INTRODUCTION
Background information
Vietnam's total natural land area is approximately 33,123 thousand hectares, with agricultural and forestry land making up about 82.4% and 54.64% of the total land area, respectively (MONRE, 2016) The agricultural sector is a vital component of Vietnam's economy, contributing 18.14% to the country's gross domestic product in 2016 (WB, 2016) Despite its significance, the agricultural sector's contribution to the economy has been on a decline, decreasing from 22.7% in 2000 to 18.14% in 2016 (WB).
The agricultural sector remains a crucial component of Vietnam's national food security strategy and is expected to support industrial development in the coming years As of 2010, agriculture employed 48% of the workforce, reflecting its significance in the economy However, over the past two decades, while agriculture has experienced steady growth, the rapid expansion of the industrial and service sectors has resulted in a relative decline in agriculture's overall contribution to the economy.
Table 1 1 Land statistics of Vietnam
Land types Area (1000 ha) Ratio (%)
Source: Ministry of Natural Resource and Environment, 2016
Agricultural exports have shown consistent growth over the years, contributing significantly to a positive trade balance Key products in this sector include rice, rubber, coffee, cashew nuts, and various fishery and forestry items Notably, the total export value surged to $25 billion in 2011, doubling the figure from 2007 (JICA, 2013).
Despite significant advancements in agriculture that have aided poverty reduction, socio-economic development, and food security in Vietnam, the sector faces numerous challenges Key issues include unstable agricultural development and limited competitiveness in the global market, primarily due to small-scale production and fragmented farming, which result in high production costs Additionally, concerns regarding food safety and low production efficiency are on the rise The agricultural support services and related industries remain underdeveloped, with most exported commodities being raw or minimally processed, leading to lower added value and product quality compared to other countries Notably, cropping constitutes over 50% of agricultural activities, with rice production being the most crucial crop.
Table 1 2 Land use structure in Northern mountainous region of Vietnam
Land types Area (1000 ha) Ratio (%)
Source: Ministry of Natural Resource and Environment, 2016
The Northern mountainous region of Vietnam boasts significant forestry resources, with approximately 2.1 million hectares dedicated to forestry production, representing 71.4% of the region's total agricultural land In addition, agricultural production land comprises about 28% of the area According to the Ministry of Agriculture and Rural Development's plan for 2011-2020, the focus will be on enhancing forestry development and cultivating advantageous industrial crops, including tea, Arabica coffee, maize, lychee, and soybean.
In Vietnam, rice production occupies a significant portion of agricultural land, comprising 59.2% of the total annual cropping area as of 2016 Over the past decade, from 2007 to 2016, the total area dedicated to rice cultivation has consistently increased, reaching around 8 million hectares in 2016 This growth highlights that the rice production area far surpasses that of other crops, including maize, peanuts, soybeans, and cotton.
Figure 1 1 Planted area of major crops in Vietnam (1000 ha)
Source: General Statistic Office of Vietnam, 2018
The northern upland region of Vietnam encompasses approximately 2.1 million hectares of agricultural land, with 77% dedicated to annual crops Rice is a crucial crop, representing 35.4% of the region's total cropping area In contrast, perennial crops, including tea, fruits, and Arabica coffee, make up about 23% of the agricultural land Notably, over 90% of the local population relies on agriculture for their livelihoods (MONRE, 2016).
Between 2007 and 2016, approximately 11 million people in rural areas relied on agricultural activities, including cropping, animal husbandry, and forest economics, as their primary source of income Rice production plays a crucial role, contributing about 25% to household income, primarily for self-sufficiency rather than commercial purposes This focus on rice is vital for food security, especially in regions with challenging transportation systems due to hilly and complex topography (Bac et al., 2013).
Table 1 3 Structure land use of NMR
Land type Area (1000 ha) Ratio (%)
Source: Ministry of Natural Resource and Environment, 2016
Over the past decade, Vietnam has seen significant changes in the production of major perennial plants The area dedicated to rubber cultivation has expanded considerably, while the planted areas for coffee and pepper have experienced moderate growth In contrast, the tea planting area has remained relatively stable throughout the years.
Figure 1 2 Planted perennial area of Vietnam Source: General Statistic Office of Vietnam, 2018
Vietnam is amongst few nations in the world that have advantages of natural and climatic conditions for tea production (SOMO, 2007) Tea production is taking place in
39 out of 64 provinces all over the country with total 130 thousand ha NMR has the largest tea production area in comparison with other four regions of Vietnam, with about
93 thousand ha accounting for 72% of total planted tea area of Vietnam
Figure 1 3 Planted tea distribution in Vietnam Source: General Statistic Office of Vietnam, 2011
Cashew nut Rubber Coffee Tea Pepper
Red river deltaNorthern mountainous areaNorth central and coastal areaCentral highland
Similarly, the region also provide the highest tea production quantity of Vietnam, accounting for 66% of total produced tea quantity
Figure 1 4 Proportion of tea production among regions in Vietnam
Source: General Statistic Office of Vietnam, 2011
Tea production is vital to Vietnam's cultural heritage and economic landscape, with a history of tea cultivation and consumption spanning approximately 3,000 years In 2012, the tea industry contributed $224.8 million to the nation's export value, with over 146,800 tons of tea products shipped abroad This sector supports around 400,000 households engaged in tea production and related activities, providing livelihoods for approximately 1.5 million people.
Red river delta Northern mountainous area
North central and coastal area Central highland
Figure 1 5 Variability of tea yield in Vietnam Source: General Statistic Office of Vietnam, 2014
Production efficiency, risk and VietGAP adoption in Vietnam
Production involves converting inputs like land, labor, and capital into outputs such as goods and services, applicable across various sectors, including agriculture Differences in production performance arise from variations in inputs and outputs, with the ultimate goals of agricultural production often being profit maximization, cost reduction, or maximizing output, which can differ by firm and over time Key concepts in this context include technical efficiency, productive efficiency, and economic efficiency.
Production efficiency consists of two key components: technical efficiency and allocative efficiency Technical efficiency refers to a farmer's ability to minimize waste during production, maximizing output from given inputs or minimizing inputs for a specified output The goal of assessing technical efficiency is to identify strategies that enhance output or reduce input usage within the existing constraints.
Red river delta Northern mountainous area North central and coastal area Central highland
Vietnam's tea yield is influenced by a combination of inputs and outputs at optimal levels, taking market prices into account Technical efficiency is assessed by analyzing input and output quantities without incorporating prices This efficiency can be broken down into three key components: scale efficiency, which refers to productivity gains from optimal firm size; congestion, where increasing certain inputs may lead to decreased output; and pure technical efficiency, as defined by Farrell in 1957.
Economic efficiency focuses on maximizing output without increasing conventional inputs, emphasizing that leveraging existing technologies can be more cost-effective than adopting new ones, especially when current practices are inefficient (Shapiro, 1977) It is categorized into two types: technical efficiency, which assesses a farmer's capability to produce the highest output using available technologies, and allocative efficiency, which evaluates the optimal application of inputs based on their prices (Farrell, 1957; Shapiro, 1977) Technical efficiency (TE) ranges from 0 to 1, where a TE of 1 indicates full technical efficiency, meaning the firm operates at its maximum production potential.
