, financial, and institutional constraints, which collectively reduce efficiency and profitability and

require comprehensive policy interventions to overcome.

DISCUSSION

The study reveals substantial differences in technical efficiency between hybrid and local cotton varieties, with

hybrid farmers achieving 92 percent efficiency compared with 81 percent for local farmers. This efficiency gap

highlights the variety-specific management advantages of hybrids, which may stem from more responsive

germplasms, stronger extension support, and better targeted input recommendations. The higher efficiency of

hybrid cotton also translated into superior profitability (BCR 2.18 vs. 1.65), demonstrating that efficiency

gains directly enhance economic performance. Nevertheless, the 19 percent potential output gain for local

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1491

www.rsisinternational.org

varieties under current input levels indicates that considerable improvements can be achieved without shifting

entirely to hybrids. Farm size significantly reduced inefficiency across both systems, but its effect was more

pronounced for local cotton, suggesting economies of scale, better access to resources, and possible

indivisibilities in input use or technology adoption. Input use patterns revealed nutrient imbalances, with

negative effects of urea and MoP indicating overapplication and poor soil testing practices, whereas positive

and significant effects of TSP and irrigation emphasized the importance of balanced fertilization and improved

water management technologies. The contrasting effects of education and training across varieties further

underscore the need for variety-specific extension strategies. For hybrid cotton, training and experience

enhanced efficiency, whereas for local cotton, they appeared to increase inefficiency, possibly reflecting

misaligned training content or conflicts with traditional knowledge systems. Moreover, the positive influence

of farmer age across both systems suggests that younger farmers are more likely to adopt improved practices,

which is consistent with innovation diffusion theory.

In addition to farm-level input and management, production outcomes are constrained by systemic challenges.

Market-related issues, such as low cotton prices, inadequate storage facilities, and weak marketing structures,

undermine profitability and limit farmers’ bargaining power. Climate variability, reported by 75% of the

respondents, poses additional risks to productivity, highlighting the urgent need for climate-smart strategies,

including drought-tolerant varieties, improved weather forecasting, crop insurance, and adjusted planting

schedules. Pest incidence, reported by 83.33 percent of the farmers, underscores gaps in pest control practices,

as indicated by insignificant or negative coefficients for insecticide use. This finding points to pesticide

resistance, poor application timing, and inadequate pest monitoring, necessitating stronger integrated pest

management (IPM) approaches and the incorporation of biological control strategies. These findings call for a

comprehensive set of policy interventions. Differentiated extension strategies must be developed to address

both hybrid and local systems, with emphasis on strengthening training content, targeting efficiency-improving

practices, and enhancing follow-up support. Input market reforms should ensure the quality assurance of seeds

and fertilizers, promote balanced fertilization through soil testing, and improve pest management advisory

services. The market infrastructure needs to be strengthened through the establishment of storage facilities,

price stabilization mechanisms, and contract farming arrangements to secure fair returns for farmers.

In addition to these measures, research priorities should focus on breeding shorter-duration and climate-

resilient varieties, optimizing fertilization strategies by soil type, and advancing integrated pest management

technologies. Expanding cotton-specific credit programs and developing crop insurance schemes will further

help farmers manage risks, while supporting farmer organizations in collective marketing and improving

access to capital for input purchases can strengthen resilience. Overall, the discussion highlights that

improving technical efficiency and overcoming production constraints in Bangladesh’s cotton sector requires

not only farm-level improvements but also systemic interventions in extension, markets, research, and risk

management.

For further clarification of technical efficiency findings, macroeconomic and policy-level factors of

agricultural production are also necessary to be considered, whereas farming-level input-output-conversions

may remain outside the farm-level input-output model. In Bangladesh, government policies on input subsidies,

credit availability, mechanization support, crop insurance and rural infrastructure investment directly affect

cotton farmers’ ability to maximize resources. Fertilization and irrigation equipment purchases, along with

state bank loans that provide seasonal agricultural loans, are key factors to farmers’ ability to make the most of

time available. The high price of hybrid seeds also reflects the structure of cotton seed markets whose imports

and limited domestic seed production depend heavily on imported grains.

At the global level, changes in international cotton prices and trade policy significantly impact domestic

production. The fabric industry in Bangladesh, one of the largest in the world, imports imported cotton,

allowing the price shocks in global markets to impact local market incentives. When the prices rise, domestic

producers experience downward pressures on farmgate prices, limiting production and investment. On the

other hand, high global prices can inspire farmers to increase acreage or to adopt better-quality varieties.

