phân tích chuỗi giá trị và hiệu quả sản xuất của các hộ nuôi cá tra ở đồng bằng sông cửu long tt tiếng anh

MINISTRY OF EDUCATION AND TRAINING UNIVERSITY
OF ECONOMICS HO CHI MINH CITY

LE THI THANH HIEU

ANALYSIS OF THE VALUE CHAIN AND
PANGASIUS FARMING HOUSEHOLDS’
PRODUCTION EFFICIENCY IN THE MEKONG
DELTA

Major: Development
Economics
Code: 9310105
SUMMARY OF DOCTORAL
DISSERTATION IN ECONOMICS

Ho Chi Minh City – 2019


Work is completed at:
University of Economics Ho Chi Minh
City
Supervisors:

Referee

1:

…………………………………………
……………...
…………………………………………

…………………………...
Referee

2:

…………………………………………
……………...
…………………………………………
…………………………...
Referee

3:

…………………………………………
……………...
…………………………………………
…………………………...
The dissertation will be defended against
the

school

dissertation

council

at:

…………………………………………
……………..

…………………………………………
…………………………
At……..O’clock……


Day…….Month…….Year


1
ABSTRACT
This dissertation combines the value chain analysis (VCA)
with the stochastic frontier production and cost function analyses to
determine actors’advantages and gaps in operation in the pangasius
value chain, especially for pangasius farming households (PFHs) and
exporting and processing enterprises (EPEs) in order to finally
develop solutions for upgrading pangasius value chain in the
Mekong Delta (MD), through the use of SWOT matrix analysis.
Research results show that PFHs still have the ability to cut
production costs from the use of inputs to enhance production
efficiency. In addition, the research results have indicated that the
issue of using fingerlings certified as disease-free has a good and
significant impact on the PFHs’ production efficiency. Also from the
research results, the author has proposed 7 solutions to upgrade
pangasius value chain in the Mekong Delta for PFHs, and 4 solutions
for EPEs.
Key words: Pangasius, Technical efficiency,
Cost efficiency, Productive efficiency, Value
chain

1.

REASONS

FOR

CHOOSING

RESEARCH THEMES
Vietnamese Pangasius product in general and MD in
particular is one of the important products ò the fishery, as it
contributes 28.6% and 21.2% of the toatl export turnover of fishery
sector, corresponding to USD 1.754 and USD 1.785 billion in 2012
and 2017.

However, in recent years, Pangasius production and


2
export situation has become more difficult due to various subjective
and objective causes. Among the subjective causes leading to this
situation, the problem of excess use of input materials (fingerlings,
aquatic feed) of Pangasius farmers was acknowledged by many
authors from their studies (Khoi, L.N.D and Son, N.P, 2012; Khoi.
L.N.D và ctv, 2008; Vo Thi Thanh Loc, 2009; Nguyen van Thuan
and Vo Thanh Danh, 2014). In fact, to measure this issue, some
authors, often scientists in the field of technology, have used
financial efficiency analysis, or some researchers in the economic
field have used Data Envelopment Analysis (DEA) to measure and
evaluate PE of

households raising aquaculture in general and

Pangasius farming in particular. Although these methods also reflect
somewhat the PE of farming households, these methods have not yet
shown how much households can save on the use of inputs, with
available technology and input prices, still maintain a constant level
of output (in the case of using the financial analysis method of
Phuong and ctv, 2007; Nguyen Thanh Long, 2015; Pham Thi Thu
Hong et al, 2015). Also, some domestic and foreign studies do not
show the true efficiency of farming households in the context of
being affected by uncontrollable non-random factors, along with
inefficiency due to the limitations of farming techniques of farming
households (in case of using DEA method of Sharma, 1999, Kaliba
and Angle, 2004; Cinamre, 2006; Bui Le Thai Hanh, 2009; Nguyen
Phu Son, 2010; Dang Hoang Xuan Huy, 2011). Therefore, there are
other authors who have used SFA to overcome the limitations of
DEA. Although this method-based analysis has been widely used
abroad for many sectors of the economy (agriculture, fisheries,