Production efficiency is crucial for enhancing agricultural output, with numerous studies highlighting its importance (Thiam, 2001) In Vietnam, the agricultural sector plays a vital role in driving economic growth, ensuring food security, promoting social stability, and alleviating poverty Consequently, both the government and researchers prioritize improving efficiency within this sector However, research on the production efficiency of crops like rice, tea, and vegetables is limited, with findings indicating that Vietnamese farmers often operate below optimal efficiency levels (Hong et al., 2015; Bac et al., 2013; Tran, 2008; Vu, 2005) This suggests significant potential for cost reduction through enhanced efficiency, particularly in the context of limited land resources Additionally, the adoption of advanced technologies necessitates greater capital investment and time, while the fragmented and small-scale nature of agricultural production in Vietnam poses further challenges to implementing higher technology solutions.
1.2.2 Linkage between agricultural risk and efficiency
Production is the process of converting inputs like land, labor, and capital into goods and services, applicable across various sectors, including agriculture This process is heavily influenced by natural conditions and environmental factors, making agricultural production particularly susceptible to uncertainties and risks Farmers face challenges beyond their control, such as weather, market fluctuations, and policy changes, all of which directly impact their returns Therefore, effective risk management is essential for farmers, involving strategies such as on-farm decisions, structural changes, market instruments, government support, and income diversification A standard risk management analysis includes three steps: identifying risk sources and variability, selecting optimal management tools, and designing supportive government policies Alternatively, a holistic approach considers the complex interrelationships among various elements, providing a broader perspective on risk management in agriculture.
Risk assessment in agricultural production has garnered significant attention from researchers, leading to a wealth of literature in this field Agricultural production faces numerous sources of risks and uncertainties, highlighting the importance of understanding these challenges for effective management.
Agricultural production in Vietnam is significantly influenced by various risky factors, which are not evenly distributed among farmers due to the complexities of natural and climatic conditions These risks can be categorized into five main sources: production risk, marketing risk, financial risk, legal and environmental risk, and human resource risk Research by Tiedemann (2013) highlights the relationship between production risk and efficiency, noting that output variability in both German organic and conventional farming is primarily driven by production risk Given the adverse effects of these risks on farmers' production outputs, it is crucial for them to identify and manage these risks effectively.
1.2.3 The situation of VietGAP adoption
VietGAP adoption in Vietnam, similar to other Asian countries, is driven by the significance of GlobalGAP, a key standard in food safety and sustainability Conventional agriculture faces challenges due to excessive pesticide and chemical fertilizer use, which negatively impacts health and the environment In response to growing food safety concerns, Good Agricultural Practices (GAPs) have been promoted GAP principles encompass standards and guidelines applicable throughout all production phases, from field selection to post-harvest To address Vietnam's unique agricultural context, the government developed the Vietnamese Good Agricultural Practice (VietGAP), based on Hazard Analysis Critical Control Points (HACCP) and ASEAN GAP principles On January 28, 2008, the Ministry of Agriculture and Rural Development (MARD) issued decree No 379/QD-BNN-KHCN to implement VietGAP, which serves as the primary standard for producing safe fruits and vegetables The adoption of VietGAP aims to minimize risks in production and post-harvest processes while providing economic benefits to farmers, such as increased revenue, reduced costs, improved market access, and decreased vulnerability to poor agricultural practices.
Despite the numerous practical benefits of VietGAP adoption, the number of certified farmers in Vietnam remains low This limited uptake can be attributed to several barriers, with the primary issue being the lower popularity of VietGAP compared to other market standards like GlobalGAP.
The lack of international recognition for the domestic VietGAP standard discourages farmers from investing in less credible certifications Additionally, the implementation of VietGAP necessitates advanced infrastructure, which poses challenges for the majority of Vietnamese farmers who typically manage small plots averaging 0.25 hectares Furthermore, the high costs associated with obtaining and maintaining VietGAP certification deter both new producers from applying and existing farmers from renewing their certifications (Nabeshima, 2015).
Problem statement
Over the past two decades, Vietnam's agriculture has seen significant growth, yet the country remains a developing nation with a low average income While the agricultural sector's contribution to GDP is declining due to the rapid expansion of industry and services, it continues to play a vital role in the economy Approximately 65.5% of the population resides in rural areas, where agriculture serves as the primary income source for many households Despite impressive advancements in agricultural development that have reduced the national poverty rate from 58.1% in 1993 to 13.5% in 2014, challenges persist, particularly among the rural poor, who heavily depend on agricultural production Ethnic minorities, comprising 35.7% of the poor, face even higher poverty rates, such as the La Hu at 84.9% and H’Mong at 82.9% The northern mountainous region exhibits the highest poverty rates in the country, prompting the Vietnamese government to prioritize agricultural development, rural growth, and support for farmers.
Over the past two decades, Vietnam's agriculture has seen significant growth, primarily driven by the expansion of planting land However, this horizontal growth is nearing its limits due to the scarcity of undeveloped agricultural land and the country's high population density, which restricts further expansion Research indicates that Vietnamese farmers lack efficiency in various cropping activities, including rice, tea, and vegetables, highlighting the need for improved production efficiency and land optimization as key factors for future growth The Vietnamese government has recognized enhancing sector efficiency as a priority goal, with initiatives such as the tea production development plan established in 1999 aimed at increasing production, export turnover, and employment opportunities for farmers reliant on tea This policy is expected to alleviate poverty in upland tea-producing areas, often characterized by small-scale farming and limited off-farm income Additional measures, such as the "Law of Private Enterprise," have also been introduced to promote the tea value chain and improve market access for rural farmers.
1990, and “the Enterprise Law” which was enacted in 1999 and revised in 2005
Agricultural activities are heavily influenced by natural conditions and environmental factors, leading to significant variability in production outcomes due to inherent risks These risks stem from unpredictable biological, climatic, and price fluctuations that are beyond the control of farmers In Vietnam, agriculture faces similar challenges, characterized by small-scale, scattered production and low technology adoption As a result, agricultural productivity and product quality suffer, making them less competitive in the market To enhance agricultural production and increase farmers' income, it is crucial to implement advanced technologies and effectively manage risk factors.
VietGAP standards were introduced as a crucial response to food safety concerns in Vietnam, driven by rising consumer awareness in both domestic and international markets Established by the Ministry of Agriculture and Rural Development (MARD) in 2008, these Vietnamese Good Agricultural Practices align with GlobalGAP, ASEAN Good Agricultural Practices, and Hazard Analysis and Critical Control Points (HACCP) Additionally, VietGAP is recognized for its eco-friendly production methods, emphasizing the extensive use of organic components in cultivation and protection (Ha, 2014).
Initially designed for vegetable production in Vietnam, the VietGAP standard package expanded to include fruit and tea production in 2009 The implementation of VietGAP standards aims to enhance farmers' economic value and mitigate production risks by offering premium prices, improved market access, and reduced average production costs However, the adoption of VietGAP has not met expectations, with varying perspectives among farmers regarding its economic benefits While numerous studies have examined production efficiency, there is a lack of research on the efficiency of crops under VietGAP and its effects on household income.