This infrastructure development, encompassing rural roads, storage facilities, and ginning/processing centers,

also has an important role to play in cutting postharvest losses and improving farmers access to competitive

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1492

www.rsisinternational.org

markets. This low-price problem was found in this study based on limited storage capacity and farmers selling

after harvest. This understanding of these larger macroeconomic and policy changes provides a clearer

explanation of why technical inefficiency persists despite relatively high input use and training participation.

CONCLUSIONS

This study provides a broad picture of technical efficiency and production challenges of hybrid and local

cotton growing in the Kushtia district of Bangladesh. Both authors imply that hybrid cotton farmers have high

average technical efficiency of 92% and whereas local cotton farmers have average efficiency of 81%. Plus,

these results suggest that a substantial advantage in productivity can be achieved by improving management,

although the use of input could not reach an increase in productivity. This largely reflect variation in

germplasm performance, input response and management practices that determines the efficiency differences

in the two production systems.

Input-specific effects indicate the importance of balanced and rational fertilizer application. Human labor, TSP,

gypsum, hybrid cotton irrigation positively correlated output while urea and MoP negatively correlated

production with problems of excess applications and imbalance in nutrients. These results suggest the need for

more precise nutrient management practices based on soil-test measurements and more specific nutrient

guidelines. The inefficiency analyses also revealed that farm size, training, and farming experience

significantly reduce inefficiency in hybrid cotton production, while only farm size increases efficiency among

local variety growers. These differing effects of training, which benefits hybrid farmers but, at higher levels,

leads to greater efficiency in local cotton, suggest that specific extension and capacity-building programs are

needed.

Farmers also face complex systemic challenges, including long crop duration, low production prices, high pest

incidence, climate change, and limited access to quality inputs and credit. The remaining challenges have

implications beyond field level management and require broad institutional and policy responses to improve

the overall productive environment for cotton production.

In policy terms, the cotton sector in Bangladesh must be strengthened in an integrated fashion through farm-

level efficiency and macroeconomic and institutional support. Prior to developing short-duration and climate

resilient varieties, improving extension services with differentiated training strategies, guaranteeing quality and

affordable seed and fertilizer markets, and adding storage capacity to discourage distress sales, priority focus

includes extending shorter-duration and climate resilient varieties, improving extension services, differentiated

training modules, improved seed and fertilizer market quality and inexpensive storage facilities. Pricing

stabilization mechanisms, cotton credit programs and crop insurance programs can further increase resilience,

encourage investment in cotton cultivation. In national terms, global prices and Bangladesh’s considerable

dependence on imported cotton, highlight the need to align domestic production strategies with market trends

and trade practices.

This study provides empirical evidence in the growing body of technical efficiency studies describing cotton in

Bangladesh and proposes strategies for improving productivity and reduced import dependency. Future

research needs to examine changes in change over time, cost-efficiency and environmental sustainability

measures, and consider value-chain bottlenecks beyond production, processing, marketing, and the potential of

Bt cotton. Identifying the efficiency gaps and systemic constraints identified in this study is essential in order

to achieve national production goals and minimize the import burden of Tk currently costing the country Tk.

12–15 thousand crore yearly.

REFERENCES

1. Aigner, D. J., Lovell, C. A. K., & Schmidt, P. J. (1977). Formulation and establishment of stochastic

frontier production function models. Journal of Econometrics, 6(1), 21-37.

https://doi.org/10.1016/0304-4076(77)90052-5

2. Bennett, R., Ismael, Y., Morse, S., & Shankar, B. (2004). Reductions in insecticide use from adoption

of Bt cotton in South Africa: Impacts on economic performance and toxic load to the environment.

Journal of Agricultural Science, 142(6), 665-674. https://doi.org/10.1017/S0021859604004745

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1493

www.rsisinternational.org

3. Biswas, P. R. (1992). An economic analysis of cotton production in some selected areas of Jessore

District [Unpublished master's thesis]. Bangladesh Agricultural University.

4. Biswas, R., Molla, M. M. U., Faisal-E-Alam, M., Zonayet, M., & Castanho, R. A. (2022). Profitability

analysis and input use efficiency of maize cultivation in selected areas of Bangladesh. Land, 12(1), 23.

https://doi.org/10.3390/land12010023

5. Debreu, G. (1951). The coefficient of resource utilization. Econometrica, 19(3), 273-292.

https://doi.org/10.2307/1906814

6. Dristy, S. A., Dhar, A. R., & Uddin, M. T. (2024). Sustainable practices for cotton production in

Bangladesh: Economic and environmental perspectives. Discover Agriculture, 2(1), 71.

https://doi.org/10.1007/s44279-024-00071-w

7. Farrell, M. J. (1957). The measurement of productive efficiency. Journal of the Royal Statistical Society

Series A (General), 120(3), 253-290. https://doi.org/10.2307/2343100

8. Garbole, J. (2025). The role of Enset-based homegarden agroforestry in agrobiodiversity conservation

and sustainable food production in the semiarid areas of Dugda dawa district, Southern Ethiopia.