3
industry, services, etc.), such as researches by MA Alam and ctv.
(2005); Kehar Singh (2008) Huy (2009); Nguyen Hong Phong
(2010); Kehar Singh and colleagues (2008); Onumah and Acquah
(2011) but in Vietnam the use of SFA is not very popular in the
fisheries sector, especially for the pangasius industry. Therefore, in
this study, the author uses SFA method to measure production
efficiency of farming households, as well as to assess factors
affecting technical and cost inefficiency to indicate The bottleneck in
the production of pangasius in order to eventually improve the PE of

farming households, and thus improve the profit for the whole
Pangasius VC in MD.
In addition, to discover the strengths and bottlenecks in the
VC, as well as the opportunities and challenges that actors involved
in VC have and encounter, researchers have used differential
methods such as VCA of German Technology Organization (GTZValuelinks, 2008), The Department for International Development,
UK (DFID-M4P), Food and Agriculture Organzation (FAO), etc. to
assess the impact of internal factors on activities of actors in the
chain, combined with some other qualitative analysis to assess the
impact of external factors on the activities of actors in VC, including
PEST analysis and Porter’s Five Forces Analysis. However, this
approach has only been applied relatively much abroad, such as Rui
Xu's study (2009); Kristina Al Farova (2011); Muzi (2014) and
Roman Anton (2015). In Vietnam, not many studies have used this
approach.
It is from the practical and theoritical context as presented
above, the author decided to choose the research topic "Analysis of


4
Value Chain and Production Efficiency of Pangasius farming
households in the Mekong Delta" as my doctoral thesis with the
desire to get a small contribution in theory and empirical research
related to the field of VCA and PE analysis, especially the
combination of these two analyzes into a study to both increase the
scientific content of the research, and to strengthen the scientific
basis for the proposed solutions, and thus expect to provide policy
makers in the study area, as well as for actors involved in VC,
especially households with useful information for giving making
decisions on policies and production and business acts to promote

the development of the pangasius industry of MD. Therefore, this
dissertation is conducted to achieve the objectives shown in the next
section.
2. RESEARCH OBJECTIVES
2.1. Overall objective
Proposing solutions to upgrade VC and improving pangasius
farming households’ PE in MD, through analyzing the pangasius VC
as well as measuring and assessing the factors affecting the
pangasius farming households’ PE in MD.
2.2. Specific objective
In order to achieve the above-mentioned overall objective,
this dissertation is conducted to meet the following specific
objectives: (i) Pangasius VCA in MD to detect bottlenecks and
advantages in the activities of actors involved in VC; (ii) Analysis of
PE and factors affecting the pangasius farming households’ PE in
MD; and (iii) Proposing solutions to upgrade pangasius VC and
improving PE of farmers raising Pangasius in MD.


5
3. METHODOLOGY
3.1. Data collection
The dissertation is done through the use of both types of
information, including secondary and primary information.
3.1.1. Secondary data
The secondary information used in this study includes
annual reports of the Department of Agriculture and Rural
Development (DARD), the General Department of Customs, the
Directorate of Fisheries, Vietnam Association of Seafood Exporters
and Processors (VASEP) and available scientific research reports,

related to research issues.
3.1.2. Primary data
Primary information used in this study was collected from
direct interviews with 227 households in the area of 3 provinces of
An Giang, Dong Thap, Vinh Long and Can Tho City. Households
were selected to interview according to the non-random, multi-stage
sampling method. Before conducting direct interviews with farming
households, the author conducted 4 group discussions with farming
households in the study area to get general information. In addition,
other actors in VC were also interviewed directly by the chain link
method, including 6 fingerling supply facilities, 6 agents / stores
providing aquatic feed, 7 SEPEs, 10 local scientists and managers in
the study area.
3.2. Analytical methods
Secondary and primary information was collected above to
meet the requirements for the two main analytical methods in this
study, namely VCA and PE analysis.


6
3.2.1. Value chain analysis (VCA)
This dissertation uses the main VCA tools of DFID-M4P to
assess the impact of internal factors on the activities of actors in VC.
These analysis tools include: Value chain mapping; Analyzing the
interaction between actors in VC; Analysis of horizontal and vertical
linkages of actors in VC; Upgrade VC; Risk analysis; Analysis of
cost distribution, added value and net added value (profit) of actors
in VC. Besides, this study also uses 2 analysis tools, PEST and
Porter's five forces analysis to analyze the impact of external factors
on VC.