Research objective
The study aims to assess the production efficiency of tea and rice farmers in northern Vietnam, highlighting the significance of VietGAP tea production for their livelihoods It specifically seeks to explore the factors contributing to inefficiencies in production and to investigate the economic implications of adopting new practices Additionally, the research will examine farmers' perceptions of risk sources and their corresponding management responses.
To achieve the overall objective, three main research questions need to be investigated:
1 Do tea and rice farmers operate at fully efficient levels or is there any potential for improving farmer’s production efficiency? And which factors have effects on improving production efficiencies of farmers?
2 How does VietGAP tea production affect household’s income in the study area? And what are determinants for shifting from conventional to VietGAP tea production?
3 What is source of risks facing by farmers and how do they respond to those risk sources?
Organization and structure of the dissertation
The study is structured around two primary objectives and consists of seven chapters Chapters 2 and 3 focus on Objective 1, while Chapters 4, 5, and 6 address Objective 2 Chapter 1, titled "Introduction," provides an overview of the agricultural sector, highlights major crops and perennial plants, and outlines the problem statement and objectives of the study A detailed analysis of the current production levels and profit efficiency of rice and tea farmers is presented in Chapter 2.
Chapters 2 and 3 will address the determinants of enhancing technical and profit efficiency for farmers Chapter 4 will analyze the factors influencing farmers' decisions to adopt VietGAP tea production The contribution of VietGAP tea production to household income will be discussed in Chapter 5 Chapter 6 will focus on farmers' risk perception and their management responses Finally, Chapter 7 will present the main findings of the research along with policy implications.
The structure of the dissertation is presented as figure 1.6 below
Figure 1 6 Overall structure of the dissertation
Objective 1 To analyze productive efficiency of tea and rice farmers
Chapter 2 Analysis of technical efficiency of rice farmers and its determinants
Chapter 3 Analysis of profit efficiency of tea farmers and its determinants
Objective 2 To determine the economics of adoption, risk sources and farmers’ risk management response
Chapter 4 Factors affecting farmers’ decision to adopt VietGAP production
Chapter 5 Assessing impacts of VietGAP production on farmers’ income
Chapter 6 Farmer’s perception of risk sources and their management response
Chapter 7 Conclusion and policy implication
Selection of study area
Northern Vietnam, encompassing the midland and mountainous regions, spans 95,222 km² and is home to approximately 11.98 million people from various ethnic minority groups This area comprises 14 provinces in the northwest and northeast, characterized by mountainous terrain that supports dominant agricultural and forestry economies due to favorable natural conditions Key crops include rice, maize, tea, rubber, and Arabica coffee The study focuses on Thai Nguyen province, which has a population of 1,227.4 thousand and an average density of 384 persons per km², divided into nine administrative units Tea and rice farming are vital for rural livelihoods, with rice primarily grown for self-sufficiency and tea cultivated for commercial purposes The research involved a random selection of farmers from representative districts, with data collected through surveys in two phases: 120 rice farmers and 116 VietGAP and 210 conventional tea farmers in 2016 A pretest survey refined the questionnaire, and trained enumerators ensured accurate data collection, with rice data analyzed in Chapter 2 and tea data utilized in subsequent chapters.
Figure 1 7 Map of study area
PRODUCTIVE EFFICIENCY OF RICE FARMERS AND ITS
Introduction
Rice is a vital staple for the Vietnamese population, significantly impacting household income in rural areas and contributing to agricultural growth and poverty reduction Despite these achievements, challenges persist, with a national poverty rate of 13.5% in 2014, particularly high among ethnic minority groups Income disparity remains pronounced between urban and rural regions, with ethnic minorities, comprising only 7% of the population, facing a poverty rate of 25.4% The Northern upland area, with 1.6 million hectares of annual cropping land, dedicates 579 thousand hectares to rice production, representing 35.4% of the region's agricultural output Farmers in this region rely heavily on agriculture, with rice contributing approximately 25% to household income, especially in rural and mountainous areas Additionally, rice self-sufficiency enhances food security in upland regions, where challenging topography complicates transportation.
While numerous studies have examined the productive efficiency of agricultural crops in Vietnam (Nguyen et al 2003; Linh, 2008), research specifically focusing on the technical efficiency of rice production remains limited Most of the existing studies are concentrated in the primary rice-producing regions, namely the Mekong River Delta and the Red River Delta Additionally, some research has attempted to assess the nationwide technical efficiency of rice production based on the assumption that significant regional differences do not exist across Vietnam (Khai and Yabe).
This study aims to analyze technical efficiency and evaluate the effects of different fertilizers on rice production in Vietnam The findings will provide valuable insights into the overall landscape of rice cultivation in the country.
Methodology
Technical efficiency (TE) is a crucial metric for production firms, as it assesses the effective use of resources like land, labor, capital, and materials Researchers focus on measuring TE to estimate the efficiency levels of farmers engaged in agricultural production This measurement helps answer a key question: Can rice farmers enhance their productivity under existing conditions? TE, along with allocative efficiency (AE), constitutes the two main components of economic efficiency (EE).
2.2.1.1 Economic, technical and allocative efficiency
Production involves converting inputs like land, labor, capital, and fertilizer into outputs such as goods and services, applicable across various sectors beyond agriculture Differences in production performance are reflected in the variations of inputs and outputs The primary goals of agricultural production can include profit maximization, revenue enhancement, cost reduction, or maximizing output, which may differ between firms and over time Key concepts in this context include technical efficiency, productive efficiency, and economic efficiency.
Production efficiency consists of two key components: technical efficiency and allocative efficiency Technical efficiency refers to a firm's ability to minimize waste during production, aiming to maximize output from given inputs or minimize inputs for a specific output The goal of assessing technical efficiency is to identify ways to enhance output or reduce inputs under existing conditions On the other hand, allocative efficiency focuses on the optimal combination of inputs and outputs based on market prices Measuring technical efficiency involves evaluating input and output quantities without considering their prices and can be further broken down into three subcomponents: scale efficiency, which reflects productivity gains from optimal firm size; congestion, where increasing some inputs may lead to reduced output; and pure technical efficiency, as defined by Farrell in 1957.
Economic efficiency focuses on increasing output without exceeding conventional input levels, emphasizing that leveraging existing technologies can be more cost-effective than adopting new ones, especially when current farming practices are inefficient (Shapiro, 1977) It is categorized into two types: technical efficiency, which assesses a farmer's capability to maximize output with available technologies, and allocative efficiency, which evaluates the optimal use of inputs relative to their prices (Farrell, 1957; Shapiro, 1977) Technical efficiency (TE) ranges from 0 to 1, where a TE of 1 indicates that a firm operates at full technical efficiency, illustrating the potential for maximum productivity.
