Impact of Fisheries Management Objective on Technical

Efficiency: Case Studies in Fisheries

Quang Van Nguyen

Submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

School of Economics and Finance

Queensland University of Technology

Principal supervisor:
Associate Professor Louisa Coglan
Associate supervisors:
Adjunct Prof. Sean Pascoe
Prof. Stephen Cox

2019
Keywords

o Australia
o Economic objective
o Fisheries management objective
o Maximum economic yield (MEY)
o Northern Prawn Fishery (NPF)
o Open access
o Social objective
o Stochastic frontier analysis (SFA)
o Technical efficiency (TE)
o Trawl fisheries
o Vietnam

i
Abstract

Fisheries management generally aims at multiple objectives, primarily including economic,

biological, social and political objectives. However, the importance or weight of each objective
is often different across each fishery both within countries and across different countries. The
objective of fisheries management creates incentives and disincentives for fishers operating
within a fishery. It is reasonable to assume that where fisheries management focuses on a strong
economic objective (i.e., Maximum Economic Yield (MEY)) this would result in a smaller
fleet, exhibiting a high level of technical efficiency, generating high boat incomes. In
comparison, the a priori assumption would be that a fishery managed under a strong social
objective would have higher boat number, with greater heterogeneity but exhibit a lower
average technical efficiency and generate lower boat income. Research on evaluating the
impact of fisheries management changes on technical efficiency of fishing vessels has attracted
fisheries economics scholars worldwide over the past years. However, the potential impact of
fisheries management objective on the technical efficiency in fisheries has received less
attention. The primary aim of this thesis is to understand the impact of differing fisheries
management objectives and potential trade off with respect to technical efficiency.

Four studies that are relevant to the main objective of the dissertation are undertaken. First, a
meta-analysis study in fisheries efficiency is conducted to identify factors affecting the level
of technical efficiency. The results reveal that fisheries managed under individual transferable
quotas (ITQs) have higher efficiency than fisheries managed under open-access regime and
input controls. By contrast, the findings also show that fisheries managed under open-access
regime yield lower mean technical efficiency than fisheries managed under input controls.
Furthermore, other factors including efficiency estimation methods used in primary study, the
number of inputs and outputs included in efficiency model, and the ranking of journal where a
primary study was published on are also found to be factors affecting efficiency. The second
study is to examine the sensitivity of technical efficiency to the choice of input variables
included in fishing efficiency, and to efficiency model specifications. The results show that
different measures of inputs and efficiency model specifications result in different levels of
technical efficiency. Efficiency scores have a lower mean and higher standard deviation as the
measure of effort becomes less precise. However, the absolute bias due to input measures is
small. The third study focuses on technical efficiency of fishing fleet managed under a strong
economic objective. The banana prawn fishery, a sub-fishery of the Northern Prawn Fishery in

ii
Australia, is managed to achieve maximum economic yield. The empirical findings indicate
that while technical efficiency varies between vessels, average efficiency is relatively constant
over the fishing season. However, average productivity of the fleet increases, with smaller
vessels (in terms of engine power) exiting the fishery earlier. Whilst, even with an economic
target in fisheries management, a wide distribution of efficiency scores in the fishing fleet is
observed, the average technical efficiency score of the fishing fleet is high. The last study
focuses on evaluating technical efficiency and its determinants of trawlers operating on open-
access trawl fisheries in Vietnam, which has a strong social objective in management where
employment and income of fishers are prioritised over economic objectives. The study finds
that the surveyed trawlers have potential room to increase their technical efficiency. Other
factors, including fisher’s age and experience, the age of the vessel, distance from fishing
ground, engine type and use of family members in the crew all affect the technical efficiency
score. Significant differences are also seen between regions, with technical efficiency
seemingly affected by the opportunity cost of labour. Poorer regions with fewer alternative
employment opportunities are associated with lower average technical efficiency, reflecting
the effects of greater capital and labour non-malleability in these areas. Policy implications for
fisheries management drawn from these studies are also discussed in the dissertation.

iii
Table of contents

Keywords .................................................................................................................................... i
Abstract ...................................................................................................................................... ii
Table of contents ....................................................................................................................... iv
List of figures .......................................................................................................................... viii
List of tables ............................................................................................................................... x
List of abbreviations ................................................................................................................ xii
Statement of original authorship ............................................................................................. xiii
List of publications and manuscripts ...................................................................................... xiv
Acknowledgements .................................................................................................................. xv
Chapter 1. Introduction .......................................................................................................... 1
1.1. Brief introduction of research topic .................................................................................... 1
1.2. Research hypothesis and questions ..................................................................................... 5
1.3. Research activities .............................................................................................................. 6
Research process ................................................................................................................ 8

1.4. Main contributions ............................................................................................................ 15

1.5. Structure of the thesis........................................................................................................ 16
Chapter 2. Fisheries management objectives ...................................................................... 20
2.1. Multiple objectives in fisheries management ................................................................... 20
2.1.1. Biological objective ............................................................................................... 20

2.1.2. Ecological objective ............................................................................................... 22

2.1.3. Economic objective ................................................................................................ 23

2.1.4. Social objective ...................................................................................................... 25

2.1.5. Political objective................................................................................................... 26

2.2. Trade-off between fisheries management objectives ........................................................ 27

2.3. Management objective and technical efficiency distribution ........................................... 31
Chapter 3. Technical efficiency measurement in fisheries and its relevant issues........... 37
3.1. Technical efficiency measurement ................................................................................... 37
3.2. Efficiency measurement approaches in fisheries .............................................................. 40

iv
3.3. Functional form of the stochastic production function in fisheries .................................. 45
3.4. Input and output measures used in fisheries efficiency literature ..................................... 48
3.5. Issues of theoretical consistency ....................................................................................... 53
Chapter 4. Meta-analysis of fisheries efficiency .................................................................. 58
4.1. Introduction ....................................................................................................................... 58
4.2. An overview of meta-analysis in efficiency ..................................................................... 61
4.3. Methods............................................................................................................................. 64
4.3.1. Meta-dataset ........................................................................................................... 64

4.3.2. Meta-regression model........................................................................................... 65

4.4. Results ............................................................................................................................... 67

4.4.1. Overview of the identified papers .......................................................................... 67

4.4.2. Meta-regression results .......................................................................................... 72

4.5. Discussion and conclusion ................................................................................................ 76

Chapter 5. Sensitivity of technical efficiency to the choice of input variables ................. 78
5.1. Introduction ....................................................................................................................... 78
5.2. Data ................................................................................................................................... 80
5. 3. Methods............................................................................................................................ 81
5.3.1. Stochastic frontier analysis .................................................................................... 82

5.3.2. Regression analysis model ..................................................................................... 85

5.4. Results and discussion ...................................................................................................... 85

5.4.1. Model specification tests........................................................................................ 85

5.4.2. Regression results .................................................................................................. 91

5.5. Conclusion ........................................................................................................................ 95

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery .................... 97
6.1. Introduction ....................................................................................................................... 97
6.2. The Northern Prawn Fishery............................................................................................. 97
6.2.1. General information on the fishery ........................................................................ 97

6.2.2. Northern Prawn Fishery Management ................................................................. 100

6.2.3. Fishery management objective in NPF ................................................................ 104

v
6.3. Current economic status of the fishery ........................................................................... 106
6.4. Previous efficiency studies in NPF ................................................................................. 108
6.5. Technical efficiency of banana prawn fishery ................................................................ 110
6.5.1. The rationale of study .......................................................................................... 110

6.5.2. The banana prawn MEY trigger .......................................................................... 111

6.5.3. Methods................................................................................................................ 113

6.5.4. Data ...................................................................................................................... 115

6.5.5. Results .................................................................................................................. 115

6.5.6. Discussion ............................................................................................................ 123

6.5.7. Conclusions .......................................................................................................... 125

Chapter 7. Social focused case study fishery: An open-access trawl fishery in

Vietnam ................................................................................................................................. 126
7.1. Introduction ..................................................................................................................... 126
7.2. Marine capture fisheries in Vietnam ............................................................................... 126
7.2.1. Overview of fisheries sector ................................................................................ 126

7.2.2. Fisheries management objective .......................................................................... 130

7.2.3. Trawl fisheries information.................................................................................. 132

7.3. Field survey..................................................................................................................... 135

7.3.1. Trawl fishery and regional conditions in study areas .......................................... 135

7.3.2. Data collection procedure .................................................................................... 139

7.4. Current socio-economic status of fishers (Activity 4) .................................................... 140

7.4.1. Introduction .......................................................................................................... 140

7.4.2. Socio-economic profile of trawlers ...................................................................... 140

7.4.3. Discussion ............................................................................................................ 145

7.4.4. Conclusions .......................................................................................................... 148

7.5. Technical efficiency of trawl fishery (Activity 5) .......................................................... 148

7.5.1. Introduction .......................................................................................................... 148

7.5.2. Methods................................................................................................................ 150

vi
 7.5.3. Data ...................................................................................................................... 150

7.5.4. Results .................................................................................................................. 154

7.5.5. Discussion and conclusions ................................................................................. 161

Chapter 8. Discussion and conclusion ................................................................................ 165

8.1. Introduction ..................................................................................................................... 165
8.2. Hypothesis, objective and research aims ........................................................................ 166
8.3. Key research results ........................................................................................................ 167
8.3.1. Meta-analysis fisheries efficiency........................................................................ 167

8.3.2. Sensitivity of the technical efficiency to the choice of input variables ............... 168

8.3.3. Economic objective focused fishery case study: Northern Prawn Fishery .......... 169

8.3.4. Social objective focused fishery study: An open-access trawl fishery in

Vietnam .......................................................................................................................... 171

8.4. Limitations and possible directions for further research................................................. 172

8.5. Conclusion ...................................................................................................................... 173
References ............................................................................................................................. 176
Appendices ............................................................................................................................. 201
Appendix 1: Human research Ethics approval certificate.............................................. 201

Appendix 2: Questionnaire for trawl fishery survey in Vietnam ................................... 202

Appendix 3: Fishing zone in Vietnam’s sea .................................................................. 211

Appendix 4: List of primary studies used in meta-analysis ........................................... 212

Appendix 5: Net economic return of the NPF from 2000/01 to 2014/15 ...................... 218

vii
List of figures

Figure 1.1. Thesis structure ...................................................................................................... 19

Figure 2.1. Maximum economic yield and maximum sustainable yield ................................. 25

Figure 2.2. Relationship between management objectives and management instrument........ 32

Figure 2.3. Distribution of technical efficiency under different fisheries management

objectives ................................................................................................................................. 36

Figure 3.1. Input-orientated measures of technical efficiency................................................. 37

Figure 3.2. Output-orientated measures of technical efficiency .............................................. 39

Figure 3.3. Illustration of a non-monotone production frontier ............................................... 54

Figure 4.1. The number of efficiency papers published in each journal.................................. 67

Figure 4.2. The number of efficiency papers undertaken in each country .............................. 68

Figure 4.3. The number of papers from 1995 to 2018 ............................................................. 69

Figure 4.4. The application of efficiency estimation approach over the past years................. 70

Figure 4.5. The number of papers and efficiency estimation approaches used by continent .. 71

Figure 4.6. The proportion of application of fisheries management tools ............................... 71

Figure 6.1. Area of the Northern Prawn Fishery ..................................................................... 98

Figure 6.2. Catch and fishing vessel for the Northern Prawn Fishery ..................................... 99

Figure 6.3. Net economic returns of the NPF over the past years ......................................... 107

Figure 6.4. Distribution of final technical efficiency scores .................................................. 117

Figure 6.5. Average technical efficiency over the fishing seasons in each year ................... 118

Figure 6.6. Number of boats fishing each week over the fishing season ............................... 118

Figure 6.7. Average hours trawled per day and engine power .............................................. 119

Figure 6.8. Stock index over the fishing season .................................................................... 120

Figure 6.9. Engine power a) production elasticity and b) productivity ................................. 120

Figure 6.10. Average catch per day from weeks 8 to 12 over the fishing seasons (The blue line
represents the 500 kg catch) ................................................................................................... 121

viii
Figure 6.11. Derived proxy cost index (i.e., the combined cost index) ................................. 122

Figure 6.12. Target catch rates factoring in the higher costs towards the end of the season (based
on 2015 price and cost parameters) ....................................................................................... 123

Figure 6.13. Target catch rates for a range of prawn prices based on 2015 cost parameters and
higher costs assumptions towards the end of the season ....................................................... 123

Figure 7.1. Total catch and CPUE over the years .................................................................. 130

Figure 7.2. The number of trawl fishing vessels in 2010 and 2017 ....................................... 133

Figure 7.3. Total catch volume of marine fisheries in 2017 by fishing gear ......................... 134

Figure 7.4. Geographical location of the surveyed areas ....................................................... 136

Figure 7.5. The proportion of males and females directly engaged in fishing ...................... 141

Figure 7.6. Catch distribution channel in Quang Ninh .......................................................... 144

Figure 7.7. Correlation among variables in the model........................................................... 154

Figure 7.8. Distribution of technical efficiency in each vessel group ................................... 159

ix
List of tables

Table 1.1. Main characteristics of the two case study fisheries ................................................. 8

Table 1.2. Summary of research activities, methodology and data used ................................. 14

Table 2.1. The potential relationship between fisheries management objective and technical
efficiency.................................................................................................................................. 33

Table 4.1. A summary of meta-analysis studies in efficiency ................................................. 63

Table 4.2. Definition of variables in the meta-regression model ............................................. 65

Table 4.3. The number of papers undertaken in each continent and country .......................... 69

Table 4.4. Summary of variables in the meta-regression model.............................................. 72

Table 4.5. Meta-regression of mean technical efficiency in fisheries ..................................... 73

Table 5.1. Weekly data summary, banana prawn fishery 2010-2015 ...................................... 80

Table 5.2. Correlation between key variables (logged) ........................................................... 81

Table 5.3. Taxonomy of the stochastic frontier models........................................................... 84

Table 5.4. Variables used in the regression model .................................................................. 85

Table 5.5. Tests of functional form (Translog versus Cobb-Douglas) .................................... 87

Table 5.6. Tests of time-invariant efficiency (η = 0) versus time-variant (η ≠ 0) efficiency under

βkj ≠ 0 ...................................................................................................................................... 88

Table 5.7. Tests of truncated normal (μ ≠ 0) versus half-normal (μ = 0) distribution, (βkj ≠ 0

and η ≠ 0) ................................................................................................................................. 89

Table 5.8. Technical efficiency scores under different models ............................................... 90

Table 5.9. Regression analysis results ..................................................................................... 95

Table 6.1. Management measures and objective in the NPF ................................................. 102

Table 6.2. Net economic return of the fishery ....................................................................... 106

Table 6.3. Initial and final parameter estimates ..................................................................... 116

Table 7.1. The total fishing boats in 2001 and 2017 regarding engine size (boats) .............. 128

Table 7.2. The number of fishing boats in 2001 and 2017 regarding fishing gear (boats) .... 129

x
Table 7.3. Fisheries management objective ........................................................................... 132

Table 7.4. Geo-economics and socio-economic characteristics in the two provinces ........... 138

Table 7.5. Socio-demographic characteristics of vessel owners............................................ 142

Table 7.6. Information about fishing household generation .................................................. 142

Table 7.7. Education level of fishers ..................................................................................... 143

Table 7.8. Livelihood activity of fishing household .............................................................. 144

Table 7.9. Definition of variables in the analysis .................................................................. 152

Table 7.10. Summary of the data variables used in the analysis ........................................... 153

Table 7.11. Initial and final model parameter estimates ........................................................ 157

Table 7.12. Summary of technical efficiency ........................................................................ 158

Table 7.13. Comparison of mean technical efficiency between different vessel groups ....... 158

xi
List of abbreviations

ABARES Australian Bureau of Agricultural and Resource Economics and Sciences

AFMA Australian Fisheries Management Authority
BRDs Bycatch Reduction Devices
CPUE Catch Per Unit of Effort
CSIRO Commonwealth Scientific and Industrial Research
CV Cheval-vapeur
DAFF Department of Agriculture Fisheries and Forestry
DEA Data Envelopment Analysis
DMU Decision-making unit
EEZ Exclusive Economic Zone
FAO Food and Agriculture Organization of the United Nations
HP Horse power
ITQs Individual transferable quotas
ITEs Individual transferable gears
MEY Maximum Economic Yield
MSY Maximum sustainable yield
NER Net Economic Return
NPF Northern Prawn Fishery
OA Open Access
OECD Organization for Economic Co-operation and Development
OLS Ordinary Least Square
SD Standard deviation
SDF Multiple output stochastic distance function
SFA Stochastic Frontier Analysis
SPF Stochastic Production Frontier
TAC Total allowable catch
TE Technical efficiency
TEDs Turtle excluder devices

xii
 Statement of original authorship

The work contained in this thesis has not been previously submitted to meet requirements

for an award at this or any other higher education institution. To the best of my knowledge and

belief, the thesis contains no material previously published or written by another person except

where due reference is made.

Signature: QUT Verified Signature

Date: July 2019

xiii
List of publications and manuscripts

Published

 Pascoe, S., Hutton, T., Coglan, L., & Nguyen, V. Q. (2017). Implications of efficiency and
productivity change over the season for setting MEY‐based trigger targets. Australian
Journal of Agricultural and Resource Economics, 59, 1-18 (Chapter 6).
 Nguyen V. Q., Pascoe, S., & Coglan, L. (2019). Implications of regional economic
conditions on the distribution of technical efficiency: Examples from coastal trawl vessels
in Vietnam. Marine Policy, 102, 51-60 (Chapter 7).

In submission

 Nguyen V. Q., Pascoe S. & Coglan L. Impact of quasi-property rights on fisheries technical
efficiency: A meta-regression analysis (Submitted to Marine Resource Economics)
(Chapter 4).
 Nguyen V. Q., Pascoe S. & Coglan L. Sensitivity of efficiency measurement in fisheries to
choice of input variables. (Submitted to Journal of Productivity Analysis) (Chapter 5).

xiv
Acknowledgements

My doctoral thesis would not have been possible without great support and assistance from
various individual and institutions. First and foremost, I would like to express my deepest
gratitude to my supervisors: Associate Professor Louisa Coglan and Dr Sean Pascoe for their
continuous support, encouragement, guidance and patience throughout my PhD journey.
Without their supervision, my dissertation would not have been possible. I have been lucky to
undertake my doctoral study under my supervisors, Louisa and Sean. They always made
themselves available when I needed their help and instructions.

I also would like to thank Dr Stephen Cox for being my Associate supervisor and for valuable
advice. I am very thankful to Dr Vincent Hoang, who introduced me to Sean and Louisa, and
Dr Nghiem Hong Son for his valuable comments and suggestions on my thesis.

This research would not have been possible without QUT-VIED Joint Scholarship. I would
like to express my great appreciation to the Ministry of Education and Training of Vietnam
(MOET) and Queensland University of Technology (QUT). This work also could not have
happened without valuable data sources and cutting-edge facilities provided by CSIRO.

I would like to acknowledge all surveyed fishers and local officers who helped me to collect
data during my fieldwork in Quang Ninh and Ben Tre provinces. A special thanks to Ha Van
Giang and Pham Ba Vu Tung for their generous support and assistance during my survey. I
also acknowledge Hong, The Anh and Binh, for their assistance and for introducing me to local
governments in Quang Ninh and Ben Tre during my data collection.

My sincere thanks go to professional editor, Jane Todd, who provided copy-editing and proof
reading services, according to the guidelines laid out in the university-endorsed national
“Guidelines for editing research theses”.

I express my very sincere thanks to staff, colleagues, friends and lecturers at School of Business
and Finance, Queensland University of Technology for all their support, assistance and
friendship during my doctoral study. Thank you Sam, Stewart, Ammar, Amar, and my
Vietnamese friends at QUT, for their friendship, sharing study and life experience. Special
thanks go to Le Lee for her unconditional support and encouragement throughout my study,

xv
and Dinh Thi Thanh Nhan and Pham Hoang Anh for their sharing valuable life, work and daily
experiences that broaden my horizon.

Importantly, I would like to thank my parents, siblings Dao, Truong and Ngot, and my lovely
nieces: Trang, Linh and Quỳnh for their endless encouragement, support and sacrifice that they
have extended to me. This dissertation is dedicated to my whole family.

xvi
 PART 1:
INTRODUCTION AND LITERATURE REVIEW
Chapter 1. Introduction

The first three sections of this chapter present the research topic, research hypothesis and
research objective (including specific research activity). The next section outlines the main
contribution of the thesis. Finally, the last section presents an outline of the remaining chapters
and thesis structure.

1.1. Brief introduction of research topic

It is undeniable that the fisheries sector plays a prominent role for many countries worldwide,
serving as a contributor to economic growth and development, source of wealth and income,
human wellbeing, food security, employment, cultural identity and livelihood sources. In 2014,
total capture fisheries were 93.4 million tonnes with a value about US$ 148 billion (FAO,
2016). There were about 40 million people engaged in capture fisheries worldwide in 2014
(FAO, 2016). Seafood consumption accounts for an estimated 17% of total animal protein of
the global population’s intake, and about 6.7% of all protein consumed (FAO, 2016). The
fisheries sector also supports the livelihood of hundreds of millions of people across the globe
(FAO, 2014), and it is reported that marine capture fisheries provide direct/indirect job
opportunities for about 200 million people (both men and women) (Sumaila & Munro, 2009).
In addition to marine capture fisheries being a source for economic development, food,
employment and human-wellbeing, they also provide cultural services (e.g., creational and
indigenous fishing) in many cultures (Coulthard, Johnson, & McGregor, 2011; Stoeckl et al.,
2017).

However, marine fish stocks have undergone a substantial decline worldwide (Chu, 2009).
According to FAO (2014), about 28.8% of fish stocks were assessed as being fished to a
biologically unsustainable level, and 61.3% were fully exploited in 2011. Only 9.9% of fish
stocks were under-fished. Most of the stocks of the top 10 species, which made up about 24%
of the total production of marine capture fisheries in 2011, were fully exploited and thus, have
no room to increase production. A number of stocks are overfished and increase in production
is only possible under effective fisheries management (FAO, 2012, 2014). Effective fisheries
management is therefore necessary to maintain the sustainable use of fisheries resources
(Gréboval, 2002; Hauge, Cleeland, & Wilson, 2009).

Chapter 1. Introduction 1
Fisheries management is generally characterised by multiple objectives that include
environmental (i.e., biological objectives), economic, social and political objectives (Cochrane,
2000; Hilborn, 2007a; Mardle et al., 2002). Fisheries that managed to achieve economic
objectives attempt to maximise incomes gained from the fishery after covering all costs of
fishing, referred to as Maximum Economic Yield ( MEY) (Kempf et al., 2016; Kompas, 2005).
In contrast, fisheries that are managed to achieve social objectives generally attempt to raise
employment opportunities, livelihoods and incomes for fishers and fishing communities
(Hilborn, 2007a). Fisheries that managed to achieve a biological objective, seek to maintain
the fish stocks at or above the level that is necessary to ensure their continuous productivity
(FAO, 2002), commonly referring to traditional maximum sustainable yield (MSY) (Hilborn,
2007a; Stouten, Heene, Gellynck, & Polet, 2011). Although fisheries management aims at
multiple objectives, sustainable fisheries are the overriding goal of management.

The relative importance of each objective (social, economic and biological objectives) in the
management of a fishery differs across fisheries both within and across countries. Some
strongly focus on economic objectives in fisheries management in which MEY is a reference
point while others are primarily driven by social objectives. An example of an economic
objective in fisheries management are fisheries managed in developed countries. For example,
Australian commonwealth-managed fisheries are managed to achieve economic objectives
(MEY).

Harvest strategies for key commercial stocks taken in Australia’s Commonwealth

fisheries will be designed to pursue maximum economic yield from the fishery and
ensure those stocks remain above levels at which the risk to the stock is
unacceptably high (DAFF, 2007).

In contrast, in undeveloped countries and developing countries, where millions of people rely
on fishing for a living and as a food source, and where alternative livelihoods for fishers are
limited, fisheries management is defined by social objectives (e.g., employment, livelihood and
wellbeing of fishing communities) (Symes & Phillipson, 2009). Clearly, the importance of each
objective in fisheries management differs from fishery to fishery.

The nature of fisheries management is to use tools available by fisheries managers to control
the activities of fishers as indicated in Hilborn (2007b) that “managing fisheries is managing
people”. The available management tools might include a diversity of types such as input and

Chapter 1. Introduction 2
output controls, open-access regime, and individual transferable quotas (ITQs), total allowable
catch, fishing season and area closures, and among many others. In reality, some fisheries are
managed under ITQs regimes while others are managed under either open-access regime or a
combination of input and/or output controls.

The choice of tool is primarily driven by management objective such as aiming at economic
objective (MEY) or social objective (open-access regime can be seen as proxy). The different
management tools applied to achieve different objectives afford different levels of quasi-
property or use rights to the fishers. This, in turn, affects the behaviour and performance of
fishers and thereby the structure of the fishing fleet. For instance, the use of tradeable use rights,
such as ITQs, has been widely advocated to improve economic performance and profitability
in fisheries (Chu, 2009). It is generally believed the application of ITQs would increase average
efficiency of the remaining vessels (Costello & Deacon, 2007; Pascoe, Coglan, Punt, &
Dichmont, 2012), with the least efficient vessels generally leaving the fishery (see e.g.,
Dresdner, Campos, & Chávez, 2010; Kompas & Che, 2005; Sharp, Castilla-Espino, & García
del Hoyo, 2004). As a result, the fishing fleet operating in ITQ fisheries consists of relatively
homogenous fishing boats regarding economic performance and efficiency.

In contrast, open-access fishery, where fishers are free to enter and exit, aligns to meeting social
objectives (e.g., employment, livelihood and income for fishing community) as it results in
more job opportunities and livelihoods for participants. Under an open-access regime, fishing
fleets are likely to involve a mix of heterogeneous vessels, consisting of commercial-oriented
fishers and livelihood-based and/or subsistence-based fishers. Under open-access regime, less
efficient fishers often have no incentives to exit the fishery as schemes such as decommission
grants, buyout behaviour practice, and consolidation of fishing effort are to be limited. This
would be reflected in economic performance and efficiency measures, with a large variation
between the relatively efficient and relatively less efficient vessels, leading to lower efficiency
on average. This is a result of newer, more efficient vessels entering the fishery without older
less efficient vessels leaving (other than eventually being forced out through economic losses)
(Holland, Gudmundsson, & Gates, 1999).

Technical efficiency (TE), originally defined by Farrell (1957), indicates the ability of a fishing
vessel to achieve maximal output without an increase in inputs used (i.e., output orientation
measure) while the former refers to the ability of a fishing vessel to minimise inputs used to
produce a given amount of output (i.e., input orientation measure). Over the past years, there

Chapter 1. Introduction 3
has been increased attention to technical efficiency measurement in fisheries. This is because
information about technical efficiency, which indicates the relationship between the amount of
inputs used (e.g., fishing effort) in harvesting and the resultant catch is essential for effective
management (García del Hoyo, Castilla Espino, & Jiménez Toribio, 2004; Pascoe, Andersen,
& de Wilde, 2001). Information on the efficiency level and its drivers have important
implications for both fishers and fisheries policy makers. For instance, fishers can improve
their performance and earnings by improving their efficiency while policymakers could use
this knowledge to formulate appropriate and effective regulations, interventions and measures
for sustainable fisheries management (García del Hoyo et al., 2004; Sharma & Leung, 1998;
Tingley, Pascoe, & Coglan, 2005).

There have been a number of studies that examine the impact of fisheries management on
efficiency within a fishery. These studies have focused on investigating the impacts of fisheries
management changes on technical efficiency, including management measures (e.g., input
controls) (Greenville, Hartmann, & MacAulay, 2006; Kompas, Che, & Grafton, 2004; Pascoe
et al., 2001); ITQs (Grafton, Squires, & Fox, 2000; Kompas & Che, 2005; Walden, Kirkley,
Färe, & Logan, 2012); fishing capacity reduction programs (e.g., buyback programs) (Fox,
Grafton, Kompas, & Che, 2006; Pascoe et al., 2012; Quijano, Salas, Monroy-García, &
Velázquez-Abunader, 2018); collective right-based management (Chávez, Quiroga, &
Dresdner, 2017; Huang, Ray, Segerson, & Walden, 2018), among others. In addition, several
empirical studies have attempted to examine the trade-off between achieving different
objectives (e.g., economic, employment and income, conservation and economic objectives) in
fisheries management. These studies focused on evaluating the trade-off between regional
employment, regional income and economic rent of fisheries (Heen, Heen, & Leung, 2014;
Leung, Heen, & Bardarson, 2001), between conservation and socio-economic objectives
(Cheung & Sumaila, 2008), and the trade-off among conservation, profitability and food
security objectives (Hilborn, Stewart, Branch, & Jensen, 2012).

However, the existing literature provides limited understanding about the association between
fisheries management objectives and technical efficiency level. Whilst it is reasonable to
assume the implication on technical efficiency under different social and economic objectives
(e.g., we would a priori expect the technical efficiency of the vessel to be higher under an
economic objective compared with a social objective) this has not been empirically tested and
the degree of the impact on technical efficiency given the objectives is not known. The primary

Chapter 1. Introduction 4
aim of this thesis to address this gap. Knowledge about the impact of management objectives
on technical efficiency and its distribution would be useful for designing and proposing
fisheries management strategies and policies. This information implies the trade-off between
pursuing different objectives in fisheries management that would be very useful to decision
makers as they also indicate the relative costs of chasing one objective instead of others
(Cheung & Sumaila, 2008; Leung et al., 2001).

1.2. Research hypothesis and questions

As previously mentioned, different fisheries often aim at different objectives in fisheries

management and the relative importance of each objective also differs. The different
management systems applied to achieve different objectives afford different levels of quasi-
property or use rights to the fishers. This, in turn, affects the behaviour of fishers and thereby
the structure of the fishing fleet. For example, fisheries management with a strong economic
objective would involve management instruments that encourage autonomous adjustment in
the fleet, enabling more efficient vessels to buy-out less efficient vessels and less efficient
vessels to exit the fisheries. In contrast, fisheries that have a strong social objective, such as
maximising regional employment or maintaining livelihoods in areas with few alternatives,
would use management instruments (or no instruments at all) that limit the incentives for less
efficient fishers to adjust out of the fishery. Consequently, we would expect to see a wider
distribution of technical efficiency in a fishery with strongly social objectives, and with a lower
average efficiency score as a result. In contrast, a fishery managed with a strong economic
objective would be expected to have a narrower distribution (i.e., few vessels with low
efficiency scores) and a higher mean efficiency score as a result.

Given this, the key broad hypothesis to be addressed in this thesis is:

Fisheries management that has an economic focus results in a more

technically efficient fleet than management that has a less economic
focus.

Assessing this hypothesis is complicated, however, for several reasons. First, most fisheries
management systems have multiple objectives, and hence technical efficiency may be
influenced by the relative importance of these objectives. Second, in most fisheries these
objectives are poorly defined, and their relative importance is also poorly defined. As a result,

Chapter 1. Introduction 5
identifying these objectives and their relative importance is often not possible without
additional research to explicitly determine these. Third, technical efficiency is a relative
measure that is estimated within a fishery. Hence, a vessel identified as efficient in one fishery
(and subsequently having a high efficiency score) may potentially be less efficient in an
absolute sense than a vessel in another fishery that is relatively inefficient in that fishery (and
hence has a lower relative efficiency score). Fourth, fisheries management changes the broader
economic environment in which the fishers operate, and in doing so influence the technical
efficiency of the fleet. However, other factors external to the fishery also influence the broader
operating environment, changing the incentives facing the industry and in doing so, potentially
influencing the overall technical efficiency of the fleet.

Given these challenges in comparing technical efficiency under different fisheries objectives,
the broad hypothesis is indirectly examined through a series of research questions:

1. Assuming that management instruments used in a fishery are related to the relative
importance of the different management objectives, can we see a relationship between
management instrument applied and technical efficiency from the existing literature? and;
2. What other factors unrelated to the vessels themselves and management objectives may
influence the estimate of technical efficiency in fisheries, obfuscating this relationship?
a. Structural form of the model used in the estimation process?
b. The broader economic environment in which the fishery operates, in particular
alternative opportunities for the vessels?

These research questions are applied to a meta-analysis of existing technical efficiency studies
as well as primary analysis of two case study fisheries as detailed in the following sections.

1.3. Research activities

As noted above, the primary objective of this dissertation is to explore the relationship between
fisheries management objectives and the level of technical efficiency, and in particular the
potential role that an economic focus in fisheries management affects technical efficiency.
Although the relationship between the objective of fisheries management and technical
efficiency is theoretically apparent, this association has not been empirically analysed in the
fisheries literature. Also as noted above, directly observing such a relationship is not possible.
However, empirical support for the broad hypothesis can be derived from a series of studies

Chapter 1. Introduction 6
addressing different aspects of the broader hypothesis. To this end, four main studies are
undertaken to address the research questions.

The first study involves a meta-analysis study based on previous analyses of technical
efficiency in fisheries. The main aim of this study is to examine whether the type of fisheries
management tools (i.e., ITQs, input controls, and open access), and study specific
characteristics have an observable impact on level technical efficiency.

As will be seen in the results of the meta-analysis, a number of different approaches to

estimating technical efficiency exist. Within these approaches, different model structures and
input combinations are also often applied that may have an influence on the measures of
technical efficiency. The second study examines the sensitivity of efficiency estimates in
fisheries to choice of input measures used in efficiency model, and to efficiency model
specifications. This is to provide a guide as to how much of the observed variation between
studies may be due to model specification rather than the operating environment of the fishery.

The third and fourth studies focus on evaluation of technical efficiency of two different
fisheries that are subject to different operating environments and different external economic
influences. The two fisheries are open-access trawl fishery in Vietnam, and the banana prawn
fishery, a sub-fishery of the Northern Prawn Fishery (NPF) in Australia. The two fisheries are
substantially different with respect to fisheries management objectives, and thereby their
management systems used to manage the fisheries. Specially, the trawl fishery in Vietnam is
open-access fishery and characterised as small-scale fishery. Social objectives including
employment, food security and livelihood for fishers and fishing communities seem to be a
priority of management in this fishery. By contrast, the banana prawn sub-fishery of the
Northern Prawn Fishery in Australia is a highly industrialised fishery. The fishery is managed
with the primary objective of maximising economic yields despite multiple objectives in
management (Pascoe et al., 2012). The main characteristics and fisheries management
objective of the two fisheries are outlined in Table 1.1.

Chapter 1. Introduction 7
Table 1.1. Main characteristics of the two case study fisheries

Open-access trawl fishery in Vietnam The Northern Prawn Fishery in Australia

o Large-scale fisheries
o Small-scale fisheries
o Trawl fisheries
o Trawl fisheries
o Individual transferable effort
o Open access fisheries
controls
o Million people rely on for food,
o Commercial fishers
livelihood and income
o Fisheries management objective:
o Fisheries management objective:
Multiple objectives but economic
Multiple objectives but social
objectives (i.e., MEY) focused
objectives focused (i.e., livelihood,
employment)

The fisheries also have different external influences in the form of opportunity costs. In the
case of the Vietnamese case study, alternative livelihoods are limited in some parts of the
fishery and less so in others. A priori, given the open-access nature of the fishery, we would
expect that technical efficiency in the areas where alternative livelihoods are available would
be higher than in the areas where the opportunity cost of remaining in the fishery is lower, as
less efficient fishers would be able to exit the fishery. In the case of NPF, which is a sequential
fishery with two sub-fisheries that overlap to a small degree, we would expect less efficient
fishers in the banana prawn sub-fishery to exit early and move to the tiger prawn fishery as
these species become available. As a result, we would expect the average efficiency of the fleet
operating in the banana prawn fishery to increase over the season as fishers aim to maximise
their own returns.

Research process

The overall objective of this thesis is to explore the impact of fisheries management objective
on technical efficiency. As noted above, the thesis has four main studies. The fourth study –
based on the Vietnamese fishing fleet – is split into two sections (referred to as Activities in
the following section). These studies and activities are as follows.

Study 1 (Activity 1): The principal objective of this study is to explore how factors such as the
fisheries management instruments and study specific attributes may have an impact on the level
of technical efficiency in fisheries. Identifying the potential influence of a management system

Chapter 1. Introduction 8
requires accounting for the effects of other study-related factors that may influence technical
efficiency.