Journal of Agriculture and Food Research, 15, 101662. https://doi.org/10.1016/j.jafr.2025.101662

9. International Cotton Advisory Committee. (2013). Cotton statistics. ICAC.

10. Jondrow, J., Lovell, C. A. K., Materov, I. S., & Schmidt, P. (1982). On the estimation of technical

efficiency in the stochastic frontier production function model. Journal of Econometrics, 19(2-3), 233-

238. https://doi.org/10.1016/0304-4076(82)90004-5

11. Koopmans, T. C. (1951). Analysis of production as an efficient combination of activities. In T. C.

Koopmans (Ed.), Activity analysis of production and allocation (pp. 33-97). Wiley.

12. Loganathan, R., Balasubramanian, R., Mani, K., & Gurunathan, S. (2009). Productivity and

profitability impact of genetically modified crops: An economic analysis of Bt cotton cultivation in

Tamil Nadu. Agricultural Economics Research Review, 22(2), 331-340.

13. Majumder, K. A., Ahmed, J. U., & Fatema, K. (2020). Comparative advantages of conservation

agriculture practice in coastal areas of Chattogram Division: A productivity and profitability analysis.

Asian Journal of Education and Social Studies, 13(2), 29-42.

https://doi.org/10.9734/ajess/2020/v13i230329

14. Meeusen, W., & van den Broeck, J. (1977). Efficiency estimation from Cobb‒Douglas production

functions with composed error. International Economic Review, 18(2), 435-444.

https://doi.org/10.2307/2525757

15. Mortuza, G. G. G. (2015). Quality cotton production for economic development of Bangladesh. The

Reflactor, 4, 37-72.

16. Nadiruzzaman, M., Rahman, M., & Pal, U. (2019). Climate resilient agriculture in Bangladesh: A value

chain analysis of cotton. Bangladesh Academy for Rural Development (BARD).

17. Nadiruzzaman, M., Rahman, M., Pal, U., Croxton, S., Rashid, M. B., Bahadur, A., & Huq, S. (2021).

Impact of climate change on cotton production in Bangladesh. Sustainability, 13(2), 574.

https://doi.org/10.3390/su13020574

18. Nadiruzzaman, M., Rahman, M., Pal, U., Hossain, M. F., Uddin, M. F., & Islam, M. K. (2019). Climate

Resilient Agriculture in Bangladesh: A Value Chain Analysis of Cotton. Action on Climate Today:

Dhaka, Bangladesh.

19. Odedokun, V.O. (2014). Economic Analysis of Cotton Production and Supply trend Estimation in

Zamfara State, Nigeria. Unpublished Ph.D. Dissertation, Department

of Agricultural Economics and

Rural Sociology, Ahmadu Bello University Zaria, Nigeria. 244p.

20. Parameswaran, C., & Cayalvizhi, B. (2020). Fertilizers, BT technology, and insecticides contributed to

60%, 23%, and 17%, respectively to the increase in BT cotton yield: An analysis from 2000 to 2014 in

India. bioRxiv. https://doi.org/10.1101/2020.05.06.081794

21. Rahman, S., Alam, M. J., & Hossain, S. T. (2013). Financial and economic profitability of selected

crops in Bangladesh. National Food Policy Capacity Strengthening Programme (NFPCSP).

22. Rahman, S., Islam, S. M. M., Prodhan, M. S. I., Hasan, M. M., & Khan, M. A. A. (2018). Productivity

and profitability of improved versus existing cropping pattern in Kushtia region. Bangladesh Journal of

Agricultural Research, 43(4), 587-598. https://doi.org/10.3329/bjar.v43i4.39157

23. Ramamoorthy, K. (1990). Economics of cotton production in India. In Cotton scenario in India,

Souvenir (pp. 103-108). Cotton Research Association.

INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)

ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XI November 2025

Page 1494

www.rsisinternational.org

24. Reddy, R., Raju, G. V. T., & Janaiah, A. (1997). The comparative economics of cotton cultivation: A

case study in Guntur district of Andhra Pradesh. Journal of Research ANGRAU, 25(1), 46-50.

25. Uddin, M. N., & Mortuza, G. (2015). Cotton production constraints and its prospects in Bangladesh.

International Journal of Business, Management and Social Research, 1(2), 62-73.

https://doi.org/10.18034/ijbmsr.v1i2.220