In microeconomic theory, a production function defines the output of a firm based on various input combinations It represents the maximum achievable output for a specific set of inputs, reflecting the most efficient use of resources Additionally, it can outline the minimum input requirements necessary to produce a given output with existing technologies Economists utilize production functions to analyze technical and allocative efficiency, identifying when a firm operates inefficiently if its output falls below the production frontier.
Two widely used methods for estimating a firm's technical efficiency are the parametric and non-parametric approaches The parametric method, known as Stochastic Frontier Analysis (SFA), relies on a defined functional relationship between inputs and outputs, utilizing statistical techniques to estimate parameters In contrast, the non-parametric method, referred to as Data Envelopment Analysis (DEA), constructs a linear piecewise function based solely on empirical data without assuming any specific relationship Comprehensive reviews of these methods have been conducted by Kalirajan and Shand (1999) and Bravo-Utera and Pinheiro.
The optimal method for assessing efficiency in agriculture remains ambiguous, as highlighted by empirical analyses comparing Data Envelopment Analysis (DEA) and Stochastic Frontier Analysis (SFA) (Sharma et al., 1999) While the efficiency scores derived from these approaches exhibit quantitative differences, the ordinal rankings of farms are notably similar Ultimately, the selection of a method should align with the study's objectives, the specific type of farms being analyzed, and the assumptions about the data generation process.
Data envelopment analysis (DEA) is a mathematical programming technique designed for evaluating production technologies with multiple inputs and outputs Developed by Charnes et al in 1978, DEA employs linear programming to create a non-parametric piece-wise frontier that assesses the efficiency of each data point in relation to this frontier The methodology requires a clear distinction between inputs and outputs, with each data point representing a decision-making unit (DMU) that aims to optimize output generation through efficient input utilization (Zheng et al., 2004).
The Stochastic Frontier Approach (SFA) employs econometrics based on a deterministic parameter frontier, as proposed by Aigner et al (1977) through the stochastic frontier production function model: lnqi = xiβ + νi – ui In this model, qi denotes the output of the i-th firm, while xi is a vector of input logarithms, β represents unknown parameters, νi accounts for statistical noise, and ui indicates technical inefficiency SFA is a stochastic and parametric method that effectively distinguishes between noise and inefficiency, although it primarily yields accurate results for single output and multiple inputs Conversely, the Data Envelopment Analysis (DEA) method is non-stochastic and non-parametric, failing to separate noise from inefficiency but proving advantageous for farms with multiple inputs and outputs The production frontier model for calculating technical efficiency is limited to single output production, and the choice of estimation techniques varies based on the data structure, whether cross-sectional or panel data.
This study employs Stochastic Frontier Analysis (SFA) to estimate technical efficiency in rice production in Vietnam, as it effectively distinguishes between noise and technical inefficiency Given that rice production involves a single output—quantity—and multiple inputs including seeds, fertilizers, labor, pesticides, and hired machinery, SFA is particularly suitable The Cobb-Douglas production function is utilized, with rice yield serving as the dependent variable, reflecting the consistent harvesting patterns among farmers in the region due to small rice areas per household, uniform technology, and similar environmental conditions The logarithmic transformation of the function leads to the development of the analytical model.
The Cobb-Douglas production function can be expressed by following equation ln yi = βo + Σ βj ln (xij) + εi (2.1)
The rice yield (yi) of the i-th farm is influenced by various inputs (xij), including seed, nitrogen, phosphorus, potassium, pesticides, labor days, and hired machinery The model incorporates an intercept (βo) and parameters (βj) that estimate the elasticity of each input Additionally, the error term (εi) consists of two components: Vi, a random variable error linked to uncontrollable factors such as measurement errors and natural disasters, and Ui, a non-negative random variable reflecting farm-specific factors affecting technical efficiency Vi is assumed to be independently and identically distributed, while Ui follows a truncated-normal distribution with a mean of à and a variance of δ².
Ui can also have other distributions, FRONTIER 4.1c computer program used in the study can only harmonize with above assumption The term ài is defined as follows ài = δo + δ1Z1j + δ2Z2j + δ3Z3j + δ4Z4j + δ5Z5j + δ6Z6j + ωi (2.2)
The inefficiency effects in agricultural production can be estimated using the two-stage estimation technique in FRONTIER 4.1c In this model, δo represents the intercept term, while δj denotes the parameters associated with the independent variables Key factors influencing inefficiency include farmers' experience (Z1j), education level (Z2j), household size (Z3j), number of land plots (Z4j), and area classification, where Z5j indicates Northeast (1) or Northwest (0) regions Additionally, access to credit is represented by Z6j, which plays a crucial role in enhancing agricultural efficiency.
In this study, farmers are categorized based on whether they have borrowed credit loans from financial agencies, indicated by a value of 1, and 0 otherwise The model incorporates an error term (ωi) representing unobservable random variables Maximum likelihood estimates (MLEs) for all parameters of both the stochastic frontier production and inefficiency model were simultaneously estimated using the FRONTIER 4.1c software developed by Coelli.
1996) This program also presented the coefficients of variance parameters σ 2 = σ 2 v + σ 2 u (2.3) γ = σ 2 u/ σ 2 v (2.4)
The γ parameter, ranging from zero to one, indicates the proportion of inefficiency in the total residual variance of rice yield A gamma value of zero signifies that all variations in rice yield are attributed to noise, while a value of one indicates that all variations are due to technical inefficiency (Coelli and Battese, 1996).
The study was conducted in Thai Nguyen province, located in the northern upland area of Vietnam, utilizing a multistage sampling technique to select 120 farmers from 18 villages across two districts Key factors considered for household selection included geographical location, rice production area, and family status Data was collected through interviews with farmers using a detailed questionnaire that covered household characteristics, farm size, and inputs and outputs such as rice yield, varieties, fertilizer applications, credit, and agricultural extension services Additionally, secondary data was obtained from the General Statistics Office of Vietnam and communal reports during the 2011 field survey.
Results and discussion
Table 2 1 Descriptive statistic of variables in the model
Variable Unit Mean St.d Min Max
Note: * 1 unit = 1000m 2 ; **vnd: monetary unit of Vietnam
Source: Author’s data was surveyed in 2011 (n0)
Table 2.1 presents summary statistics for production function variables and farm-specific characteristics influencing technical inefficiency The average rice yield for the sampled farmers was 427 kg per unit, with an average seed application of 6.91 kg per unit Farmers utilized various types of chemical fertilizers, including nitrogen, phosphorus, and potassium, while none reported using organic fertilizers on their rice farms Additionally, chemical pesticides were commonly used in the study area to mitigate damage from pests and insects.
Annual crop production varies due to specific pest statuses, and the low level of mechanization in rice production is attributed to small-scale, scattered, and sloping land Machinery is primarily utilized for land preparation The analysis of technical efficiency includes descriptive statistics of output and input variables, alongside summary statistics of determinants on technical inefficiency The average age of the household head is approximately 44 years, with educational attainment primarily at the primary (30%) and secondary school (59.17%) levels On average, each family comprises about 4.78 members, with a range from 1 to more.