Empirical studies in technical efficiency undertaken in agriculture, aquaculture and other

maritime sectors have shown that technical efficiency scores were affected by the choice of
efficiency estimation techniques, data types, and functional forms, amongst other factors
(Bravo-Ureta et al., 2007; Iliyasu, Mohamed, Ismail, & Abdullah, 2014; Odeck & Bråthen,
2012; Thiam, Bravo-Ureta, & Rivas, 2001). The efficiency literature elsewhere also shows
mixed results and conflicting views in regard to the merits of the various methodologies
(Bravo-Ureta et al., 2007). Similarly, empirical studies in fisheries efficiency have also
reported different findings regarding efficiency scores due to the use of different approaches
(Herrero, 2005; Orea, Alvarez, & Paul, 2005).

Furthermore, fisheries are fairly unique in terms of resource management as the management
approach itself may also impact on the level of efficiency. Different fisheries are often managed
by adopting different management tools (e.g., input and output controls, open-access regime,
and ITQs, among others). These different management systems afford different levels of quasi-
property or use rights to the fishers1. The use of ITQs has been widely advocated to improve
economic performance in fisheries (Chu, 2009) and it is generally believed to increase average
efficiency of the vessels (Costello & Deacon, 2007), with the least efficient vessels generally
leaving the fishery (see e.g., Dresdner et al., 2010; Kompas & Che, 2005; Pascoe et al., 2001;
Sharp et al., 2004). In contrast, open-access fisheries are likely to involve a larger mix of
heterogeneous vessels, with some being relatively efficient and others relatively less efficient,
leading to lower efficiency on average.

Despite the increasing popularity of technical efficiency studies in fisheries, to our knowledge,
no meta-analysis has been conducted in the field of efficiency studies in fisheries to provide a
broad overview of how efficiency scores vary with different attributes of a study, and in regard
to fisheries management system in particular. To fill this gap, study 1 (Activity 1) addresses
the following questions: (i) to what extent do fisheries management tools affect the average
level of efficiency? (ii) have technical efficiency indexes changed over time (i.e., year of data

1
While ITQs are often described as providing property rights in fisheries, this description is strongly opposed by
some economists (for example, Bromley, 2009). In this study, we will refer to these rights as either quasi-property
rights or use rights.

Chapter 1. Introduction 9
used in primary study)? (iii) do the mean technical efficiency values vary according to different
analytical methods used? (iv) does the number of inputs and outputs used in frontier model
have an impact on average efficiency levels?; and (v) does the relative ranking of the journals
where such studies are published (a rough indicator of the robustness of the study) have a
relationship with the average efficiency levels found in the studies?

Study 2 (Activity 2): The primary aim of this study is to examine the sensitivity of efficiency
estimates in fisheries to choice of input measures and efficiency model specifications. While
the impact of the broad estimation process could be identified in the meta-analysis (study 1),
the different studies applied the models in different ways, using different functional forms and
different measures of inputs. The effects of these differences on the technical efficiency can
only be determined by applying a similarly wide range of approaches on a common data set.

Furthermore, the measurement of technical efficiency requires the estimation of an appropriate

production frontier that is based on a bundle of inputs and outputs available in which the former
are assumed to affect the latter (Pascoe, Hassaszahed, Anderson, & Korsbrekke, 2003).
However, building up models to illustrate such relation is not straightforward, and the choice
of the right bundle of inputs in estimation of production frontiers in fisheries can be seen as
one of the difficulties faced (Andersen, 2005; García del Hoyo et al., 2004; Pascoe & Mardle,
2003). The literature interrogated showed that misspecification of production function, through
the use of inappropriate input measures, may result in a bias in the measures of efficiency
(Pascoe, Hassaszahed, et al., 2003). Consequently, management decisions based on biased
measures of efficiency could lead to ineffective management strategies, and thereby
unsustainable fisheries. In empirical studies, a wide range of input measures has been used in
fisheries efficiency studies, and different studies use different types of input measures. This
may stem from the fact that fisheries data are limited in many cases. The choice of input
measures for efficiency analysis in fisheries is, thus, often based on available data,
characteristics of fisheries (e.g., gillnet, longline and trawl fisheries), and the selection of input
used is relatively different among studies (Andersen, 2002; Greenville et al., 2006).

Policies that are based on biased results and/or fragile inference may cause inappropriate
regulations and mismanagement, thereby unsustainable fisheries (Andersen, 2002, 2005;
Valdmanis, 1992). Technical efficiency estimates are important for informing fisheries
management (Yang, Lou, Matsui, & Zhang, 2016), and there are no prior reasons in favour of
the use of input measures in fisheries efficiency. It is, thus necessary to examine whether

Chapter 1. Introduction 10
different input measures included in efficiency estimation, and different model specifications
would result in different results of efficiency scores. Activity 2 (study 2) examines the effect
of using different fishing effort measures (including fixed input and variable input measures),
and efficiency model specifications on technical efficiency scores. The key question considered
is to what degree technical efficiency estimates are sensitive to the choice of input variables in
fisheries efficiency and efficiency model specifications.

Study 3 (Activity 3): The key aim of this study is to assess the extent of efficiency and
productivity change over the banana prawn fishing season given the change in opportunity
costs that occur over the season. While these results are relevant for the general thesis
hypothesis, the results also have potential implications for the management of the fishery, and
in particular on the setting of the trigger and target catch rate used with the aim of achieving
the economic objective.

The banana prawn sub-fishery of the NPF is currently managed to achieve MEY by using a
target trigger catch rates. As the fishery proceeds over the season, catch rates decrease. When
the trigger catch rate is reached, the fishery is closed. The estimation of these target catch rates
relies on the assumption of relative homogeneity of the fishing fleet regarding productivity,
cost structure and efficiency. This study seeks to determine the extent of efficiency and
productivity change over the banana prawn fishing season, and the implications of these
changes, if any, on the setting of the MEY trigger/target catch rate. The key questions addressed
are: (i) Given the strong economic objective of management, does the average technical
efficiency of the fishing fleet change over the season in response to changes in economic
conditions? And, as a result, (ii) what are the implications of the assumptions regarding
homogeneity of the fleet on MEY trigger and target catch rate?

The fourth study (study 4) focuses on an open-access trawl fishery in Vietnam. This study
involves two research activities (Activities 4 and 5) that are presented as follows.

Activity 4: The primary aim of this activity is to provide a socio-economic status quo profile
of trawlers in Vietnam.

Existing literature indicates that socio-economic characteristics such as age, gender,

educational level, employment status and demographic characteristics (e.g., family size,
generation and tradition among others) are factors affecting individual behaviour in fisheries
(Aldon, Fermin, & Agbayani, 2011; Pascoe, Cannard, Jebreen, Dichmont, & Schirmer, 2015;

Chapter 1. Introduction 11
Tzanatos et al., 2006). Thus, information on socio-economic status of fishers can provide
important insights into designing and developing appropriate fisheries policies and measures
for sustainable fisheries. This is particularly important for small-scale fisheries management in
developing countries, like Vietnam, where marine fisheries have played an important role in
the socio-economic development of the country. Millions of people directly/indirectly rely on
fisheries resources for food security, poverty alleviation, and job creation, and many other
things.

Despite the importance of information on socio-economic status of fishers to sustainable

fisheries management, the data on fisheries, including socio-economic status of fishers, are
relatively scarce. To have a comprehensive understanding about the socio-economic profile of
small-scale vessel owners, the following questions are addressed: (i) What is the current socio-
economic profile of trawl fishery (e.g., demographic profile, education level, and fishing
tradition)?; (ii) What are the main livelihood activities of fishers?; and (iii) How is the fishers’
catch distributed to market?

Activity 5: The main aims of this research are (i) to estimate the level of technical efficiency
and its determinants of trawl fishing vessels; and (ii) propose policy recommendations for
managing fishing capacity in the trawl fishery in Vietnam.

In Vietnam, marine capture fisheries are open-access and classified as small-scale. Over the
past few years, the number of fishing boats has increased dramatically, leading to overfishing
and raising concerns about the sustainability of the fisheries, especially in nearshore coastal
areas. Actions have been taken to reduce the fishing pressure on coastal fisheries, through a
reduction of the number of fishing vessels operating in coastal areas (i.e., small-scale trawlers
in particular) and encouraging the development of larger fishing vessels to fish in offshore
areas. However, these decisions have been made in the absence of information on technical
efficiency within these fisheries.

It is well recognised that the level, distribution of technical efficiency and its determinants are
helpful for designing fishing effort reduction programs and sustainable fisheries management.
This is because the success of fishing capacity reduction programs is largely affected by
variations in efficiency between boats (Pascoe & Coglan, 2000). Given this variation, the
efficacy of management measures to reduce fishing effort primarily depends on which fishing
boats to remove from the fishing fleet (efficient or less efficient vessels), and this is likely to

Chapter 1. Introduction 12
be more important than how many fishing boats are removed (Herrero & Pascoe, 2003).
Eliminating inefficient boats from a fishing fleet would result in less proportional decrease in
an impact on the total fishing capacity of the fishing fleet (Herrero & Pascoe, 2003; Pascoe &
Coglan, 2000; Pascoe, Hassaszahed, et al., 2003). As a result, the development of an effective
fishing capacity reduction policy needs to be informed by technical efficiency at the level of
the fishing vessel. Furthermore, identifying factors that affect the efficiency of fishing vessels
is also useful to formulate appropriate regulations and policies for fisheries management
(Sharma & Leung, 1998; Tingley et al., 2005). The research undertaken in this study will
address the question: what is the level of technical efficiency and its determinants of trawl
fishing vessels?

Table 1.2 provides overall information about studies (including research activities) as follows.

Chapter 1. Introduction 13
Table 1.2. Summary of research activities, methodology and data used

Activity Activity 1 (study 1) Activity 2 (study 2) Activity 3 (study 3) Activity 4 (study 4) Activity 5 (study 4)
Investigate factors Examine the sensitivity Assess the extent of Develop status quo of Estimate the level and
affecting technical of input choice and efficiency and socio-economic profile its determinants of
efficiency (i.e., model specifications on productivity change of trawl vessel owners technical efficiency of
Summary
management systems technical efficiency over the banana prawn in Vietnam open-access trawl
and study’s estimates fishing season fisheries in Vietnam
characteristics)
Provide an overview of Determine the Implication for setting Identification of socio- Identify stochastic
technical efficiency sensitivity of technical of the MEY economic production frontier and
studies in fisheries in a efficiency that may be trigger/target catch rate characteristics of trawl the role of fisher’s and
systematic way. affected by the choice in the banana prawn vessel owners vessel characteristics
Benefit
Identify factors of different inputs in fishery of Northern and environmental
affecting technical production function, prawn fishery in factors on explaining
efficiency and model Australia the variation of the
specifications technical efficiency
Database searches Production efficiency Production efficiency Fisheries survey Production efficiency
Meta-analysis approach analysis using the analysis using the Descriptive analysis analysis using the
Main
by using meta- stochastic production stochastic production stochastic production
approach
regression model analysis. analysis analysis
Regression analysis
Available published Panel data from CSIRO Panel data from CSIRO Primary data through Primary data through
papers through (i.e., catch, engine and
(i.e., catch, engine and fisheries survey fisheries survey
scholarly search length of vessel, fishing
length of vessel, fishing
Data engines including time and fish stock time and fish stock
Scopus, Web of information) information)
Science and Google
scholar, and others.
The overall objective of this dissertation is to examine the impact of fisheries management objective on the level and distribution
Research
of technical efficiency. Several studies (Study 1 to Study 4), including Activity 1, Activity 2, Activity 3 and Activity 5 are
integration
undertaken.

Chapter 1. Introduction 14
1.4. Main contributions

There have been a number of technical efficiency studies investigating the impact of fisheries
management changes (e.g., input controls, ITQs, and buyback programs etc.) on technical
efficiency. However, understanding the impact of fisheries management objective on technical
efficiency has been limited, and no study has addressed this gap. This thesis is the first study
empirically investigating this relationship. The overall contribution of the thesis is to provide an
understanding of the relationship between management objectives and technical efficiency in
fisheries. This result would provide policy implications for fisheries management regarding
pursuing different fisheries management objectives, especially social objectives (e.g., open-access
fishery being a proxy) and economic objective (e.g., MEY reference point). Understanding the
impact of fisheries management objective on technical efficiency would provide implications for
the trade-off of technical efficiency in achieving different objectives in fisheries management.

This thesis consists of four main studies. Findings generated from each study provide significant
contributions to the existing literature on both technical efficiency and its application in fisheries
management. Furthermore, the thesis also contributes valuable information for practical fisheries
management for two fisheries: the Northern Prawn Fishery in Australia and the open-access trawl
fishery in Vietnam.

First, regarding meta-analysis study (study 1 – Chapter 4), there has been a growing interest on
technical efficiency of fisheries over the past two decades. However, to our best of our knowledge,
there is a lack of literature that undertakes meta-analysis based on the available studies in
efficiency fisheries to provide a broad overview of how efficiency scores vary with different
attributes of a study, and in regard to management system in particular. This is the first meta-
analysis of efficiency in fisheries that provides a better understanding of the major determinants
of efficiency estimates in fisheries, especially the impact of fisheries management tools (i.e.,
impact quasi-property rights on fisheries technical efficiency). It is also hoped that this meta-
analysis study would be useful reference for relevant researchers, fisheries managers, and
newcomers who enter technical efficiency research in fisheries.

Chapter 1. Introduction 15
The second study (study 2 - Chapter 5) of this thesis examines the sensitivity of efficiency
estimates to choice of input variable measures in fisheries efficiency, and efficiency model
specifications. This is also the first study looking at this topic in the field of fisheries efficiency.
Thus, another contribution of this thesis to existing academic knowledge is to enhance knowledge
about the sensitivity of technical efficiency estimates to the choice of input measures (e.g., fishing
effort measure), and efficiency model specifications in fisheries efficiency study.

The third study (study 3 - Chapter 6) focuses on banana prawn fishery, a sub-fishery of the
Northern Prawn fishery in Australia, currently managed to achieve MEY. Findings about technical
efficiency and productivity changes over the season provide policy implications for the fishery
management, especially aiming at MEY.

Fourth, the last study (study 4 - Chapter 7) also provides a broader socio-economic status-quo of
the trawlers in Vietnam; and information about technical efficiency level and its determinants of
trawlers. These findings would have useful implications for sustainable trawl fishery management
in Vietnam.

1.5. Structure of the thesis

This thesis consists of three parts, and a total of eight chapters as shown in

Figure 1.1. Part 1 comprises three chapters: introduction, fisheries management objectives and
technical efficiency in fisheries. Chapter 1 presents the research problems being addressed, how
the research problems are to be examined, the research hypothesis, research objective and
activities, and main contributions of the thesis. Chapter 2 provides a brief overview of information
about fisheries management objectives, trade-offs between fisheries management objectives, the
relationship between management objectives and management instruments, and the potential
impact of management objective on technical efficiency. Chapter 3 provides an overview of the
process of estimating technical efficiency. This includes the concept of technical efficiency and
approaches for estimating technical efficiency in fisheries. This chapter also presents topics related
to functional forms used in parametric approaches, the choice of input and output measures in
efficiency estimation in fisheries and theoretical consistency issues in estimating production
function.

Chapter 1. Introduction 16
Part 2 includes the four main studies that correspond to four chapters (from Chapter 4 to Chapter
7) as follows. In Chapter 4, a meta-analysis is undertaken to explore the impact of fisheries
management instruments and specific study characteristics on technical efficiency in fisheries. A
meta-regression model is applied. Results from this study show that there is room to increase
efficiency level in fisheries. Furthermore, a number of factors including fisheries management
(i.e., open access, input controls, and ITQs), year of data used in the primary study, efficiency
estimation approaches, the number of inputs and outputs included in efficiency model, and the
ranking of journal where primary study published on, affect the reported level of technical
efficiency.

In Chapter 5, the sensitivity of technical efficiency to the choice of input variables in efficiency
estimation is considered as this may also influence the results of the meta-analysis (Chapter 4).
This study utilised data from the banana prawn fishery of the NPF in Australia for analysis (the
case study is also considered in the next chapter). Stochastic frontier analysis approach is applied
to estimate technical efficiency scores using different combinations of factors, including the choice
of input variables and efficiency model specifications. The results show that the choice of input
variable in production function affects the level of technical efficiency. Specifically, when the
measure of fishing effort gets less precise (either day fished or hour fished or hour trawled),
technical efficiency scores have a lower mean and higher standard deviation. However, absolute
bias is generally small. Additionally, model specifications (i.e., functional form of production
function, distributional efficiency and time-varying assumptions of inefficiency) also have
significant impact on technical efficiency.

Chapter 6 looks at a fishery with a strongly focused economic objective (i.e., MEY) for
management, namely the banana prawn fishery (a sub-fishery of the NPF in Australia). This study
aims to assess the extent of efficiency and productivity change over the banana prawn fishing
season, and the implications of these changes due to changes in opportunity cost facing the fleet,
if any, on the setting of the MEY trigger/target catch rate. Stochastic frontier analysis method is
used with the data of the banana prawn fishery (the same data set used in the Chapter 5). Our
research results have several important findings. Firstly, given the changes in fleet productivity
over the season, and the observed changes in catch rates as a result, equating the MEY target and
trigger catch rates is a reasonable approach for pragmatic fisheries management. However, the

Chapter 1. Introduction 17
likelihood of higher fishing costs at the end of the season suggests that the current MEY trigger
estimates are too low. As the estimated target levels in recent years have been below the lower
precautionary buffer, they have not affected the actual implemented target and trigger. Secondly,
even with an economic objective in fisheries management, there is a wide distribution of efficiency
scores in the fishery. The model score, however, is high.

Chapter 7 focuses on an open-access trawl fishery in Vietnam. Fisheries management in this

fishery is strongly focused on achieving a number of social objectives. This study aims at providing
a description of the socio-economic status quo (Activity 4), and estimating technical efficiency
and its determinants (Activity 5) of the trawlers. In the absence of any data for this fishery, a field
survey was undertaken to collect data for the study’s purposes. A socio-economic profile of
trawlers is provided based on surveyed data. Furthermore, a stochastic frontier model is used to
estimate technical efficiency and its determinants of the surveyed trawlers. The empirical results
indicate that there is potential room to improve performance of fishers. Several factors have been
identified which affect technical efficiency of fishers, including age of fishing vessel, vessel
owners’ experience and age, distance from shore, engine type, and using family members in crew
size. Significant differences are also found between two surveyed regions, with technical
efficiency seemingly affected by the opportunity cost of labour. Poorer regions with fewer
alternative employment opportunities are also associated with lower average technical efficiency,
reflecting the effects of greater capital and labour non-malleability in these areas.

Finally, the discussion and conclusions are presented in Chapter 8.

Chapter 1. Introduction 18
 Impact of fisheries management objective on technical efficiency: Case studies in fisheries

Part 1 Part 2
Chapter 1: Introduction Studies: Four studies (study 1 to study 4) Part 3
Chapter 2: Fisheries management objective with five research activities (Activity 1 to Chapter 8:
Chapter 3: Technical efficiency in fisheries Activity 5) Discussion and Conclusions

Activity 2 (Chapter 5): Activity 4, 5 (Chapter 7):

Activity 1 (Chapter 4): Activity 3 (Chapter 6):
Sensitivity of TE to the Socio-economic profile; TE and
Meta-analysis of technical TE of banana prawn fishery:
choice of input variables, its determinants of open-access
efficiency in fisheries and efficiency models NPF fisheries in Australia
fisheries in Vietnam

Research approaches

Stochastic production
Meta-analysis approach frontier Stochastic production frontier Stochastic production frontier
Regression analysis

Figure 1.1. Thesis structure

Chapter 1. Introduction 19
Chapter 2. Fisheries management objectives

This chapter examines the broad economics literature related to fisheries management and the
potential impact of fisheries management objective on technical efficiency and its distribution. The
first section of this chapter provides an overview of fisheries management objectives. Section 2.2
examines the trade-offs between different fisheries management objectives. The last section of this
chapter discusses the potential impact of fisheries management objective on technical efficiency,
and how these objectives affect the distribution of technical efficiency and its level of fishing fleet
is emphasised.

2.1. Multiple objectives in fisheries management

It is widely documented that fisheries management generally aims at multiple objectives (Charles,
1989; Mardle et al., 2002; Wattage, Mardle, & Pascoe, 2005). These objectives include a wide
range of issues related to marine resource conservation; provision of food, incomes and wealth for
fishers and community; employment opportunity for fishermen, and sustaining the wellbeing and
viability of coastal communities (Charles, 1989). However, these objectives are mainly classified
as biological (e.g., ecological and environmental objectives), economic, social and political
objectives2 (Hilborn, 2007a; Mardle et al., 2002; Pascoe et al., 2014; Pascoe et al., 2009; Stouten
et al., 2011).

2.1.1. Biological objective

Generally, a biological objective aims at maintaining targeted fish stocks at or above the levels
necessary to ensure their continuous biological productivity (FAO, 2002). This objective
commonly refers to traditional maximum sustainable yield (MSY) (Hilborn, 2007a; Stouten et al.,
2011). The MSY determines the maximum level of catch (e.g., tons/kg of fish) at which fisheries

2
Some studies indicate that fisheries management objectives include biological, economic and social objectives in
which ecological objectives (including environmental objectives) are included in biological objective, meanwhile
political objectives are included in social objectives. However, in this section we present biological and political
objectives as individual objectives.

Chapter 2. Fisheries management objectives 20

resources can regenerate and be sustained over time (at equilibrium) (Diop et al., 2018).
Theoretically, a biological objective can be set through imposing controls on quotas (e.g., total
allowable catch) or fishing effort (an input-controlled fishery) (Guillen et al., 2013; Ye et al.,
2013). Given multiple objectives in fisheries management, a biological objective can be seen as a
prominent objective that also underpins other objectives, as without a biologically sustainable fish
stock, other objectives are hardly achievable (Mardle et al., 2002). If fish stocks are exploited at
the level of MSY, catches are maximised at levels where fish stocks can reproduce (Diop et al.,
2018; Legović, Klanjšček, & Geček, 2010). When the fish stocks are harvested at a higher level
than can be replaced through reproduction, the fish stocks begin to decline. If the fish stocks are
continuously exploited at a rate that is higher than the reproductive rate of the fish stocks, the
possibility of collapse of the fish stocks is likely, assuming an absence of recovery or effective
management measures. Once the fish stocks are depleted and the fishery has collapsed, this would
result in detrimental effects on ecosystem, environment, food source and socio-economic,
wellbeing of those that rely upon the fishery both directly and indirectly. Historically, a number of
fisheries have collapsed, caused mainly by overfishing, and this has been documented. Examples
include Peruvian anchovy, Atlanto-Scandian herring, and Northern cod in Newfoundland, Canada
(Grafton & Kompas, 2014; Hannesson, 2015). The collapse of these fisheries led to serious
biological, ecological, and socio-economic problems. For instance, the collapse of the Northern
cod fishery on the Eastern coast of Canada in 1992 resulted in approximately 19,000 fishers and
plant workers losing their jobs and an ecosystem in a complete state of decay (Schrank & Roy,
2013). After two decades a number of social, economic, and environmental issues have been
implicated in the collapse of the cod stocks in this fishery (Mason, 2002).

The long-term biological sustainable use of fisheries resources is an important goal of fisheries
management. In reality, many fish stocks worldwide are currently managed with the aim of
maximum sustainable yield in fisheries management policies (Diop et al., 2018; Ghosh & Kar,
2013; Legović et al., 2010; Sumaila & Munro, 2009). For example, in the European Union the
primary objective of fisheries management is to maintain high sustainable yields of fish stocks
(Holma, Lindroos, Romakkaniemi, & Oinonen, 2018; Kempf et al., 2016; Mardle et al., 2002; The
European Parliament and The Council of The European Union, 2013). In the USA, maximum
sustainable yield is also the principal aim of fisheries management (NOAA, 2017).

Chapter 2. Fisheries management objectives 21

Managing fisheries under MSY objective may lead to maximal catch from the fish stocks (cited in
Giron-Nava, Johnson, Cisneros-Montemayor, & Aburto-Oropeza; Holma et al., 2018). However,
fisheries management that aims only to sustain the highest catch possible may not be desirable
(Hart & Reynolds, 2008). Other objectives such as ecological, economic and political objectives
in fisheries management are also considered, as the sustainability of a fishery is also based on these
objectives rather than relying solely on biological objective (cited in Asche et al., 2018; Boström,
2012; Rindorf, Dichmont, et al., 2017).

2.1.2. Ecological objective

According to FAO (2002) ecological objectives in fisheries management aim to minimise the
detrimental impacts of harvesting activities on the environment and marine ecosystems (in the
narrowest sense this may include a subset of non-targeted species and dependent species).

The ecological objective in fisheries management used to receive little attention (Brewer et al.,
2006). However, over the past decades, the negative effects of fishing activities on marine habitats,
ecosystems and environments (e.g., protection of non-target species, ecosystems, habitats) have
been a major concern for fisheries managers, environmentalists, conservatism, economist,
ecologist and biologists and other interested bodies, and the wider public (M. R. Clark et al., 2015;
Gilman, Passfield, & Nakamura, 2014; Hilborn, 2007a). It is widely recognised that fishing
activities have significant direct and indirect impacts on fish stocks, marine ecosystem and marine
habitats (Jones, 1992; Sciberras et al., 2018). Fishing affects not only targeted species, but also
non-target species (Jennings & Kaiser, 1998; Torres-Irineo et al., 2014). A number of fishing gears,
bottom trawl fisheries for instance, target only a few commercially valuable species while many
non-target species are caught incidentally (Walker, Maxwell, Le Quesne, & Jennings, 2017). A
large number of non-target species and vulnerable and/or protected species such as sea snakes, sea
turtles, seabirds, and marine mammals, among others, are components of bycatch in trawl fisheries
(Brewer et al., 2006; Torres-Irineo et al., 2014). Additionally, fishing gear (e.g., bottom trawling)
has a direct physical effect on seabed and seafloor, damaging coral and seagrass species, imposing
stress for benthos whenever fishing areas are swept. Fishing activities also affect the seabed
through waste dumped from fishing vessels, abandoned or lost fishing gear etc. (Jones, 1992).
Aquatic populations are co-dependent on each other (FAO, 2002). The negative effects of fishing
through bycatch and physical effects on seabed cause economic loss and ecological effects on key

Chapter 2. Fisheries management objectives 22

species which are important for ecosystem functioning and the structure of marine ecosystems and
habitats (Torres-Irineo et al., 2014), for example changes in total biomass, species composition
and predator-prey relationship (Jennings & Kaiser, 1998; Torres-Irineo et al., 2014).

To limit the detrimental impacts of fishing on ecological and environmental systems, non-
target/bycatch species and dependent species, there are a number of fishing regulations, measures
and policies used in fisheries management in different fisheries and different countries. These may
include gear regulations, season and area closures, species and fish size regulations, economic and
market mechanism among others. The application of these measures and regulations varies among
countries, fisheries and regions. Moreover, in response to the detrimental impacts of the fishing
activities, a number of governments and international organisations worldwide have cooperated to
create policies and strategies to reduce the negative influences of fishing on marine fisheries
resources and ecological systems, and manage shared fisheries sustainably. For example, FAO
(1995) introduced a Code of Conduct for responsible fisheries. A set of international standards and
principles of fishing practices for responsible fisheries are introduced for sustainable fisheries and
maintaining ecosystems and biodiversity.

2.1.3. Economic objective

It is widely recognised that an efficiently managed fishery could provide continuous economic
benefits for resources users (Gordon, 1953, 1954). Besides the aim of the biological sustainability
of fish stocks, how to manage a fishery to achieve maximal economic benefit from the fishery is
also an important objective of fisheries managers. Improving the economic performance of
fisheries is increasingly getting attention in fisheries management (Norman-López & Pascoe,
2011). Fishing, like other production activities, is also driven by the economic maximisation of
fishing activity (Holma et al., 2018). Unlike traditional MSY management objective, which
focuses on maximising sustainable catch of a fishery, maximum economic yield, a management
target, is to maximise economic returns (the difference between total revenue and total cost) from
the fishery (Hoshino, Pascoe, Hutton, Kompas, & Yamazaki, 2017; Kompas, 2005; Kompas et al.,
2010). Previous studies indicated that exploiting at MEY level could lead to a ‘win-win’ outcome
for both fishery industry and environment because MEY realises profit for the industry (a fishery
as a whole), maximal economic profit for fishers, larger fish stocks and lower impacts on the rest

Chapter 2. Fisheries management objectives 23

of the ecosystem (Grafton, Kompas, Chu, & Che, 2010; Kompas et al., 2010; Norman-López &
Pascoe, 2011).

Theoretically, the demonstration of MEY, MSY and open-access target reference points is
presented in Figure 2.1. Note that at MSY, the catch of a fishery is sustainably maximal, but at this
point maximising economic yield from the fishery is not realised (Holma et al., 2018; Hoshino et
al., 2017). In contrast, fishing at MEY level is both biologically sustainable and generates the
highest profits for the fishery. Fishing effort and cost at MEY level are lower than those at open-
access and MSY levels. MEY is achieved through a reduction in fishing effort compared with
MSY and open-access fisheries. However, reducing the number of fishing boats to reduce fishing
effort would result in decrease in employment. Existing literature shows that moving toward MEY
fishery, a substantial reduction in the number of fishing boats and employment is acknowledged
(see e.g. Waldo et al., 2016). In reality, defining and implementing MEY in practical management
is not easy due to the complexities that arise (see e.g. Dichmont, Pascoe, Kompas, Punt, & Deng,
2010 for detail; Wu, Wang, Hu, Hong, & Wang, 2019). For further information about MEY, a
number of references are available (e.g. Christensen, 2010; Grafton et al., 2010; Guillen et al.,
2013; Kompas, 2005; Sumaila & Hannesson, 2010).

Chapter 2. Fisheries management objectives 24

 Maximum sustainable yield
Benefits
(Cost, yield, Maximum economic yield (MSY)
profit, (MEY)
revenue)
Open access (OA)

Total cost

Sustainable yield
curve (TR)

0 Fishing efforts
EMEY EMSY EOA
(E)

Figure 2.1. Maximum economic yield and maximum sustainable yield

Conventional fisheries resource management approaches have focused on achieving maximum

yield of the fish stocks, and on ecological aspect (Cheung & Sumaila, 2008; Stephenson et al.,
2018). However, the importance of economic objectives in fisheries management is increasingly
being considered (Brooks et al., 2015; Pascoe et al., 2014; Rindorf, Dichmont, et al., 2017; Wu et
al., 2019). More recently, achieving MEY as a management objective in fisheries management has
received more attention. MEY has been used as management objective in Commonwealth-
managed fisheries in Australia, and is considering in other fisheries in other countries (Dichmont
et al., 2010; Pascoe, Dichmont, Vieira, et al., 2013; Waldo et al., 2016).

2.1.4. Social objective

Fisheries are important resources to provide socio-economic development, biodiversity benefits

and employment opportunities. It is widely admitted that sustainable fisheries largely depend on
three pillars: economic, social and environmental (e.g., biological and ecological) aspects (Asche
et al., 2018; Ye et al., 2013). However, social aspects had been relatively neglected or integrated
into fisheries management (Rindorf, Dichmont, et al., 2017). Social considerations are now
increasingly important and having an influence on fisheries management policies (Pascoe et al.,
2014).

Chapter 2. Fisheries management objectives 25

Unlike other objectives, social objective is arguably the hardest to define or identify due to its
complexity (OECD, 2011), and a lack of broadly agreed and associated indicators (Pascoe,
Plagányi, & Dichmont, 2016). Existing literature shows that a number of indicators have been used
as social objectives in fisheries (Pascoe et al., 2014). These include a variety of issues associated
with industry community, local/regional communities (Brooks et al., 2015), cultural and political
considerations (Hilborn, 2007a), family income and livelihood of fishers, maintaining or
maximising employment (Hilborn, 2007b; Pascoe, Dichmont, Brooks, Pears, & Jebreen, 2013),
maintaining communities and social capital (Brooks, 2010; Mardle & Pascoe, 2002; Mardle,
Pascoe, & Herrero, 2004), recreational access and traditional activities (Urquhart, Acott, & Zhao,
2013), reducing conflicts between alternative users (Andalecio, 2011; Leung, Muraoka,
Nakamoto, & Pooley, 1998; Mardle & Pascoe, 2002; Mardle et al., 2004), safe and healthy
working conditions for fishers (Heen et al., 2014) among others.

Despite a wide range of social considerations in fisheries management (see e.g. Pascoe et al.,
2014), employment, livelihood and income for participants who depend on the fishery for their
livelihoods, food sources and maintenance of traditional communities and equities are primary
elements of social objectives (Hilborn, 2007a; Pascoe et al., 2014). In this thesis, regarding social
objectives we follow this concept for consideration. In addition, the open-access regime in fisheries
management can be seen as a proxy of social objectives as this might allow more fishers to
participate.

Social objectives are easily contradictory with other objectives, especially economic and biological
objectives in fisheries management. This topic will be discussed in section 2.2.

2.1.5. Political objective

Political considerations are also objectives of fisheries management. These objectives are
primarily relevant to retaining social peace and reducing conflict among user groups (Hilborn,
2007a; Stouten et al., 2011). Fisheries-related conflicts are diverse (Pomeroy, Parks, Mrakovcich,
& LaMonica, 2016) and the reason behind conflicts is varied (Khondker, Belton, & Viswanathan,
2014; Khondker, Salayo, & Kanagaratnam, 2009; Muawanah, Pomeroy, & Marlessy, 2012).
Examples of conflicts include: between fisher groups (e.g., industrial fishers/commercial fishers
and small-scale fishers), between fishing gears, and between user groups (aquaculture, fisheries,

Chapter 2. Fisheries management objectives 26

recreation and tourism, offshore oil and ocean mineral users among others) (Brown, 2016; Charles,
1992; Kearney, 2001, 2002; Pomeroy et al., 2016).

The political issues (i.e., conflicts in fishing) not only occur between different users within a
country or fishery but also arise between countries in international waters/shared fishing zones.
Historically, a number of “fish wars” between countries have been documented (Gänsbauer,
Bechtold, & Wilfing, 2016; Spijkers et al., 2018). More recently, there have been well-reported
international fishing ground disputes in the East Sea among regional countries (Duy, Flaaten, &
Long, 2015; Schofield, Sumaila, & Cheung, 2016; Spijkers et al., 2018), and in the Yellow Sea
(Kim-Hyun, 2019). The degree of conflicts between countries varies. While sometimes migrant
fishers and foreign fishing boats are forced to expel, crews are sometimes imprisoned, and vessels
are destroyed (Pomeroy et al., 2007; Schofield et al., 2016; Spijkers et al., 2018).

There is broad agreement that desired objectives in fisheries management are higher fish stock,
less impact of fishing on marine ecosystems, higher economic profitability and more social
sustainability (Hilborn, 2007c). However, these objectives are commonly competing and
conflicting so the trade-offs between them are easily recognised. Generally, pragmatic fisheries
management aims at multiple objectives that include biological, ecological, economic, social and
political objectives. The focus of this thesis is on social and economic objectives.

The next section presents and discusses in detail the trade-offs between fisheries management
objectives.

2.2. Trade-off between fisheries management objectives

Fisheries management generally aims at a number of objectives but these objectives are often
incompatible and conflict (Hilborn, 2007a; Mardle et al., 2002). Therefore, fisheries management
primarily involves a complex balance or trade-offs between objectives (e.g., ecological,
sustainable fish stock, social and economic objectives) (Leung et al., 2001; Mardle & Pascoe,
2002). It is unlikely to maximise net income from a fishery as a whole while maximising
employment opportunity (e.g., an indicator of social objectives) as shown in Figure 2.1 (MEY
point and open-access point). This is relatively intuitive, as the more fish stocks and marine
ecosystem (i.e., indicating biological and ecological objectives) are protected the less fish

Chapter 2. Fisheries management objectives 27

resources that are available for utilisation such as yield and impacting on either economic rents
(economic objective) or maximising employment in the short, medium term and long term
(Hilborn, 2007a). The existing literature shows that a fishery that is managed under MSY
objectives achieves maximum biological production, but it is likely to fail to achieve maximum
economic yield (Hilborn, 2007a; Rindorf, Mumford, et al., 2017). Furthermore, in most cases when
a fishery moves from an open-access fishery to MEY fisheries or right-based fisheries such as
ITQs fisheries, a reduction in fishing capacity is necessary. As a result, a reduction in the number
of fishing boats, which in turn would result in a drop in employment, is observable (Cheung &
Sumaila, 2008; Hilborn, 2007a; Norman-López & Pascoe, 2011).