In addition, the author has also used the method of stochastic
frontier production and cost analysis to measure and evaluate PE of
farming households. Finally, use SWOT matrix analysis to develop
solutions to upgrade pangasius VC in MD, using the results obtained
from the above analysis.
3.2.2. Production efficiency analysis
This study uses the stochastic frontier production and cost
function analysis to measure pangasius farming households’PE that
includes technical efficiency (TE) and cost efficiency (CE), and to
identify factors that affect pangasius farming households’ technical
and cost inefficiency.
3.2.2.1. Choosing the suitable stochastic
frontier production function
Through using Likelihood Ratio Test – LR test (Coelli,
1996), form of the stochastic frontier production function in this
study is determined. This form of production function may be


7
Translog or Cobb-Douglas depending on a set of available data. This
statistic test is conducted based on the following formula
LR = -2[L0 – L1]

(3.11)

If the statistic value of LR is greater than the critical value of
Chi-square distribution with the degree of fredom k (difference
between the number of independent variables used in Cobb-Douglas
and Translog models) at statistically significal level of α%, then the
hypothesis that the apropriate function form of Cobb-Douglas is

rejected and vice versa. In which L0 is LR statistic value in the case
of appropriate

Cobb-Douglas form, and

L1 for the case of

appropriate Translog form. The test results show that the appropriate
form of production function is Translog. Thus, Do vậy, the model of
Translog stochastic frontier production function look like as follows
(3.14)
In which,
yi : yield of the ith farming household;
β : regression parameter;
xni : nth input used by the ith farming household.
ui: errors due to technical inefficiencies of the ith farming
household
vi: random errors of the ith farming household.
Then, the model of Translog stochastic frontier cost function
is as follows:
++
With the subject to
αnm = αmn for all n and m (i)
(m=1,…,N)

(ii)

The subject to (i) is set up to ensure the symmetry

8
The subject to (ii) is set up to ensure the homogenity at the
level 1 for all inputs’ prices
At where,
Ci : total cost of the ith farming household
wli : price of labour paid by the ith farming household (equal
to the average price of family labour price and hired labour price)
wni : prices of inputs (aquatic feed and fingerling) used by
the ith household (n=1,2,…N).
α : regression parameter
yi : output yield of the ith farming household
ui : errors due to cost inefficiencies of the ith farming
household
vi : random errors of the ith farming household
3.2.2.2. Analysis of factors affecting
technical and cost inefficiency
In order to determine the impact of the elements of socioeconomic characteristics on the inefficiency (u i) of farmers, the
following model of inefficiency assessment is used.
ui = δ0 + δ1Z1i + δ2Z2i +……+ δhZhi (3.16)
(h is the number of variables belonging to the socioeconomic characteristics of the HSX)
At where,
ui: errors due to technical or cost inefficiency of the ith
farming households
Zhi:

variables

belonging

to

characteristics of the ith farming households
δ: regression parameter.

the

socio-economic


9
From the literature reviews, the author proposes to include
the following social and economic characteristics into the control
model of technical inefficiency:
Z1i: Education level of main farmers in the
ith farming households (number of years of
schooling)
Z2i: Number of years of experience of
main farmers in the ith farming households
i (number of years)
Z3i: Squares the number of years of
experience of the ith farming households i
(number of years)
Z4i: Rate of hired labor in the total number
of used employees (%)
Z5i: Fingerling source is certified to be
free of disease (which is equal to 1 when
the household uses a certified disease free
fingerling; 0 in the opposite case)
Z6i: Join in the input and output link (equal
to 1 when the farm is connected with the

input suppliers and either with the buyers
of raw fish products; 0 in the opposite
case)
Z7i:

Participating

in

technical

and

economic training courses (valued at 1
when the household attended technical and


10
economic training courses; 0 in the
opposite case)
Z8i: Areas for raising pangasius of the ith
farming households (1000 m2)
3.3. Analytiacal framework
Research conducted based on the
analytical framework is shown in Figure
2.2. This analytical framework shows the
impact of external environmental factors
on actors involved in VC, using PEST
analysis tools and Porter’s five forces
analysis.

In

addition,

the

analytical

framework also shows the impact of the
internal factors of VC on the activities of
actors in VC, using qualitative analysis
tools (CGT diagram; analysis of interaction
between station agents in VC, analysis of
horizontal and vertical linkages of actors in
VC, VC upgrade and risk analysis) and
quantitative

tools

(cost

distribution

analysis, value-added and net value-added
(profit) of actors in VC). The outputs of
these analyzes become the inputs of SWOT
matrix analysis. Finally, combining the
results of the SWOT matrix analysis and
the results of analysis of PE, using the

method of stochastic frontier analysis


11
(SFA) to develop solutions to upgrade VC
and improve the PE of fish farmers raising
pangasius in MD.