The study analyzes 11 laborers across an average of 5.55 land plots per household, incorporating two key dummy variables: area and credit access The area variable differentiates between the distinct natural conditions of the Northwest and Northeast regions, while the credit access variable evaluates the impact of credit utilization on rice yield improvement in the study area.
2.3.2 Estimation of stochastic frontier production function
The MLEs of stochastic frontier production function is shown in Table 2.2 below Gamma values (γ) of model are 0.863 and its value is statistically significant at 1% level
The analysis reveals that 86.3% of the variation in rice yield is attributed to technical inefficiency, confirming the appropriateness of the stochastic frontier function model for this research The presence of technical inefficiency was evaluated using the Likelihood Ratio (LR) test, where the null hypothesis (H0) posits that the gamma value is zero, indicating that yield variations are merely random noises Conversely, the alternative hypothesis (H1) suggests a non-zero gamma value, supporting the model's adequacy The LR test, which follows a mixed Chi-square (χ²) distribution, produced a value of 35.86, exceeding the critical Chi-square value of 20.09 at the 1% significance level, leading to the rejection of the null hypothesis This outcome indicates that farmers in the study area are not fully technically efficient.
Table 2 2 Estimated parameters of stochastic frontier production function
Note: *, **, *** are significant level at 10%, 5%, 1% respectively; χ 2 (1%, 8) = 20.09
Source: author’s surveyed data in 2011 (n0)
The analysis revealed that the estimated coefficients for seed, nitrogen, phosphorus, potassium, labor, and hired machinery were positive, indicating a favorable relationship with rice yield However, only nitrogen, potassium, and hired machinery showed statistical significance at the 1% and 5% levels, respectively This suggests that rice farmers can enhance yields by increasing nitrogen and potassium use, while employing hired machinery can lead to improved land preparation and higher yields Conversely, the coefficients for seed, phosphorus, and labor were not statistically significant, and the pesticide variable had a negative coefficient, indicating that excessive pesticide use may harm rice growth, although this result was not statistically significant.
2.3.3 Input elasticity and its responsiveness to rice yield
The elasticity of various inputs is assessed to quantify their impact on rice yield, utilizing the Cobb-Douglas production function and natural logarithms The parameters derived from this function serve as elastic coefficients, reflecting the relative significance of each input on output in percentage terms Detailed coefficients of the Cobb-Douglas production function can be found in Table 2.2.
The study reveals that nitrogen input has the highest elasticity in rice yield among sampled farmers, with a response rate of 0.031 Potassium input follows with an elasticity of 0.011, while machinery input ranks third at 0.003 This indicates that a 1% increase in nitrogen, potassium, and machinery inputs results in corresponding increases in rice yield of 0.031%, 0.011%, and 0.003%, respectively Other factors such as seed, phosphorus use, pesticide costs, and working days also show a relationship with rice yield, but their coefficients are not statistically significant, suggesting that farmers may have overutilized these inputs or reached their production limits.
2.3.4 Frequency distribution of technical efficiency
Frequency distribution of technical efficiency in Northern upland area is presented in Table 2.3 Average technical efficiency scores (TE) of rice production were
Farmers have the potential to enhance their technical efficiency by an average of 12% with the current inputs and technology, achieving a technical efficiency rate of 88% This finding is lower than the results reported by Hassan and Ahmad (2005) but slightly higher than those of Shehu et al (2007) and Nguyen et al (2003).
The study reveals a significant variation in technical efficiency among farmers, ranging from 45% to 98% Notably, the most efficient farmers achieved an average of 98%, while the least efficient operated at only 45% Additionally, approximately 80% of farmers attained a technical efficiency above 80%, demonstrating that more than three-fourths of the farmers in the area functioned at a fairly efficient level.
Table 2 3 Frequency distribution of technical efficiency
Efficiency level No of farmer Percentage (%)
Source: author’s surveyed data in 2011 (n0)
Achieving the highest level of technical efficiency at 98 percent could enable farmers to reduce their production costs by an average of 9 percent This significant cost saving is calculated using the formula: Cost saving = 1 – TEmean/TEmax.
2.3.5 Analysis of determinants of technical inefficiency
The study not only analyzed technical efficiency in rice production but also identified factors contributing to technical inefficiency The Maximum Likelihood Estimates (MLEs) for the technical inefficiency model are detailed in Table 2.4 Notably, the coefficients for experience, education level, household size, and region variables are negatively correlated with technical inefficiency, indicating a positive relationship with technical efficiency Specifically, educational level and household size significantly reduce technical inefficiency, with p-values of 10% and 5%, respectively This suggests that enhancing the education of household heads and increasing labor availability can mitigate inefficiency, aligning with findings from Bravo-Ureta and Pinheiro's research.
(1997) This might be because more educated farmers are more likely to access better-
Table 2 4 Determinants affecting technical inefficiency
Note: *, **, *** are significant level at 10%, 5%, 1% respectively
The study reveals that household size positively influences timely rice cultivation, aligning with Shehu et al (2007), as larger labor forces are advantageous, especially with increased rice area due to rotation and multiple seasons Conversely, the region coefficient indicates a significant negative relationship with inefficiency, likely due to more favorable natural conditions in Northeast compared to Northwest Additionally, the number of land plots shows a statistically significant positive correlation with inefficiency; more plots necessitate increased working days and costs, thereby reducing technical efficiency in rice production This finding is consistent with research by Hung (2007) and Rahman et al (2009) on land fragmentation in Vietnam and Bangladesh While credit generally enhances technical efficiency, in this study area, its impact is negative as farmers often use loans for non-rice-related purposes, incurring interest while mismanaging inputs like pesticides and fertilizers This aligns with Battese and Broca (1997) but contrasts with Idiong (2007) in Nigeria.
2.3.6 Estimation of potential rice yield
As defined in method section, technical efficiency (TE) is ratio of actual yield (observable yield) to frontier yield (potential yield) So it could be expressed by equation below
TE = Actual yield/Potential yield (2.5)
From above concept, we could estimate potential yield for each rice farmer by the following equation:
Potential rice yield = Actual rice yield/TE (2.6)
In case all farmers could reach the potential rice yield in given set of inputs and technology, average rice yield could be reached 480.98 kg per unit
This research evaluates the technical efficiency of rice farmers in northern upland Vietnam, revealing an average technical efficiency score of 88%, indicating significant potential for improvement in production practices The hypothesis test confirmed the presence of technical inefficiency in rice production Key inputs such as nitrogen, potassium, and machinery were found to positively and significantly impact rice yields, suggesting that their increased use could enhance productivity However, the study also identified a considerable imbalance in fertilizer application among farmers, which may contribute to inefficiency Furthermore, the analysis highlighted that the educational level of the household head and family size are crucial determinants of technical efficiency, while an increasing number of land plots negatively affects efficiency.
To enhance the technical efficiency of rice production, it is essential for agricultural extension services to encourage farmers to use fertilizers more effectively, emphasizing a balanced application of different types Additionally, improving the educational level of household heads can positively impact efficiency, suggesting that increased investment in public education will benefit both rice production and the agricultural sector as a whole in the long term Furthermore, public interventions should focus on mitigating the adverse effects of small-scale production and land fragmentation in agriculture.