Trade-offs between long run and short run objectives are also known in fisheries management. For
instance, most management strategies and policies used to restore over-exploited fisheries and
improve economic profit of fisheries require a reduction in the level of fishing effort (e.g.,
reducing fishing boats) (Cheung & Sumaila, 2008). Whilst this would result in long run benefits
(i.e., both economic and conservation) it would cause a short run decrease in employment
opportunity, regional incomes, food security and livelihood options of fishers (Asche et al., 2018;
Mardle & Pascoe, 2002; Sale, 2002).

Theoretically, as shown in Figure 2.1, fisheries managed under an open-access regime would result
in zero profit for the fishery (Cheung & Sumaila, 2008). To achieve MEY (a proxy of economic
objective) or MSY (biological objectives) levels, fishing efforts (e.g., number of fishing boats)
need to decrease from the open-access level. Optimising all objectives simultaneously is
impossible due to conflict and incompatibility among objectives (cited in Mardle et al., 2002).
Moreover, the literature notes that failure in fishery management often comes from the root of
conflicting objectives such as social, economic and conservation objectives (Sale, 2002).

Understanding the trade-off resulting from opposing objectives is important for fisheries
management as this enables fisheries managers and stakeholders to design policies and measures
to manage or restore fisheries resources that align to key stakeholders (Cheung & Sumaila, 2008;
Leung et al., 2001). Empirically, there are a number of studies that investigated the trade-offs
between fisheries management objectives (e.g., economic, employment and income, conservation
and economic objectives). Specifically, Leung et al. (2001) and Heen et al. (2014) analysed
conflicting goals: economic rent and employment, using a multi-objective programming model

Chapter 2. Fisheries management objectives 28

with data of the North Norwegian cod fisheries. These studies indicate that there are trade-offs
between employment and economic rent of the fisheries.

Cheung and Sumaila (2008) examined the trade-offs between socio-economic and conservation
objectives, using the Northern South China sea as an example. The authors showed that fishing
capacity reduction is necessary to improve both economic and conservation benefits. However,
the reduction in the number of fishing boats would result in a reduction in the number of fisheries-
related jobs while there were limited alternative livelihoods for displaced fishers. The study
therefore confirms the existence of incompatible objectives: social objectives (e.g., job
opportunities for fishers) and conservation objectives. Moreover, creating alternative livelihoods
for fishers can be seen as a feasible measure to improve marine ecosystem conservation.

Hilborn et al. (2012) explored the trade-off among conservation, profitability and food security
objectives in the bottom-trawl fishery in the USA. The authors show that there are trade-offs
achieving these objectives in fisheries management. Specifically, aiming at conservation objective
by reducing catch rate that could protect all fish stocks from overfishing but this also would result
in losses of yield and profit of the fishery.

At a global scale, Ye et al. (2013) applied a bio-economic model to evaluate the benefits and costs
of restoring overfished stocks. They reported that there is a trade-off between costs and benefit of
restoring the overfished stocks. Specifically, fishing capacity would have to reduce by 36-43%,
making 12-15 million fishers unemployed, with a cost of US$ 96-358 billion in buybacks.
However, annual fishery production and rent generated from the fishery would increase by 16.5
million tons and US$ 32 billion respectively if meeting the goal of the restored plan. The authors
indicate that a rebuilding plan would cause socio-economic consequences, so the success of the
plan depends also on willingness to accept the short term losses.

Norman-López and Pascoe (2011) analysed the short term and long term effects of achieving MEY
in some Australian fisheries using an input-output model. The authors found that overall losses
were recognised in the short run but in the long run a net economic benefit to society was generated
when pursuing MEY as management objective.

Due to multiple objectives in fisheries management, in reality the relative importance of each
objective in fisheries management may differ considerably from fishery to fishery and country to

Chapter 2. Fisheries management objectives 29

country (Coglan & Pascoe, 1999; Hilborn, 2007a; Mardle et al., 2002). However, it seems that in
developed or high-income countries, fisheries management is more likely to focus on maximum
economic yield and/or maximum sustainable yield objectives than social objectives (i.e.,
employment and livelihoods of fishers). This is understandable, as marine capture fisheries provide
a modest employment source in these countries. For example, in Australia, maximum economic
yield is identified as a primary management objective in the Commonwealth-managed fisheries
(Dichmont et al., 2010). In the European Union, the primary objective of fisheries management is
to maintain high sustainable yields of fish stocks (Mardle et al., 2002; The European Parliament
and The Council of The European Union, 2013), as stated below:

The principal aim of fisheries management under the Common Fisheries Policy
(CFP) is to ensure high long-term fishing yields for all stocks by 2015 where
possible, and at the latest by 2020. This is referred to as maximum sustainable yield.

Similarly, in the USA fisheries aim at multiple objectives that include domestic seafood supply,
recreational and subsistence fishing opportunities, ecosystem health and sustainability. However,
maximum sustainable yield is the principal aim of fisheries management (NOAA, 2017).

In contrast, fisheries in developing/less developed countries are a key source of employment,

livelihoods and food (Béné et al., 2016; Sumaila & Munro, 2009) and fisheries management is
largely focused on social objectives. Small-scale fisheries dominate in developing countries
providing key employment, incomes and subsistence food sources to those that live in coastal
communities (Allison & Ellis, 2001). It is widely recognised in the fisheries literature that small-
scale fishers are frequently characterised as “the poorest of the poor”, and small-scale fishing
activities can be seen as “the occupation of the last resort” when there are no other alternatives for
fishers (Allison & Ellis, 2001; Béné, 2009; Onyango, 2011).

The conflict between social objectives (employment, livelihoods, food) and biological
(conservation, production), and economic objectives (profit generated from the fishery, economic
performance of fishery) are well understood and documented. However, whilst fisheries
management is generally characterised by the pursuit of opposing and conflict management
objectives, there is a clear dichotomy between the “key objectives” in developed versus developing
countries fisheries.

Chapter 2. Fisheries management objectives 30

The following section discusses the potential impact of fisheries management objective on
technical efficiency in which management objectives (focusing on economic and social objectives)
affect the distribution of technical efficiency and its level of fishing fleet is emphasised.

2.3. Management objective and technical efficiency distribution

As detailed above, fisheries objectives are many and varied. However, in this section, the analysis
is simplified to focus on two diametrically opposed objectives (i) social objectives (i.e., open-
access fishery as a proxy) and (ii) economic objective (i.e., MEY fishery as a proxy) and the impact
of each respective management objective on the technical efficiency of fishing fleet.

FAO (2002, p.3) proposes an overview of the task of fisheries management as follows:

The integrated process of information gathering, analysis, planning, consultation,

decision-making, allocation of resources and formulation and implementation, with
enforcement as necessary, of regulations or rules which govern fisheries activities in
order to ensure the continued productivity of the resources and the accomplishment of
other fisheries objectives.

Generally, fisheries management might require different tools to achieve different objectives (i.e.,
biological, economic and social objectives) (Anderson et al., 2019). In other words, the objectives
of fisheries management would determine what are the management approaches (i.e., policies,
regulations and measures) to meet the objectives. Available tools used by fishery managers to
manage fisher’s activities might include open-access regime, regulated open-access regime, input
controls, output controls (total amount of fish allowed to fish), a number of forms of quota systems
(e.g., ITQs), fishing season and area closures, etc.

Figure 2.2 broadly presents potential links between management objectives and management
approaches (instruments or regimes). Fisheries that focus on economic objective in which
maximum economic yield is the reference point, employing rights based management approaches,
are supported by robust theoretical and empirical research that demonstrate that economically and
biologically sustainable fisheries are achievable (Squires & Vestergaard, 2016). For example,
existing fisheries literature has shown that the use of transferable rights (e.g., catch quota or effort),

Chapter 2. Fisheries management objectives 31

reduce excess capacity and raise economic returns of fisheries and sustain the marine resource
(Hentrich & Salomon, 2006; Pearse, 1992; Soliman, 2014). Individual transferable quota, a type
of rights based fisheries regime, can be seen as one of the effective instruments to obtain the
economic objective (Kompas, 2005).

In contrast, open-access regime in fisheries management is often used as a way of achieving social
objectives. This is understandable, as open-access fisheries allow more fishers to participate in the
fisheries, potentially ensuring food, employment opportunities and livelihoods.

The link between fisheries management objective and fisheries management approach, however,
is not always clear. While individual transferable quotas (and other rights-based systems) closely
align to economic objectives, and open access aligns to social objectives, many fisheries are
managed through input controls, which may have a mix of economic or social objectives as shown
in Figure 2.2. The instruments broadly map to the main objectives, although there is considerable
variability in the strength of these objectives.

Figure 2.2. Relationship between management objectives and management instrument

The hypothesised link between fisheries management objective(s) and distribution of technical
efficiency is summarised in Table 2.1. Generally, different management approaches and
instruments (e.g., regulations, tools, measures) applied to achieve different objectives afford
different levels of quasi-property or user rights to the fishers. This creates different incentives

Chapter 2. Fisheries management objectives 32

and/or constraints for fishers to operate, thereby affecting the economic performance and fishing
activities of fishers.

Table 2.1. The potential relationship between fisheries management objective and technical
efficiency

Indicators Economic objective Social objective

Employment, livelihood,
Main aims/reference
Maximum economic yield (MEY) income, food security, among
point
others
Management approach Rights-based approach Competitive-based approach
Individual transferable quotas or Open access, Competitive
Instruments
effort quotas quotas
o Incentives for more efficient
fishers to buy out less efficient
o Race to fish
fishers’ quota and/or fishing
o Incentives to
efforts
overcapitalise
Impact on behaviour of o Incentives for less profitable
o No means for less efficient
fishers and fishing fleet fishers to leave
vessels to leave
o Incentive to reduce cost of
o No mechanisms for
fishing
autonomous adjustment
o Incentives to align cost
structure with catch structure
Homogenous fleet: Heterogeneous fleet:
Impact on technical
o High average TE and tight o Low average TE and wide
efficiency
distribution of TE distribution of TE

Individual transferable quotas approach is an example of rights-based management in fisheries

(Squires, Kirkley, & Tisdell, 1995) that serve as a management approach to pursue economic
objectives in fisheries management. Operating in ITQs fisheries, fishers are given rights to fish a
given amount of catch (i.e., quota), and no one has rights to fish without holding a quota. Fishing
quota holder can sell or lease their quota for other fishers who want to buy or lease it (see e.g. Chu,
2009; Grafton, 1996). Under ITQ system, the behaviour of competitive fishing is eliminated, and
fishers can have economic incentives to make rational economic options about their operations
(i.e., when and where to fish) (Beddington, Agnew, & Clark, 2007). The use of ITQ system in
fisheries management allows fishers who are more efficient (i.e., more profitable fishers) to have
economic incentives to buy quota from less efficient vessel owners (i.e., less profitable fishers).

Chapter 2. Fisheries management objectives 33

Previous studies has shown that less profitable fishers generally leave the fisheries while harvesters
that are more profitable would stay in the ITQs fisheries (Dresdner et al., 2010; Kompas & Che,
2005; Pascoe et al., 2001). Furthermore, previous studies indicate that the application of ITQ in
fisheries management would improve the economic efficiency of the fishery or fishing fleet as a
whole, and allows the industry to stay on a fishing effort level that optimises individual economic
returns to fishers (Anderson et al., 2019; Beddington et al., 2007). It is reasonable to expect that
fishing fleets operating in ITQs fisheries are quite homogeneous fleets regarding technical
efficiency as less efficient fishers exit the fisheries due to economic incentives and/or autonomous
adjustment.

In addition, fisheries literature shows that buyback3 has been an important policy tool to tackle
overcapacity, overexploitation, and to provide a way to move forward to a more rationalised
fishery (Squires, 2010). When the buyback program is introduced, inefficient fishers generally
have strong incentive to participate in the buyback and leave the fishery first while efficient fishers
would stay in the fishery as investigated in a number of studies (e.g. Dresdner et al., 2010; Kompas
& Che, 2005; Pascoe et al., 2001; Pascoe et al., 2012; Squires, 2010). This directly affects the
structure of the fishing fleets. Specifically, the fishing fleet would comprise a majority of efficient
vessels that are likely to be homogenous in terms of economic performance and efficiency. Thus,
it is likely that the distribution of technical efficiency of the fishing fleet managed under economic
objectives is tight, and the fleet has a high average level of technical efficiency.

Fisheries that are managed with a strong social objective often have relatively few controls on
fishing activities, with the possible exception of closures or basic regulations on fishing techniques
(e.g., dynamite) to maintain resource sustainability. The extreme case of this are open-access
fisheries, where fishers can enter or exit the fishery without restrictions imposed on the fishery.
Open-access regime can serve as a proxy to achieve social objectives as open-access fisheries
allow fishers and fishing communities to make a living.

3
For further information about buyback in fisheries, a number of references are available (e.g. Groves & Squires,
2007; Holland et al., 1999; Squires, 2010; Weninger & McConnell, 2000).

Chapter 2. Fisheries management objectives 34

Under open access, the phenomenon of “the race to fish” is, thus, inevitable. The open-access
fisheries provide no mechanisms for autonomous adjustment, and instead provide incentives for
new, more efficient vessels to enter a fishery without providing a mechanism (other than
bankruptcy) for less efficient vessels to exit. Such management still prevails as a result of
insufficient management resources, and in some cases a deliberate policy to support social
objectives such as livelihood, welfare and employment opportunities for fishing communities
practice. Unlike rights-based management regimes such as the ITQs system, less profitable and
inefficient fishers who are operating under an open-access regime, have limited incentives and
means to leave the fishery. This is because their fishing, to some extent, is not only driven by return
and profit from fishing but by subsistence, traditional cultures, employment, and livelihood.
Fisheries, especially small-scale fisheries in developing and “the Third World” countries are
examples (Andrew et al., 2007; Belhabib, Sumaila, & Pauly, 2015). As stated by Pomeroy (2012),
due to limited mobility and lack of non-fishing alternative livelihoods, many small-scale fishers in
open-access fisheries utilise whatever resources are available (e.g., technology, labour and skills,
and capital) to catch fish for their living. As a result, the structure of a fishing fleet operating under
open-access fisheries may include heterogeneous fishing boats regarding efficiency.

It is reasonable, therefore, to hypothesise that moving from socially focused objectives to

economic objectives in fisheries management, the variation of technical efficiency would decrease
and the average technical efficiency of the fleet would increase. The hypothesised distribution and
mean of technical efficiency under different fisheries management aiming at different objectives
are presented in Figure 2.3. Whilst this result appears obvious, pragmatic fisheries management is
not necessarily aware of this explicit trade-off. This can lead to a mix of non-compatible fisheries
management approaches in any given fishery, where it is assumed that efforts to make a fleet more
technically efficient will not undermine social objectives. The current push towards eco-systems
based management is predicated on the belief that it is possible to achieve economic, social and
environmental objectives simultaneously. In developing countries, there is a similar paradox,
where social objectives are necessitated because of a lack of alternative employment opportunities
but a strong desire to maximise economic outcomes from fishing as global consumption of fish
increases.

Chapter 2. Fisheries management objectives 35

 Frequency

Strong economic
objective
Strong social
objective

0
TE

Figure 2.3. Distribution of technical efficiency under different fisheries management objectives

Chapter 2. Fisheries management objectives 36

Chapter 3. Technical efficiency measurement in fisheries and its
relevant issues

The main purpose of this chapter is to provide general information about technical efficiency and
efficiency measurement approaches mainly applied in fisheries. Related issues including
functional forms used for stochastic production frontier approach, input and output measures used
in fisheries efficiency, and the issues of theoretical consistency are also discussed in this chapter.
The first section of this chapter provides the concept of technical efficiency and efficiency
measures. Some approaches including stochastics frontier analysis (SFA), Data Envelopment
Analysis (DEA) and multiple output stochastic distance function (SDF) that are commonly applied
in fisheries efficiency are outlined in section 3.2. In section 3.3, we briefly outline some common
functional forms used for stochastic production frontier in fisheries. Finally, the two last sections
present issues related to input and output measures used for efficiency estimation in fisheries, and
the issues of theoretical consistency in fisheries efficiency estimation.

3.1. Technical efficiency measurement

Technical efficiency, originally defined by Farrell (1957), indicates the ability of a firm to generate
the maximal output from a given set of inputs (output orientation measure) or the ability of a firm
to utilise minimal inputs to generate a given amount of output (input orientation measure).

Figure 3.1. Input-orientated measures of technical efficiency

Chapter 3. Technical efficiency in fisheries 37

Figure 3.1 presents an example of a firm using two inputs (x1 and x2) to produce a single output
(y), assuming constant return to scale (CRS) technology.

The fully efficiency frontier is represented by SS’. Firms that lie on the curve SS’ are considered
fully technical efficiency (C, B’) and any firms that are not on the curve SS’ then are considered
technically inefficient (B, D). Reducing the inputs used (x1 and x2) radically by moving to firm C;
firm D would have generated the same quantity of output while using less amount of inputs (x 1
and x2). Thus, the distance from firm D to C represents the technical inefficiency of firm D. An
input-orientated measure of technical efficiency of firm D (Farrell, 1957) is defined by the ratio:

TE = OC/OD

This ratio takes a value between zero and unity. A firm is technically efficient when this value
equals to one, and a firm is technically inefficient if this value is less than one. If input price
information is available, represented by the iso-cost curve AA’, it would be possible to measure
cost efficiency (CE)4 of the firm (D). Although firm (C) and firm (B’) are technically efficient
firms, firm (B’) gives the least-cost combination of inputs to produce the same amount of output
(y). To obtain the same level of cost as firm (B’), the input used by firm (D) needs to be reduced
to firm (B). Cost efficiency (firm D) can be defined as ratio:

CE = OB/OD

If the input price ratio measured by the slope of the iso-cost curve AA’ is available, then allocative
efficiency (AE) of firm (D) can be calculated as ratio:

AE = OB/OC

The distance from firm (B) to firm (C) represents production costs that would be saved if firm (C)
moves to firm (B’) to be allocative and technically efficient firm instead of moving to point (B) to

4
This terminology also refers to economic efficiency or profit efficiency (Coelli, Battese, Donnell, & Rao, 2005).

Chapter 3. Technical efficiency in fisheries 38

be technically efficient but not allocative efficient firm. A combination of technical efficiency and
allocative efficiency produces a measure of cost efficiency as follows.

TE x AE = (OC/OD) x (OB/OC) = OB/OD = CE

The output-orientated measure is graphically depicted in Figure 3.2. Two outputs (y1 and y2) and
a single input (x1) are involved in production technology.

Figure 3.2. Output-orientated measures of technical efficiency

The curve EE’ represents maximal possible output frontier. Any firms operating below the curve
EE’ are technically inefficient firms. In this case, firm (B) is inefficient firm. The output produced
by firm (B) can be expanded radically to firm (C) without increase in input used. The distance
from firm (B) to firm (C) represents technical inefficiency of firm (B). Hence, technical efficiency
of firm (B) can be measured as ratio:

TE = OB/OC

When output price information is available, represented by the curve AA’, it would be possible to
measure revenue efficiency (RE) of the firm (B). Despite being technically efficient, firm (C) could
achieve higher revenue by operating at firm (D). However, the same level of revenue can be
obtained if firm (D) that is revenue efficiency but technical inefficiency, moves to firm (D’) to

Chapter 3. Technical efficiency in fisheries 39

become simultaneously technical and revenue efficient firm. Hence, the revenue efficiency of firm
(B) can be defined as:

RE = OB/OD

The measure of allocative efficiency of firm (B) is defined by ratio:

AE = OC/OD

Given the allocative and technical efficiency measures, revenue efficiency can be measured by a
combination of these two measures as follows:

RE = OB/OD = (OB/OC) x (OC/OD) = TE x AE

Note that the value of TE, RE, AE and CE is bounded between zero and one. Furthermore, under
assumption of constant return to scale, the input orientation and output orientation measures are
equivalent (Färe & Lovell, 1978).

3.2. Efficiency measurement approaches in fisheries

Over the past decades, efficiency measurement has been an increasingly interesting topic in many
fields of production, economics and management, including fishing sector. Efficiency analyses
can be seen as important tools for evaluating the performance of a decision-making unit (DMU)
(e.g., individual, group, firm, plant, organisation, and fisher) and indicating its potential
improvement (Lampe & Hilgers, 2015; Michaelides, Vouldis, & Tsionas, 2010). Information on
efficiency of a DMU level and/or an industry is thus very important for producers, policy makers
and stakeholders.

Empirical research on efficiency in fisheries have become a topic of interest for scholars worldwide
for decades. Existing literature indicates that information about the relationship between the
amount of inputs used in fishing and the resultant catch is necessary for effective management
(Pascoe et al., 2001; Tingley, Pascoe, & Mardle, 2003). The understanding of technical efficiency
level and its drivers could provide directions for fisheries management, leading to a considerable
resource savings i.e., input source saving and/or output expansion. These have important

Chapter 3. Technical efficiency in fisheries 40

implications for both fishers and fisheries policy makers. For instance, fishers can improve their
performance and earnings by improving their efficiency while policymakers could use this
information to formulate appropriate and effective regulations and policies for sustainable fisheries
management (Sharma & Leung, 1998; Tingley et al., 2005).

There have been a number of existing different approaches applied to measure technical efficiency
in a number of different production sectors including fisheries. However, common approaches in
fisheries are Stochastic Frontier Analysis (SFA) and Data Envelopment Analysis (DEA). SFA, a
parametric approach, relies on econometric techniques independently proposed by Aigner, Lovell,
and Schmidt (1977), and Meeusen and Broeck (1977). The method assumes that output is defined
as a function of a bundle of inputs used, random noise and inefficiency. SFA thus has two error
terms: one representing normal random error and the other, a one sided error term, representing
technical inefficiency in production (Coelli et al., 2005). On the other hand, DEA is a non-
parametric approach first proposed by Charnes, Cooper, and Rhodes (1978), drawn on the earlier
work of Farrell (1957). This approach uses mathematical programming techniques to build a
frontier that envelops the observed data and thereby indicating the best practice. Based on this
frontier, efficiency scores are then measured. Fishing boats that are on the frontier are considered
technically efficient, and fishing boats that are beneath the frontier are classified as inefficient
boats.

Both approaches are able to estimate technical efficiency from an output or input orientation.
However, the output orientation approach is the most commonly adopted approach in efficiency
assessments in fisheries since this is more consistent with the assumption that (unless constrained
by quotas) fishers aim to maximise their revenue each trip (Pascoe & Tingley, 2007; Tingley et
al., 2005). That is, once a plan for a fishing trip has been made, necessities such as fuel, food, water
and ice for the fishing trip are prepared. The goal of the fishers is then to fish as much catch as
possible during the trip (Walden, 2006).

Chapter 3. Technical efficiency in fisheries 41

DEA and SFA methods have drawbacks and benefits5. The major advantages of DEA approach
are that this method can easily accommodate multiple-outputs and multiple-inputs in production,
and no functional forms are needed to predetermine. However, its primary drawback is that it does
not account for random error. Therefore, the efficiency estimate embeds inefficiency and random
error. In contrast, the SFA can include a measure of random error. Nevertheless, the primary
limitation of SFA is that it requires the production technology to be specified in advance (through
a production function). While multiple outputs can be modelled using variants of SFA (e.g., a
distance function version of the model, described in more detail below), it is best suited to a single
output production technology.

Previous studies has shown that there is no consensus as to which method is best (Herrero, 2005;
Murillo‐Zamorano, 2004), and both DEA and SFA have emerged as preferable approaches to
estimate efficiency in fisheries (Walden, 2006). The choice of method applied in each study largely
depends on the main purpose of the study, the availability of data, the characteristics of the
productions process, number of outputs and the degree of stochasticity (Herrero, 2005; Pascoe &
Tingley, 2007). However, given the relative importance of random variation in fisheries (e.g.,
weather, storms, and environmental factors), more research, which focus on technical efficiency
and its determinants, have favoured the application of the SFA approach, whereas DEA has tended
to be more used for studies of capacity and capacity utilisation in fisheries (cited in García del
Hoyo et al., 2004; Pascoe & Tingley, 2007).

Many fisheries are characterised by multiple output production due to multiple species fisheries in
nature. However, each of DEA and SFA has benefits and drawbacks in dealing with the multiple
output production technology. Another feasible option is to use dual models such as profit, cost or
revenue functions when price data is available (Coelli et al., 2005; Géraldine, Henningsen, &
Jensen, 2015; Kumbhakar & Lovell, 2000). However, information on input and output prices is
not always available in many fisheries worldwide (Felthoven, Hiatt, & Terry, 2002; Pascoe,
Hassaszahed, et al., 2003; Walden, 2006). In such cases, the only remaining attractive option is a

5
For further information about benefits and drawbacks, distinction between DEA and SFA, review and the
applications of DEA and SFA approaches, see Lampe and Hilgers (2015), Mardani, Zavadskas, Streimikiene, Jusoh,
and Khoshnoudi (2017) and Liu, Lu, Lu, and Lin (2013).

Chapter 3. Technical efficiency in fisheries 42

primal approach using a distance function (Géraldine et al., 2015). In this regard, multiple output
stochastic distance function model (SDF) has attracted much attention to deal with the issue of
multiple-output production technologies and statistical noise within a primal framework. The main
advantage of using the multiple output stochastic distance frontier models is that it enables the
inclusion of both random error and multiple outputs in the analysis which cannot be addressed by
using either DEA or SFA approach individually (Herrero, 2005). Although some difficulties arise
when using this approach in fisheries (e.g., issues around multiple-output production function,
multi-collinearity), the approach can be seen as justifiable (Orea et al., 2005; Pascoe & Tingley,
2007). The application of this approach in fisheries efficiency study has been limited to date but
can be found in a number of studies (Herrero, 2005; Orea et al., 2005; Pascoe, Punt, & Dichmont,
2010; Thøgersen & Pascoe, 2014).

Furthermore, to handle the issue of multi output technologies and overcome the limitations of DEA
approach that do not account for random error, bootstrapped-DEA models have been proposed to
take account for a stochastic element into non-parametric DEA models (Simar & Wilson, 1998;
Simar & Wilson, 2000). These approaches have been used in a number of efficiency studies in the
context of fisheries (see e.g. Ceyhan & Gene, 2014; Chen-Lei, Gupta, Mukherjee, & Wanke, 2016;
Hoff, 2006; Quynh, Hailu, Schilizzi, & Iftekhar, 2018; Tidd, Reid, Pilling, & Harley, 2016;
Walden, 2006). To deal with the issue of multiple outputs, other approaches have been applied in
fisheries efficiency such as stochastic ray production function (Löthgren, 1997), and directional
distance function approach (Färe & Grosskopf, 2000). These approaches have been hitherto
limited in the field of fisheries efficiency although the application of these approaches in the
fisheries context can be seen in several studies (Färe, Kirkley, & Walden, 2006, 2011; Fousekis,
2002).

Efficiency analyses have been undertaken in a wide range of industries (e.g., agriculture, banking,
healthcare and education) since the 1980s (see e.g. Lampe & Hilgers, 2015; Liu et al., 2013)
although the measurement of efficiency in fisheries was still relatively limited until the mid-1990s
(Herrero, Pascoe, & Mardle, 2006). Earlier analyses undertaken prior to 1980, noted variations in
the performance and productivity of fishing vessels (e.g. Gulland, 1956; Houghton, 1977; Robson,
1966; Zijlstra & de Veen, 1963). However, these studies were largely conducted by non-
economists, and considered variations of individual “fishing power” of vessels to largely be a

Chapter 3. Technical efficiency in fisheries 43

function of the physical inputs. Relatively few of these studies attributed any of these differences
to skipper or crew characteristics (e.g. Houghton, 1977). While Hannesson (1983) was the first to
estimate a production frontier, albeit a deterministic one, Kirkley, Squires, and Strand (1995) were
the first to apply the stochastic frontier in fisheries (Fousekis & Klonaris, 2003; Kompas et al.,
2004; Susilowati et al., 2005), and this is now regarded as a seminal paper in the field (Lampe &
Hilgers, 2015).

So far, the number of studies on fisheries efficiency has increased considerably, and many aspects
related to technical efficiency in fisheries have been investigated. Empirical efficiency and/or
efficiency-related studies in fisheries have primarily focused on some main relevant streams as
follows:

(i) Estimate technical efficiency level (e.g., sometimes allocative and economic efficiency are also
estimated in a number of studies) and its determinants (e.g., age and experience of skippers, engine
age and power), and impacts of fisheries regulations and managements on efficiency (e.g. Chávez
et al., 2017; Fousekis & Klonaris, 2003; Greenville et al., 2006; Huang et al., 2018; Pascoe &
Coglan, 2002; Pascoe et al., 2012; Pascoe, Innes, Courtney, & Kienzle, 2017; Sharma & Leung,
1998; Viswanathan et al., 2001);

(ii) Evaluate fishing capacity and capacity utilisation in fishing (e.g. Färe, Grosskopf, & Walden,
2015; Fissel, Felthoven, Kasperski, & O'Donnell, 2015; Kirkley, Morrison Paul, Cunningham, &
Catanzano, 2004; Lee & Rahimi Midani, 2015; Oliveira, Gaspar, Paixão, & Camanho, 2009;
Walden, Fissel, Squires, & Vestergaard, 2015; Walden et al., 2012);

(iii) Estimate productivity change in fisheries (e.g. Färe et al., 2015; Fissel et al., 2015; Kirkley et
al., 2004; Lee & Rahimi Midani, 2015; Oliveira et al., 2009; Walden et al., 2015; Walden et al.,
2012);

(iv) Compare of different efficiency measure approaches in evaluating fisheries efficiency (e.g.
Fousekis, 2002; García del Hoyo et al., 2004; Herrero, 2005; Herrero et al., 2006; Tomberlin &
Holloway, 2010).

Chapter 3. Technical efficiency in fisheries 44

3.3. Functional form of the stochastic production function in fisheries

As abovementioned, it is not necessary to predetermine functional form when using non-

parametric method, including DEA approach. However, when it comes to SFA, a parametric
approach for efficiency analysis, a functional form of production function needs to be
predetermined.

A number of functional forms are available (see, among others, Griffin, Montgomery, & Rister,
1987). However, the most commonly used functional forms in fisheries efficiency are translog
(Duy & Flaaten, 2016; Kirkley et al., 1995; Pascoe, Innes, et al., 2017; Tingley et al., 2005) and
Cobb-Doulas (Duy & Flaaten, 2016; Innes & Pascoe, 2008; Kompas et al., 2004; Ngoc, Flaaten,
& Kim Anh, 2009). The translog is considered to be a relatively flexible functional form as it does
not require assumptions about constant elasticities of production nor elasticities of substitution
between inputs (Coelli et al., 2005; Pascoe, Kirkley, Gréboval, & Morrison, 2003). Nevertheless,
a large number of variables in the estimation process may cause problems such as degree of
freedom problems and multi-collinearity (Coelli et al., 2005). The application of translog
functional form has been applied to estimate technical efficiency for a wide range of fisheries
worldwide (e.g., Greenville et al., 2006; Pascoe et al., 2001; Pascoe & Coglan, 2002; Pascoe &
Robinson, 1998; Sharma & Leung, 1998; Squires, Grafton, Alam, & Omar, 2003; Tingley et al.,
2005).

On the other hand, Cobb-Douglas, a special case of the translog functional form, is another
common functional form of production function when using parametric approaches for efficiency
estimation. The main advantage of the Cobb-Douglas functional form is that it is easy to estimate
as a logarithmic transformation provides a model which is linear in the logarithms of the inputs
(Coelli, Rao, & Battese, 1998). However, the Cobb-Douglas forms are less flexible than the
translog forms as Cobb-Douglas forms are first-order flexible while translog forms are second-
order flexible. The second-order flexible functional forms are, ceteris paribus, preferable to the
first-order flexible functional forms (Coelli et al., 2005). The use of Cobb-Douglas forms can be
found in a number of fisheries efficiency studies (e.g. Kompas et al., 2004; Sharp et al., 2004;
Vinuya, 2010; Yang et al., 2016).

Chapter 3. Technical efficiency in fisheries 45

Other functional forms such as Constant Elasticity of Substitution, Generalised Leontief and
normalised quadratic are also available (see e.g. Coelli et al., 2005; Griffin et al., 1987 for further
information). However, the application of these functional forms in fisheries is relatively limited.
The use of these functional forms can be found in a number of efficiency studies applied in other
industries (e.g. Gong & Sickles, 1992; Kopp & Smith, 1980; Zhu, Ellinger, & Shumway, 1995).

Choosing an appropriate functional form is important when parametric approaches (e.g., SFA) are
applied (Battese & Broca, 1997; Giannakas, Tran, & Tzouvelekas, 2003). Estimates obtained from
inappropriate functional forms, which may be biased and inaccurate, could result in misleading
recommendations in designing/proposing policies and strategies for management or decision
making (Giannakas et al., 2003). Thus, efficiency estimates produced from appropriate functional
form are fundamental for proposing appropriate policies and management strategies. However, the
choice of functional form is generally not an easy task (Giannakas et al., 2003; Michaelides et al.,
2010; Thompson, 1988). As it is impossible to determine true functional form that depicts a certain
relationship, so choosing the best appropriate functional form is generally considered (instead of
building the true functional form) (Griffin et al., 1987). Existing literature has proposed a set of
criteria that is commonly taken into account to judge how a functional form is more appropriate
than others. These criteria are not exhaustive although they provide some general preference for
the choice of an appropriate functional form. The preferred criteria generally include theoretical
regularity, flexibility, linear in parameter, parsimonious specification (see e.g. Coelli et al., 2005),
and other criteria such as ease of interpretation, computational ease, interpolative robustness,
extrapolative robustness (Fuss, McFadden, & Mundlak, 1978). For further information on this
topic, one can be find Coelli et al. (2005), and others (e.g. Fuss et al., 1978; Griffin et al., 1987;
Sauer, 2006; Sauer, Frohberg, & Hockmann, 2006).

The translog production frontier (Aigner et al., 1977; Meeusen & Broeck, 1977), the most flexible
functional form, is specified by:

𝑙𝑛𝑦𝑖 = 𝛽0 + ∑ 𝛽𝑘 𝑙𝑛𝑥𝑘,𝑖 + 0.5 ∑ ∑ 𝛽𝑘,𝑗 𝑙𝑛𝑥𝑘,𝑖 𝑙𝑛𝑥𝑗,𝑖 − 𝑢𝑖 + 𝑣𝑖

𝑘 𝑘 𝑗

where y is the amount of output produced, x is a vector of inputs, ui is a one-sided error term (u
≥0) that indicates the level of inefficiency of the vessel i, and vi is a random error term that is

Chapter 3. Technical efficiency in fisheries 46

assumed to be independently and identically distributed (iid), 𝑁(0, 𝜎𝑣2 ). Regardless of functional
form, a distributional assumption about inefficiency term has to be made to separate the stochastic
and inefficiency effects in the model (Pascoe, Hutton, Coglan, & Nguyen, 2017). Several
distributional assumptions have been proposed but the most distributional assumption is either a
normal distribution truncated at zero, 𝑢𝑖 ~ │𝑁(𝜇𝑖 , 𝜎𝑢2 )│ (Aigner et al., 1977) or a half-normal
distribution truncated at zero, 𝑢𝑖 ~ │𝑁(0, 𝜎𝑢2 )│ (Jondrow, Knox Lovell, Materov, & Schmidt,
1982). The technical efficiency (TE) of the i-th fishing boat is specified by 𝑇𝐸𝑖 = exp(−𝑢𝑖 ). The
technical efficiency score ranges from zero to one, with the higher score implying greater
efficiency. When ui = 0, the i-th vessel lies on the stochastic frontier, being technically efficient.
The production lies below the frontier and hence can be considered technically inefficient if ui >
0.

In some cases, an inefficiency model may be assumed and estimated as a function of known
characteristics and exogenous effects (Battese & Coelli, 1995) given by:

𝑢𝑖 = 𝛿0 + ∑ 𝛿𝑚 𝑧𝑖𝑚 + 𝑤𝑖
𝑚=1

where 𝑢𝑖 is in efficiency, z is a vector of vessel-specific variables and exogenous factors which

may affect the vessel’s inefficiency, δ is a vector of estimated parameters, and wi is an iid random
error term.