12


13

4. RESEARCH RESULTS
4.1. Value chain analysis
4.1.1. Value chain map
The VC map of Pangasius in MD shows that there are 6
stages, including the stages: input ,

production, collection,

processing, trade and consumption. There are 3 distribution channels
in this VC, in which the main distribution channel is the pangasius
product channel that goes directly from the production households to
the SEPEs, then the product is exported. go abroad. This distribution


14
channel accounts for 91.1% of total Pangasius production (Figure

4.1)
4.1.2. Impact of macro factors on the activities of actors involved
in VC
The analysis results obtained from PEST analysis and
Porter’s five foorces analysis show that macro factors have an impact
on the activities of the actors involved in the pangasius VC. These
impacts include:
Commercial Pangasius farms are required to have VietGap
certification and SEPEs must reduce the rate of weight gain below
20%. In the short term, this impact is considered a challenge for both
farmers and SEPEs. However, in the long term, this is considered an
opportunity by improving the quality of Vietnamese pangasius
products in the international market.
Lack of supply of market information about the quantity and
price of pangasius. This is considered a challenge for farming
households An indirect way also creates certain challenges for
SEPEs due to the shortage of raw materials.
The State does not yet have a strict quality management
mechanism for fingerling production, thus contributing to the
reduction of production efficiency of farming households due to a
lack of clean seed sources, and thus a high loss rate. Therefore, this
is considered a challenge for pangasius farming households.America
canceled Vietnam's Catfish Monitoring Program. This is considered
an opportunity for both farmers and seafood companies because it
facilitates better export of pangasius products.


15
There is no regional link & horizontal link among SEPEs.
This impact is considered a challenge for both farmers and SEPEs.

Technical and commercial barriers from pangasius importing
countries increased. As a result, these barriers become challenges for
farming households and SEPEs.
Prices of exported and material pangasius are unstable. This
price instability has created a decrease in profit for both farmers and
SEPEs. Thus, this is considered a challenge for theses two actors.
The demand for pangasius consumer markets abroad is high,
both in quantity and quality. This impact is both an opportunity and a
challenge for farming households and SEPEs.
The pangasius farming households are supported by programs and
projects of the State and non-governmental organizations on
production techniques according to safety standards such as ASC,
BMP, GlobalGap, VietGap. This is considered an opportunity for
farmers.

4.1.3. The impact of micro factors on the activities of actors
involved in VC
In addition to the impact of macro factors on the activities of
actors in the VC, there are micro factors that affect the performance
of the actors themselves involved in VC. These impacts include:
Farming households are aware of the application of farming
procedures according to safety standards. This is considered a strong
point of farming households.


16
The high experience of pangasius farming households. this is
also considered one of the strong points of pangasius farmers and
SEPEs participating in VC.
Vertical

linkages

between

farming

households

or

organizations and SEPEs have not yet sustainable, as shown by the
situation of a breach of trust that often happens between farmers and
SEPEs, especially when market prices fluctuate. , this is considered a
weakness of the actors involved in VC.
Farmers increase farming scale not based on market
planning and conditions. this is considered one of the weaknesses of
pangasius farming households.
The cooperative quality of cooperative groups and
cooperatives has not been extensive. Survey results show that
farmers only stop at sharing experiences; building canals; production
and market information. This is considered a weakness of pangasius
farming households.
Small scale of production. Through survey of farming
households, the average pond area of each household is only 0.4 ha;
the average number of family workers directly involved in pangasius
farmnig is only 2 people. This is also considered one of the
weaknesses of pangasius farming households in the study area.
The production and business level of farmers is still limited and the
awareness and knowledge of farming households in the use of inputs

is still limited. these are also considered the other weaknesses of
households raising pangasius in the study area.
SEPEs have the capacity to build their own raw material
areas and develop a form of association with farming households


17
through the form that farming households raise pangasius raw
materials for SEPEs. This is considered a strong point of SEPEs.
SEPEs have been investing in developing value-added
products from pangasius. this is considered a strong point of actors
participating in VC (including farmers and SEPEs).
Fingerling quality is low while farmers’ production
behaviors towards using fingerlings with cheap prices in order to
compensate for the death of fish, leading to an average loss of 23%.
Therefore, this is considered one of the challenges for farming
households.
Input prices tend to increase. Through surveying 227 farming
households, all farming households said that although the price of
raw fish products fluctuated strongly (at increase and decrease), the
prices of most inputs were variable. moving in the direction of
increase. As a result, this

affects the profitability of farmers.