Conclusions and recommendations
Tea is a vital component of Vietnam's culture and economy, with a rich history spanning over 3,000 years As a significant cash crop, tea farming supports around 400,000 households and generates approximately 1.5 million jobs, particularly in the northern provinces In 2012, Vietnam exported about 146,700 tons of tea, valued at USD 224.6 million Despite the steady increase in tea consumption and exports since the 1990s, the industry faces challenges, including reliance on conventional farming practices and excessive use of pesticides, which pose risks to health and the environment In response to growing concerns about food safety, Vietnam is transitioning to VietGAP tea production, which emphasizes eco-friendly practices and the use of organic components, aiming to enhance the quality and safety of tea products for both domestic and international markets.
Analyzing various tea production practices is crucial for comprehending farmers' decision-making processes (Tran, 2009) These practices significantly influence production efficiency, making it essential to examine the efficiency of VietGAP tea farmers (Nguyen et al., 2015; Tran, 2008) Additionally, the outcomes of GAP impact evaluations vary across countries and crops.
PROFIT EFFICIENCY OF TEA FARMERS AND ITS
Introduction
Tea is a vital part of Vietnamese culture and economy, with a history spanning over 3,000 years It serves as a significant cash crop for farmers in northern Vietnam, contributing to the livelihoods of approximately 400,000 households and generating around 1.5 million jobs In 2012, Vietnam exported about 146,700 tons of tea, valued at USD 224.6 million Despite the steady increase in tea consumption and exports since the 1990s, challenges persist, particularly regarding conventional farming practices and the excessive use of pesticides and chemicals, which pose risks to health and the environment In response to growing consumer concerns about food safety, Vietnam has initiated a shift towards "safe or clean" tea production through VietGAP certification, promoting eco-friendly practices that minimize chemical residues and support a transition to organic farming.
Analyzing various tea production practices is crucial for understanding farmers' decision-making (Tran, 2009), as different methods can lead to varying production efficiencies Studying the production efficiency of VietGAP tea farmers is essential (Nguyen et al 2015; Tran 2008), given that the impact of Good Agricultural Practices (GAP) differs across nations and crops While some research indicates that GAP adoption improves technical efficiency (Taraka et al 2011; Ha 2014b), other studies suggest that farmers may not benefit from premium pricing (Calvin et al 2004; Subervie and Vagneron 2012; Pongvinyoo et al 2015), though some have reported positive effects on price, yield, or income (Kariuki et al 2012; Islam et al 2012) In Vietnam, there is limited information on the economic efficiency of VietGAP tea production due to its relatively recent implementation Previous studies have explored aspects of tea farming in Vietnam, such as Tran (2008) estimating the economic efficiency of organic tea farmers, Saigenji (2011) assessing the impact of contract farming on production efficiency and household income, and Hong and Yabe (2015) investigating profit efficiency without focusing on specific production practices To fill this research gap, we compared the profit efficiency of VietGAP and conventional tea farmers using propensity score matching to control for selection bias.
Methodology and data collection
3.2.1 Measurement of production and profit efficiency
Over the past two decades, empirical studies on agricultural production efficiency have primarily focused on two areas: one that estimates productive efficiency based on input demand price responses and another that examines production inefficiency without considering these price responses (Arnade and Trueblood, 2002) The profit function estimation technique, developed by Kumbhakar in 1996, addresses both production inefficiency and price response Production inefficiency is typically analyzed through three components: technical, allocative, and scale inefficiency A production unit is deemed technically inefficient if its output falls below the maximum feasible output, known as the frontier output Additionally, allocative inefficiency occurs when a production unit fails to use inputs in optimal proportions based on observed input prices and output levels In the context of profit maximization, a firm is not considered scale efficient if it does not produce at an output level that aligns the product price with marginal cost (Kumbhakar et al.).
Econometrics integrates multiple measurements into a unified system for more efficient estimation, particularly through a profit function framework (Ali and Flinn 1989; Kumbhakar et al 1989; Wang et al 1996) The frontier production function is a common method for assessing efficiency, though it may not be suitable for farmers with varying prices and factor endowments (Ali and Flinn 1989) Consequently, the stochastic profit function is employed to estimate firm-specific efficiency (Kumbhakar et al 1989; Ali and Flinn 1989; Wang et al 1996), merging concepts of technical, allocative, and scale inefficiency into the profit relationship Two primary methods for measuring production efficiency are the econometric approach and the mathematical programming approach (Lovell 1993) The stochastic frontier model consists of a symmetric component for random variations and measurement errors, along with a one-sided component that accounts for inefficiency relative to the stochastic frontier (Aigner et al 1977) The stochastic frontier approach (SFA) effectively distinguishes between noise and technical inefficiency The stochastic profit function is expressed as πi = f(Pi, Zi) exp(ξi), where normalized profit is calculated for each farm, incorporating input price variables and fixed factors, along with an error term comprising independent components related to inefficiency (Ali and Flinn, 1989) The profit efficiency (PE) of each farm is derived from this stochastic frontier profit function.
PE = E[exp(-ài) | ξi], (3.3) where E is an expectation operator that can be estimated by obtaining the expressions for the conditional expectation ài upon the observed value of ξi (0 ≤ PE ≤ 1)
Over the past decade, various impact evaluation approaches and econometric methods have been utilized, with the selection of the most appropriate method often debated in empirical economic analysis For instance, treatment effects can be assessed through regression model coefficients or by incorporating dummy variables related to crop cultivation or technology adoption Ideally, impact evaluations yield accurate results when comparing the same farmers before and after adoption, necessitating a baseline survey to collect data on potential adopters While this is feasible in small-scale trials, it poses challenges at a regional level Notably, much of the existing literature relies on non-randomized observational studies instead of randomized trials.
In 2002, it was noted that participants in certain activities often differ from nonparticipants, even without treatment For instance, individuals with high skills are more likely to achieve specific goals (Caliendo and Kopeinig, 2008) This indicates that inherent differences among farmers can significantly influence their production choices and overall performance Therefore, when comparing the profit efficiency of tea practices, failing to account for these differences may lead to biased results.
Tea producers have various options for selecting production inputs and marketing their products, resulting in price variations influenced by location and product quality It is assumed that these producers allocate inputs optimally by aligning their ratios with marginal productivity In economic theory, a farm is considered to maximize profit under conditions of perfectly competitive markets and established output technology Profit efficiency refers to a production unit's capacity to achieve the highest possible profit given the prevailing prices and fixed factors Conversely, profit inefficiency arises from not operating on the profit frontier, leading to potential profit losses.
The profit efficiency of a tea farmer is determined by the profit attained while operating on the profit frontier, factoring in variable input prices and quasi-fixed input quantities Specifically, a farm's profit is calculated by subtracting total variable costs from total revenue.