Maximum likelihood estimation method can be used to estimate the unknown parameters of the
stochastic frontier model and the model for inefficiency effects, with variance parameters defined
(Battese & Coelli, 1995) by 𝜎 2 = 𝜎𝑣2 + 𝜎𝑢2 and γ = 𝜎𝑢2 /(𝜎𝑢2 + 𝜎𝑣2 ). The value of γ lies between
zero and one. If γ=1 indicates that all the deviations from the frontier are attributable to the
technical inefficiency while if γ=0 means all deviations from the frontier are due to noise. If 0< γ
<1 implies the deviations from the frontier are due to both noise and technical inefficiency (Coelli
et al., 2005).

However, the frontier model above is applied for cases when cross-sectional data are used. When
panel data (i.e., multiple phenomena taken over some periods for the same vessel) are available,
technical efficiency of DMUs may also be seen to vary over time. Battese and Coelli (1992)

Chapter 3. Technical efficiency in fisheries 47

introduced a time-varying inefficiency model for panel data that incorporates the time-varying
firm effects (assuming technical efficiency to be an exponential function of time). This is defined
by 𝑢𝑖𝑡 = 𝑢𝑖 𝑒 −𝜂(𝑡−𝑇) , where 𝑢𝑖𝑡 is inefficiency of i-th vessel at time t, T represents time periods
(i.e., 𝑢𝑖𝑡 = 𝑢𝑖 , when t=T), and η is an unknown scalar parameter, measuring the rate of change in
technical inefficiency over time. If η > 0, η = 0 and η < 0 indicate that 𝑢𝑖𝑡 decrease, remain
unchanged or increase, respectively when t increases.

Generally, a number of hypotheses about the model specifications (e.g., functional form,
inefficiency distribution and others) are tested to determine the most appropriate model
specifications (Pascoe, Hassaszahed, et al., 2003). Likelihood ratio (LR) test is widely used. The
test is defined by 𝐿𝑅 = −2 [ln[𝐿(𝐻0 )] − ln[𝐿(𝐻1 )]], where ln[𝐿(𝐻0 )] and ln[𝐿(𝐻1 )] indicate the
log-likelihood function values under the null (H0) and alternative (H1) hypotheses, respectively.
This follows a chi-squared (χ2) distribution with the number of degrees of freedom given by the
number of the restrictions imposed (Coelli et al., 2005).

3.4. Input and output measures used in fisheries efficiency literature

Measuring technical efficiency in fisheries requires both information on output (e.g., catch) and
input elements (e.g., fishing vessel, labour, fish biomass).

Regarding output measures in efficiency analysis, there are three common measures of output: a
physical measure of volume (catch weight), value (total revenue of catch) and other measure
(aggregated quantity measure weighted by revenue share). The measure of catch weight is widely
used for single-species fisheries while the use of total revenue of catch is commonly applied when
fishers operate in multi-species fisheries (Pascoe & Tingley, 2007). Regarding DEA approach, the
issues of output measures are not challenged as this approach can handle multi outputs. In this
case, the choice of output measure would be quite flexible (e.g., multiple output, single output
measured in term of total catch weight or total revenue of catch). However, when it comes to SFA
method that assumes a single output only. In this case, the requirement of output aggregation is
necessary (Orea et al., 2005). Single aggregated output can be made from either aggregated-weight
catch or aggregated-revenue catch. When the measure of aggregated-weight catch is used, multiple
species are simply aggregated into one single variable measured by total catch (e.g., ton or kg of

Chapter 3. Technical efficiency in fisheries 48

fishes) irrespective of the relative importance of each species. When the aggregated revenue of
catch is used, the value each species measured by revenue (a combination of the quantity and price
of the species) is aggregated into one single output. However, the use of aggregated-value measure
has some issues for efficiency analysis (Pascoe & Coglan, 2002; Pascoe & Tingley, 2007; Sharma
& Leung, 1998). First, aggregated-value measure is a function of price and quantity of species
caught. Thus, the measurement of technical efficiency is affected when price of fish changes.
Fishers who catch a large quantity of catch (i.e., but primarily low-value species composition) may
not be efficient if aggregated-value of catch is used. However, they are efficient if aggregated-
weight of catch is used as output measure. Economic efficiency is composed of both technical
efficiency and allocative efficiency (Coelli et al., 2005). As a result, the use of aggregated-value
of catch as output measure may lead to problems in efficiency analysis, thereby policy and
management recommendations. Second, profit maximisation behaviour of fishers is generally
assumed (Herrero & Pascoe, 2003) so a change in relative prices of fish would lead to a change in
their fishing strategy. However, empirical studies have shown that the use of aggregated-weight
of catch and aggregated-value of catch produces quite consistent results in terms of technical
efficiency (Andersen, 2002; Herrero & Pascoe, 2003; Orea et al., 2005).

Another measure of output used in efficiency studies is to use value that is obtained from an
aggregated quantity measure weighted by revenue share (Pascoe & Tingley, 2007). This measure
considers the relative importance of each species in terms of monetary values (Andersen, 2002).
The application of this approach can be seen in a number of studies (Duy & Flaaten, 2016;
Greenville et al., 2006; Pascoe et al., 2001; Squires et al., 2003). Generally, it can be seen that
choosing output measures primarily depends on several factors including characteristic of
fisheries, fisher’s behaviour, available data, and efficiency estimation approach (Herrero & Pascoe,
2003).

As far as input is concerned, although different studies used a different combination of input
bundles when evaluating fishing vessel’s efficiency, the inputs used in efficiency estimation in
fisheries generally include physical measure of fixed inputs used (e.g., engine size, hull length and
wide, fishing gear and equipment), labour (i.e., crew), fishing time and stock size.

Chapter 3. Technical efficiency in fisheries 49

The fixed inputs are primarily expressed in physical measures such as engine size and boat size
(e.g., hull length, width or combined length and width measure, gross registered tonnage). These
input measures likely refer to capital invested in fisheries. However, the correlation between
physical measures of boat size/engine size (hull’s length and width, gross registered tonnage,
engine size) is often high. Therefore, including two or more inputs that are highly correlated into
efficiency model might cause problems for econometrics models. In fisheries literature, the choice
of input used in efficiency studies mostly depends on available data and characteristics of fisheries
(e.g., gillnet, trawl fisheries), among others (Andersen, 2002). Some studies used engine size (Duy
& Flaaten, 2016; Greenville et al., 2006; Kompas et al., 2004; Oliveira, Camanho, Walden, &
Gaspar, 2016; Pascoe, Hutton, et al., 2017) while some others used hull’s length and/or width, and
deck area (Fousekis & Klonaris, 2003; Ngoc et al., 2009; Pascoe, Innes, et al., 2017; Squires &
Kirkley, 1999), or GRT as fixed inputs (Jeon, Ishak Haji, Kuperan, Squires, & Susilowati, 2006;
Squires et al., 2003; Susilowati et al., 2005; Viswanathan et al., 2001), or fishing gear (Duy &
Flaaten, 2016; Fousekis & Klonaris, 2003; Kompas et al., 2004; Lokina, 2009; Rust, Yamazaki,
Jennings, Emery, & Gardner, 2017). Furthermore, some studies used a combined index that is
mixed between engine size and boat size (i.e., length and width of vessel’s hull) or fishing gears
(Greenville et al., 2006; Pascoe & Coglan, 2002; Vestergaard, Squires, & Kirkley, 2003).

In addition to physical measure of inputs, economic measure of inputs that captures capital
invested in fishing (e.g., fishing vessel, gear and equipment) is sometimes used (Fousekis &
Klonaris, 2003; Pascoe, Hassaszahed, et al., 2003). However, the use of economic measures of
fixed input has some limitations. First, economic measure of fixed inputs is not routinely collected,
thereby the economic information is not available and limited in most fisheries. Second, economic
measures of fixed inputs (i.e., vessel, engine and equipment) in reality are often based on book
values where the depreciation rate is fixed. However, in reality the level of depreciation varies
between vessels within a fleet and between fleet groups (Pascoe, Hassaszahed, et al., 2003). In an
attempt to examine the effect of using different input measures (i.e., economic and physical input
measures), Pascoe, Hassaszahed, et al. (2003) found that the use of physical measures of inputs is
not better or worse than that of economic measures of inputs.

Labour primarily represented by crew size is also another important input that is included in
efficiency fisheries models (see e.g. Jeon et al., 2006; Kirkley et al., 1995; Kirkley, Squires, &

Chapter 3. Technical efficiency in fisheries 50

Strand, 1998; Ngoc et al., 2009; Quynh et al., 2018; Squires, 1987; Viswanathan et al., 2001).
Similar to neoclassical economic theory, where capital and labour are two essential elements of
the production, labour is an important input of fishing activities. Nevertheless, despite being an
important input, increasing crew size would result in higher catch is only reasonable for some
fisheries that require intensive labour (e.g., pole and lines) (Pascoe, Hassaszahed, et al., 2003).
With the recent advance in fishing technology in some fisheries, larger crew size may not help to
increase catch, instead only a given size of crew is necessary. Furthermore, crew size is often
highly related to vessel size/engine size (i.e., bigger vessel have more crewmembers) (Pascoe et
al., 2001; Pascoe & Robinson, 1998). This may be a reason why crew size is not a subject of
fisheries management controls in reality.

Fishing time is commonly included into efficiency models in fisheries (see e.g. Duy & Flaaten,
2016; Ngoc et al., 2009; Pascoe, Hutton, et al., 2017; Pascoe, Innes, et al., 2017). Theoretically,
fishing time is not a production input per se in neoclassical production theory but this input is an
important element in fishing activities. This variable is considered to represent capital utilisation,
meaning higher fishing time refers to higher utilisation of capital (fishing vessel/gear). The
measure of fishing time used in fisheries efficiency research is different among studies that are
partly dependent on fisheries characteristics and available data. However, fishing time is normally
either measured in days fished (total days per fishing trip or per month/year) (Duy & Flaaten, 2016;
Kirkley et al., 1995, 1998; Kompas et al., 2004; Oliveira, Camanho, & Gaspar, 2010) or total hours
fished per trip/month (total fishing hours per trip or per month) (Jeon et al., 2006; Lokina, 2009;
Pascoe, Innes, et al., 2017; Susilowati et al., 2005; Viswanathan et al., 2001). Fishing time
measured by total fishing days primarily includes time for fishing (e.g., the time when fishing gear
is in use), and time used for other activities such as travelling, sorting and handling catch, searching
for fish stocks and others. On the other hand, the fishing time measured by the time when fishing
gear is used reflects the exact time for fishing. Clearly, there is a difference between the two
measures of fishing time. It is advised that the latter measure would be a more suitable measure of
fishing time to use, and comparing technical efficiency when using different measures of fishing
time is also an interesting research (Andersen, 2002). However, to the best of our knowledge, no
such study has been conducted in fisheries efficiency literature.

Chapter 3. Technical efficiency in fisheries 51

Furthermore, some other variable inputs such as fuel costs and other operating costs (e.g., ice,
miscellaneous costs), are sometimes used as input measures. These input measures have been
applied in fisheries efficiency analysis in a number of studies (Quynh et al., 2018; Sharma &
Leung, 1998; Truong, Terje, Ngoc, Kim Anh, & Thanh Thuy, 2011). However, these variable costs
are often highly correlated to engine’s size or boat’s size. Moreover, these variables are also
functions of fishing time; higher fishing time meaning higher variable costs.

An important element in fisheries production is fish stock level. Existing literature shows that the
amount of fish caught is generally assumed a function of fishing effort used and the abundance of
fish stock (C. W. Clark, 1973; Gordon, 1954). Fish stock is considered not a theoretically
conventional input per se in neoclassical production theory, as fish biomass is a non-discretionary
and is out of control of fishers (Andersen, 2005). However, fishing is quite different from other
production activities in which fish biomass, a non-discretionary and out of fisher’s control, serves
as an important input in a neoclassical production function (Hannesson, 1983; Squires, 1992). It is
widely acknowledged that fishing under different levels of fish abundance would result in a
different catch given all other things equal, thereby affecting the performance and efficiency of
fishers. It sounds therefore unreasonable to compare vessels and/or fishing fleets under different
biological stocks. Excluding a variable that represents the difference in fish stock from efficiency
analysis would bias the level of technical efficiency, thereby leading to inappropriate regulations
and management strategies in fisheries management (Andersen, 2005).

If data used is cross-sectional and single species fishery is considered, fish stock variable is not
necessary as all vessels are operating under the same stock level within a period of time considered
(García del Hoyo et al., 2004). However, if fishery is characterised by multi-species where vessels
target different species, the difference in the level of fish abundance between fishing vessels
irrespective of data type (i.e., cross sectional data or panel data) needs to be considered (Andersen,
2005). Several ways can be applied to account for stock biomass variable or a proxy measure of
fish stock into production frontier. Some methods are the stock biomass measure itself that are
independent stock estimates produced by independent research institutions. However, some are
not truly biomass index but may capture the fish biomass indicators such as catch per unit of effort
(CPUE) or a proxy measure of fish biomass (see Pascoe & Herrero, 2004 for detailed approach).
Some used CPUE as stock information in their production frontier models (e.g. Andersen, 2005;

Chapter 3. Technical efficiency in fisheries 52

Duy & Flaaten, 2016; Greenville et al., 2006; Kim, Seo, Kim, & Lee, 2012) while others used
stock index obtained from independent sources (e.g., Oliveira et al., 2010; Oliveira, Camanho, &
Gaspar, 2013; Pascoe, Hutton, et al., 2013; Rust et al., 2017; Walden et al., 2012). Also, a number
of studies used dummy variables to account for different stock effects on production that stem
from difference in fishing grounds, areas and seasons (i.e., month or years) (e.g. Innes & Pascoe,
2008; Orea et al., 2005; Quijano et al., 2018).

Regarding input measure in efficiency analysis in fisheries, it seems that different studies use
different measures of input. The selection of inputs used in fisheries efficiency analysis is largely
based on available data, comparison of different models, characteristics of fisheries, among others.
Although the choice of different input measures might affect estimates of technical efficiency
(Pascoe, Hassaszahed, et al., 2003), understanding the sensitivity of technical efficiency scores to
the choice of different input measures employed in fisheries efficiency is little known. This
uncovered topic will be further examined in Chapter 5.

3.5. Issues of theoretical consistency

The use of parametric efficiency measurement techniques such as stochastic production frontier
has been widely applied in production economics. When it comes to parametric approaches used
in efficiency measurement, functional form needs to be predetermined. It is widely recognised that
using an appropriate functional form is crucial for efficiency estimation, and then for policymaking
as biased efficiency measure might result from model misspecification such as using inappropriate
functional forms. Management decisions and strategies based on incorrect measures of efficiency,
resulted from inappropriate functional forms, could lead to mismanagement and ineffective
policies (Giannakas et al., 2003).

Production theory suggests that it is necessary to estimate functional form that requires fulfilling
assumption of monotonicity as well as quasi-concave (Lau, 1978). Nevertheless, violating these
properties in production function estimation is commonly recognised in empirical studies
(Michaelides et al., 2010; O’Donnell & Coelli, 2005). Once the violation of the properties dictated
by economic theory is a case, it is impossible to interpret the results properly (O’Donnell & Coelli,
2005). For instance, the monotonicity property indicates that an increase in quantity of any inputs

Chapter 3. Technical efficiency in fisheries 53

will not decrease the level of output. The use of DEA implicitly imposes while the application of
SFA with flexible functional forms often ignores this condition (Henningsen & Henning, 2009).
Although the importance of monotonicity condition is widely known most empirical studies using
SFA did not satisfy this condition (Sauer, 2006; Sauer et al., 2006).

The problem of efficiency estimates produced from a non-monotone production frontier is

depicted in Figure 3.3 below. As can be seen, firm A is on the production frontier so this firm is
considered technically efficient while firm B lies below the frontier and hence this firm is
inefficient. Both firm A and B produce the same amount of outputs (OE) while firm B uses less
inputs than that of firm A. This means that firm B would be more efficient than firm A in terms of
input-orientated measure but firm B is inefficient as it is not on the frontier. Hence, it seems that
efficiency scores based on non-monotone production frontier are unreasonably interpretable in this
case. Furthermore, it is notable that a non-monotone production frontier might result in misleading
in not only interpreting efficiency estimates but in identifying factors affecting technical efficiency
level (Henningsen & Henning, 2009).
Output

A
E
B

O C D Input

Figure 3.3. Illustration of a non-monotone production frontier

Source: Henningsen and Henning (2009).

Chapter 3. Technical efficiency in fisheries 54

Microeconomic theory also suggests that for profit maximisation, the production function should
be quasi-concave in all inputs (Lau, 1978). However, imposing quasi-concave constraints when
estimating production frontiers is not necessary. This is because measuring technical efficiency
generally assumes that a producer aims to maximise output for a certain bundle of inputs used
rather than maximise their profit per se. While not imposed, convexity can be tested ex post
(Henningsen & Henning, 2009).

There have been some approaches proposed in literature to impose monotonicity and quasi-
concave conditions. These approaches include Bayesian techniques (O’Donnell & Coelli, 2005)
and three-step procedure (Henningsen & Henning, 2009). However, the former approach is
complex to apply and demands requirements in econometrics and computer programming skills
(Henningsen & Henning, 2009) while the latter is considered to be a much simpler method.
Therefore, the three-step procedure method is increasingly used in empirical efficiency studies.
The application of this technique can be found in a number of efficiency studies in fisheries (see,
among others, Gedara, Wilson, Pascoe, & Robinson, 2012; Kularatne, Pascoe, Wilson, &
Robinson, 2018; Pascoe et al., 2012; Pascoe, Hutton, et al., 2017; Quijano et al., 2018).

The first step of the approach proposed by Henningsen and Henning (2009) is to estimate the
stochastic translog production frontier (as shown in section 3.3) and extract the unrestricted
parameters 𝛽̂ of the production frontier and their covariance matrix Σ̂𝛽 from the estimation results.

In the second step, a set of restricted ˆ 0 parameters is estimated by using a minimum distance
approach, defined as:

𝛽̂ 𝑜 = 𝑎𝑟𝑔𝑚𝑖𝑛(𝛽̂ 𝑜 − 𝛽̂ ) Σ̂𝛽−1 (𝛽̂ 𝑜 − 𝛽̂ )

subject to:

𝜕𝑓(𝑥, 𝛽 𝑜 )
≥ 0 ∀ 𝑖, 𝑥
𝜕𝑥

Chapter 3. Technical efficiency in fisheries 55

To find the revised set of coefficients 𝛽̂ 𝑜 that conform to the monotonicity assumption, a quadratic

programming can be used. In the third step, the stochastic frontier model is determined as:

𝑙𝑛𝑦𝑖 = 𝛼𝑜 + 𝛼1 𝑙𝑛𝑦̃ − 𝑣𝑖 + 𝜀𝑖

where the only input is the estimated frontier output based on the restricted parameters of the
restricted model 𝑦̃ = 𝑓(𝑥, 𝛽̂ 𝑜 ). The parameters α0 and α1 indicate final adjustments to the
parameter estimates. An advantage of the three-step approach is that the parameter values
estimated in the first step provide appropriate starting values while the variance-covariance matrix
limits the degree to which these parameters are altered when imposing monotonicity in the
nonparametric component (Pascoe, Hutton, et al., 2017).

The three-step method was used for imposing monotonicity condition on the frontier models in
this thesis. In particular, we applied this approach for technical efficiency studies undertaken in
open-access trawl fishery in Vietnam (Chapter 7) and banana prawn fishery in the NPF in Australia
(Chapter 6).

Chapter 3. Technical efficiency in fisheries 56

 PART 2:
FOUR CASE STUDIES

Part 2: Four case studies 57

Chapter 4. Meta-analysis of fisheries efficiency6

This chapter presents case study 1 (Activity 4) of the thesis. The main aim of chapter is to identify
factors, especially fisheries management tools and study specific characteristics that may affect
the variation of technical efficiency estimates across empirical studies conducted in fisheries. To
this end, a meta-regression model that is a subset of a meta-analysis approach is used. The first
two sections present the introduction and an overview of meta-analysis in efficiency. Methodology
including data collection and meta-regression model used is outlined in section 4.3. Section 4.4
presents results. Finally, the last section of this chapter provides a discussion and conclusion.

4.1. Introduction

Since the first technical efficiency study of fisheries by Kirkley et al. (1995), there has been
increasing attention to the role variations in technical efficiency within a fishery make to the
effectiveness of different management options, and also how management affects the overall level
of efficiency in a fishery. Technical efficiency is a necessary condition for economic efficiency
(Kumbhakar & Lovell, 2000), so changes in technical efficiency provide information about
changes in the relative economic performance of the industry. Many studies also consider factors
affecting technical efficiency, either management related or individual fisher related, providing
information to managers as to how, and the possible extent to which, efficiency may be improved
through management changes.

Technical efficiency, originally defined by Farrell (1957), can be measured in two ways: either as
input or output orientated measures. The latter indicates the ability of a fishing vessel to achieve
maximal output from a given set of inputs used while the former refers to the ability of a fishing
vessel to use minimum inputs to produce a given amount of output. Existing literature indicates
that information about the relationship between the level of inputs used in fishing and the resultant
catch is necessary for effective management (Pascoe et al., 2001; Tingley et al., 2003), as

6
The manuscript of this chapter has been submitted to Marine Resource Economics.

Chapter 4. Meta-analysis of fisheries efficiency 58

information on efficiency level and its drivers have important implications for both fishers and
fisheries policy makers. For instance, fishers can improve their performance and earnings by
improving their efficiency while policymakers could use this information to formulate appropriate
and effective regulations and policies for sustainable fisheries management.

However, technical efficiency estimates may also vary across many dimensions unrelated to vessel
characteristics or management. Such variation may also be an artefact of the methodologies used,
data used and other study-specific characteristics. Empirical studies in technical efficiency
undertaken in agriculture, aquaculture and other maritime sectors have shown that technical
efficiency values were affected by efficiency estimation techniques used, data types, and
functional forms among other factors (Bravo-Ureta et al., 2007; Iliyasu et al., 2014; Odeck &
Bråthen, 2012; Thiam et al., 2001). The efficiency literature elsewhere also shows mixed findings
and opposing ideas in regard to the merits of the various methodologies (Bravo-Ureta et al., 2007).
Similarly, empirical studies in fisheries efficiency have also reported different findings regarding
efficiency scores due to the use of different approaches (Herrero, 2005; Orea et al., 2005).

Fisheries are fairly unusual in terms of resource management as the management approach itself
may also impact on the level and measure of efficiency. Different fisheries are often managed by
adopting different management tools (e.g., input and output controls, open-access regime, and
ITQs, among others). These different management systems afford different levels of quasi-
property or use rights to the fishers. The use of ITQs has been widely advocated to improve
economic performance in fisheries (Chu, 2009) and it is generally believed to increase average
efficiency of the vessels (Costello & Deacon, 2007), with the least efficient vessels generally
leaving the fishery (see e.g., Dresdner et al., 2010; Kompas & Che, 2005; Pascoe et al., 2001;
Sharp et al., 2004). In contrast, open-access fisheries are likely to involve a larger mix of
heterogeneous vessels, with some being relatively efficient and others relatively less efficient,
leading to lower efficiency on average. This is a result of newer, more efficient vessels entering
an open access fishery without older less efficient vessels leaving (other than eventually being
forced out through economic losses). Direct comparison of efficiency scores across studies,
however, is difficult as the empirical estimates of technical efficiency are likely to also vary across
different studies due to differences in estimation methods used, the type of data used, number of
input and output used in the frontier models, and so on.

Chapter 4. Meta-analysis of fisheries efficiency 59

Within this context, meta-analysis seems to be an appropriate approach to identify reasons for the
variation of efficiency estimates across studies. The main idea behind the meta-analysis approach
is to combine results from an individually empirical study with the purpose of synthesising the
main findings (Glass, McGaw, & Smith, 1981). Over the past decades, meta-analyses have been
increasingly conducted in a wide range of subjects (Gurevitch, Koricheva, Nakagawa, & Stewart,
2018). These topics include water utility (Carvalho, Marques, & Berg, 2012); the valuation of
forest (Barrio & Loureiro, 2010; Lindhjem, 2007); the effect of institutions on output (Efendic,
Pugh, & Adnett, 2011); education cost (Colegrave & Giles, 2008); labour demand and income
elasticity (Gallet & Doucouliagos, 2014; Lichter, Peichl, & Siegloch, 2015); consumer behaviour
(Brons, Nijkamp, Pels, & Rietveld, 2008; Espey, 1998), impact of aid on growth (Mekasha & Tarp,
2013); relationship between debt and growth (Moore & Thomas, 2010); and many others. There
has also been a growing interest in meta-analysis studies of the efficiency literature, including
aquaculture (Iliyasu et al., 2014), agricultural activities (Bravo-Ureta et al., 2007; Mareth, Thomé,
Cyrino Oliveira, & Scavarda, 2016; Ogundari, 2014; Thiam et al., 2001), agribusiness (Djokoto &
Gidiglo, 2016), public transport operator (Brons, Nijkamp, Pels, & Rietveld, 2005), banking and
finance sectors (Aiello & Bonanno, 2016; Fall, Akim, & Wassongma, 2018; Irsova & Havranek,
2013), hospital (Kiadaliri, Jafari, & Gerdtham, 2013; Nguyen Kim & Coelli, 2009; Varabyova &
Müller, 2016), and seaport industries (Odeck & Bråthen, 2012).

Despite the relative popularity of technical efficiency studies in fisheries and increased interest in
meta-analysis in economics, to our knowledge no meta-analysis research has been done to date to
provide a broad overview of how efficiency scores vary with different attributes of a study, and in
regard to fisheries management system in particular. To fill this gap, we have undertaken a meta-
analysis of efficiency in fisheries to synthesise systematically the efficiency literature in this area,
and provide comprehensive knowledge about the major determinants of efficiency estimates. With
the increasing number of technical efficiency studies in fisheries, a meta-analysis study would be
helpful and timely for researchers, newcomers or early-career researchers and those whose
research is relevant to this field. Furthermore, as discussed above the overall aim of this thesis is
to examine the impact of fisheries management objective on the level and distribution of technical
efficiency. Therefore, information about the impact of fisheries management tools, which can be
seen as a proxy measure of fisheries management objectives, on technical efficiency would be
helpful to spotlight in examining the primary objective of the thesis.

Chapter 4. Meta-analysis of fisheries efficiency 60

Meta-regression model is used to address the following questions: (i) to what extent do fisheries
management tools affect the average level of efficiency? (ii) have technical efficiency indexes
changed over time (i.e., year of data used in primary study)? (iii) do mean technical efficiency
values vary according to different analytical methods used? (iv) does the number of inputs and
outputs used in frontier model have an impact on average efficiency levels?; and (v) does the
relative impact factor of the journals where such studies are published (a rough indicator of the
robustness of the study) have a relationship with average efficiency levels found in the studies?

4.2. An overview of meta-analysis in efficiency

Meta-analysis, first introduced by Glass (1976) is an approach for the systematic quantitative
summary of evidence from previously empirical studies (Nelson & Kennedy, 2009). The meta-
analysis approach has been widely applied in a variety of sectors, and its application to economic
area, including efficiency study, is not exceptional. Within the field of economics, there are about
200 meta-analysis studies conducted annually (Stanley et al., 2013). In economics, most meta-
analysis studies use a method referred to as meta-regression analysis (MRA) (Nelson & Kennedy,
2009; Stanley et al., 2013). The meta-regression analysis is an approach that uses all the estimates
of previous studies to explain the variation of these estimates based on differences across the
studies (Stanley et al., 2013; Stanley & Jarrell, 1989). It is the systematic review and quantitative
synthesis of a set of empirical studies on a given hypothesis, phenomenon, or effect 7 (Stanley et
al., 2013).

Most meta-analyses of efficiency have used meta-regression analysis to investigate the relationship
between the dependent variable (i.e., the efficiency estimate of the studies) and a set of specific
characteristics of the studies (e.g., efficiency estimation method, year of data used, year of
publication, data type: either cross-sectional data or panel data or time series). Thiam et al. (2001)
presented the first meta-analysis study in efficiency that focused on efficiency in agriculture sector
of developing countries. The study used 51 technical efficiency estimates retrieved from 32

7
For further information on meta-analysis, one may find it in a number of studies (e.g., Moher, Liberati, Tetzlaff, &
Altman, 2009; Nelson & Kennedy, 2009; Stanley & Doucouliagos, 2012; Stanley et al., 2013).

Chapter 4. Meta-analysis of fisheries efficiency 61

primary studies that evaluated agricultural efficiency in developing countries. A Tobit model was
used to identify factors affecting variation in technical efficiency levels among empirical studies.
Their results suggested that primal approach in efficiency estimation, the number of fixed inputs
and variable inputs used in the efficiency model increase average technical efficiency scores while
the use of cross-sectional data and Cobb-Douglas functional form produce lower TE scores.

The number of meta-analysis studies in efficiency has considerably increased since 2010. This
may stem from the increase in the number of efficiency studies that cover a wide range of research
fields ranging from agriculture to hospital to aquaculture to banking and many other sectors.
Efficiency meta-analysis research has been undertaken in a number of sectors, including banking
and finance, hospital, seaport, transportation, agribusiness and aquaculture (see Table 4.1). Among
those, meta-analysis studies that focused on efficiency of agriculture and farming sector are
probably dominant, following by banking and finance sector. This is understandable as the survey
of DEA application study conducted by Liu et al. (2013) found that the five-top industries where
most applications of DEA approach have been used were banking, health care, agriculture,
transportation, and the education sectors. Lampe and Hilgers (2015) also found that banking and
agriculture sectors were the most common application areas of both DEA and SFA approaches.

A summary result of meta-analysis studies conducted in productive efficiency field in a number of

sectors is presented in Table 4.1. It is notable that research on efficiency in fishery sector has
attracted great attention of researchers worldwide as mentioned above, thereby leading to an
increase in the number of studies in this sector. However, there has been no meta-analysis study
conducted in this field so far. As far as we are aware, this present study is the first meta-analysis
study conducted in the field of fisheries efficiency.

Chapter 4. Meta-analysis of fisheries efficiency 62

 Table 4.1. A summary of meta-analysis studies in efficiency

No. No. Time period8 Efficiency scores

Authors Industry/sector
studies obs. Mean TE Min Max
1 Thiam et al. (2001) Agriculture 32 51 1983-1998 0.680 0.17 1.00
2 Brons et al. (2005) Public transportation 33 93 1986-2002 0.825 0.31 1.00
3 Bravo-Ureta et al. (2007) Agriculture 167 569 1979-2005 0.766 0.17 1.00
4 Moriera Lopez and Bravo-Ureta (2009) Dairy farms 65 329 1986-2006 0.784 0.39 1.00
5 Nguyen Kim and Coelli (2009) Hospital 95 253 1987-2008 0.861 0.52 0.99
6 Ogundari and Brümmer (2011) Agriculture 124 164 1999-2010 0.724 0.21 0.99
7 Odeck and Bråthen (2012) Seaport industry 40 127 1995-2011 0.687 n/a9 n/a
8 Kiadaliri et al. (2013) Hospital in Iran 29 43 2000-2012 0.846 0.44 1.00
9 Irsova and Havranek (2013) Banking 14 119 1993-2003 0.670 0.34 0.91
10 Iliyasu et al. (2014) Aquaculture 36 55 1998-2012 0.690 0.24 0.88
11 Ogundari (2014) Agriculture in Africa 442 612 1984-2013 0.682 0.04 0.99
12 Djokoto and Gidiglo (2016) Agribusiness 33 70 Upon 2012 0.618 n/a 0.90
13 Mareth et al. (2016) Dairy farms 85 443 1987-2013 0.791 0.35 0.96
14 Aiello and Bonanno (2016) Banking 120 1661 2000-2014 0.699 n/a n/a
15 Fatima and Yasmin (2016) Farming production 43 71 1971-2014 0.730 0.11 0.96
16 Fall et al. (2018) Microfinance 38 262 2003-2016 0.611 n/a n/a
17 Aiello and Bonanno (2018) Banking 120 1661 2000-2014 0.700 n/a n/a
18 Present study Fisheries 103 326 1995-2018 0.717 0.291 0.98

8
The time when primary study is published or undertaken.
9
Not available.

63
4.3. Methods

4.3.1. Meta-dataset

Meta-analysis is considered as “the analysis of analysis” (Glass, 1976), with the collection
relevant studies in efficiency in fishery is, therefore, an important first step (Fall et al., 2018).
To this end, an intensive search for fisheries technical efficiency studies was conducted using
various scientific search engines such as Scopus, Web of Science and Google Scholar, and
other online database using related keywords. The search was used to find original research
papers published in English language journals up to August 2018. All English peer-reviewed
journal papers related to technical efficiency in fisheries were selected for review in the first
stage. Further, relevant papers were also identified from the bibliographic reference lists of
those research papers identified in the first stage.

Our dataset for the meta-analysis study included book chapters, working papers, conference
papers, PhD and Master theses that were not subsequently published in a journal. A diversity
of studies with larger geographical areas (e.g., country and continent) would provide a better
knowledge about the association between study specific characteristics and efficiency scores
obtained from the studies (Thiam et al., 2001).

In the second stage, we sorted these papers and eliminated those that did not provide
information needed for the current study. Only papers that were written in English were
retained. Only studies in which technical efficiency scores were provided or could be calculated
from the reported values in the primary studies (see e.g. Odeck & Bråthen, 2012), and other
information needed for the present study were included in the current analysis. That is, if
technical efficiency values were not presented or only presented graphically then these studies
were eliminated from the data set. Where authors reported the same results of a given efficiency
analysis in more than one journal article the technical efficiency scores were taken from only
one of the published studies.

From each paper, a range of data was collected for use in the meta-regression model. These
included year of publication; the technical efficiency values; fisheries management system (i.e.,
open access, ITQ, and input and TAC controls); efficiency estimation methods (DEA, SFA or
SDF); the number of input and output variables used in the model; mean of data year used;
journal ranking where the paper was published, and so on. A number of papers applied several

Chapter 4. Meta-analysis of fisheries efficiency 64

approaches in a given study, or considered multiple fleets, and hence multiple values were
extracted from some papers. However, only technical efficiency values that were generated by
DEA, SFA and SDF were included in the study. While other approaches such as stochastic ray
production function, bootstrapped DEA, and slack-based measure model approaches were
among those used in fisheries efficiency analysis (see e.g., Fousekis, 2002; Herrero, 2005;
Herrero et al., 2006; Walden, 2006), relatively few cases of these were applied in fisheries
efficiency estimation resulting in too few observations to provide a meaningful analysis. As a
result, the basic comparison regarding methodologies used was limited to the three main
methods, namely DEA, SFA and SDF approaches.

4.3.2. Meta-regression model

The meta-regression model is expressed as TEi  o   j X ij   i where 𝑇𝐸𝑖 is the dependent

variable, representing the mean technical efficiency score provided in study i; 𝑋𝑖𝑗 represents
study specific characteristics. The errors are represented by 𝜀𝑖 . The definition used in the model
is provided in Table 4.2.

Table 4.2. Definition of variables in the meta-regression model

Variables Description
TE Continuous Technical efficiency estimate
Management Dummy Whether open access, ITQs, or other measures (i.e.,
input and/or competitive TAC controls) used in
fishery management. “Others” was reference
variable.
Data year Continuous Year of the data used in a primary study.
Method Dummy Whether DEA or SFA or SDF approach applied in
primary study. DEA was the reference variable.
Input Continuous Number of production inputs incorporated in the
model
Output Continuous Number of production outputs incorporated in the
model.
Journal ranking Continuous Journal ranking in 2017 based on SCImago Journal
Rank (SJR)10

10
“SCImago Journal Rank (SJR) is a measure of the number of times an average paper in a particular journal is
cited, and as such is conceptually similar to the Impact Factor. A major difference is that instead of each citation
being counted as one, as with the Impact Factor, the SCImago Journal Rank assigns each citation a value greater
Chapter 4. Meta-analysis of fisheries efficiency 65
The data used in the studies included both cross-sectional data and panel data. If data used in
primary study was cross-sectional, the data year is the year of period covered by the data. This
is not necessarily exactly the same year when the data was collected or the primary study was
published (Djokoto & Gidiglo, 2016). However, for the studies that use time series and panel
data or several years of survey data, the data year is the mid-point of the years of the data
(following previous studies e.g. Djokoto & Gidiglo, 2016; Ogundari, 2014).

In terms of journal ranking, for papers published on journals that have no journal ranking,
conference papers, book chapters, working papers and PhD dissertations, the journal ranking
of those papers was set to zero (e.g. Aiello & Bonanno, 2018; Ogundari, 2014).