Therefore, this is considered a challenge for farmers.


18

19

4.2. Pangasius farming households’ production efficiency
4.2.1. Measure and analyze pangasius farrming households’
production efficiency
the pangasius farming households’ TE and CE are measured
based on the formulas of 3.14 và 3.15. The measured results of TE
are presented in Table 5.3.
The estimation of TE coefficient of farming households, using model
3.14, is presented in Table 5.3, showing that the average TE of
farming households reached 80.6% with a standard deviation of
20.4%. That is, farmers can simultaneously reduce 19.4% of all
inputs of labor, fingerlings and aquatic feed, but still maintain a
constant level of production. This shows that farmers are still
technically limited, especially in the use of combination of inputs. In


20
other words, for pangasius farmers in MD, there is still an
opportunity to improve PE through the reduction of production costs.
Table 5.3. Allocating frequency of TE coefficients of pangasius
farming households
Eficiency
TE
coefficient (%)
No of households
Rate (%)
<= 50
32

14
51-78
34
15
79-90
38
16
=> 90
123
55
Total
227
100
Mean
80,6
Minimum
28,1
Maximum
97,6
Standard Deviation
20,4
For pangasius farming, according to industry experts, 19.4%
reduction in input, especially for aquatic feed, has a very big
financial significance because investments for pangasius farming are
very high (about 5-6 billion VND / ha / crop). The results of this
study are similar to many research results of other authors in the
fisheries sector. For example, the study results of Huy (2009) and
Phong (2010). In general, from the research results as mentioned
above, the farmers in MD have opportunities to improve their PE
through reducing the amount of inputs. In addition, the reduction of

production costs will contribute to stabilizing the input materials for
SEPEs, and thus contributing to stabilizing the supply of pangasius
fillets for export. Therefore, it will also make the income of farming
households more stable. In addition, having good competitive prices
will create opportunities for the linkage between farmers and SEPEs
to become more sustainable.


21
The results of TE estimation as analyzed once again confirm
that farmers should reduce stocking density and thus reduce the
amount of aquatic feed to improve TE. In addition, the data in Table
5.3 also show that the lowest and highest TE achieved between farms
is very high, indicating that the farming techniques are not uniform
among households. Moreover, the analysis also showed that nearly
30% of households reached TE below the mean TE. This shows that
the farming techniques of farming households are still limited.
The results in Table 5.8 show that the mean CE of farming
households is 78.1% with a fluctuation of 21.5%, meaning that
farmers can cut the cost of using inputs by 21.9%, but still maintain a
constant output. The results of these households are relatively high,
however, the CE difference between households is quite high,
indicating that the production level among households is not equal.
Up to 34% of farmers achieved CE below the mean CE.
Table 5.8. Allocating frequency of cost efficiency coefficients of
pangasius farming households
Efficiency
CE
coefficiency (%)
No of households

Rate (%)
<= 50
34
15
51-78
44
19
79-90
35
16
=> 90
114
50
Total
227
100
Mean
78,1
Minimum
17,9
Maximum
97,3
Standard Deviation
21,5
In short, the research results show that farmers can improve
their production level to increase their production efficiency. Like the


22
above analysis, this reduction in production costs has a great

significance to the survival and development of the pangasius
industry in MD in general and of pangasius farming households in
particular.
4.2.2. Analyze factors affecting the production efficiency of
farming households
Through the literature review as well as the survey process,
it is recognized that, in addition to the factors related to the use of
production factors that affect the TE of farming households, factors
that belong to socio-economic characteristics of farmers also has a
certain influence on their TE. The impact of these factors on the TE
of the farmersis presented in Table 5.5.

Table 5.5: The results of regression analysis on the effect of socioeconomic variables of farm households to technical inefficiencies
Sign
Name
of Coefficien St.Dev P>[Z]
variables
t
Z1
Education level
-0.0187 0.0806 0.816
Z2
Experience
-0.0311 0.0526 0.554
Z3
Squares
the
0.0004 0.0019 0.838
number of years
of experience

Z4
Rate of hired
-0.0128 0.0051 0.012**
labour in the total
number of used
employees
Z5
Fingerling source
-4.2325 1.9636 0.031**
is certified to be
free of disease