The Cobb-Douglas production function can be expressed in logarithmic form as ln νn/p = α0 + Σαi ln Pi/p + Σαq ln zq + νi – ài, where νn/p represents the normalized profit frontier, Pi/p indicates normalized input prices, and zq refers to quasi-fixed input quantities In this equation, αi and αq are unknown parameters, while Pi denotes the price of the i-th input used by each tea farm, normalized by the farm's tea price (p) This includes costs related to chemical fertilizers (converted to NPK), organic compounds, pesticides, labor, and other expenses.
The variable zq represents the amount of fixed inputs utilized by a tea farm, which includes the size of the farm measured in hectares Additionally, νi accounts for statistical noise, while ài reflects the impact of profit inefficiency The parameter α is an unknown value that requires estimation.
Maximum likelihood estimation is employed to estimate unknown parameters, with the likelihood function defined by variance parameters, σ² (σ²v + σ²u) and γ = σ²u/σ²v (Battese and Coelli, 1995) Profit efficiency levels of specific tea farms were predicted using statistical software, and a regression model was utilized to identify factors influencing the profit efficiency of tea farmers.
Profit efficiency (PE) for the i-th tea farmer is determined by the equation PE = βo + ΣβjZj + ω In this model, Zj encompasses various socioeconomic and farm characteristics impacting profit efficiency, such as gender, education level, family labor, farming experience, access to irrigation and credit, the ratio of tea income, cooperative membership, and machinery status The term ω accounts for unobserved factors that may influence the results.
In randomized experiments, the mean effect of a treatment is typically estimated by comparing the outcome variable's mean values between treatment and control groups However, this method is unsuitable for assessing VietGAP tea farmers due to their non-random selection Consequently, we utilized propensity score matching (PSM) to evaluate the impact of VietGAP application on tea farmers, relying on cross-sectional data for our analysis.
The Propensity Score Matching (PSM) method is utilized to align individuals with comparable characteristics across two groups, aiming to establish an experimental framework where participants and non-participants are randomly chosen This two-step mathematical procedure, as outlined by Becker and Ichino (2002), begins with the estimation of a farmer's propensity score through logit or probit models.
The model used in the study is represented by the equation Y(1,0) = β0 + β1X1 + β2X2 + … + βnXn, where Y indicates the dependent variable distinguishing between VietGAP and conventional farmers The estimated coefficients (β) and covariates (Xn) are selected based on economic theory and existing research, as omitting significant variables can lead to biased results Following the approach of Noltze et al (2012), the study includes covariates such as household size, the formal education of the household head, credit access, and extension access, while also introducing irrigation and machinery use to reflect mechanization in tea production, as suggested by Tran (2008) Additionally, factors like gender, farming experience, and farm size are incorporated to assess their impact on the adoption of agricultural innovations, as outlined by Kersting and Wollni (2012) Lastly, the model includes the variable "ratio of tea income" to emphasize the significance of tea income in the region.
Then, the propensity score was estimated using the following equation:
In the second step, farmers with similar propensity scores between the groups were matched to estimate the average treatment effect for the treated (ATT), denoted as
ATT = E(Y1-Y0)|x, D=1) = E(Y1| x, D=1) - E(Y0| x, D=1), (3.8) where D is an indicator equal to one if the farmer applies VietGAP, and zero is otherwise,
In evaluating the impact of VietGAP adoption, Y1 represents the outcome for adopters while Y0 indicates the outcome for non-adopters, with x as a vector of control variables To ensure comparability, single nearest neighbor matching (NNM) was implemented to align similar observations, aiming for statistically identical variable means between treatment and control groups Propensity score matching (PSM) relies on the assumptions of conditional independence and common support, yet unobservable variables may influence both VietGAP adoption and its outcomes, potentially introducing hidden biases These biases can lead to either an overestimation or underestimation of treatment effects, depending on the nature of the unobserved selection Consequently, a balancing test is necessary to assess whether the matching procedure effectively equalizes the distribution of relevant variables, as significant differences should not persist after conditioning on the propensity score Achieving adequate matching quality is critical, as it should result in significantly reduced standardized bias, supported by statistically insignificant likelihood ratio tests on the joint significance of all regressors and low pseudo R-squared values.
After matching, it is essential to achieve common support by visually inspecting the propensity score densities of both treatment and control groups Alternatively, the Kolmogorov-Smirnov nonparametric test can be employed for comparison If significant differences are observed between the maxima and minima of the density distributions, it is necessary to remove cases that fall outside the support of the other distribution.
Results and discussion
3.3.1 Socio-economic characteristics of tea farmers
The descriptive statistic method was employed to analyze the current conditions of tea farms in the study area As shown in Table 3.2, the table highlights key variables utilized in the model alongside specific characteristics of the farms.
Table 3 2 Descriptive statistics of tea production practices
Variables All samples VietGAP CON Diff t-stat
Organic (kg/ha) 2025.3 2557.3 1724.3 832.9 *** 3.8233 Pescost (liter/ha) 175.1 157.2 185.2 -28.04 * -2.0963 Labor (man-day) 1088.5 1236.3 1004.8 231.6 *** 3.8239 Ocost (K.vnd) 25,566.9 24,839.7 25,978.4 -1138.7 -0.8610 Input variables of profit function
Variables Mean Std.dev Min Max
for 5%, and * for 10% Additionally, the monetary unit used is K.vnd, with an exchange rate of approximately 1 USD equating to 21 K.vnd.
A comparative analysis of VietGAP and conventional tea farmers was conducted using t-statistics, revealing a strong correlation between "experience" and other variables Consequently, "experience" was retained for regression analysis The findings showed that the average tea yield is approximately 8,151 kg per hectare, with VietGAP tea farmers achieving significantly higher yields than their conventional counterparts at a 5% significance level.
VietGAP tea farmers incur higher production costs compared to conventional farmers, primarily due to their increased use of organic fertilizers, which they recognize as beneficial for sustainable tea production and farmland health While VietGAP farmers have lower pesticide costs, variability in expenses may arise from factors such as farm size, pest attacks, and individual farmer attitudes towards pesticide use Technical training on pesticide application under the VietGAP program has positively influenced farmers' perceptions regarding chemical use in tea cultivation Additionally, VietGAP tea farms require more labor, averaging around 1,100 labor days per hectare annually, especially during the labor-intensive harvesting season Despite these differences, the average size of tea farms remains consistent at approximately 0.35 hectares per farmer, with no significant variation between VietGAP and conventional farming practices.
Table 3.3 presents comparative statistics for the key variables utilized in the regression analysis The difference between the two groups is calculated by subtracting the mean of conventional tea farmers from the mean of VietGAP tea farmers Additionally, the t-statistic value reflects the significance level of the observed differences.
Table 3 3 Comparative statistics of model variables
Variables All samples VietGAP CON Diff t-stat
In a 2016 survey conducted by the author, data revealed significant differences between VietGAP adopters and non-adopters, with distinctions marked at the 1%, 5%, and 10% significance levels The analysis shows that the difference (Diff.) is calculated as the mean of adopters minus the mean of non-adopters, highlighting the impact of VietGAP practices on farm performance The study included a sample size of 116 VietGAP farms and 210 conventional farms, providing a robust comparison of agricultural outcomes.