As mean technical efficiency values are truncated between zero and one, the application of
Tobit model (Greene, 2008a) is widely used for limited and truncated distribution of the
technical efficiency variable. This approach is commonly applied in other meta-analysis studies
of technical efficiency undertaken in other industries (e.g., Bravo-Ureta et al., 2007; Brons et
al., 2005; Odeck & Bråthen, 2012; Thiam et al., 2001). However, although TE is truncated at
one, average TE is less than one (unless every vessel is equally efficient) and Tobit models are
not necessarily appropriate. Ordinary least squares (OLS) is also applied for comparison
purpose. Stanley and Jarrell (1989) suggested that meta-regression errors are likely to be
heteroscedastic. This is because a number of papers reported multiple efficiency values while
sharing the same variables such as inputs, outputs, fisheries management, data year, and other
factors. A common approach widely used in meta-regression analysis to address the problem
of potential heteroscedasticity is to estimate robust standard errors. This technique has been
intensively applied in a number of previous meta-regression studies in technical efficiency (see,
amoung others, Bravo-Ureta et al., 2007; Brons et al., 2005; Odeck & Bråthen, 2012) and in
other meta-regression studies in other subjects (e.g., Barrio & Loureiro, 2010; Lindhjem,
2007). This approach to address the problem of heteroscedasticity in meta-regression model is
also applied in our study.

or less than 1.00 based on the rank of the citing journal. The weighting is calculated using a three-year window
of measurement and uses the Scopus database”.
Journal ranking is available at https://www.scimagojr.com/journalrank.php (accessed on 11 July 2018).

Chapter 4. Meta-analysis of fisheries efficiency 66

4.4. Results

4.4.1. Overview of the identified papers

In total, 103 papers from 1995 to 2018 were identified that met the criteria for the present meta-
analysis. These included 85 (82.5%) research articles and 18 (17.5%) other studies (i.e., book
chapters, conference papers, working papers, and PhD theses). The meta-data used in our study
consists of 326 observations for meta-regression analysis as a number of papers provided more
than one efficiency estimate.

The identified journal articles were published in 47 different journals. However, there are seven
journals that attracted the greatest number of identified papers (i.e., more than four papers per
journal). These include Marine Resource Economics, Marine Policy, Fisheries Research, ICES
Journal of Marine Science, Applied Economics, Environment and Development Economics,
and Journal of Productivity Analysis (Figure 4.1).
Journals

Others 18

Other journals 35

Journal of Applied Science 2

Journal of Agricultural Economics 2

Fisheries Science 2

European Journal of Operational Research 2

Applied Economics Letters 2

Journal of Productivity Analysis 4

Environment and Development Economics 4

Applied Economics 4

ICES Journal of Marine Science 5

Fisheries Research 6

Marine Policy 8

Marine Resource Economics 9

0 5 10 15 20 25 30 35 40
Papers

Figure 4.1. The number of efficiency papers published in each journal

The identified papers were undertaken in fisheries from 30 countries across all continents (i.e.,
Africa, America, Asia, Europe and Oceania continents as shown in Figure 4.2). This indicates
that the efficiency research in fisheries is of increasing interest and concern to diverse

Chapter 4. Meta-analysis of fisheries efficiency 67

researchers throughout the world. Among these countries, the USA, Australia and Spain
attracted the most studies with 19, 11 and 7 papers, respectively. This is followed by Denmark
and Indonesia with six papers each. Malaysia, Nigeria, UK and Vietnam had the same number
of papers undertaken in their fisheries with five papers each.

Turkey 1
Country

Sri Lanka 1
Portugal 1
Philippines 1
Norway 1
New Zealand 1
Japan 1
India 1
Ghana 1
Georgia 1
Fiji 1
Ecuador 1
Chile 1
Tanzania 2
Sweden 2
Mexico 2
Bangladesh 2
South Korea 3
Italy 3
Iran 3
Greece 4
Vietnam 5
UK 5
Nigeria 5
Malaysia 5
Indonesia 6
Denmark 6
Spain 7
Australia 11
USA 19
0 2 4 6 8 10 12 14 16 18 20

Number of papers

Figure 4.2. The number of efficiency papers undertaken in each country

The thirty countries are also categorised into four income groups including high-income
countries (HICs), low-income countries (LICs), lower-middle-income countries (LMICs), and
upper-middle-income countries (UMICs) (World Bank, 2018). Although fisheries efficiency
studies were conducted in a wide range of countries, ranging from high-income countries to
low-income countries, the number of efficiency fisheries studies in high-income and lower-
middle-income countries is dominant, accounting for 63.1% and 22.3%, respectively (Table
4.3). Fisheries efficiency studies in Europe, Asia and America were also dominant with 30.1%,

Chapter 4. Meta-analysis of fisheries efficiency 68

27.2% and 22.3% respectively, while efficiency studies in fisheries in Africa contributed least
to the total number of papers (about 7.8%).

Table 4.3. The number of papers undertaken in each continent and country11
Continent/
HICs LICs LMICs UMICs Total Proportion (%)
Income level
Africa 0 2 6 0 8 7.8
America 20 0 0 3 23 22.3
Asia 4 0 16 8 28 27.2
Europe 29 0 1 1 31 30.1
Oceania 12 0 0 1 13 12.6
Total 65 2 23 13 103 100
Proportion (%) 63.1 1.9 22.3 12.6 100

35
No. papers

30
29
30
26
25

20
15
15

10

5 3

0
1995-2000 2001-2005 2006-2010 2011-2015 2016 onward
Years

Figure 4.3. The number of papers from 1995 to 2018

Since the application of stochastic frontier in fisheries conducted by Kirkley et al. (1995) and
now regarded as a seminal paper in the field (Lampe & Hilgers, 2015), the number of efficiency
studies in fisheries has increased over the past two decades (as shown in

11
The information about income country group is available at https://data.worldbank.org. (Accessed 14 August
2018).
High-income countries: Australia, Chile, Denmark, Greece, Italy, Japan, Korea, New Zealand, Norway, Portugal,
Spain, Sweden, UK and USA.
Low-income country: Tanzania.
Lower-middle income countries: Bangladesh, Georgia, Ghana, India, Indonesia, Nigeria, Philippines, Sri Lanka
and Vietnam.
Upper-middle-income countries: Ecuador, Fiji, Iran, Malaysia, Mexico and Turkey.
Chapter 4. Meta-analysis of fisheries efficiency 69
Figure 4.3). There were only three efficiency studies before 2000. However, after 2000 there
were more than 20 papers produced in every five-year period.

35
No. papers

SDF DEA SFA

30

25
18 15
19
20

15

10 8
12
10 12
5
7
3 3 4
0 2
1995-2000 2001-2005 2006-2010 2011-2015 2016 onward
Years

Figure 4.4. The application of efficiency estimation approach over the past years12

Prior to 2000, all studies used only the conventional stochastic production frontier approach,
SFA. The applications of DEA and SDF in fisheries started later. While both DEA and SFA
are common approaches for efficiency estimation in fisheries the number of papers that used
SFA approach (63 papers) is higher than the number of studies that used DEA (41 papers)13
(Table 4.4 and Figure 4.4). The stochastic nature of harvesting fisheries such as weather and
resource conditions may explain why SFA is a preferable approach. In contrast, while also
stochastic, the application of SDF approach is relatively limited (only nine papers used this
approach). This may be because it is subject to difficulties such as multi-output function and
problems of multi-collinearity (Pascoe & Tingley, 2007). There has been no study used SDF
approach for fisheries efficiency study over the past two decades in Africa and Asia (Figure
4.5). Possible reasons may primarily stem from data-poor fisheries in these continents and the
complexity of the SDF approach compared to DEA and SFA approaches.

12
Note: A paper may have used more than one approach.
13
A number of studies used more than one approach so the total number of applications of the efficiency
estimation is not necessary equal to total selected papers (103).
Chapter 4. Meta-analysis of fisheries efficiency 70
 40
SDF DEA SFA
No. papers
35

12
30

25
17
20
17
15 20

10
9
13
5 5
8
5 3
3
0 1

Africa America Asia Europe Oceania

Continent

Figure 4.5. The number of papers and efficiency estimation approaches used by continent

Regarding fisheries management tools, the majority of fisheries in the studies were managed
by using input controls and /or comparative TAC (75.2%), reflecting the dominance of this
management system (Table 4.4). ITQs and open-access regimes are less commonly used in
fisheries in the reviewed papers. Interestingly, the application of open access, others (i.e., input
and/or comparative TAC controls) and ITQs likely depend on country-income status.
Specifically, low-income countries tend to use open-access (e.g., no property rights) regime
while ITQs (strong use rights or quasi-property rights) are only used in high-income countries
(Figure 4.6).

Input and TAC controls ITQ Open access

100%
34
90%
80% 9
70%
60% 35
50% 3
225
40%
30% 13
20%
10% 7
0%
LICs LMICs UMICs HICs
Income level

Figure 4.6. The proportion of application of fisheries management tools

Chapter 4. Meta-analysis of fisheries efficiency 71
A summary of variables used in the meta-regression model is given in Table 4.4. Summary of
variables in the meta-regression model. Overall, the mean technical efficiency over all studies
was 0.717, suggesting that there is room to increase efficiency levels in fisheries.

Table 4.4. Summary of variables in the meta-regression model

No. of Technical efficiency

Variables TE est. Prop. (%)
studies Min Max Mean
TE 326 100 103 0.291 0.980 0.717
Fisheries management 326 100 108a
- ITQs 34 10.4 12 0.414 0.955 0.743
- Open access (OA) 47 14.4 28 0.360 0.902 0.679
- Input & TAC controls 245 75.2 68 0.291 0.980 0.721
Data year 1988 2016 2002
Estimation method 326 100 113a
- DEA 172 52.8 41 0.291 0.741 0.960
- SDF 29 8.9 9 0.394 0.589 0.850
- SFA 125 38.3 63 0.355 0.714 0.980
Number of inputs 1.000 9.000 3.700
Number of outputs 1.000 7.000 1.580
Journal ranking 0.000 3.521 0.721
a
: A number of papers used more than one efficiency estimation approaches, and more than one fisheries
management tools can be found in a given fishery. Hence, total number of estimation approaches and
fisheries management tools is not necessarily equal to total number of studies (103 papers).

4.4.2. Meta-regression results

The meta-regression results using the OLS and the Tobit approaches are presented in Table
4.5. Both models produce similar patterns in term of both magnitude and sign of the
coefficients14, although the Tobit model is generally considered the most appropriate given the
censored dependent variable (Bravo-Ureta et al., 2007). Following Bravo-Ureta et al. (2007)
and Mareth et al. (2016), the interpretation of the findings is primarily based on the results
produced by the Tobit model.

14
In our dataset the dependent variable (technical efficiency score of empirical study) is not censored at zero and
1 as no empirical study has average efficiency score equal to zero or 1.
Chapter 4. Meta-analysis of fisheries efficiency 72
Table 4.5. Meta-regression of mean technical efficiency in fisheries

OLS model Tobit model

Variables
Robust Std. Robust
Coefficient Coefficient
error Std. error
Constant 8.4894** 2.8452 8.4894** 2.8056
ITQs 0.0736** 0.0201 0.0736** 0.0198
Open access -0.0345 0.0259 -0.0345 0.0255
Data year -0.0039* 0.0014 -0.0039* 0.0014
SDF -0.1715*** 0.0268 -0.1715*** 0.0264
SFA -0.0133 0.0191 -0.0133 0.0188
Number of outputs 0.0211* 0.0060 0.0211* 0.0059
Number of inputs 0.0237** 0.0063 0.0237** 0.0062
Journal ranking -0.0367* 0.0162 -0.0367* 0.0160
R2 0.171
F-statistic 8.156***
Log-likelihood 171.7
Wald-statistic 67.1***
Number of obs. 326 326
***, **, * Significant at 0.1%, 1%, and 5% level respectively

Dummy variables were used in the model for ITQ and open-access management, with the
effects of input and TAC control management captured within the constant. From the results,
the coefficient relating to ITQs was positive and significant, suggesting that ITQ fisheries are
characterised by higher levels of mean efficiency than fisheries with other management types,
as expected. The variable relating to open access was negative but non-significant. The
negative coefficient of open access implies that fisheries under open access achieved lower
mean efficiency than input controls fisheries. The category “input and/or TAC controls” cover
a wide range of management measures with varying strengths of use rights, from weak rights
under limited access to reasonably strong rights under individual transferable fishing effort
system. Hence, the negative but non-significant coefficient for open-access management may
suggest that mean efficiency increases with the strength of use rights.

Chapter 4. Meta-analysis of fisheries efficiency 73

The data year variable negatively affected mean TE and was statistically significant at the 5%
level. While this suggests that the reported TE values in fisheries have decreased over time, it
may also be an artefact of the shift in studies from mainly industrial fleets in the earlier years
to more artisanal fleets in later years.

With regard to the effect of data year (either data year or the year of the study published) on
efficiency, existing efficiency meta-analyses found mixed results. Some found a significant
positive relationship between efficiency and data year used in primary studies or year of
publication while others found negative or no relationship between mean efficiency and data
year. For instance, Ogundari (2014) conducted a meta-analysis on agricultural efficiency in
Africa using 442 studies with 612 efficiency estimates from 1984 to 2013, and found that mean
efficiency in the selected studies decreases significantly over time. Similarly, Odeck and
Bråthen (2012) found that average efficiency scores have decreased over time when they
conduct efficiency meta-analysis of seaport industry from 1995 to 2011. By contrast, the
positive relationship between efficiency and year of data used in primary studies can be found
in a number of efficiency meta-analysis studies, including in agriculture and microfinance (Fall
et al., 2018; Ogundari, Amos, & Okoruwa, 2012; Ogundari & Brümmer, 2011). Further,
positive correlation between year of data and efficiency scores are also explored by previous
meta-analysis studies undertaken in other industries such as aquaculture efficiency (Iliyasu et
al., 2014) and agriculture in developing countries (Thiam et al., 2001).

The efficiency estimation techniques used were divided into three categories: DEA, SFA and
SDF, with the former treated as the reference group in the meta-regression model. From the
results, the negative and statistically significant coefficient of the SDF variable indicates that
studies using multiple output stochastic approaches produce lower mean TE than non-
parametric DEA models, all else being equal. However, there was no significant difference
between mean TE generated by SFA and DEA approaches. Similar findings have been shown
in other meta-analysis studies in agriculture and hospital sectors (e.g., Nguyen Kim & Coelli,
2009; Thiam et al., 2001), while others have found a statistically significant difference between
estimates of TE from both SFA and DEA. For example, Iliyasu et al. (2014) found that SFA
models yield lower TE scores than their non-parametric DEA counterparts in the aquaculture
industry, while similar results have been found in studies in agriculture (Bravo-Ureta et al.,
2007); dairy sector (Moriera Lopez & Bravo-Ureta, 2009); seaport industry (Odeck & Bråthen,
2012); and banking (Aiello & Bonanno, 2016).

Chapter 4. Meta-analysis of fisheries efficiency 74

As efficiency also captures the effects of inputs not fully captured in the production function
(Pascoe & Coglan, 2002), a priori there may be a relationship between the number of inputs
and outputs in the frontier production function models and the mean TE. The results suggest
that the number of inputs and outputs both positively affect mean TE at significance of 0.1%
and 1% level, respectively. These findings are in line with those other studies (Bravo-Ureta et
al., 2007; Brons et al., 2005; Djokoto & Gidiglo, 2016; Thiam et al., 2001). However, in
fisheries, many inputs are correlated (e.g., larger boats usually have larger engines, larger
crews, tow (or carry) more fishing gear and often fish for a longer period), so including too
many inputs may produce other issues for TE estimation.

With regard to the association between journal ranking and efficiency estimates, we included
journal-ranking variable into the meta-regression model to investigate whether the efficiency
values depend upon the kind of journals in which papers appear. Interestingly, the results show
that there was significant difference between efficiency values from studies published on
different journals. Results from Table 4.5 imply that studies published in journals with higher
journal ranking reported significantly lower efficiency scores. Other things being equal, studies
with lower efficiency scores tend to be published in high-ranking journals, and vice versa. This
finding is in line with other studies such as meta-regression study in banking (Aiello &
Bonanno, 2016, 2018), and agriculture efficiency meta-regression study (Ogundari, 2014).
Specifically, Aiello and Bonanno (2018) conducted a meta-analysis in banking efficiency by
using a data of 1661 efficiency estimates retrieved from 120 studies between 2000 and 2014.
The authors found that, with regard to non-parametric studies, the level of banking efficiency
decreases when journal ranking (i.e., impact factor of journal where the study was published)
increases. However, this finding is not held when it comes to studies that applied parametric
approaches. Specifically, parametric studies published in higher impact factor journals would
yield higher level of efficiency scores, meaning that there is a positive relationship between
efficiency score and journal ranking. Furthermore, Ogundari (2014) investigated that the level
of efficiency of papers that are published in journals had higher efficiency scores than papers
that are not published in journals (i.e., conference papers, working papers, and Master and PhD
theses). However, the authors also found that efficiency scores of studies that are published in
journals with impact factor are lower than papers published in journals with no impact factor.

Chapter 4. Meta-analysis of fisheries efficiency 75

4.5. Discussion and conclusion

The use of meta-regression analysis can identify trends that are not apparent in a single study
primarily as each fishery is subject to a limited set of conditions (e.g., a particular management
type or technology type). Of key interest to this study was the impact of strengthening use
rights on technical efficiency. However, no single study has been able to examine a fishery
transitioning from open access (no property rights) to ITQs (strong use rights or quasi-property
rights) to assess the impact of this on the efficiency of the fleet. An assumption underlying the
adoption of management systems such as ITQs is that it provides incentives for more efficient
fishers to buy out less efficient fishers, which should manifest itself in the form of an increase
in average fishery efficiency (as the tail end is removed).

The use of ITQs has been widely advocated as a means of achieving both economic and
biological sustainability objectives. It is generally believed to increase the average efficiency
of the fishery as the least efficient vessels would leave the fishery through the economic
incentives created by the improved quasi-property rights associated with the quota use right
(see, among others, Dresdner et al., 2010; Kompas & Che, 2005; Pascoe et al., 2001; Sharp et
al., 2004). Further, as each vessel or fisher is allocated a given catch, there is an incentive to
catch the quota at least cost (Dupont, 2000; Eggert & Tveterås, 2007). In our study, we found
that the estimated coefficient for ITQs variable was positive and significantly different from
other management tools including input and TAC controls, and open access, as expected.

By contrast, from our study, non-significant but negative sign of the parameter on the open-
access dummy variable may imply that fisheries managed under open access yield lower mean
TE than fisheries managed under input and TAC controls, as would be expected given that
input controls afford greater (but imperfect) rights than open access. It is well documented that
open-access fisheries result in the situation of overcapacity (“too many boats and fishers
chasing too few fish”) (Eggert & Tveterås, 2007; Ward, Kirkley, Metzner, & Pascoe, 2004).
With no mechanism for autonomous adjustment, non-malleability of capital (C. W. Clark,
Clarke, & Gordon, 1979) results in older, less efficient vessels remaining active in the fishery
as newer and more efficient vessels are introduced, resulting in a lower average efficiency than
a more efficient set of boats would reflect. As a result, open access fisheries are likely to involve
a large mix of vessels, with some relatively efficient and others relatively less efficient. In some
countries, less efficient vessels may also stay locked into an open-access fishery due to
subsistence, tradition, and employment (or lack of alternative employment) reasons, while in
Chapter 4. Meta-analysis of fisheries efficiency 76
others, open access is seen as a means of achieving broader livelihood objectives. The results
also may reflect the diversity of use rights embedded in input controls, some strong (for
example, under individual transferable effort units) and others weak (for example, license
limitations only). Many input control-based fisheries use a range of controls that may, in
combination, also affect the strength of the quasi-property rights afforded to the fishery.
Identifying the strength of quasi-property rights was not possible from the descriptions of the
management systems provided, although our results suggest that the mean TE may be a useful
measure of the strength of these rights.

The results of the meta-analysis suggest that the estimated mean efficiency scores and their
distributions is not independent of the approach used, so simple comparisons between studies
is not appropriate. Efficiency scores are relative measures; there will always be at least some
vessels that are more efficient than others within a fishery and it is these vessels that define the
frontier. In this regard, it is the distribution that provides the most information about the
effectiveness of management. A long tail suggests the persistence of inefficient vessels, most
likely reflecting the strength of any quasi-property or use rights in the fishery. Such a tail also
reduces the mean efficiency score, so this is an indicator also of the effectiveness of fisheries
management in achieving potential economic and sustainability objectives. Relatively few
studies reported information on the distribution of efficiency scores in the fishery. From our
analysis, this may be a further useful indicator as to the performance of management, and
should be encouraged in future studies.

The increase in the number of studies looking at technical efficiency in fisheries over the last
two decades is testament to the perceived usefulness of this information for informing fisheries
management. Through our meta-regression analysis of fisheries sector studies over the past
two decades, we have attempted to integrate the effects of study-specific attributes and fisheries
management tools on mean TE scores. The overview of these results may assist selection of
appropriate approaches for future measuring and modelling TE (Mareth et al., 2016) in the
fisheries sector, as well as aid in their interpretation.

Chapter 4. Meta-analysis of fisheries efficiency 77

Chapter 5. Sensitivity of technical efficiency to the choice of input
variables15

This chapter presents case study 2 (Activity 2) out of four case studies in this thesis. The main
objective of this chapter is to examine the sensitivity of technical efficiency to the choice of
input measures in fisheries efficiency model and efficiency model specifications. The first part
of this chapter provides an introduction. Section 5.2 and section 5.3 present data and
methodology used, respectively. Section 5.4 presents findings and discussion. Finally, the last
section of the chapter provides conclusions.

5.1. Introduction

The evaluation of technical efficiency needs the estimation of an appropriate production

frontier that is based on the set of inputs and outputs available in which the former are assumed
to influence the latter (Pascoe, Hassaszahed, et al., 2003). However, building up models to
illustrate such relation is not straightforward. The choice of right bundle of inputs in estimation
of production frontiers in fisheries can be seen as a challenging task faced by fisheries
economic researchers while this concern have been rarely addressed in fisheries literature
(Andersen, 2005; Pascoe & Mardle, 2003). The use of inappropriate input measures in the
production function may results in a bias in the measures of efficiency so management
decisions based on biased efficiency measures could lead to mismanagement and inappropriate
management strategies (Pascoe, Hassaszahed, et al., 2003).

Existing literature in fisheries efficiency shows that a wide range of inputs has been used in
fisheries efficiency studies. Generally, different studies use different types of input measures.
This may stem from the fact that fisheries data are limited in many cases. In particular, data on
the level of capital and labour used in fishing are relatively limited, and physical inputs such
as engine or length of vessel tend to be used in analysis of fisheries production and efficiency
(Pascoe, Hassaszahed, et al., 2003). Measures of boat capital are also difficult to obtain. Often
information on the physical size of the boat is available, along with the main engine power.

15
The manuscript of this chapter has been submitted to Journal of Productivity Analysis.
Chapter 5. Sensitivity of technical efficiency to the choice of input variables 78
However, these two variables are often highly correlated (bigger boats need bigger engines) so
using both seems to be problematic.

Common measures of fishing effort range from hours fished to just the presence/absence of
fishing on a particular day. Generally, the choice of input variables for efficiency analysis in
fisheries is, thus, often based on available data, characteristics of fisheries (e.g. gillnet, trawl
fisheries), and is relatively different among studies (Andersen, 2002; Greenville et al., 2006).
Specifically, some studies used engine size (Duy & Flaaten, 2016; Greenville et al., 2006;
Kompas et al., 2004; Oliveira et al., 2016; Pascoe, Hutton, et al., 2017) while some other studies
used hull’s length and/or width, and deck area (Fousekis & Klonaris, 2003; Ngoc et al., 2009;
Pascoe, Innes, et al., 2017; Squires & Kirkley, 1999), or GRT as fixed inputs (Jeon et al., 2006;
Squires et al., 2003; Susilowati et al., 2005; Viswanathan et al., 2001), or fishing gear (Duy &
Flaaten, 2016; Fousekis & Klonaris, 2003; Kompas et al., 2004; Lokina, 2009; Rust et al.,
2017). Some studies used a combined index that often mixed engine power and boat size (i.e.,
length and width of vessel’s hull) and/or fishing gear to capture the level of capital stock
invested and to avoid the problem of multi-collinearity in the models (Greenville et al., 2006;
Pascoe & Coglan, 2002; Vestergaard et al., 2003). It is noted that while there is no consensus
about which type of measures of inputs to include in efficiency analysis (Andersen, 2002), it is
important to examine the sensitivity of technical efficiency level when choosing different
measures of inputs used in technical efficiency analysis.

Additionally, when stochastic production frontier approach is used for efficiency estimation,
assumptions about functional form of the production function, the distribution of inefficiency
term, and time-variant efficiency need to be predetermined. Previous studies indicate that these
assumptions might affect the final measures of efficiency (e.g. Andersen, 2002; Herrero &
Pascoe, 2003; Sharp et al., 2004). Furthermore, previous results found in Chapter 4 indicated
that the number of inputs and outputs were also found to have a positive impact on technical
efficiency.

It is important to understand whether different input measures included in efficiency

estimation, and different model specifications would result in different results of efficiency
scores. Policies that are based on biased results and/or fragile inference may cause
inappropriate regulations and mismanagement, thereby unsustainable fisheries (Andersen,
2002, 2005; Valdmanis, 1992). The main objective of this study is to examine whether the

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 79

choice of input variables and assumptions imposed on the model may affect the estimates of
technical efficiency, to see if this had any bearing on the results in the Chapter 4.

5.2. Data

For the purposes of this study, data of the banana prawn fishery that is a sub-fishery of the
Northern Prawn Fishery in Australia were used. The data are again used for the technical
analysis in Chapter 6 with a different purpose. The main purpose in their use in this chapter is
to examine how input measures and different model specifications can affect the TE scores.
Detailed information about the fishery such as general information, fishery management and
fishery management objective is further described in Chapter 6.

The data were derived from logbook and vessel registry information covering the period 2010-
2015. The original data were daily catch and effort for each individual vessel, with the latter
recording total hours (including searching) and hours trawled separately. This was aggregated
into weekly values for each vessel, with a further measure of effort as days fished each week.
Other vessel characteristics such as engine power (kW) and length (m) were also available and
merged with the catch and effort data. While 52 licences are available for the fishery at any
one time, over the period examined 62 different vessels reported fishing activity. A summary
of the data is given in Table 5.1.

Table 5.1. Weekly data summary, banana prawn fishery 2010-2015

Min. 1st Qu. Median Mean 3rd Qu. Max.
Catch (kg) 111.0 4084.0 7704.0 9368.0 13075.0 37814.0
Length (m) 16.8 21.8 22.3 22.7 24.1 29.2
Engine power (kW) 262.0 342.0 410.0 397.5 450.0 526.0
Hours fished 11.0 66.0 94.0 92.5 117.0 167.0
Hours trawled 11.0 17.0 26.0 32.3 39.8 161.0
Days fished 1.0 5.0 6.0 5.9 7.0 7.0

All data were logged and normalised, such that their mean value was equal to zero (i.e. ln(𝑥) =
0). Correlation between the logged variables is given in Table 5.2. Engine power and vessel
length are highly correlated, so including both measures into a production function would be
inappropriate. Different models including these two fishing effort measures would be
examined. Correlation between the different effort measures is generally low, with only a

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 80

moderate correlation (i.e., 0.59) between the number of days fished per week and the total
number of hours per week (including searching time).

Table 5.2. Correlation between key variables (logged)

Ln(Trawl
Ln(Engine) Ln(Length) Ln(Hours) Ln(Days) Ln(Catch) Ln(Stock)
Hours)
Ln(Engine) 1.00 0.76 0.01 -0.01 0.06 0.09 -0.15
Ln(Length) 0.76 1.00 0.08 -0.03 0.10 0.04 -0.17
Ln(Hours) 0.01 0.08 1.00 0.59 0.28 0.05 -0.08
Ln(Days) -0.01 -0.03 0.59 1.00 0.24 0.22 0.12
Ln(Trawl
0.06 0.10 0.28 0.24 1.00 -0.02 -0.19
Hours)
Ln(Catch) 0.09 0.04 0.05 0.22 -0.02 1.00 0.62
Ln(Stock) -0.15 -0.17 -0.08 0.12 -0.19 0.62 1.00

As output from fishing is generally a function of the inputs employed by fishers and biomass
of fish available, so both information on fishing effort and fish biomass are ideally required
when modelling production functions (Hannesson, 1983; Pascoe & Tingley, 2007). Estimates
of biomass change over the season are not routinely undertaken for the banana prawn
component of the fishery, although are integral to the tiger prawn bio-economic model (Punt
et al., 2011). Zhou, Buckworth, Ellis, Deng, and Pascoe (2015) estimated biomass changes in
the banana prawn stock over the period 1981-2011 using a population depletion analysis which
also simultaneously estimated natural mortality and changes in fishing power. In this analysis,
we employ a productivity based approach, namely the Malmquist productivity index (Caves,
Christensen, & Diewert, 1982) and more specifically the technical change component of the
Malmquist index. This was shown by Pascoe and Herrero (2004) to provide a useful stock
index for the purposes of subsequent estimates of fisheries production frontiers. The index is
not purely an index of stock size, but also captures changes in availability due to factors such
as luminosity.

5. 3. Methods

Methods applied in this study comprise two steps. In the first stage, technical efficiency scores
were derived from different stochastic production models in which each model includes input
variables (i.e., one variable input, one fixed input and fish stock index) under different model
specifications. These efficiency scores were then regressed on a number of factors in the second

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 81

step to examine how the choice of input variables and model specifications can affect technical
efficiency scores.

5.3.1. Stochastic frontier analysis

Stochastic frontier analysis approach presented in detail in Chapter 3 was used for efficiency
estimation. A number of models were estimated in the first step. In this stage, the use of output
and input variables in the stochastic frontier models as well as model specifications are
discussed as follows.

In terms of output used in the stochastic model, the output measure in this study is catch
volume. Generally, two measures of output are widely used that include catch weight and catch
revenue. However, according to Pascoe and Tingley (2007) the use of catch weight is widely
used for a single-species fishery while catch revenue is commonly applied when fishers operate
in multi-species fisheries. Further, a number of previous studies have found that output variable
measured by whether catch volume or catch value produced quite similar technical efficiency
levels (Andersen, 2002; Herrero & Pascoe, 2003). Therefore, catch volume was used as the
measure of output in this study.

Regarding input variables, the input measures used can be economic measure of capital
invested in fishing (e.g., vessel, fishing gear), labour (e.g., crew size), physical measure of fixed
input used (e.g., engine power, gross registered tonnage, vessel length, onboard equipment),
and fishing time (e.g., fishing hour, trawled hour, fishing day, trip). However, the information
on economic measures of capital invested in vessel and fishing gear are often unavailable. Crew
size is often used as a measure of labour in fishing, but this variable is often highly correlated
to boat size as bigger boats have more crew members employed. Instead, common physical
measures of inputs used are either engine power or length of vessel or gross registered tonnage.
This is because these measures are often readily available for most of the fleet (Pascoe,
Hassaszahed, et al., 2003). The measure of fishing time can be either days fished (days at sea)
or hour fished or fishing trip (e.g. Duy & Flaaten, 2016; Jamnia, Mazloumzadeh, & Keikha,
2015; Kirkley et al., 1995; Ngoc et al., 2009).

In this study, information about fixed inputs including engine and vessel length that can be
seen as measures of capital invested is available. The fishing time that includes day fished,
hour fished (including searching time) and hours trawled is also available. It is notable that
these three measures of fishing time are different. The day fished is the number of days at sea

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 82

that includes the time for fishing and other activities such as travelling to fishing ground and
return to the port. The difference between hour fished and hour trawled is the time for searching
fish stock. Hour trawled is factual time when the trawl gear is in use while hour fished includes
the time for actual trawling time and time to search for fish stock. Furthermore, fish stock
biomass index is available in this fishery, so this variable was included in the efficiency
analysis.

Due to high correlation between the input variables, each model included only a fixed input
and a fishing time variable as well as a stock index variable. The fixed input in this analysis
was either engine power or vessel length. The fishing time would be either hours trawled or
hour fished (including searching time) or days fished.

Different model specifications using both translog against Cobb-Douglas under different
assumptions (i.e., time-variant efficiency and time-invariant efficiency, truncated normal
distribution and half normal distribution of inefficiency component) were estimated. A
combination of different input used under different functional forms and model specifications
(inefficiency distribution and time-invariant inefficiency assumptions) resulted in 48 models,
named Model 1 (M1) to Model 48 (M48) as given in Table 5.3 below. As a result, 48 efficiency
estimates retrieved from the 48 different models were treated as dependent variables in the
regression analysis model (stage 2). There were 24 translog models and Cobb-Douglas models.
On the translog side, there were 12 models with time-variant specifications and 12 models with
time-invariant specifications. Under translog functional form with time-variant and truncated
distribution of inefficiency, six models with a different combination of different inputs used
(either engine or hull length and either hour trawled or hour fished or day fished and fish stock)
were estimated. Note that in all models, stock index was included as input variable. The
“taxonomy” of the combination different model specifications using different inputs, Model 1
(M1) for an example, is explained as follows. The production function of Model 1 included
engine size, hour trawled and fish stock index under translog form with time-variant and
truncated distribution assumptions of efficiency.

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 83

 Table 5.3. Taxonomy of the stochastic frontier models

Translog Cobb-Douglas
Input Variables Time-variant Time-invariant Time-variant Time-invariant
Half- Half- Half- Half-
Truncated Truncated Truncated Truncated
Normal Normal Normal Normal
Hour trawled M1 M7 M13 M19 M25 M31 M37 M43

Engine power Hour fished M2 M8 M14 M20 M26 M32 M38 M44

Day fished M3 M9 M15 M21 M27 M33 M39 M45

Hour trawled M4 M10 M16 M22 M28 M34 M40 M46

Hull length Hour fished M5 M11 M17 M23 M29 M35 M41 M47

Day fished M6 M12 M18 M24 M30 M36 M42 M48

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 84

5.3.2. Regression analysis model

In order to examine the variation of technical efficiency estimates due to the use of different
input variables under different assumptions regarding functional form, inefficiency distribution
and time-invariant inefficiency, regression analysis model was applied. The regression model
is expressed as 𝑇𝐸𝑖 (𝑆𝐷𝑖 ) = 𝛽0 + 𝛽𝑗 𝑋𝑖𝑗 + 𝜀𝑖 where 𝑇𝐸𝑖 and 𝑆𝐷𝑖 are dependent variables,
representing the technical efficiency score and the standard deviation of technical efficiency
produced by model i, respectively. 𝑋𝑖𝑗 represents the matrix of covariates implying the model
specific characteristics. The errors are presented by 𝜀𝑖 . Table 5.4 presents the definition of
variables used in the regression model.

Table 5.4. Variables used in the regression model

Variable Description
TE: Technical efficiency; SD: Standard deviation
TE, SD Continuous
of TE
Whether Engine power or Vessel length (Vessel
Fixed inputs Dummy
length is reference)
Whether Hour trawled or Hour fished or Day
Fishing time Dummy
fished (Hour trawl is reference)
Whether translog or Cobb-Douglas (translog is
Functional form Dummy
reference)
Whether time-variant or time-invariant efficiency
Time (in)variant efficiency Dummy
(Time-variant specification is reference)
Whether truncated normal distribution or half-
Efficiency distribution Dummy normal distribution (Truncated distribution is
reference)

5.4. Results and discussion

5.4.1. Model specification tests

This section describes the results of model specification tests conducted to examine different
hypotheses imposed on the models evaluated. Several tests were undertaken on the functional
forms of the model by imposing restrictions on parameters. The restrictions were tested by
applying the likelihood ratio (LR) test. The test is defined by 𝐿𝑅 = −2 [ln[𝐿(𝐻0 )] −
ln[𝐿(𝐻1 )]], where ln[𝐿(𝐻0 )] and ln[𝐿(𝐻1 )] indicate the values of the likelihood function under
the null (H0) and alternative (H1) hypotheses, respectively. The LR value has a chi-squared (χ2)
Chapter 5. Sensitivity of technical efficiency to the choice of input variables 85
distribution with the number of degrees of freedom given by the number of the restrictions
imposed (Coelli et al., 2005).

Translog functional forms were tested against Cobb-Douglas specifications. The results of the
test are presented in Table 5.5. The results show that the most appropriate functional forms in
all models are translog functional forms. Using the translog form (βkj ≠ 0), the test of time-
variant efficiency against time-invariant efficiency was also conducted. The results of the test
on null hypothesis of time-invariant efficiency was rejected (as shown in Table 5.6). This
implies that under translog functional forms the most appropriate assumption for inefficiency
component is time-variant inefficiency (η ≠ 0) for all models that used different inputs. The
assumption of half-normal distribution against truncated distribution of inefficiency was also
tested. The results presented in Table 5.7 show that half-normal distribution assumption of
inefficiency was found to be the most appropriate (μ = 0) for models that used engine power
and either hour trawled or hour fished or day fished. This assumption is also the most
appropriate for models in which hull length and hour fished were included as input variables
in efficiency models. However, truncated normal distribution is the most appropriate for
models in which hull length and hour trawled or day fished were used for analysis.
Furthermore, it is notable that an assumption about existence of a frontier was also tested for
all models (i.e., M1 to M48). In all cases, the hypotheses that γ = 0 was rejected16 (i.e. p <
0.01), indicating that a frontier and inefficiency existed in all models irrespective of used inputs
and model specifications. Technical efficiency scores and its standard deviation derived from
all models (i.e., M1 to M48) are given in Table 5.8. These efficiency scores were then regressed
to examine the sensitivity of efficiency scores to the choice of different input variables and
different model specifications.

16
The test for existence of a stochastic frontier was conducted with null hypothesis being specified as γ = 0, where
γ = 𝜎𝑢2 /(𝜎𝑢2 + 𝜎𝑣2 ) and the parameter, γ, lies between zero and one. If the null hypothesis (γ = 0) is not rejected,
this indicates that 𝜎𝑢2 equals zero and then the term 𝑢𝑖 should be removed from the model. The stochastic
production frontier then becomes the production function in which parameters can be consistently estimated using
ordinary least squares (Coelli, 1996).
Chapter 5. Sensitivity of technical efficiency to the choice of input variables 86
 Table 5.5. Tests of functional form (Translog versus Cobb-Douglas)

AIC Log-likelihood value Mean TE Standard deviation Test for functional form

Likelihood Degree
βkj ≠ 0 βkj = 0 Accept/
Input used βkj ≠ 0 βkj = 0 βkj ≠ 0 βkj = 0 βkj ≠ 0 βkj = 0 ratio test of P-value
(H1) (H0) Reject H0
(χ2) freedom
Engine power (Kw)
 Hour trawled 3242.535 3315.637 -1607.268 -1649.819 0.814 0.801 0.144 0.155 85.102 6 0.00 Reject H0
 Hour fished 3282.799 3316.509 -1627.400 -1650.255 0.799 0.786 0.147 0.158 45.710 6 0.00 Reject H0
 Day fished 3203.269 3253.782 -1587.635 -1618.891 0.790 0.782 0.158 0.167 62.512 6 0.00 Reject H0
Hull length (m)
 Hour trawled 3240.304 3321.682 -1606.152 -1652.841 0.802 0.768 0.152 0.170 93.378 6 0.00 Reject H0
 Hour fished 3284.530 3324.080 -1628.265 -1654.040 0.783 0.752 0.158 0.173 51.550 6 0.00 Reject H0
 Day fished 3194.097 3256.349 -1583.049 -1620.175 0.784 0.762 0.165 0.178 74.252 6 0.00 Reject H0

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 87

 Table 5.6. Tests of time-invariant efficiency (η = 0) versus time-variant (η ≠ 0) efficiency under βkj ≠ 0

AIC Log-likelihood value Mean TE Standard deviation Tests of time-invariant efficiency

Likelihood
η ≠ 0 η =0 Degree of
Input used η ≠ 0 η =0 η ≠ 0 η =0 η ≠ 0 η =0 ratio test P-value Accept/reject
(H1) (H0) freedom
(χ2)

Engine power
 Hour trawled 3242.535 3251.865 -1607.268 -1612.933 0.814 0.776 0.144 0.137 11.330 1 0.001 Reject H0
 Hour fished 3282.799 3290.302 -1627.400 -1632.151 0.799 0.765 0.147 0.142 9.502 1 0.002 Reject H0
 Day fished 3203.269 3215.960 -1587.635 -1594.980 0.790 0.747 0.158 0.152 14.690 1 0.000 Reject H0
Engine power
 Hour trawled 3240.304 3250.005 -1606.152 -1612.003 0.802 0.763 0.152 0.146 11.702 1 0.001 Reject H0
 Hour fished 3284.530 3292.256 -1628.265 -1633.128 0.783 0.746 0.158 0.154 9.726 1 0.002 Reject H0
 Day fished 3194.097 3207.292 -1583.049 -1590.646 0.784 0.748 0.165 0.159 15.194 1 0.000 Reject H0

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 88

 Table 5.7. Tests of truncated normal (μ ≠ 0) versus half-normal (μ = 0) distribution, (βkj ≠ 0 and η ≠ 0)
Standard
AIC Log-likelihood value Mean TE Test for efficiency distribution
deviation

Likelihood
μ≠0 μ=0 Degree of
Input used μ≠0 μ=0 μ≠0 μ=0 μ≠0 μ=0 ratio test P-value Accept/reject
(H1) (H0) freedom
(χ2)

Engine power
 Hour trawled 3242.535 3244.303 -1607.268 -1609.151 0.814 0.783 0.144 0.142 3.766 1 0.052 Accept H0
 Hour fished 3282.799 3283.499 -1627.400 -1628.749 0.799 0.768 0.147 0.146 2.698 1 0.100 Accept H0
 Day fished 3203.269 3204.111 -1587.635 -1589.055 0.790 0.760 0.158 0.157 2.840 1 0.092 Accept H0
Hull length
 Hour trawled 3240.304 3242.345 -1606.152 -1608.173 0.802 0.772 0.152 0.150 4.042 1 0.044 Reject H0
 Hour fished 3284.530 3284.685 -1628.265 -1629.342 0.783 0.750 0.158 0.156 2.154 1 0.142 Accept H0
 Day fished 3194.097 3195.994 -1583.049 -1584.997 0.784 0.755 0.165 0.162 3.896 1 0.048 Reject H0

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 89

 Table 5.8. Technical efficiency scores under different models

Translog Cobb-Douglas
Time-variant Time-invariant Time-variant Time-invariant
Input Variables
Truncated Half-Normal Truncated Half-Normal Truncated Half-Normal Truncated Half-Normal
TE SD TE SD TE SD TE SD TE SD TE SD TE SD TE SD
Hour
0.814 0.144 0.783 0.142 0.776 0.137 0.749 0.134 0.801 0.155 0.766 0.155 0.757 0.148 0.723 0.147
trawled
Engine power Hour 0.799 0.147 0.768 0.146 0.765 0.142 0.738 0.140 0.786 0.158 0.753 0.157 0.746 0.152 0.715 0.151
fished
Day fished 0.790 0.158 0.760 0.156 0.747 0.152 0.725 0.150 0.782 0.167 0.751 0.167 0.742 0.161 0.711 0.159
Hour
0.802 0.152 0.772 0.150 0.763 0.146 0.733 0.144 0.768 0.170 0.735 0.168 0.722 0.162 0.688 0.161
trawled
Hull length Hour
0.783 0.158 0.750 0.156 0.746 0.154 0.714 0.158 0.752 0.173 0.718 0.171 0.704 0.165 0.675 0.164
fished
Day fished 0.784 0.165 0.755 0.162 0.748 0.159 0.718 0.156 0.762 0.178 0.730 0.175 0.717 0.169 0.683 0.167

Note: SD means standard deviation of technical efficiency

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 90

5.4.2. Regression results

Results from regression analysis are presented in Table 5.9. The results indicate that different
inputs used in production function and different model specifications had significant impact on
technical efficiency and its standard deviation.

Regarding fixed input variable (either engine power or vessel length), fishing effort measured
in engine power resulted in higher efficiency and lower standard deviation of TE compared
with models that used vessel length as measure of input in production function. In an attempt
to investigate whether the different measures of fish stock inclusion in production function
would give different results, Andersen (2005) showed that in the short term, models with both
engine and length of vessel as input measures produce lower efficiency scores than models
with only length of vessel as input measure. However, in the long run the difference in
efficiency scores between those models does not exist (Andersen, 2005). In other study,
Andersen (2002) used different input and output measures to estimate technical efficiency of
Danish seiners operating in the North Sea and Skagerrak. Based on the results, the author
concluded that the level of efficiency scores might differ due to the use of different output
measures used (i.e., weight or revenue or revenue-weighted weight). Furthermore, the authors
also found that how inputs are included into the efficiency model seem unlikely to be important.
However, Pascoe, Hassaszahed, et al. (2003) used different measures of input in production
frontier to examine the impact of the use of different input measures on efficiency. Data on
different fishing fleets that operate in the North Sea areas are used for the analysis. The authors
concluded that different input measures used could produce considerably different efficiency
measures. Moreover, the authors argued that there is no superior measure between physical
measure and economic measure of input used in the production function that can produce better
or worse estimates than the other.

In respect to fishing time (either hour fished or day fished or hour trawled) the use of hour
trawled as fishing time produced higher technical efficiency and lower standard deviation of
TE than the use of the other measures (i.e., day fished and hour fished in this study). As
previously discussed, although fishing time is not a theoretical production input per se in
neoclassical production theory, this input is a common variable used in efficiency models in
fisheries, representing for capital utilisation (Pascoe, Hassaszahed, et al., 2003; Pascoe, Innes,
et al., 2017). Fishing time measures used in fisheries efficiency studies are different among

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 91

studies, and the selection of fishing measure mostly relies on available data. Fishing time can
be measured in the number of hours, the number of fishing days, the number of days per fishing
trip, trip length (e.g., week or day), and number of trips per season or month/year, and so forth.
However, most efficiency studies in fisheries use fishing day (measured by the number of days
at sea) as a measure of fishing time. The number of days at sea used as a fishing time measure
can be found in a number of fisheries efficiency studies. These include the Danish seine fishery
in the North Sea (Andersen, 2002, 2005); Dutch beam trawl fishery (Pascoe et al., 2001); small-
scale trawl fishery in Vietnam (Ngoc et al., 2009); offshore hand-line and gillnet fisheries in
Vietnam (Duy & Flaaten, 2016); the Northern prawn trawl fishery in Australia (Pascoe, Hutton,
et al., 2017); the Moreton Bay prawn trawl fishery (Pascoe, Innes, et al., 2017); and many
others. Detailed data on actual time of fishing (i.e., the time when fishing gear is used) is
unavailable in most fisheries. Instead, the number of days at sea or the number of days per
fishing trip is a common measure of fishing time. It is suggested that fishing time measured by
the time when the fishing gear is in use can be seen to be more appropriate (Andersen, 2002).

Furthermore, the results from Table 5.9 show that models with Cobb-Douglas functional form
produced lower efficiency scores, and higher standard deviation of technical efficiency scores
than those of translog models. Nguyen Kim and Coelli (2009) argued that the application of
more flexible functional form (i.e., translog model) tends to have higher efficiencies, ceteris
paribus, as more flexible functional form fits the data better. Existing literature shows that no
functional forms are superior to other functional forms, and there are no empirical reasons in
favour of a given functional form (Giannakas et al., 2003; Sauer, 2006). Thus, research on the
sensitivity of technical efficiency to the selection of functional form is crucial (Giannakas et
al., 2003). Research on the sensitivity of efficiency to the choice of functional forms is little
known in fisheries. However, there are a number of research undertaken in other fields, which
examine the effect of the choice of functional forms on efficiency scores (see for example
Battese & Broca, 1997; Giannakas et al., 2003; Zhu et al., 1995). These studies conclude that
technical efficiency is sensitive to the choice of functional form. Especially, Giannakas et al.
(2003) used data from olive growers in Greece to examine the effect of functional choice on
efficiency scores and the sources of technical inefficiency. The authors show that the choice of
functional form not only have significant impacts on efficiency score but affects significantly
the potential sources of inefficiencies so inappropriate policies or decision making could be
made if biased efficiency estimates retrieved from inappropriate functional forms are in use.
However, these findings conflict with other research by Ahmad and Bravo-Ureta (1996) who

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 92

evaluate technical efficiency of dairy farms in the USA using different functional forms and
model specifications. They find that Cobb-Douglas models produce the same efficiency
estimates (i.e., mean, minimum and maximum) as translog models do when all other things are
equal. The authors conclude that functional forms do not affect technical efficiency measures.

A number of meta-analysis research in efficiency explore the relationship between functional

forms and technical efficiency levels. However, the findings on this issue are opposing. For
example, Thiam et al. (2001) conducted a meta-analysis on technical efficiency of agricultural
sector, and found that the application of Cobb-Douglas functional form had lower average
technical efficiency values than those using translog functional form. The authors implied that
the use of more flexible functional form would produce higher average technical efficiency
scores. Several other meta-analysis studies on efficiency also confirmed this finding (e.g.,
Bravo-Ureta et al., 2007; Mareth et al., 2016). Nevertheless, other research show that the
impact of functional form on efficiency estimates is inconsistent and unclear. Some studies
explored that using Cobb-Douglas functional form yielded higher technical efficiency scores
compared to using translog functional form (e.g. Aiello & Bonanno, 2018; Iliyasu et al., 2014),
while some research concluded that there is no significant impact of functional form on
technical efficiency levels (see e.g. Ogundari, 2014; Ogundari & Brümmer, 2011).

The results in Table 5.9 also indicate that the assumption on distribution of efficiency has
significant impact on technical efficiency scores. Specifically, half-normal distribution of
efficiency produced lower efficiency scores and lower standard deviation of technical
efficiency scores than the truncated distribution, other things being equal. Regarding
assumption on distribution of technical inefficiency, the assumption on distribution of technical
inefficiency have no generally accepted criteria so estimating efficiency is often based on
several common forms of inefficiency distribution (Schmidt, 1985; Sharp et al., 2004). Several
specification models should be made and then the preferred model is selected using likelihood
ratio tests (Coelli, 1996). It is helpful to examine the effect of different assumptions on
efficiency as it can give signposts for researchers undertaking efficiency studies (Yane & Berg,
2013). So far, there have been a number of papers that examine the consistency and robustness
of technical efficiency levels under different assumptions on the distribution of inefficiency.
However, findings from previous studies that focus on examining the sensitivity of efficiency
to the assumptions of distributional inefficiency are mixed.

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 93

Sharp et al. (2004) evaluated technical efficiency of the rock lobster fishery in New Zealand
using a stochastic frontier analysis under three different assumptions about the distribution of
technical efficiency (i.e., half-normal, exponential and gamma distributions). They found that
the level of technical efficiency scores estimated using the different distributional assumptions
varied, from an average of 0.735 for the half-normal distribution, 0.797 for the exponential
distribution, and 0.814 using a gamma distribution. Other research on technical efficiency of
dairy farms in USA by Ahmad and Bravo-Ureta (1996) also found that technical efficiency
scores were influenced by the distributional assumption. Using stochastic frontier models, they
found that applying an assumption of a half-normal distribution resulted in about a 10% higher
average technical efficiencies (about 0.86) than those models with assumptions of truncated
normal distribution (about 0.76).

However, Greene (1990) estimated efficiency of the U.S. electricity sector by applying
stochastic cost frontiers. In his study, four models corresponding to four different distributions
of inefficiency (i.e., half-normal, truncated, gamma, and exponential distributions) are used.
The author concluded that all models provide the relatively similar mean efficiencies although
the gamma-distributed model results in a different distribution of inefficiency. Using the same
data with Greene (1990) but using a production function frontier model, Greene (2008b) again
found that mean inefficiency estimates are quite identical under four assumptions on
distribution of inefficiency, including half-normal distribution (0.0790), truncated-normal
distribution (0.0742), exponential distribution (0.0748), and gamma distribution (0.065).
Similarly, Chakraborty, Biswas, and Lewis (2001) measured technical efficiency of public
education in the U.S. by using stochastic production frontier model under two assumptions of
the distribution of inefficiency (i.e., half-normal and exponential distributions). The authors
found that efficiency estimates are not sensitive to the choice of distributional assumptions of
inefficiency. Yane and Berg (2013) applied a stochastic production frontier model to examine
the sensitivity of technical efficiency score rankings to the distributional assumptions of
inefficiency. A translog functional form is specified under four assumptions on distributional
inefficiency, including truncated normal, half-normal, gamma, and exponential distributions.
The authors found efficiency estimates generated from truncated-normal, exponential, and
gamma models are roughly similar (about 0.867) while only half-normal model provides a
slightly lower efficiency estimate (0.8121). The authors conclude that efficiency rankings are
consistent when four distributional assumptions of inefficiency are considered.

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 94

Concerning assumptions on time-invariant efficiency, results from Table 5.9 indicate that
models with time-invariant efficiency yield significantly lower efficiency than models with
assumptions of time-variant efficiency. Literature shows that there is no generally accepted
criteria to choose between time-variant or time-invariant efficiencies. Instead, a preferable
model should be chosen based on the results of likelihood ratio tests of variant models (Coelli,
1996). The finding in this analysis is inconsistent with other studies. For instance, Ahmad and
Bravo-Ureta (1996) compared technical efficiency measures yielded from different model
specifications using data from dairy farms in the U.S. The authors found that average technical
efficiency measures produced by models where technical efficiency was assumed time variant
are slightly lower than efficiency measures retrieved from models with time-invariant
inefficiency specifications.

Table 5.9. Regression analysis results

Dependent variable (TE) Dependent variable (SD)

Variables
Coefficients t-ratio Coefficients t-ratio
Intercept 0.796 291.077*** 0.154 186.480***
Engine power 0.022 10.615*** -0.011 -17.105***
Hour fished -0.015 -5.961*** 0.005 6.444***
Day fished -0.016 -6.133*** 0.012 15.113***
Cobb-Douglas -0.025 -11.976*** 0.012 19.489***
Time invariant -0.040 -19.285*** -0.006 -10.094***
Half-normal -0.031 -14.957*** -0.002 -2.429***
R square 0.956 0.961
Observations 48 48
Note: *** significant at 1% level.

5.5. Conclusion

The main purpose of this chapter is to examine the sensitivity of technical efficiency estimates
in fisheries to choice of input variables. Additionally, how technical efficiency scores may
differ due to the application of different model specifications (i.e., functional form,
assumptions on distributional inefficiency, and time-varying inefficiency) is also the focus of
this chapter. To this end, data set from the Northern Prawn Fishery for the 2010-2015 period
was used. Two measures of fixed inputs (i.e., either engine size or hull length) and three
measures of fishing time (either hour trawled or hour fished or day fished), and fish stock index
Chapter 5. Sensitivity of technical efficiency to the choice of input variables 95
were included into stochastic production frontier model. Different models were estimated using
a combination of different measures of inputs under different model specifications. Different
estimated efficiency scores derived from the different models were then regressed on a number
of factors (i.e., input variables and model specifications) in order to find the variation of
estimated efficiency scores due to different measures of fixed inputs and fishing time as well
as different efficiency model specifications.

A number of statistical hypotheses were tested. Given the specification of stochastic frontier
models, the results from the tests reveal some conclusions as follows. First, the hypothesis
favoured Cobb-Douglas functional form is rejected and the translog functional form is accepted
as a more appropriate functional form in all models. Second, the results indicate that the
stochastic frontier model is time-variant efficiency. Third, the distribution of inefficiency
follows a truncated distribution for models that include engine power, and models that include
hull length and hour fished, while the other models follow a half-normal distribution.

The results from this study also show that the choice of different measures of input included in
production function in fishery significantly affect the levels of technical efficiency scores and
its standard deviation. Especially, when the measure of fishing time gets less precise (ranging
from hour trawled (a measure of exact fishing time when fishing gear is in use) to hour fished
to day fished), have a lower mean efficiency score and higher standard deviation. Furthermore,
model specifications such as functional form and model specifications (i.e., assumption on
efficiency distribution and on time-variant efficiency) also impact the estimated efficiency
scores.

Chapter 5. Sensitivity of technical efficiency to the choice of input variables 96

Chapter 6. Economic focused fishery case study: Northern Prawn
Fishery17

6.1. Introduction

This chapter focuses on banana prawn fishery, a sub-fishery of the Northern Prawn Fishery,
which is managed to aim at maximum economic yield (i.e., MEY). The main purpose of this
chapter is to estimate the technical efficiency of fishing boats operating in the banana prawn
fishery, and examine how average efficiency changes over the season in the banana prawn
fishery. In addition to the main topic, information about fisheries management, fisheries
management objectives, and economic performance of this fishery are also outlined. The
remainder of this chapter is structured as follows. Section 6.2 provides general information of
the fishery that includes overall information on the fishery, fisheries management arrangements
and management objective. The current economic performance of the fishery is presented in
section 6.3. Previous efficiency studies related to the Northern Prawn Fishery are briefly
reviewed in section 6.4. The main section 6.5 focuses on technical efficiency estimation of the
banana prawn fishery.

6.2. The Northern Prawn Fishery

6.2.1. General information on the fishery

The Northern Prawn Fishery (NPF) is situated off the northern coast of Australia, ranging from
Cape York in Queensland to Cape Londonderry in Western Australia (ABARES, 2016b) and
covering approximately one million km2 (Kompas et al., 2010). The NPF is considered the
largest, and one of the most valuable fisheries in terms of total landed value in Australia, with
annual production value at $107 million in 2014-15 (ABARES, 2016b).

17
The main part of this chapter has been published in Australian Journal of Agricultural and Resource
Economics (see Pascoe, Hutton, et al., 2017). Reproduction of the work in the thesis has been done with the
expressed consent of the other authors of the paper.
Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 97
 Figure 6.1. Area of the Northern Prawn Fishery

Source: AFMA (2016a) and AFMA (2017).

The fishery aims at high valued prawn species that are mainly sold within Australia and
exported to Asian countries such as Japan and China (Kompas et al., 2004; Kompas et al.,
2010). The prawn species groups include tiger, banana and endeavour prawns, and each group
includes at least two species (Dichmont, Punt, Deng, Dell, & Venables, 2003). The biological
characteristics of the three species groups differ considerably. Specifically, the common banana
prawn is a highly fecund and short lifespan species. Its abundance is primarily driven by
environmental factors (i.e., rainfall level) rather than influenced by fishing pressure (Pascoe et
al., 2012). On the other hand, the endeavour and the tiger prawns are longer lifespan species
(Dichmont, Deng, Punt, Venables, & Haddon, 2006).

The NPF has two sub-fisheries, including a tiger prawn fishery and a banana prawn fishery
although the different species are caught by the same set of vessels (Pascoe, Hutton, et al.,
2017). The banana prawn fishery is dominated by white banana prawn (Fenneropenaeus
merguiensis) and generally starts from March/April to June but the length of the fishing season
also depends on catch rates (AFMA, 2017).

The tiger prawn fishery generally operates from August/September to November/December,

occurring during the second fishing season. However, the start and end date also depends on
catch rate within season (AFMA, 2017). Key species caught in the tiger prawn fishery are
brown tiger prawns (Penaeus esculentus), grooved tiger prawns (Penaeus semisulcatus) and

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 98

two endeavour prawn species (Metapenaeus endeavouri and M. ensis). However, other species
of banana prawns such as white leg species and red leg species, king prawns, fish and other
crustaceans are also caught as by-products during both seasons in the NPF (Pascoe et al., 2012;
Pascoe et al., 2010; Pascoe, Vieira, Dichmont, & Punt, 2011). Some other species such as
scampi, squid, scallops and bugs are also taken in the NPF (Laird, 2018).

Information about total catch and the number of total fishing boats is presented in Figure 6.2.
It can be seen that over the past decades, the total catch has fluctuated but banana prawn has
been dominant. In 2017, the total catch was 6,545 tons of which banana prawn accounted for
the majority of the total catch (77%), tiger prawn made up 16% and the other species occupied
the remaining 7%. The total catch value of the NPF in 2014-2015 fishing year was at AU$ 107
million (ABARES, 2016b). The number of fishing vessels considerably decreased from 191 in
1970 to 52 vessels in 2017 albeit with a fluctuation in the early 1980s. The number of fishing
vessels was highest in 1981 with 286 vessels. The dramatic increase in fishing vessels was
considered to stem from the liberal entry together with subsidy policy for boat building from
the Government in this period (AFMA, 2009; Dichmont, Loneragan, Brewer, & Poiner, 2007).
From 1981 to 2008, the number of fishing boats decreased significantly due to both voluntary
and compulsory buyback schemes. Since 2012, the number of fishing vessels has remained the
same, at around 52 fishing boats.

16000 350

No. of vessels
Banana Prawns Tiger Prawns Endeavour Prawns
Catch (tons)

14000 300

12000
250
10000
200
8000
150
6000
100
4000

2000 50

0 0
1992
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990

1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016

Year

Figure 6.2. Catch and fishing vessel for the Northern Prawn Fishery

Source: Laird (2018)

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 99

6.2.2. Northern Prawn Fishery Management

The Northern Prawn Fishery can be seen as a global model fishery management praised by the
United Nations Food and Agriculture Organization (CSIRO, 2018). The Australian Fisheries
Management Authority (AFMA), an agency of the Australian Commonwealth government, is
now responsible for managing the fishery (Pascoe et al., 2012).

Historically, the NPF has had long experiences in management since its establishment in 1966
(Kompas et al., 2010). Over the past five decades, management strategies in the NPF have
considerably changed regarding management objectives and thereby measures used (Kompas
& Gooday, 2007). Table 6.1 shows the major changes in the NPF management over the past
decades.

In the early years of establishment in the 1960s, the banana prawn species (Penaeus
merguiensis) were the targeted species (Dichmont et al., 2007). The number of fishing vessels
rapidly increased to over 200 boats in a short period since the establishment of the fishery
(Kompas et al., 2010). In 1971, seasonal closure measures were first applied for banana prawn
fishery (Kompas & Gooday, 2007). From 1970s to the early 1980s, the number of fishing boats
climbed and reached a peak in 1981 with 286 fishing vessels (Laird, 2018). As a result, fishing
effort exceeded fishing effort levels that were required for the long term sustainable yield of
the fishery (AFMA, 2009). Due to the excessive fishing effort and the decline in the catch of
tiger prawns from 1983, the first buyback program was introduced in mid-1980s to reduce the
overcapacity and overcapitalisation in the fishery in this period (AFMA, 2009; Pascoe,
Dichmont, Vieira, et al., 2013). Management measures in this period were mainly the use of
measures to control boat replacement, and technical measures (i.e., seasonal and areas
closures). The use of the former aims to strictly control fishing efforts while the application of
the latter measures (i.e., seasonal and area closures) to conserve stocks (i.e., juveniles and
spawning stocks).

From 1990s to early 2000s, due to overfishing in the tiger prawn species and excessive fishing
capacity, most of the management measures applied during this period aimed to reduce fishing
pressure on these stocks and to recover depleted stocks (Pascoe et al., 2012). Therefore, the
fishery again had experienced several buybacks and reduction in fishing capacity. Specifically,
the number of vessels declined from 127 boats in 1993 to 95 boats in year 2004 (see Figure
6.2). In 2004, maximum economic yield was defined as the target level of catch for tiger prawn
fishery. However, the number of fishing boats in 2004 (95 boats) was considered higher than
Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 100
optimal fleet size (56 boats) to achieve MEY (Kompas & Gooday, 2007). Further reduction in
fishing capacity (i.e., fishing boats) was implementing during the period 1993 and 2004. The
most recent buyback program took place in 2006-07 with the aim of moving the fishing fleet
size toward achieving the maximum economic yield in the fishery (Pascoe, Dichmont, Vieira,
et al., 2013).

Since 2007 the MEY target has formally been in place for the NPF (Kompas et al., 2010). To
achieve MEY, the two sub-fisheries including banana prawn fishery and tiger prawn fishery
are managed by different management systems with different approaches. Specifically, the
tiger prawn fishery has an explicit biomass of MEY, and a bio-economic model is applied to
determine annual fishing effort required to achieve MEY target (ABARES, 2016b). On the
other hand, the banana prawn fishery is managed by using an MEY-based catch rate trigger
that has been implemented since 2013 (Laird, 2018). As the species in the banana prawn fishery
are short-lived species, and highly susceptible to environmental factors (i.e., rainfall), so a
strong stock-recruitment relationship model for this fishery is not appropriate to apply
(Richard, Kevin, & Mihaela, 2017). The trigger level is estimated in season on the basic
information of prawn price, catch and fishing cost (ABARES, 2016b). The MEY target is the
point where weekly marginal revenue equals marginal cost so marginal profit is zero (i.e.,
revenue from catch is equal to daily costs of fishing) (ABARES, 2016b). Since the fishing
season 2014, the MEY trigger for the banana prawn has been set between a minimum value
at 425kg/boat/day and maximum value at 575kg/boat/day (ABARES, 2016b). During the
banana prawn-fishing season, catch data are used to determine whether closures should occur
based on the trigger (ABARES, 2016b).

Additionally, a number of measures have been used to manage the NPF. These measures
include the use of a combination of input controls such as a system of seasonal and area
closures, limited entry, individual transferable gear units (a form of individual transferable
effort quota, or ITE), gear restrictions and bycatch restrictions (ABARES, 2016b; Laird, 2018;
Pascoe et al., 2012; Pascoe et al., 2011). A summary of major management changes applied in
the NPF is presented in Table 6.1. Further information about historical management changes
in the NPF can be found in a number of references (e.g., AFMA, 2009; Dichmont et al., 2007;
Kompas & Gooday, 2007; Laird, 2018).

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 101
 Table 6.1. Management measures and objective in the NPF
Main management
Period/Year Main measures/tools used in fisheries management Reference
objective
Free entry criteria;
Dichmont et al. (2007),
Biological sustainable Seasonal closures applied for banana prawns in 1971;
1960s-1980 Kompas and Gooday
Fishing effort controls Boat building subsidy from government;
(2007), AFMA (2009).
Controls on boat replacement applied.
Unitisation of fishery applied: A management structure based on class A units (a Kompas and Gooday
Biological sustainable
1984 combination of vessel hull and engine power) and class B units (the number of boats); (2007), (Dichmont et al.,
Fishing effort controls
Transferable Class A and Class B unit systems. 2007), AFMA (2009).

First buyback program in mid-1980s: recommended a 25% reduction in effort (1985);

Kompas and Gooday
Further effort reduction in 1987 (targeted 30% reduction);
Mid-1980s to Biological sustainable (2007), Pascoe, Dichmont,
Daylight trawling ban during the tiger prawn season;
1990 Fishing effort controls Vieira, et al. (2013),
Controls on maximum size of net (i.e., 9 fathom headrope net for vessels under 375 A unit
Dichmont et al. (2007).
capacity and 14 fathom headrope net for vessels larger than 375 A unit capacity).

Voluntary buy-back scheme in 1990: Reducing vessel numbers from 216 to fewer than 132
vessels by 1993;
Opening date and mid-season closure introduced to protect spawn;
Compulsory buy-back scheme in 1993: targeted reducing 30% remaining A units;
Biological sustainable
Controls on maximum size of net removed in 1993. Dichmont et al. (2007),
1990 to mid- Fish stocks recovery
Kompas and Gooday
1990s Eliminating
(2007), AFMA (2009).
overcapacity

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 102
 Main management
Period/Year Main measures/tools used in fisheries management Reference
objective
A Management Plan 1995 (amended in 2001) and Transferable Statutory Fishing Rights (i.e.,
Biological sustainable determining the number of trawlers and gears used in the fishery) introduced;
Fish stocks recovery Class A and Class B units system abolished; Kompas and Gooday
1995-2002
Eliminating Gear-based management introduced (i.e., limiting headrope length); (2007), Laird (2018)
overcapacity Reducing fishing effort by reduction in headrope across the fishery and reducing fishing
season length.
MEY introduced in 2004 as the target catch reference point;
Reduced fishing efforts: effort reduction through reduction in total headrope length and
Kompas and Gooday
Maximum economic shortening fishing season;
2004-present (2007), Laird (2018),
yield Buy-back scheme applied in 2006/07: 43 vessels and 18365 gear SFRs removed;
AFMA (2009)
Individual transferable effort units implemented in 2006;
NPF harvest strategy under input controls introduced in 2007.
Implementing MEY catch-rate trigger for banana prawn since 2014;
Maximum economic Laird (2018)
2014-present Using a combination of input controls: tradeable gear units and seasonal closures, area
yield
closures, gear restrictions and operational controls to manage the NPF.

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 103
6.2.3. Fishery management objective in NPF

Over the years, the management objective in NPF has changed considerably, thereby so have
fisheries management strategies. As aforementioned, in 1971 seasonal closures for banana were
introduced to protect spawning stocks and nursery habitats.

From 1980-1990, the noticeable increase in fishing efforts and the decline in fish stocks in the
early 1980s raised concern about the sustainability of the fishery. The aims of management in
this period were thus to control and reduce the fishing efforts in the fishery. The introduction
of buyback programs implemented in the mid-1980s and controls on boat replacement were
implemented to achieve the objectives.

From 1990s to 2000s, the primary objective of management was to recover depleted stocks and
to limit fishing capacity. Further buyback schemes were also introduced to achieve the
objective (Table 6.1). It can be seen that prior to 2004, the fisheries management objective
mainly aimed at biological sustainable fisheries, referring to the Maximum Sustainable Yield
(MSY) (Dichmont, Jarrett, Hill, & Brown, 2014). However, in 2004, MEY was defined as the
target point for tiger prawn fishery as a whole (ABARES, 2016b). In 2007, the Commonwealth
Fisheries Strategy Policy (the Policy) was introduced by DAFF (2007). The Policy provides a
legislative framework and objective for management of the Commonwealth-managed fisheries.
The Policy proposes guidelines and aims for fisheries management as follows.

Harvest strategies for key commercial stocks taken in Australia’s

Commonwealth fisheries will be designed to pursue maximum economic yield
from the fishery and ensure those stocks remain above levels at which the risk to
the stock is unacceptably high.

Since 2007, MEY becomes an explicit objective of fisheries management in the NPF and the
fishery is considered the first commercial fishery in the world to apply MEY as the main
management target (Dichmont et al., 2010; Pascoe et al., 2011). In 2014, the Northern Prawn
Fishery Harvest Strategy was introduced that is in line with the Commonwealth Fishery
Harvest Strategy (DAFF, 2007; Dichmont et al., 2014). According to the Harvest Strategy,
objectives of the fishery management in NPF are as follows:

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 104
  Ensure the utilisation of the fishery resources within the Northern Prawn Fishery is
consistent with the principles of ecologically sustainable development and the exercise
of the precautionary principle.
 Maximise economic efficiency in the utilisation of the fisheries resources within the
Northern Prawn Fishery.
 Implement efficient and cost effective management of the fishery.
 Effectively communicate and consult with AFMA, the fishing industry, other marine
resources users and the broader community.
 Ensure that incidental catch on non-target commercial and other species in the Northern
Prawn Fishery is minimal.

Apart from biological and economic objectives in this fishery, the ecologically sustainable
issues have been also objectives in management of this fishery for a long time. These objectives
indicate that the impact of fishing on non-targeted species and protection of ecosystems, by-
catch, associated and dependent species should be minimal. In fact, the effect of fishing
especially demersal trawling for prawn of commercial fishers on non-targeted species and
unwanted catch or by-catch has been of great concern to Australia’s community (e.g., including
conservationists, environmentalist, fisheries managers, scientists and politicians) (Kennelly,
1995). This is because prawns are not the only species caught in the fishery but a number of
non-target species are caught incidentally due to unselective trawl gear (Dichmont et al., 2007).
Concerned with the problems of bycatch and ecological problems related to trawling, fisheries
managers actively have introduced and carried out by-catch avoidance initiatives over the past
years (Dichmont et al., 2007). The NPF is thus considered one of the first fisheries in Australia
to explicitly focus on bycatch research and minimisation (AFMA, 2014). Important measures
for bycatch management are the mandatory applications of turtle excluder devices (TEDs) and
bycatch reduction devices (BRDs) that have been applied since 2000. As a result, the
introduction of TEDs and BRDs into the NPF has significantly mitigated the negative impacts
of trawling on non-target species such as sea turtle, sea snakes, sharks and rays (Brewer et al.,
2006). The fishery managers also aim to reduce the capture of 30% of bycatch and small fish
during a period 2015-2018 which is stated in the Bycatch Strategy from 2015 to 2018 (AFMA,
2017).

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 105
6.3. Current economic status of the fishery

This section presents the economic performance of the Northern Prawn Fishery as a whole.
The purpose of this section is to observe how net economic return (NER) of fishery changes
according to the changes in fishery arrangements, and present the economic performance of
the fishery at whole fishery level. According to ABARES (2016b) evaluating economic
performance of the fishery and identifying reasons for change in the economic performance
trends are important, especially when key management arrangement changes are applied. In
fact, the NPF management has changed significantly to aim at different objectives.