Most of the tea farmers (85%) have a basic education at a secondary or high school level
A limited number of tea farmers possess higher education, highlighting a lack of attraction for educated individuals, especially among the youth, which may hinder the adoption of technology and access to premium tea markets However, these farmers have significant experience in tea cultivation, averaging around 22 years, and generate approximately 62% of their total income from tea, indicating that tea production is a dominant economic activity in the region.
Farmers adopting VietGAP standards have significantly more family labor available compared to conventional tea farmers, with a notable difference at the 1% level While VietGAP farms are not intensive, they require more labor, particularly during the harvesting period The stringent requirements for VietGAP certification necessitate greater investment in production systems, including irrigation, to ensure water quality Certified organizations regularly verify water sources and other inputs, and only tea products that meet all regulatory standards receive the VietGAP trademark Consequently, the presence of an active irrigation system is a key factor for adopting these new practices, leading to more actively irrigated VietGAP tea farms compared to conventional ones Additionally, there is a significant difference in the tea income ratio between the two farming groups.
Farmers adhering to VietGAP standards experience higher tea production incomes compared to their conventional counterparts, indicating that conventional farmers rely less on tea for their earnings The disparity between VietGAP and conventional tea farms is further highlighted by their participation in local agricultural cooperatives and investment in machinery, with VietGAP farmers demonstrating a greater commitment to these practices However, there were no statistically significant differences in factors such as gender, education, farming experience, and access to credit between the two groups.
3.3.2 Estimated result of profit frontier function
Economic efficiency is crucial for both farmers and policymakers, and the dual method was employed to assess the profit efficiency of tea farmers The analysis, detailed in Table 3.4, utilized the dummy variable “adop” to compare VietGAP and conventional farms Results indicated that the variable “Fsys” is positive and significant, revealing that tea farmers engaged in the VietGAP program achieve higher profit efficiency compared to their conventional counterparts However, this increased efficiency may stem from superior farm and farmer characteristics rather than the VietGAP program itself To further investigate this, a more detailed analysis controlling for selection bias will be conducted in the subsequent section Additionally, the significant positive effect of farm size suggests that tea producers with larger landholdings experience greater profit efficiency, aligning with findings from previous studies by Ali and Byerlee (1991), Kolawole (2006), Abdulai and Huffman (2000), and Tran and Yamagida (2015).
Table 3 4 Estimation result of profit efficiency among tea farmers
Variables Coefficient Std.dev z-stat p>|z|
Note: *** , ** , and * denote significance at the 1%, 5%, and 10% levels, respectively
Source: author’s surveyed data in 2016 (n26)
VietGAP tea production employs chemical components and pesticides, but farmers minimize their use and adhere to strict harvesting intervals while increasing organic fertilizers and biological pest control methods The resulting tea products are free from pesticide and chemical residues, certified by authorized agencies However, rising production input costs, particularly for chemical and organic fertilizers, negatively impact profit efficiency, as supported by previous studies A strong dependence on fertilizer inputs is evident, while labor costs significantly affect profit efficiency, with increases in labor prices leading to decreased profitability Additionally, costs related to hired irrigation, machinery, and processing positively influence profit efficiency, indicating that farmers utilizing machinery in production and post-harvest stages achieve greater profitability.
3.3.3 Factors explaining the profit efficiency of tea farmers
Understanding the factors influencing the profit efficiency of tea farmers is crucial for policy development Using a Tobit model, we analyzed the determinants of profit efficiency, with the profit efficiency score as the dependent variable The results, detailed in Table 3.5, reveal that access to an irrigation system significantly enhances profit efficiency, as regular irrigation minimizes yield losses, particularly during dry seasons and for high-yield tea varieties Additionally, the ratio of tea income, referred to as "Ritea," significantly affects profit efficiency; higher income from tea farming leads farmers to invest more time and resources into their farms, resulting in improved performance This finding aligns with previous research by Hong and Yabe (2015), Ali and Flinn (1989), and Wang et al (1996), indicating that tea production serves as a vital income source for farmers and their families.
Farmers who rely more on off-farm income tend to exhibit lower efficiency levels However, involvement in production cooperatives significantly enhances profit efficiency for tea farmers Participation in these groups provides farmers with better access to new information, technical training, and opportunities for experience exchange, ultimately fostering improved agricultural practices.
Table 3 5 Factors affecting profit efficiency of tea farmers
Education junior level -0.024 0.021 -1.12 0.262 senior level -0.023 0.023 -1.01 0.313 higher level -0.008 0.029 -0.29 0.773
Note: *** , ** , and * denote significance at the 1%, 5%, and 10% levels, respectively
Source: author’s surveyed data in 2016 (n26)
Joining a production group is appealing for tea farmers as it helps reduce costs related to machinery, processing, packaging, and trademark registration Technical support from extension services positively influences profit efficiency by providing essential information on disease prevention and new farming techniques However, factors such as gender, education level, household size, credit access, machinery, and experience do not significantly impact profit efficiency While higher education levels generally enhance profit efficiency, most farmers in the study area possess only basic education, which limits their farming knowledge This aligns with previous findings by Collie et al (2002), indicating that education has minimal influence on efficiency Additionally, the effects of credit access and machinery on profit efficiency are consistent across farm types, though the negative correlation with credit access is unusual and not statistically significant, as few tea farmers utilize credit from agencies.
3.3.4 Distribution of profit efficiency and average treatment effect
Frequency distribution of the profit efficiency of both farmer groups are presented in table 3.6 Most farmers (72%) operate with profit efficiency scores ranging from 0.70 to 0.89
Table 3 6 Frequency distribution of profit efficiency (PE)
Source: author’s surveyed data in 2016
The average profit efficiency score of tea farmers is approximately 74%, with a range from 29% to 94% Notably, VietGAP tea farmers demonstrate a higher mean profit efficiency of 76.4%, compared to 73.4% for conventional tea farmers This indicates a significant opportunity to enhance profit efficiency among tea farmers by 26%.
3.3.5 Propensity score for VietGAP tea adoption
Conclusions and recommendations
Tea production in northern Vietnam is primarily small-scale, with an average farmland size of 0.35 hectares, and it significantly contributes to household income in the region This study utilized a stochastic profit frontier function to analyze profit efficiency in tea production and employed a propensity score matching approach to address self-selection biases Findings revealed that tea farmers are not achieving full profit efficiency, with safe tea producers having the potential to enhance their efficiency by approximately 24%, while conventional farmers could improve by around 27% Additionally, the results indicate that transitioning to safe tea production practices leads to higher profit efficiency for farmers.
The study's findings suggest several key policy implications to enhance the profit efficiency of tea farms Supporting the development of irrigation systems and improving the operational efficiency of cooperatives can significantly boost productivity Additionally, promoting larger production scales is crucial for encouraging farmers to adopt VietGAP standards, as this allows for better utilization of machinery and production inputs Therefore, public policies should focus on promoting eco-friendly practices through innovations that address the challenges of small production scales Finally, providing suitable labor-saving machinery and enhancing extension services could serve as effective incentives for the conversion in the study area.