One of the key indicators measuring economic performance of the fishery is net economic
return, which is calculated and reported at the fishery level. The net economic return of the
fishery is the difference between the total cash receipts from the fishery and the costs
(ABARES, 2016b). These costs include fuel, repairs, family and owner labour costs, fishery
management costs, crew costs, depreciations, and the opportunity cost of capital. The net
economic returns of the NPF for a given period can be estimated as follows (ABARES, 2016b).

Table 6.2. Net economic return of the fishery

Total cash receipts from the fishery

- Operating costs
= Fishery cash profit
- Owner and family labour
- Opportunity cost of capital
- Depreciation
+ Interest, quota/permit leasing and recovered management fees
= Net economic return (excluding management costs)
- Management costs
= Net economic returns
For economic performance analysis, we primarily used annual data for the period 2000/01 to
2014/15 that was collected by the Australian Bureau of Agriculture and Resource Economics

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 106
and Sciences. This data is available on the Australia fisheries economic indicators report 2015:
Financial and economic performance of the Northern Prawn Fishery18 (ABARES, 2016b).

100
Actual prices (2014-15$)
Prices held constant at 2014-15 level
Linear (Prices held constant at 2014-15 level)
80

60
Net economic returns ($m)

40
y = 0.7814x - 8.15
R² = 0.206
20

-20
Input controls, MSY target Buyback ITEs, MEY target Years

-40

Figure 6.3. Net economic returns of the NPF over the past years

Source: ABARES (2016b)

The results from Figure 6.3 show that the net economic returns of the fishery varied
significantly from 2000/01 to 2014/15. From more than $80 million in 2000/01, the net
economic returns dropped sharply in the following years to around -$16.7 million in 2004/05
(actual price). From 2004 to 2006, the remarkable reduction in NER is mainly due to decrease
in unit prices, effort and catch volume (due to buyback programs introduced in this period).
From 2007 onwards, the NER of the fishery experienced positive return except in 2011/12
where total catch and price of shrimp decreased significantly. The NER reached its peak in
2009/10 at around $13.2 million. The positive return from the fishery is mainly driven by the
increase in catch of banana prawn and increase in efficiency of fishing fleet that is a result of

18
The data is available on https://data.gov.au/dataset/pb_afsnpd9abme20161221
Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 107
structural arrangements of the fleet (ABARES, 2016b). However, it is notable that when the
price of catch is held constant at 2014/15 price, the fishery witnessed negative NER from the
year 2000 to 2007. From 2007 onwards, the NER of the NPF is positive in which the NER of
the fishery reached the peak in the year 2010-2011 (Figure 6.3).

It is notable that the net economic returns of the fishery varied considerably according to fishery
management objectives and management changes. However, the remarkable change was the
time when MEY was introduced as management target reference in 2004. After a negative
economic return period, from 2004 to 2007 due mainly to decline in catch, stemming from
reducing the number of fishing boats, the fishery has remained positive NER from 2007
onwards.

6.4. Previous efficiency studies in NPF

Being one of the most valuable fisheries in Australia’s Commonwealth fisheries and becoming
a model for marine fisheries management (ABARES, 2016a), the NPF has thus received great
attention from a diverse range of researchers, including fisheries economists, scientists,
conservationists, biologists, fisheries technologist, among many others. Thus, research topics
related to the NPF are also diverse, ranging from economic topics to fisheries management to
biology areas, to many others.

Within the economic topics, technical efficiency and efficiency-related scopes are also
increasingly interesting areas for researchers. The NPF has a long history of management
arrangements since its establishment as presented above. Thus, a number of studies have been
conducted in the fishery that directly/indirectly examined the impact of fisheries management
changes on technical efficiency and efficiency-related topics. Both tiger prawn and banana
prawn fishery components are the focus of those studies. The main aim of each study may
differ although most of the efficiency studies look at how fisheries management measures (e.g.,
input controls, buyback programs, and responding of fisher to management changes and so on)
affect/relate to efficiency.

The direct efficiency studies can be found in a number of studies undertaken by Kompas et al.
(2004) and Pascoe et al. (2012), while efficiency-related studies include Pascoe et al. (2010)
and Pascoe et al. (2011). Note that efficiency studies in the NPF have been undertaken since
the early 2000s. Kompas et al. (2004) estimated the impact of input controls (i.e., vessel size

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 108
and engine capacity restrictions) on technical efficiency of fishing boats operating banana
prawn fishery. By applying stochastic production frontiers based on two different datasets (i.e.,
1990-1996 data set and 1994-2000 data set) of the banana prawn fishery, the authors showed
that technical efficiency decreased during the study period due to input restrictions (i.e., engine
power and hull size). The explanation for this finding was that when input controls were
introduced (i.e., engine size and vessel size controls), fishers may have economic incentives to
substitute unrestricted inputs (i.e., headrope length) for restricted inputs. The technical
efficiency of fishing boats increased with a measure of vessel size and engine power that was
restricted by the fishery regulators while the technical efficiency declined with an unrestricted
input (i.e., gear headrope length).

Pascoe et al. (2012) examined the impact of a buyback program on technical efficiency of
fishing fleet operating in the NPF. A multi-output production function approach was used to
estimate the technical efficiency and its determinants, based on data set from 1995 to 2007 of
the tiger prawn fishery. The findings implied that an average of scale efficiency in the fleet
increased because the remaining vessels approached closer to optimal scale. The remaining
fishing vessels have higher average efficiency than that of the exited vessels. Additionally, the
authors found that due to reducing crowding in fishing (e.g., the number of boats fishing at any
point in time) and removing less efficient vessels, the buyback program led to an increase in
the individual technical efficiency of the remaining fishing boats. Furthermore, factors
affecting efficiency of the fishing boats were also examined. Results also showed that company
structure, the number of vessels fishing and vessel’s vintage significantly affected technical
efficiency of the fishing boats.

In addition, efficiency topics are not of primary interest in some studies although efficiency
aspects are also discussed in some papers. Specifically, Pascoe et al. (2010) used a multi-output
distance function to examine the ability of fishers to control output mix in the NPF if ITQs
management strategies would be applied in the fishery. The study used a data set from 1995 to
2005 of tiger prawn component for the analysis. Findings indicated that fishers have relative
inability to target their output mix, and proposed that a single quota might be sufficient to
control catches of the key species. The authors also found that technical efficiency of the fleet
was relatively high, most vessels operating at between 70% to 90% efficiency. Similarly,
Pascoe et al. (2011) applied a restricted profit function model to determine the optimal vessel
characteristics and output levels of fishing vessels operating in NPF under ITQ management

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 109
system scenario. The study was based on a data set that covers a 12-year period from 1994/95
to 2005/06. Again, efficiency analysis is not the focus of the study but efficiency topic is also
analysed. The authors showed that the level of economic efficiency of most fishing boats was
between 60% and 70%.

6.5. Technical efficiency of banana prawn fishery

6.5.1. The rationale of study

As previously discussed, the NPF has two sub-fisheries that are spatially separate, and require
different fishing methods. Given the shorter nature of the banana prawn season, fishing effort
usually focuses on this component at the start of the year, with boats progressively moving
onto the tiger prawns as banana prawn catch rates decline (Pascoe, Sharp, & Buckworth, 2015).

Since 2007, the fishery has had an objective of achieving maximum economic yield (MEY)
(Dichmont et al., 2010). The two sub-fisheries are subject to different management systems,
with different approaches to achieve MEY. Tradeable gear units and seasonal closures are used
to manage the tiger prawn component, with the target effort level (expressed in terms of both
season length and gear units available) determined by a bioeconomic model of the fishery (Punt
et al., 2010; Punt et al., 2011). The banana prawn component, in contrast, is managed through
a catch rate trigger and seasonal closures. The banana prawn sub-fishery is characterised by a
large recruitment at the start of the year that is largely environmentally driven (Vance, Staples,
& Kerr, 1985). This is fished down over the season until a predefined catch rate triggers the
fishery closure, at which time fishers are able to move to the tiger prawn component if they
have not already done so (or the mid-season closure is in place which covers the entire fishery).
An MEY-based trigger was introduced in 2013; prior to this the trigger was mainly based on a
biological escapement target, with some limited consideration by the industry at the time as to
what was profitable at boat level.

The MEY-based trigger for the banana prawn season is estimated largely using average catch,
price and fishing cost information. As will be expanded upon in the following section, this
effectively assumes that the average cost is equal to the marginal cost, which in turn implies
that the structure and fishing behaviour of the fleet does not change over the season. The key
aim of this section is to assess the extent of efficiency and productivity change over the banana

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 110
prawn fishing season, and the implications of these changes, if any, on the setting of the MEY
trigger/target catch rate.

To this end, we estimate the technical efficiency of the boats operating in the banana prawn
season, and examine how average efficiency changes over the season. We also examine how
fishing behaviour, which may affect fishing costs, also changes over the season. Implications
of these changes for the estimation of the trigger are discussed.

6.5.2. The banana prawn MEY trigger

The MEY trigger is based on traditional microeconomic theory of the firm. Assume the catch
function can be given by 𝐶 = 𝑞𝑥0 𝐸 𝜆 , where q represents the (constant) proportion of the stock
removed by one unit of fishing effort, E, and x0 is the starting stock size with zero effort. The
stock decreases as effort is applied, so that the marginal catch rate declines as effort increases,
that is 𝑑𝐶/𝑑𝐸 = 𝜆𝑞𝑥0 𝐸 𝜆−1 . Given such a catch function, and assuming constant prices and
costs per unit of effort, the profit function is given by 𝜋 = 𝑝𝑞𝑥0 𝐸 𝜆 − 𝑐𝐸, where p is the
constant price and c is the constant cost per unit effort. Profits are maximised when is
𝜆𝑝𝑞𝑥0 𝐸 𝜆−1 = 𝑐, which is essentially when marginal revenue (𝑝𝑑𝐶/𝑑𝐸) is equal to marginal
cost.

Deriving the appropriate parameters to implement this relationship in the fishery is

complicated. Foremost of the complications is that the size of the starting stock is not known
with any accuracy at the start of the season (Buckworth et al., 2014; Venables et al., 2011), and
only few attempts have been made to estimate stock sizes ex post (Zhou et al., 2015; Zhou et
al., 2007). Recruitment to the fishery appears to be driven largely by environmental factors,
particularly rainfall (Vance et al., 1985). Second, abundance is not constant over the season, as
factors such as the moon phase affects the behaviour of the prawns and hence the catch per unit
of fishing effort. The schooling behaviour of prawns also obfuscates the relationship between
fishing effort and the rate at which the stock is reduced (Die & Ellis, 1999; Zhou et al., 2007).
Finally, heterogeneity in the fishery in terms of fishing costs and also technical efficiency of
the individual vessels results in potential changes in the measure of fishing effort and also costs
over the season. Few boats fish the full season, with weekly effort levels declining largely as a
consequence of vessels ceasing fishing, reflecting differences in costs (either opportunity costs
or financial costs) (Pascoe, Sharp, et al., 2015).

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 111
Given this, a proxy estimate of the appropriate profit maximising level of fishing activity is
derived based on a simple stopping rule for fishing. Assuming marginal revenue is represented
by the observed average catch per day in a given week, CPD𝑡 , times the average price (assumed
constant over the season), 𝑝, and assuming average costs per day fished, 𝑐, also a constant such
that the average cost is equal to the marginal cost, then the “optimal” stop fishing condition is
𝑝𝐶𝑃𝐷𝑡 = c, which gives a “target” catch rate of 𝐶𝑃𝐷𝑡 = c/p.

For the purposes of estimating the target catch rate, only variable costs are considered. These
include fuel costs, f ($/day); a portion of repairs and maintenance, rm, $/day (Pascoe, Vieira,
& Thebaud, 2015); crew costs, cs, as a proportion of revenue; and marketing costs, m, $/kg
catch. The latter two cost components are related to the quantity of catch, while the first two
are related to the number of days fished. To factor these into the target, the price component is
modified to reflect the catch related cost items, and the proxy target catch rate given by 𝐶𝑃𝐷𝑡 =
[𝑓 + 𝑟𝑚]/[𝑝(1 − 𝑐𝑠) − 𝑚] where the denominator effectively represents the net price
received after crew and marketing costs have been removed.

Information on the appropriate cost and prices are estimated in collaboration with the industry
at the start of the season. Fuel costs per day are estimated based on average fuel use per day
(derived from a survey of fishers in previous seasons) and the price of fuel paid by fishers at
the start of the season. Repairs and maintenance costs are derived from actual costs in the
previous season, and assumed to stay the same for the current season. Similarly, crew share
and marketing costs are based on values derived from a survey of fishers in the previous season.
The average prawn price used in the analysis is provided by industry based on expectations for
the current season.

Prior to the introduction of the MEY target catch rate, the fishery was subject to a minimum
escapement-based trigger catch rate of 500 kg/day. This was a precautionary limit agreed to by
industry reflecting the uncertainty in the stock-recruitment relationship. While no apparent
relationship can be identified (Vance et al., 1985), concern that such a relationship may exist
has led to a precautionary limit on harvesting the species. With the introduction of a MEY
target catch rate, industry was concerned that a lower catch rate than the precautionary catch
rate may result in longer term costs to the industry, while a substantially higher catch rate may
provide lost opportunities to parts of the industry if the target was inappropriate (i.e., lower
cost or more efficient vessels were still able to fish profitably at lower catch rates than

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 112
“average” costs would imply). Hence, a buffer was placed on the MEY target to be no more
(or less) than 15% above (or below) the original precautionary catch rate.

The average catch rate in the fishery is monitored over the season, with critical time periods at
6, 8 and 10 weeks into the season (AFMA, 2016b). The fishery is opened in early April (usually
the first, but this varies). Catch and effort data relating to weeks 4 and 5 are assessed and if the
average of these values over this period is below the target catch rate the fishery is closed in
week 6. If the target catch rate is exceeded on average over this period, fishing continues and
a further assessment is made for week 8 relating to catch and effort in weeks 6 and 7. Similarly,
if the target catch rate is exceeded at this point fishing continues and a further assessment is
made for week 10 relating to the preceding two weeks. If the target catch rate is still exceeded,
the fishery remains open until the pre-set closure date (in mid-June) (AFMA, 2016c). Hence,
that target catch rate is also the trigger that leads to closing the fishery for the season.

The initial derivation of the trigger catch rate rules (Buckworth et al., 2013) included an
allowance for further declines in the catch rate between the two week monitoring period and
the following decision week. That is, the trigger catch rate was initially proposed to be 1.63
times higher than the target to allow for the decline in catch rates over the two-week period,
based on observed changes in the fishery over recent previous years. This was opposed by
industry who argued that lower cost and more efficient vessels could continue to fish profitably
at lower average catch rates than the average trigger value. As a result, no adjustment was made
to increase the trigger above the target catch rate.

6.5.3. Methods

The study involves two key questions: first, does the average technical efficiency of the fleet
change over the season; and second, are variable costs per day fished likely to be constant over
the season? The first of these is assessed using a stochastic production frontier approach, from
which estimates of individual vessel technical efficiency can be derived. The second of these
is more complex, as ideally a cost function should be estimated. In the absence of appropriate
information to estimate such a function, a simpler approach is adopted that is indicative of cost
changes over the season. These approaches are outlined below.

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 113
6.5.3.1. Stochastic production frontier and efficiency estimation

Stochastic production frontiers are econometric models that estimate the output of a vessel as
a function of a number of inputs (fixed and variable) as well as a measure of the relative
inefficiency of the vessel. In this study, the production frontier was estimated as a translog
production function. Both a half normal and truncated normal distribution was tested (Battese
& Coelli, 1988), along with the potential for time variant efficiency (Cornwell, Schmidt, &
Sickles, 1990). Theoretical consistency was imposed using the approach proposed by
Henningsen and Henning (2009). The approach and stochastic production frontier model were
previously described in Chapter 3.

6.5.3.2. Changes in fishing costs

As noted above, the target catch rate in the fishery is given by 𝐶𝑃𝐷𝑡 = [𝑓 + 𝑟𝑚]/[𝑝(1 − 𝑐𝑠) − 𝑚]
where f are fuel costs ($/day); rm is a portion of repairs and maintenance ($/day); cs are the
crew share of revenue and m are marketing costs ($/kg catch). These parameters are derived as
averages from an economic survey of the industry. Fuel and repairs and maintenance costs are
estimated as an average cost per day. However, fishing activity within a day varies over the
season. At the start of the season, catch rates are high and nets are filled with relatively short
sets/hauls. However, towards the end of the season, catch rates are lower and hours trawled per
day tend to be longer. As a result, fuel use per day is likely to be higher towards the end of the
season. Changes in fleet composition over the season may also result in changes in average
fuel and repairs/maintenance cost per day fished.

The data available are insufficient to estimate the change in cost per day, but an assumption
was made that these would be proportional to the relative number of hours trawled per day and
the average engine power of boats fishing in each week.19 An index (Iw) was developed based
on the product of the average number of hours trawled in a given week relative to the average
number of hours trawled over all weeks and the average engine power in a given week relative
to the average engine power over all weeks, given by

19
Ideally, a cost function or frontier would be estimated to test this assumption. While economic data are collected
biannually on the fishery (e.g. ABARES, 2016b), only aggregated fleet level information is made available.
Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 114
 ∑𝑖∈𝑖(𝑤) ℎ𝑖,𝑤 /𝑛𝑤 ∑𝑖∈𝑖(𝑤) 𝑘𝑊𝑖 /𝑛𝑤
𝐼𝑤 = ,
∑𝑤 ∑𝑖 ℎ𝑖,𝑤 / ∑𝑤 𝑛𝑤 ∑𝑤 ∑𝑖 𝑘𝑊𝑖 / ∑𝑤 𝑛𝑤

where i(w) is the set of boats operating in week w, hi,w is the hours trawled in week w by boat
i, kWi is the engine power of boat i, and nw is the number of boats operating in week w. This
index was applied to the currently used fuel and repairs and maintenance costs, such that a
modified minimum catch rate for a given week could be estimated as 𝐶𝑃𝐷𝑡𝑤 =
𝐼𝑤 [𝑓 + 𝑟𝑚]/[𝑝(1 − 𝑐𝑠) − 𝑚].

6.5.4. Data

Data used for this analysis is previously presented in Chapter 5 (section 5.2) that used the same
data set to examine the sensitivity of input variable selection on technical efficiency. For
detailed information on the data, we refer to Chapter 5.

6.5.5. Results

6.5.5.1. Production function and efficiency analysis

Econometric analysis

The results of the production frontier estimation are given in Table 6.3. Initial and final
parameter estimatesThe initial model (Step 1) included assumptions of truncated normal and
time variant distributions, both of which were found to be significant. Most of the other input
parameters were also found to be significant, including key cross-product coefficients. Given
the significance of the cross-product terms, a Cobb-Douglas production function was not
tested, and the translog production function was accepted as the more appropriate functional
form. The model was also tested against the assumption of no technical inefficiency and found
to be significant, indicating that heterogeneity in inefficiency exists, and hence a frontier rather
than a standard production function is a more appropriate specification.

The initial model did not fully satisfy the monotonicity requirements, especially in relation to
engine power, so the second and third stages described in the methodology were implemented.
The coefficients estimated through the minimum distance approach (second stage) were not
statistically different from the initial estimates (based on the standard errors of the initial
coefficients). The final estimates were derived by the scaling coefficient estimated in the final
step. As expected, the intercept term in the final step was not significantly different from zero,

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 115
whereas the scaling coefficient was not significantly different from 1 suggesting that no
substantial bias has been introduced through the three-step process (Henningsen & Henning,
2009).

Table 6.3. Initial and final parameter estimates

Step 1 Step 2 Step 3
Min.
Initial Std. distance Final
Est. Error Sig coeff. t stat Pr(>|t|) est.
Constant 0.333 0.043 *** 0.299 0.781 0.198 0.307
ln(Engine) 0.697 0.236 ** 0.988 -1.234 0.126 0.986
ln(Days) 0.690 0.104 *** 0.640 0.477 0.259 0.639
ln(Stock) 1.157 0.037 *** 1.151 0.144 0.312 1.149
ln(Engine)^2 -5.323 2.011 ** -2.462 -1.423 0.105 -2.457
ln(Engine)*ln(Days) 0.262 0.387 0.170 0.238 0.301 0.170
ln(Engine)*ln(Stock) -0.399 0.200 * -0.347 -0.259 0.298 -0.346
ln(Days)^2 -0.342 0.292 -0.507 0.565 0.241 -0.506
ln(Days)*ln(Stock) 0.407 0.105 *** 0.323 0.802 0.194 0.322
ln(Stock)^2 0.100 0.083 0.104 -0.052 0.317 0.104

Sigma^2 2.246 0.547 ***

Gamma 0.790 0.057 ***
Mu -2.664 1.119 *
Time -0.015 0.004 ***

Log Likelihood -1587.6 -1589.8

Monotonicity (%)
ln(Engine) 76.4 100
ln(Days) 99.7 100
ln(Stock) 99.9 100
Quasi-concavity (%) 99.9 99.9
Mean efficiency 0.79 0.78
Note: *** significant at 0.1%, ** significant at 1% and * significant at 5%

Efficiency distributions

The distribution of technical efficiency over the whole data period is shown in Figure 6.4.
Distribution of final technical efficiency scoresA high proportion of the fleet had relatively
high technical efficiency scores (i.e., greater than 0.9), although some vessels had very small
efficiency scores (i.e., less than 0.5) which impacted on the overall average. Factors driving
the level of inefficiency were not examined as the focus of the study was on the distribution of

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 116
the efficiency scores, and may be an area for further investigation. In the tiger prawn fishery,
the use of quad gear, vessel age, company size and the number of other boats fishing were all
found to have a significant impact on efficiency (Pascoe et al., 2012), so may have similar
impacts on efficiency in the banana prawn fishery.

Figure 6.4. Distribution of final technical efficiency scores

The average technical efficiency over the fishing season in each year is shown in Figure 6.5.
In 2010, 2011 and 2014, average efficiency was relatively constant over most of the year with
a small increase towards the end of the year. In the other three years, average efficiency
declined sharply at the end of the season, although in the case of 2012 this was preceded by an
increase in average efficiency. Both 2013 and 2015 were characterised by lower than average
total catches, 2013 more so, which may have resulted in different (more efficient) vessels
leaving the fishery earlier to fish for tiger prawns. However, the total number of vessels fishing
each week was fairly similar in each year of the data as seen in Figure 6.6.

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 117
 Figure 6.5. Average technical efficiency over the fishing seasons in each year

Figure 6.6. Number of boats fishing each week over the fishing season

6.5.5.2. Productivity impacts on catch rates

While no clear trends in efficiency were observed, both average hours trawled and average
engine power increased over the season (as shown in

Figure 6.7). The former of these is not directly captured in the production function, which is
specified with days fished as the effort measure, but is likely to influence the catch per day for
a given biomass level. Engine power is specifically captured in the production function.

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 118
 Figure 6.7. Average hours trawled per day and engine power

From the production function, the responsiveness of catch to engine power – the production
elasticity20 – varies with the average level of engine power; days fished each week and stock
size. While stock size declines over the season, the engine power elasticity remains relatively
constant for most of the season, increasing in the last few weeks (as seen in Figure 6.8 and
Figure 6.9). The effect of this on vessel productivity (ignoring the stock decline) is that vessel
productivity is roughly 10-20% higher in the last couple of weeks that the fishery is open
(Figure 6.9). This reduces the effect of declining stock levels of catch rates, resulting in average
catch rates generally staying above the 500 kg level in most years from weeks 8 to 12 (Figure
6.10).

20
The production elasticity is given by 𝜕𝑙𝑛𝐶 ⁄𝜕𝑙𝑛𝑘𝑊.
Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 119
 Figure 6.8. Stock index over the fishing season

Figure 6.9. Engine power a) production elasticity and b) productivity

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 120
 Figure 6.10. Average catch per day from weeks 8 to 12 over the fishing seasons (The blue
line represents the 500 kg catch)

6.5.5.3. Changes in cost per day

The higher productivity of the fleet in the last two weeks resulting from the greater input use
is also likely to have implications for the cost per day, and consequently the appropriate trigger
catch rate. The cost data used in the estimation of the trigger value is based on the average costs
for the fleet in total over the whole season. The marginal crew cost and marketing costs per
day are linked to the level of catch, which will decline as catch rates decline. However, these
declines are implicit in the estimation of marginal revenue when deriving the trigger. In
contrast, fuel and repairs and maintenance costs are assumed constant per day, but in reality
may change as a result of changes in fleet structure over the season (as seen in Figure 6.6) and
the change in the average hours per day and engine power of the vessels (Figure 6.7).

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 121
 Figure 6.11. Derived proxy cost index (i.e., the combined cost index)

As noted in the methodology, the cost index is assumed to be proportional to average engine
power and hours trawled each day, with the combined cost index assumed to be the product of
the two input indexes, given in Figure 6.11. From this, average variable running costs are
expected to be below the annual average for the first eight weeks of the season, and above this
average for the remainder of the season. In weeks 11 and 12, running costs are potentially 45-
55% higher than the annual average. Given this, target catch rates are also expected to be higher
if based on the average costs towards the end of the season. For example, based on the price
and cost parameters used for the estimation of the 2015 target, the target catch rate in week 12
given the higher costs is estimated to be 351 kg/day compared with the value based on average
costs of 221 kg/day (ignoring the buffer discussed previously) (Figure 6.12).

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 122
 Figure 6.12. Target catch rates factoring in the higher costs towards the end of the season
(based on 2015 price and cost parameters)

The effect of increasing the costs on the target catch rate for a range of prices is shown in Figure
6.13. The average cost index in week 12 (1.54, see Figure 6.11) was used to inflate the running
costs (fuel and repairs and maintenance). As expected, the target catch rate increases as prawn
prices decrease, but do not reach the lower buffer level (425 kg/day) even with higher cost
assumptions until prices fall to $10.75/kg, and does not reach the upper buffer level (575
kg/day) until prices fall lower than $8/kg.

Figure 6.13. Target catch rates for a range of prawn prices based on 2015 cost parameters and
higher costs assumptions towards the end of the season

6.5.6. Discussion

The key aim of the study was to assess if changes in fleet composition over the year affects the
estimation of the target and trigger MEY catch rate used to currently manage the banana prawn
fishery. In principle, the trigger catch rate should be higher than the target, as there is a delay
between when the trigger is observed and when fishing ceases. However, industry suggested
that only the more efficient vessels fished at the end of the season, and the (assumed) lower
cost of these vessels (compared with the average of the fleet as a whole) means that they can
still fish profitably at lower catch rates. Given the uncertainty around the fleet and cost structure
over the season, the trigger and target catch rates were considered to be the same (i.e., no
allowance was made to reflect the decline in the stock over the period between when the trigger
rate was observed and the fishing ceased).

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 123
Uncertainty around stock sustainability resulted in further modification to the MEY trigger.
Prior to the implementation of the MEY trigger, an escapement target of 500 kg/day was set as
a biologically oriented trigger catch rate. This was a precautionary target and not based on a
detailed biological modelling of the fishery. Industry concern that catching at lower rates than
this may affect future stock levels resulted in a catch rate buffer being implemented, such that
the minimum acceptable MEY target catch rate (and trigger) would be no less than 15% below
the biological trigger rate. The estimated MEY target rate has fallen below this buffer since
2013, so the precautionary 425 kg/day has been implemented as the management trigger.

The results of this study found no clear pattern around efficiency changes over the season.
While the number of vessels generally declined towards the end of the season, the average
efficiency of those remaining was generally similar to that of the larger fleet earlier in the
season. The model analysis used fishing days as the measure of variable fishing effort as this
corresponds to the management system in place. However, days is potentially a poor measure
of fishing effort as the amount of trawling time within the day varies over the season, and this
may have affected the efficiency scores. The use of hours trawled as an alternative effort
measure is also potentially problematic. At the start of the season when stocks and catch rates
are high, low hours trawled per day may reflect a capacity constraint of the vessel in terms of
its ability to process the catch. Towards the end of the season when catch rates were lower,
longer trawling hours may be required to reach the capacity constraint.

While average efficiency did not change substantially in most years, productivity of the
remaining vessels at the end of the season was generally greater than the fleet in the earlier
weeks. The remaining vessels were generally larger in terms of engine power, and trawled
more hours per day. This had the effect of maintaining catch rates at higher levels, and on this
basis equating the target and trigger catch rates to be the same is a reasonable assumption.
However, higher engine power and hours trawled also has implications for the costs of fishing.
An assumption was made that running costs would be proportional to these inputs, and on this
basis running costs at the end of the fishing season may be up to 55% higher than the average
values used in the current setting of the MEY trigger. While no cost functions have been
estimated for the fishery, Zhou et al. (2010) found that, generally, fuel costs increased almost
linearly with vessel length (which is highly correlated with engine power in the NPF) and at a
diminishing rate with days fished, while repairs and maintenance costs were estimated to be
linearly proportional to vessel length.

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 124
Even assuming higher costs at the end of the season, the MEY target catch rate is likely to be
below the current buffer lower limit, and is likely to remain so unless banana prawn prices
decrease substantially. However, if this buffer is to be removed or modified (subject to
biological sustainability constraints), more detailed consideration will need to be given to
changes in cost structure over the season. In such circumstances, estimating a cost function for
the fishery may be beneficial.

The analysis has focused on the effects of changing efficiency and productivity levels on the
existing MEY trigger estimation process. This process is based on the general economic
principle that profits are maximised when the marginal cost of fishing is equal to marginal
revenue. Studies of similar fisheries elsewhere, however, have found that the pattern of fishing
effort earlier in the season can also have a substantial impact on fishery profitability as this
affects the rate of change in marginal revenue (Huang & Smith, 2014). Consequently, while
the MEY trigger catch rate provides a proxy for implementing MEY in the fishery, it may not
fully achieve MEY. Other factors may also affect MEY in the fishery, particularly the
interaction with the tiger prawn fishery. For example, the level of fishing effort in the banana
prawn fishery is affected by the tiger prawn price (Pascoe, Sharp, et al., 2015). Ideally, given
that vessels operate across both fisheries, MEY in each fishery should be assessed accordingly
taking into consideration conditions in the other.

6.5.7. Conclusions

Given the changes in fleet productivity over the season, and the observed changes in catch rates
as a result, equating the MEY target and trigger catch rates is a reasonable approach for
pragmatic fisheries management. However, the likelihood of higher fishing costs at the end of
the season suggest that the current MEY trigger estimates are too low. As the estimated target
levels in recent years have been below the lower precautionary buffer, they have not affected
the actual implemented target and trigger. Ideally, a fishery specific cost function should be
estimated to refine these costs and improve the estimate of the trigger and target.

Chapter 6. Economic focused fishery case study: Northern Prawn Fishery 125
Chapter 7. Social focused case study fishery: An open-access
trawl fishery in Vietnam21

7.1. Introduction

This chapter focuses on an open-access trawl fishery in Vietnam that can be seen as a strongly
social economic objective fishery. This chapter presents two main topics including providing
the status quo of trawlers’ socio-economic profile in surveyed areas, and presenting the
analysis of technical efficiency and its determinant of surveyed trawlers in Vietnam. Before
focusing on the main parts of the chapter, this chapter also provides general information about
fisheries sector in Vietnam including overall information about fisheries sector, fisheries
management objective, and trawl fishery in Vietnam. The remainder of the chapter is structured
as follows. The general information on fisheries in Vietnam is presented in section 7.2 that
contains several parts including overview of fisheries, fisheries management objectives and
trawl fishery in Vietnam. Field survey to collect primary data is presented in section 7.3.
Information on trawl fishery in surveyed areas and data collection procedures are also given in
section 7.3. Results of socio-economic profile of the surveyed trawlers are provided in section
7.4. Section 7.5 presents the main results on technical efficiency and its determinants analyses.

7.2. Marine capture fisheries in Vietnam

7.2.1. Overview of fisheries sector

Vietnam is a coastal country, situated in Southeast Asia. With a coastline of 3260 km and an
Exclusive Economic Zone (EEZ) of more than one million km2, Vietnam has great potentials
for marine fisheries development. Marine fisheries resources have been estimated at about 4.1
to 4.6 million tons, with total allowable catch from 1.2 to 1.4 million tons annually (Directorate
of Fisheries, 2018; Long, Flaaten, & Anh, 2008; Pho Hoang Han, 2007).

Marine capture fisheries play an important role in socio-economic developments of Vietnam.

In 2012, the contribution of fisheries sector (including aquaculture) to the national economy

21
The main part of this chapter (technical efficiency and its determinants of small-scale trawlers) has been
published in Marine Policy (Van Nguyen, Pascoe, & Coglan, 2019).
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 126
was around 4% of the GDP (Nam, 2013). In 2017, total marine capture was about 3.2 million
tons and seafood export value accounted for US$8.31 billion (Directorate of Fisheries, 2018).
Vietnam ranks as the eighth marine-capture-fishing nation in the world (FAO, 2016). Further,
the sector also acts as an important source for employment opportunities, food security,
livelihood, poverty alleviation, and nutrient sources, especially in coastal fishing communities.
Almost one million people are directly engaged in the capture fisheries (Directorate of
Fisheries, 2018). For many centuries, “Rice and Fish” have been daily important foods for
Vietnamese households (Quang & Leung, 2009). Fish can be seen as a major source of protein
for the Vietnamese diet (FAO, 2014). The average annual level of fish and fish product
consumption in Vietnam was 10.5 kg/capita in 1990, accounting for one third of the animal
protein (Pho Hoang Han, 2007; Pomeroy, Kim Anh, & Thong, 2009; Raakjær et al., 2007).
This figure dramatically increased to 14.6 kg/capita in 2011, representing 8.5% of protein
consumed (FAO, 2014).

Marine capture fisheries in Vietnam are open-access (Long et al., 2008; Pomeroy et al., 2009;
Raakjær et al., 2007). Due to open-access regime the number of fishing boats has increased
dramatically in parallel with the increase in fishing effort (i.e., number of fishing boats and
total engine power) over the past years (Directorate of Fisheries, 2013). Since 1990, the number
of fishing vessels has increased from 41,226 vessels with a total of 727,500 CV22, to 109,622
fishing boats on 31 December 2017, with a total engine power over 10 million CV (Directorate
of Fisheries, 2018; Pham, Gert, Chu, & Goethals, 2014). However, fishing fleets in Vietnam
are comprised primarily of small-scale fishing vessels, and fishing technology is limited (The
Prime Minister, 2013b). In 2017, the number of fishing boats equipped with engines less than
90 CV was 73,167 vessels, accounting for 66.7% (Table 7.1). Hence, the focus of fishing
activity has largely been coastal water areas23 (Pham, De Laender, Everaert, Chu, & Goethals,

22
Prior to 2010, unit measure of engine power was horsepower (HP). However, according to Decree
No.33/2010/ND-CP dated on 31 March 2010 by the Prime Minister of Vietnam, the unit measure CV
has been used officially in legal documents since 2010.
1 HP = 0.7457 kW; 1CV=0.7354 kW
23
Vietnam’s sea area is divided into three fishing zones. These include coastal areas, inshore areas and
offshore areas. Fishing vessels with engine capacity less than 20CV or vessels without engine are only
allowed to operate in the coastal areas. Fishing vessels with engine capacity from 20-90 CV can only
be operated in the coastal areas and offshore areas but are not allowed to operate in high sea areas.
Fishing boats with engine power greater than 90 CV are only allowed to fish in the offshore areas and
high sea areas (i.e., the water beyond national jurisdiction).
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 127
2014; Pomeroy et al., 2009) as most fishermen are considered to be poor as well as having
limited ability for capital investment (Long et al., 2008; Nguyen Long, 2003). Consequently,
coastal water areas are considered to be over-exploited, leading to concern about sustainability
of the fishing activities in coastal waters (The Prime Minister, 2013b). Although over the past
few years the proportion of offshore vessels (engine ≥90CV) has increased considerably, from
8.1% in 2001 to 33.3% in 2017, the proportion of small boats has still been dominant (Table
7.1). The increase in the number of offshore vessels over the past decades is mainly in response
to policies and supports introduced by the Vietnamese government (FAO, 2005a). These
policies and measures primarily include subsidy schemes (i.e., low interest rate on loan) for
new offshore vessel constructions, fuel and insurance subsidies among others. However, the
offshore fisheries are considered an underdeveloped and infant industry (Duy, 2016)

Table 7.1. The total fishing boats in 2001 and 2017 regarding engine size (boats)

2001 2017
Engine size %
(CV) No. of Proportion No. of Proportion (2017/2001)
vessels (%) vessels (%)
Engine <90 68490 91.9 73167 66.7 106.8
Engine ≥ 90 6005 8.1 36455 33.3 607.1
Total boats 74495 100 109622 100 147.2
Source: Directorate of Fisheries (2012) and Directorate of Fisheries (2018)

There is a variety of fishing gears deployed by fishers (Table 7.2). However, they can be
grouped into seven primary gear types that include trawl, gillnet, hand line, purse seine, lift-
net with lights, trap and others (Directorate of Fisheries, 2012). Among these fishing gears,
trawl and gillnet fishing are dominant. Nevertheless, the proportion of trawl fishing vessels has
decreased over the past decade while the proportion of gillnet has experienced an increasing
trend.

In this study, the term “offshore vessels” refers to fishing vessels that are equipped with an engine ≥ 90
CV.
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 128
Table 7.2. The number of fishing boats in 2001 and 2017 regarding fishing gear (boats)

2001 2017 %
Fishing gear
No. vessels % No. vessels % (2017/2001)
Trawl 16761 22.5 19574 17.9 116.8
Gillnet 18251 24.5 36518 33.3 200.1
Purse seine 5736 7.7 5231 4.8 91.2
Longline & Handline 14676 19.7 18462 16.8 125.8
Others24 19071 25.6 27697 25.3 145.2
Mother ship - - 2140 2.0 -
Total 74495 100 109622 100 147.2
Source: Directorate of Fisheries (2012) and Directorate of Fisheries (2018).

Regarding total landings, the volume of marine capture fisheries in Vietnam has increased
considerably since the 1990s. From 700 thousand tons in 1990, the total catch went up to 3.2
million tons in 2017 (Directorate of Fisheries, 2018; Pham, De Laender, et al., 2014). However,
it is notable that the increase in the total landings is mainly attributable to the increase in total
fishing effort (i.e., total number of fishing vessels). Despite the significant increase in total
catches, catch per unit of fishing effort-CPUE (i.e., horsepower – HP) has declined noticeably
over the past decades. Specifically, the average catch per fishing effort was about 1.1 ton/HP
in 1980s, but this figure decreased considerably to around 0.35 ton/HP in recent years (Pham,
De Laender, et al., 2014). The decrease in CPUE, to some extent, might imply there is a decline
in fish stocks in Vietnamese marine waters, especially in coastal areas.

24
Other fishing vessel groups include lift net with light, trap and other gears.
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 129
 3,000 1.2

CPUE (ton/HP)
Total catch ('000 tons) Total catch
2,500 CPUE 1

2,000 0.8

1,500 0.6

1,000 0.4

500 0.2

0 0

Year

Figure 7.1. Total catch and CPUE over the years

Source: Pham, Gert, et al. (2014)

It is undeniable that marine capture fisheries occupy an important role in Vietnam’s economy
and socio-economic development. However, the sustainability of the fisheries sector,
especially in near-shore fisheries is increasing the concern of fisheries managers, researchers,
fishers and relevant stakeholders. The problem of over-capacity and dwindling fisheries
resources in coastal areas seems to be the most challenging for fisheries management in
Vietnam.

7.2.2. Fisheries management objective

The Fisheries Law 2003 is the highest legal framework for fisheries management in Vietnam.
Under the Law, there are a number of legal documents issued to manage and regulate fishing
activities and fisheries resource management. The objective of fisheries management is not
implicitly defined in the Fisheries Law 2003 but some general principles for fishing activities
are introduced in Article 4 of the Fisheries Law 2003. The principles are as follows:

(i) Economic efficiency in fishing activities have to be in line with fisheries resources
protection and preservation; biodiversity; and environmental and natural resource
protection.
(ii) Planning and developing fisheries sector must be based on fisheries development plan
of the entirely country and each region.
(iii) Safety for fishers, fishing vessels, gears and other infrastructures related to fishing
activities (e.g., fishing ports, markets) needs to be given attention.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 130
 (iv) The development of fisheries sector and fishing activities must be incorporated with
border security, defence and sovereignty of the nation.
Furthermore, the most recent fisheries development strategy is the Master plan for fisheries
development to 2020 and vision toward 2030 approved by the Central government on 1 March
2013. The Master plan not only sets the targets for fisheries sector (e.g., total number of fishing
boats and total catch among others) but also provides a framework and directions for fisheries
development strategies in the future. The specific targets for the total catch of the whole
country and each fishing ground, number of fishing vessels and fishing fleet structure are also
determined in the Master plan. According to the Master plan, some general objectives of the
Master plan and future directions for fisheries development are as follows:

 Fisheries sector (including aquaculture) would become an industrialised sector in 2020

and modernised industry in 2030;
 Fisheries industry would develop sustainably and become a mass commercially
oriented production sector with rational production structure;
 Fisheries sector would become a highly productive and competitive sector;
 Increasing income and living standards of fishers and fishing communities, and creating
employment opportunities and livelihoods for fishers and fishing communities;
 Fishing activities have to incorporate ecological environment and marine resources
protection and conservation;
 Fisheries development must incorporate national defence, sovereignty and security of
the nation.
In addition, a number of specific objectives of the Master plan are also defined including
fisheries production volume and value, fishing gear structure, fishing ground and the number
of fishing vessels among others. For instance, according to the Master plan, the total capture
fisheries production (including inland fisheries) in 2020 would be at 2.4 million tons in which
marine capture fisheries would account for 2.2 million tons. The total number of fishing boats
in 2020 will be kept at 110 thousand boats and at 95 thousand boats at 2030. The Master plan
also proposes a plan to restructure the fishing fleets by reducing the number of destructive
fishing gears such as trawl gears and trap techniques and increasing the number of friendly and

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 131
highly selective fishing gears including offshore purse seine, handline and longline are
priorities.

Table 7.3. Fisheries management objective

Objectives Description

Economic Economic efficiency in fishing sector

objectives A mass commercially-oriented production sector with rational
production structure
Competitive and modernised fishery sector

Biological Sustainable fisheries sector

objectives Ecological environment and marine resources protection and
conservation

Social objectives Increasing income and living standard of fishers and fishing
communities,
Livelihoods for fishers and fishing communities, and safe
conditions for fishing vessel and fishers

Political objectives National defence, sovereignty and security at sea

Generally, fisheries management in Vietnam aims at several objectives including economic,

biological, social and political objectives (albeit poorly defined objectives). However, in
reality, social objectives such as livelihoods, food security, nutrition source, and employment
creation for fishers and fishing communities have the highest priorities as fisheries in Vietnam,
and contribute to fisheries still being open-access.

7.2.3. Trawl fisheries information

Trawl fisheries play an important role in marine capture fisheries in Vietnam in terms of both
the number of fishing boats and catch volume. Trawl-fishing technique is one of the most
common gears after gillnet. Trawlers come from all 28 coastal provinces in Vietnam, stretching
from the North to the South.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 132
 Engine <90 CV Engine ≥90 CV
25000

Fishing boats 20000

8442

15000
13173

10000

14112
5000
6401

0
2010 2017 Year

Figure 7.2. The number of trawl fishing vessels in 2010 and 2017

Source: Directorate of Fisheries (2012, 2018).

In general, from 2010 to 2017, the number of trawlers decreased relatively with a significant
change in the fleet structure. From 22,554 fishing vessels in 2010, the total number of trawl
fishing vessels decreased to 19,574 fishing boats, accounting for 17.4% total fishing vessels in
Vietnam (Figure 7.2 and Table 7.2).

The structure of trawl fishing fleet also changed considerably between 2010 and 2017. The
proportion of offshore trawlers (engine ≥ 90 CV), accounting for 37.4% in 2010 increased to
67% in 2017 while the share of coastal and inshore trawl vessels (engine <90 CV) decreased
noticeably, from 62.6% in 2010 to 32.7% in 2017 (Figure 7.2). It is worth noting that the
increase in the proportion of offshore trawl vessels is mainly due to policies introduced by the
government to promote the development of offshore fisheries, and limit and discourage the
increase in coastal and inshore vessels over the past years.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 133
 Others
11%
Handline
3%

Gillnet
6%

Trawl
Pure sein 56%
24%

Figure 7.3. Total catch volume of marine fisheries in 2017 by fishing gear

Source: Directorate of Fisheries (2018)

Regarding total catch volume, trawl fleet play an important role for total marine capture volume
in Vietnam. In 2017, the total production volume of trawl fishing fleet was about 1.7 million
tons, accounting for 56% total marine capture volume that was greater than that of other fishing
fleets altogether.

In term of fisheries management, no specific legal frameworks for management of trawl fishery
in Vietnam exist (Thong, 2013). Instead, general legal frameworks are applied to all fishing
gears.

Due to the important role of trawl fishery to the fisheries sector (e.g., production volume,
employment opportunities), the sustainable development of trawl fisheries is a great concern
of fisheries managers, fishers and stakeholders. Overfishing capacity (i.e., the number of trawl
vessels), especially coastal and inshore trawl vessels can be seen to be the biggest concern of
fisheries managers. While fisheries resources in coastal and inshore areas have been
overexploited, the number of small trawlers are higher than an acceptable level for sustainable
development (Thong, 2017). According to the Master plan, the total number of trawlers in 2015
would be limited to below 15% total fishing vessels in Vietnam (The Prime Minister, 2013a).
Nevertheless, in reality the number of trawl fishing vessels in 2015 was 20201 boats,
accounting for 17% (2% higher than defined number) of total fishing boats (Department of
Capture Fisheries and Resources Protection, 2015). Further, trawl-fishing method is considered

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 134
unfriendly to coastal ecosystems and marine habitat (e.g., non-selective gear, harmful
technique to seagrass, coral and seabed) (Pham, De Laender, et al., 2014). The proportion of
trash fish species of trawl fishing method is relatively high, accounting for from 40% to 80%
of total catch (Pomeroy et al., 2009; Thanh Long, 2015). In addition, the post-harvest loss of
trawl is also high compared to other fishing methods, around 15% to 25% (Directorate of
Fisheries, 2018).

Due to the excessive number of trawl vessels and the environmentally unfriendly fishing
method of the trawl vessel to marine ecosystems, reducing the number of small vessels (engine
<90CV) is a priority of the government. Recently, the government has introduced a number of
measures and policies to mitigate the number of trawl fishing vessels. For instance, no new
fishing license for trawl fishing gear has been issued since 2016. Further, other existing fishing
gears are not permitted to switch to trawl fishing gears (Ministry of Agriculture and Rural
Development, 2015). Instead, the government encourages small trawlers to switch to other
fishing techniques such as gill net, purse seine, handline and longline that are considered
environmental-friendly and high-selective fishing gears.

7.3. Field survey

7.3.1. Trawl fishery and regional conditions in study areas

The study considers trawl fisheries in two provinces of Vietnam: Quang Ninh in the north and
Ben Tre in the south of Vietnam (Figure 7.4).

Quang Ninh is one of the largest fishery centres in Vietnam, and includes fisheries operating
using trawl, gillnet, line (handline and longline), and other fishing gears. In 2015, Quang Ninh
was home to 8,032 fishing vessels, of which 871 were trawlers (around 11%) (Department of
Agriculture and Rural Development of Quang Ninh, 2016). Quang Ninh is the largest province
in Northern Vietnam in terms of the number of trawl vessels (Department of Agriculture and
Rural Development of Quang Ninh, 2016). Although fishers come from several districts in the
Quang Ninh province, most trawl vessel owners are from Quang Yen and Van Don Districts,
where fishing activities have been long recognised as traditional livelihood activities. Fishers
fish year round and only stay onshore when weather conditions are unfavourable for fishing
(e.g., storms and rough sea). The catch of trawl vessels comprises a wide number of species
that can be broadly categorised into fish, crabs, octopus, cuttlefish and squids, shrimps, and
trash fish species (e.g., low-valued species and/or bycatch). Most catch is handled and

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 135
preserved in ice, although some species (e.g., shrimp, crab and high valued species) are kept
alive by using an aerator as they command a higher price. Fishers may opt to sell their catch
either directly to middlemen or local fish markets or both (i.e., some quantity of landings to
middlemen and the rest at the local fish market).

Figure 7.4. Geographical location of the surveyed areas

Quang Ninh is one of the largest fishery centres in Vietnam, and includes fisheries operating
using trawl, gillnet, line (handline and longline), and other fishing gears. In 2015, Quang Ninh
was home to 8,032 fishing vessels, of which 871 were trawlers (around 11%) (Department of
Agriculture and Rural Development of Quang Ninh, 2016). Quang Ninh is the largest province

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 136
in Northern Vietnam in terms of the number of trawl vessels (Department of Agriculture and
Rural Development of Quang Ninh, 2016). Although fishers come from several districts in the
Quang Ninh province, most trawl vessel owners are from Quang Yen and Van Don Districts,
where fishing activities have been long recognised as traditional livelihood activities. Fishers
fish year round and only stay onshore when weather conditions are unfavourable for fishing
(e.g., storms and rough sea). The catch of trawl vessels comprises a wide number of species
that can be broadly categorised into fish, crabs, octopus, cuttlefish and squids, shrimps, and
trash fish species (e.g., low-valued species and/or bycatch). Most catch is handled and
preserved in ice, although some species (e.g., shrimp, crab and high valued species) are kept
alive by using an aerator as they command a higher price. Fishers may opt to sell their catch
either directly to middlemen or local fish markets or both (i.e., some quantity of landings to
middlemen and the rest at the local fish market).

Ben Tre province is situated in the Mekong Delta in the south of Vietnam. As with Quang
Ninh, there are a number of different fishing gear types used in the province (e.g., trawl, gill
net, purse seine, and longline among others), although trawl is the dominant activity,
constituting 70% of the total number of fishing vessels (3.626 vessels) and about 87% of the
total annual landing volume of the province in 2012 (Thong, 2015; Tinh & Thanh, 2014). In
2015, the total number of nearshore trawl vessels (i.e., engine power less than 90 CV) was 992
vessels, accounting for 39% of total trawlers (Department of Capture Fisheries and Resources
Protection of Ben Tre, 2016). Most small-scale trawl vessel owners live in Binh Dai district
which is considered one of the poorest coastal districts in the province (Kim Anh & Flaaten,
2011). Fishers fish all year round when weather conditions are favourable. The catch consists
of a number of species due to the multi-species fisheries nature of the fishery and nonselective
trawl fisheries, including fish, crabs, squids, shrimps, and trash fish species. Catches are
normally sorted out and stored in plastic containers with ice, and then primarily sold to local
middlemen.

The two regions are substantially different also in terms of opportunity cost of labour. Ben Tre
is one of the poor provinces in Vietnam (Kim Anh & Flaaten, 2011) while Quang Ninh is
considered to be a developed industrial province (General Statistics Office of Vietnam, 2016a).
Table 7.4 illustrates the differences between the two provinces regarding geo-economics,
natural resources and socio-economics.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 137
 Table 7.4. Geo-economics and socio-economic characteristics in the two provinces
Quang Ninh Ben Tre

 Natural conditions  Natural conditions

 The richest coal mines in the country  One of the most vulnerable provinces to
(providing 95% of the total hardcoal climate change and sea level rise
production in Vietnam)  Saline intrusion and severe weathers (i.e.,
 Border economic zone with China: flood, drought etc.)
Important import and export hub  Large number of people whose
 Attractive tourist destinations for livelihoods depend strongly on natural
international and domestic tourists: i.e., conditions
Ha Long Bay (a UNESCO World
Heritage Site), island and beach tourism
 Developed ship building and seaport
industries
 Socio-economic indicators  Socio-economic indicators
 Monthly income per capita in 2014: 3,056  Monthly income per capita in 2014: 2,161
thousand VND (1) thousand VND (1)
 GDP structure (2014): Agriculture,  GDP structure (2014): Agriculture,
forestry and fishery (5.8%), industry forestry and fishery (42.8%), industry
(50%) and service (44.2%) (2) (34.5%) and service (24.3%) (2)
 Proportion of poor households in 2015:  Proportion of poor households in 2015:
4.56% (3) 12.01% (2)
 Proportion of monthly income per capita  Proportion of monthly income per capita
by sources of income in 2014: by sources of income in 2014:
Salary/wage (51.8%), agriculture, Salary/wage (32.6%), agriculture, Forestry
Forestry and Fishery (10.1%); non- and Fishery (30%); non-agriculture,
agriculture, forestry and Fishery (26%); forestry and Fishery (20.5%); others
others (12.1%) (3) (16.9%) (2)
 Percentage of employed workers at 15  Percentage of employed workers at 15
years of age and above compared years of age and above compared
provincial population: 54.7% (4) provincial population: 62.8% (4)
 Percentage of literate population at 15  Percentage of literate population at 15
years of age and above: 96.8% (5) years of age and above: 94.6% (5)

Note: Data sources: (1) General Statistics Office of Vietnam (2016b); (2) Quang Ninh Statistics
Office (2016) and Ben Tre Statistics Office (2016); (3) Ministry of Labour - Invalids and Social
Affairs (2016); (4) General Statistics Office of Vietnam (2018a); and (5) General Statistics
Office of Vietnam (2018b).

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 138
7.3.2. Data collection procedure

Primary data collection through survey was approved by the Research Ethics Committee –
Queensland University of Technology (Ethics Approval Number 1600000916) (Appendix 1:
Human research Ethics approval certificate). The Ethics category of the research was low-risk.

Primary data used in this case study were collected through a designed questionnaire (following
Thuy, 2013). The questionnaire covered a wide range of information that are relevant to trawl
fishing. These include vessel physical characteristics (e.g., engine size and vessel length);
fishing operating costs (e.g., fuel, provision, ice, labour and other operating costs); information
about catch and revenue; vessel owner and/or skipper’s characteristics and their demographic
characteristics (e.g., age, experience and education level). The questionnaire was pre-tested,
and then corrected to examine the applicability and appropriateness of the questionnaire before
using it for the official survey. The corrected questionnaires were used to collect data relating
to fishing activities for the year 2015 through a face-to-face survey. The respondents were
informed about the purpose the research. If they agreed to participate in the survey then the
survey took place with their consent. The face-to-face survey was conducted in Vietnamese.
The survey was undertaken between November 2016 and January 2017. Totally, 165 trawl
vessel owners were surveyed of which 89 fishers were from Quang Ninh province and 76 came
from Ben Tre province (see Figure 7.4 for survey areas).

In Quang Ninh, the survey was conducted in three different fishing harbours (Hon Gai (36
fishers), Van Don (34 fishers) and Ben Do (19 fishers)) where most fishing vessels land their
catch. The survey was undertaken while fishers were anchoring their boats on the fishing ports
to sell their fish and/or make any necessary repairs to their boat and/or fishing gear and resupply
the boat with groceries for their next trip. Fishers generally stay on their boats after returning
from sea because their home is generally far from the fishing harbour.

In Ben Tre province, as previously mentioned the majority of small-scale trawlers are
concentrated in the poorest coastal district, that is, Binh Dai district (Department of Capture
Fisheries and Resources Protection of Ben Tre, 2016; Kim Anh & Flaaten, 2011). Hence, Binh
Dai district was chosen for the survey. Trawl vessel owners were questioned about their fishing
in 2015 either at their home or fishing ports. In total, 76 fishing vessel owners were surveyed
in Ben Tre province.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 139
7.4. Current socio-economic status of fishers (Activity 4)

7.4.1. Introduction

Existing literature indicates that socio-economic characteristics such as age, gender,

educational level, employment status and demographic characteristics (e.g., family size,
generation and tradition) have been shown to be major factors affecting individual behaviour
in fisheries (Aldon et al., 2011; Pascoe, Cannard, et al., 2015; Tzanatos et al., 2006). Further,
fisheries management is about managing the people who fish (Hilborn, 2007b; Wilson &
McCay, 2001). Thus, understanding information on socio-economic status and livelihood
activities of fishers can provide important insights for designing and developing appropriate
fisheries policies and regulations (Perret & Yuerlita, 2014).

This is particularly important for small-scale fisheries in undeveloped countries as small-scale

fisheries in developing countries play an important role for food security, poverty alleviation,
and job creation, socio-economic and so on (Andrew et al., 2007; Béné, Macfadyen, & Allision,
2007). Like other developing countries, small-scale marine capture fisheries in Vietnam play
an important role to the country’s social and economic development as the industry provides
directly/indirectly job opportunities, food and livelihood for millions of people (Ministry of
Agriculture and Rural Development, 2012; Pomeroy et al., 2009). However, understanding the
socio-economic status of fishers is limited due to insufficient human and other necessary
resources despite its importance for forming and designing effective fisheries management
policies.

The main aim of this section is to provide the status quo of socio-economic profile of trawlers
in Vietnam. The following sections will present surveyed results, discussion and conclusion
around socio-economic status of trawlers in Vietnam.

7.4.2. Socio-economic profile of trawlers

Surveyed results show that all vessel owners were male and they all were skippers. Both males
and females were directly involved in fishing activities (i.e., being crewmembers) although
male fishers were dominant (accounted for 84.4%) (Figure 7.5). In Quang Ninh, the proportion
of females who engaged directly in fishing was relatively higher than that of Ben Tre, averaging
21.1% and 7.8% females involved in Quang Ninh and Ben Tre, respectively.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 140
 100 92.2

%
90 84.4
78.9
80
70
60
50
40
30 21.1
20 15.6
7.8
10
0
All sample Quang Ninh Ben Tre
Province
Male Female

Figure 7.5. The proportion of males and females directly engaged in fishing

The age of fishers in both areas appeared to be quite similar, at around 39 years old but vessel
owners in Quang Ninh had slightly more fishing experience than those in Ben Tre, 17.9 and
16.4 years, respectively. However, the vessel ownership experience of trawlers in Quang Ninh
and Ben Tre appears to be quite similar, at around 11 years (as seen in Table 7.5).

Empirical results also reveal that family size of fishers were 4.3 people on average, ranging
from 2 to 9 people. Family size between the two regions were also quite similar, about 4.3
people per household. In general, each fishing household had about 1.5 people who were
involved directly to fishing (i.e., crewmember). However, the number of fishers’ family
members in Quang Ninh tends to be higher participation in fishing than that of Ben Tre,
averaging 1.8 and 1.2 people in Quang Ninh and Ben Tre, respectively. Additionally, vessel
owners in Quang Ninh had more children in school than vessel owners in Ben Tre had, around
1.6 and 1.1 children in Quang Ninh and Ben Tre province, respectively.

Results from Table 7.6 show that only a small proportion of surveyed vessel owners was the
first generation, about 38%. By contrast, the majority of fishers came from traditional fishing
families (i.e., those families who had more than one generation involved in fishing activities).
The proportion of vessel owners who are the second, third and above fourth generations were
30.3%, 26.7% and 4.8%, respectively.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 141
Table 7.5. Socio-demographic characteristics of vessel owners

Whole sample (N=165) Min 1st Qu Median Mean 3rd Qu Max

Age (years) 21 32 38 38.8 45 61
Fishing experience (years) 2 12 17 17.1 20 40
Trawling experience (years) 2 8 12 12.4 15 30
Vessel owner experience (years) 1 6 10 11.3 15 31
Fishing involvement (persons) 1 1 2 1.6 2 3
Family size (persons) 2 4 4 4.3 5 9
Children schooling (persons) 0 1 1 1.4 2 4
Quang Ninh (N=89)
Age (years) 22 33 39 39.0 45 61
Fishing experience (years) 2 13 17 17.9 21 40
Trawling experience (years) 2 9 13 12.6 15 30
Vessel owner experience (years) 1 6 10 11.8 15 30
Fishing involvement (persons) 1 1 2 1.8 2 3
Family size (persons) 2 4 4 4.5 5 6
Children schooling (persons) 0 1 2 1.6 2 4
Ben Tre (N=76)
Age (years) 21 31.8 37 38.6 45 61
Fishing experience (years) 5 10 15.5 16.4 20 35
Trawling experience (years) 2 8 10 12.1 15.3 30
Vessel owner experience (years) 1 6 10 10.8 15 31
Fishing involvement (persons) 1 1 1 1.3 1.25 3
Family size (persons) 2 3 4 4.2 5 9
Children schooling (persons) 0 0 1 1.1 2 3
Note: N indicates the number of surveyed vessel owners

Table 7.6. Information about fishing household generation

1st generation 2nd generation 3th generation 4th and above
Province No. No. No. No.
% % % %
fishers fishers fishers fishers
Whole sample 63 38.2 50 30.3 44 26.7 8 4.8
Quang Ninh 39 43.8 13 14.6 29 32.6 8 9.0
Ben Tre 24 31.6 37 48.7 15 19.7 0 0.0

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 142
A summary of educational level25 of surveyed vessel owners is presented in Table 7.7. The
findings indicate that the vast majority of trawlers only finished primary and secondary school.
These proportions were 50.3% and 41.8% for primary and secondary school levels,
respectively. Only a few fishers obtained high school level, making up only 3%. About 5% of
fishers reported that they did not have formal schooling and no one attained higher education
level (i.e., college or university education). It is also notable that there is a difference in
educational level between the two areas. Fishers in Quang Ninh had a higher educational level
than vessel owners in Ben Tre. A majority of respondents, accounted for 65% in Quang Ninh,
stated they obtained secondary school level whereas most trawlers, about 75%, in Ben Tre only
achieved primary school level. The proportion of fishers who had high school level in Quang
Ninh also outnumbered that of those in Ben Tre, accounted for 4.5% and 1.3%, respectively.
Furthermore, the proportion of fishers who had no schooling in Ben Tre was relatively higher
than that of Quang Ninh, averaging 9.2% and 1.1%, respectively.

Table 7.7. Education level of fishers

Whole
Quang Ninh % Ben Tre % %
sample
No schooling 1 1.1 7 9.2 8 4.8
Primary school 26 29.2 57 75.0 83 50.3
Secondary school 58 65.2 11 14.5 69 41.8
High school 4 4.5 1 1.3 5 3.0
Total 89 100 76 100 165 100

Surveyed trawlers were also asked about whether their families have other alternative
livelihoods apart from fishing. The surveyed results indicate that the livelihood of vessel
owners was mainly categorised into three groups: agriculture, aquaculture and other activities
(e.g., trading and self-employment). Specifically, almost half of the surveyed fishers (46.1%)
depended solely on fishing for their living, without non-fishing livelihood sources (Table 7.8).
This means that these fishing households had no other income-generating activities. Apart from
fishing, agricultural activities (i.e., mainly rice production), was the most prevalent livelihood
of the fishing households in both regions. The farming activities attracted about 50% and 38%
of fishing households engaged in this livelihood in Quang Ninh and Ben Tre, respectively. By

25
Formal education in Vietnam consists of 12 years of basic education, dividing 3 levels namely primary,
secondary and high school levels. Primary level lasts for 5 years, beginning with grade 1 to grade 5. Secondary
level lasts for 4 years, from grade 6 to grade 9. High school starts from grade 10 to grade 12.
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 143
contrast, aquaculture and other activities were much less common livelihoods of the surveyed
fishing households in both areas, making up only 4.8% each. However, it is notable that fishers
in Quang Ninh are more likely to engage in non-fishing livelihoods than those in Ben Tre.

Table 7.8. Livelihood activity of fishing household

Quang
Livelihood activities26 % Ben Tre % All %
Ninh
Agriculture 44 49.4 29 38.2 73 44.2
Aquaculture 2 2.2 6 7.9 8 4.8
Others 6 6.7 2 2.6 8 4.8
* Note: One household may have more than one livelihood activity.

Regarding catch distribution channel, all fishers in Ben Tre responded that they sold all their
catch directly to middlemen whereas fishers in Quang Ninh province sold their catch to either
midddlemen or local fish markets (i.e., end consumers, restaurants and tourists) or both
channels (i.e., middlemen and fish markets). Regarding Quang Ninh area, about 60% of fishers
sold their catch through both channels inlcuding middlemen and local fish markets.
Approximately 25% of fishermen sold all their catch at local fish markets while only 16%
vessel owners sold all their catch directly to middlemen as seen in Figure 7.6.

25%

59%
16%

Middlemen and fish market Middlemen Fish market

Figure 7.6. Catch distribution channel in Quang Ninh

26
Some fishing households had more than one livelihood activity. However, those kinds of households were few.
In total, only 4 fishing households (3 households in Ben Tre and 1 household in Quang Ninh) were involved in
both aquaculture and agriculture activities, and only one fishing household was involved in both agriculture and
others.
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 144
7.4.3. Discussion

The findings from the survey show that fishing vessels are owned and operated by fishing
households. Both male and female fishers are directly involved in fishing activities but often
in different roles and activities. Male are dominant and masters at sea while the main tasks of
female fishers primarily include catch sorting, net repair, provision preparation and fish
marketing (i.e., selling fish at local fish markets). The participation and role of women in
capture fisheries are widely documented (Harper, Grubb, Stiles, & Sumaila, 2017; Harper,
Zeller, Hauzer, Pauly, & Sumaila, 2013). In Vietnam, women who directly engage in marine
capture fisheries in Vietnam were recognised in a number of fisheries (Harper et al., 2017).
The high proportion of males involved in fishing is understandable as fishing activities are
likely to be physically demanding and a risky occupation (FAO, 2016; Pfeiffer & Gratz, 2016).
Fishers often face the harshness and vagaries of the weather, remoteness of fishing ground, and
unexpected events at sea, especially in offshore areas (Duy, Flaaten, Anh, & Ngoc, 2012; Long
et al., 2008).

The age of fishers on average is about 39 years27. Surveyed fishers stated that they often
followed their parents to sea at an early age (e.g., from 12 to16 years old) after dropping out of
school due to some obstacles such as demand for labour forces of their family and/or the
financial constraints of schooling. At first, they served on the fishing boat to help their parents
to do some work onboard such as sorting and grading fish, fishing gear preparation, cooking
and housework. Through the years, they acquired fishing skills and experiences from their
parents, peers and other crewmembers. They, then, often become independent fishers and
vessel owners when they get married. Young fishers, to some extent, would be a great
opportunity for the marine capture sector to boost its productivity and optimise the country’s
human capital through skills training, education and technology transfer. Especially, this is
quite important when over the past years the Vietnamese government has provided tremendous
efforts to shift its fisheries from coastal and small-scale fisheries to modernised offshore
fisheries that require advanced technology and skills in fishing.

The results also show that fishing is a traditional occupation for most surveyed fishing
households. The majority of trawlers come from traditional fishing families while a small

27
In Vietnam, working age is between 15 and 60 for men and between 15 and 55 for women.
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 145
number of fishers (38.2%) are not traditional fishing households. For traditional fishing
households, fishers tend to be handed fishing activities from previous generations, and parents
often hand fishing vessels and fishing gears to their sons when they get old and/or exit fishing
activities. It is noteworthy that fishing expertise and experience of fishers are partly obtained
through accumulated experiences that are handed down from generation to generation within
fishers’ family, social interactions among fishers/peers and “learning by doing”.

It is notable that family size of the surveyed vessel owners, on average 4.35 people per
household, is higher than the average level of nation (averaging 3.81 people per household)
(General Statistics Office of Vietnam, 2016b). The family size of surveyed fishing households
in both provinces is also higher than the average family size of households in Quang Ninh (3.7
people) and Ben Tre (3.4 people) (General Statistics Office of Vietnam, 2016b). A possible
reason is that most fishing households are living in coastal and/or rural areas. The fact that a
large number of families in rural areas still keep in mind with cultural perceptions that
“Happiness is having many children – Có phúc thì có nhiều con”, or “Many children are better
than many possessions – Nhiều con hơn nhiều của” or “son preference” are still common
perceptions (Gallup, 1995). Furthermore, as the education level of the surveyed fishers is
generally low that possibly leads to fishing households being likely to have more children than
other households (Kim Anh & Flaaten, 2011).

As far as educational attainment of fishers is concerned, it is noticeable that educational level

of fishers is relatively low. Only 3% of fishers attained high school level, and there were about
4.8% of illiterate trawlers. This proportion is relatively higher than the national average level
(about 2%) (Kim Anh & Flaaten, 2011). Most small-scale fishing households live in rural-
coastal areas, and are considered to be poor. Due to financial constraints and demand for labour
forces, fishers are unlikely to have the opportunity to attain high education levels. The low
education level of vessel owners may cause hardship for fishers not only to adopt advanced
technology for fishing, but also to find alternative professions and/or non-fishing livelihoods.
This is particularly challenging for fishers when they have to cope with declines in fish stocks,
market fluctuations, natural shocks and vagaries (e.g., environmental variability, and harsh and
irregular weather events), and other external shocks (Béné, 2009).

The surveyed results indicate that fishing plays an important role by providing livelihoods and
employment opportunities for a large number of the surveyed households. Almost half of
respondents solely reply on fishing to make a living. Apart from fishing, agricultural practice

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 146
is the most common livelihood of the fishing households. However, fishers stated the product
of their agricultural production is mainly rice that is primarily used for their families’
consumption. Although aquaculture and other livelihood options are also other income-
generating activities of some fishers, only some the surveyed fishing households participated
in these as these production activities require much capital investment, knowledge and
experience as well as high risk. The existing literature indicates livelihood diversification is
important to fishing households to both maximise earning and leave households less vulnerable
(Béné, 2009). However, both aquaculture and agriculture practices embody various risks
(Betcherman & Marschke, 2016). In fact, small-scale fishers are faced with a number of
constraints that may hinder them finding alternative livelihood options. These include a lack
of investment capital (e.g., aquaculture practice requires high capital investment and necessary
technical knowledge), the availability of land, regional and economic conditions, and other
constraints. For example, a number of interviewed fishers in Ben Tre province responded that
they had neither a piece of land nor the necessary educational level so fishing was a last resort.
Some have cropland for farming but natural conditions such as saline intrusion and drought are
not favourable for farming as Ben Tre province is one of Vietnam’s most vulnerable provinces
to climate change (Le, Nguyen, & Jolly, 2016; Nguyen, Jolly, Bui, & Le, 2015). Furthermore,
the high production cost of farming stemming from high cost of fertilizers, seeds and related
operating costs, and market access all combined to make crop farming ineffective and
unprofitable (CCAFS-SEA, 2016; Kim Anh & Flaaten, 2011). The linkage between livelihood
diversification and fishing has important implications for fisheries management strategies.
When fishing is a last resort of fishers for their living, creation of non-fishing livelihood options
for fishers might be seen as a key response to reducing fishing capacity in coastal areas by
helping a number of small-scale fishers exiting the fisheries (Martin, Lorenzen, & Bunnefeld,
2013).

Of particular concern is the catch distribution channel (selling fish after return from sea) in the
surveyed areas. All fishers in Ben Tre province sold all their catch to wholesaler/middlemen
who then sold the catch to seafood processing companies and/or restaurants and/or end-
consumers at local fish markets. The reason is that Binh Dai district is one of the poorest coastal
districts in Ben Tre province so poor infrastructure and market access constraints may prevent
fishermen selling their fish at local fish markets and/or city with a high price. On the other
hand, fishers in Quang Ninh have different options to sell their fish. Quang Ninh province is
one of the most famous tourism destinations as well as an industrial province in Vietnam with

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 147
high living standard and favourable conditions for economic development (Broemme & Stolpe,
2007; Mai, Rahtz, & Shultz, 2014). Fishing harbours in Quang Ninh are primarily located near
central cities, tourism centres, and local fish markets. Thus, fishers have a variety of options
for selling their catch.

7.4.4. Conclusions

Understanding the socio-economic and demographic characteristics of fishers is important

information for managing fisheries. The main aim of this section is to provide a socio-economic
characteristic profile of trawlers in Vietnam. The results from the surveyed data suggest that
the educational level of trawlers is rather low and would possibly prevent the fishers from both
adopting advanced technologies in fishing and also finding alternative non-fishery livelihoods.
The majority of vessel owners depended solely on fishing to make a living, especially in the
poor region: Ben Tre province. Although some fishers have other non-fishery livelihoods such
as aquaculture and agriculture production but these livelihood activities are mainly rice crop
farming for their own family consumption purpose, and also embody risk in production (i.e.,
aquaculture). Furthermore, crop production is considered unprofitable due to high costs of
inputs and unfavourable natural conditions. Non-fishing livelihoods are important to small-
scale fishers but the low education level of fishers, unfavourable natural and regional economic
conditions, and a lack of investment capital probably deter fishers from finding and/or
diversifying alternative livelihood options. Furthermore, market access constraint is also
difficulty for small-scale fishers in a poor region such as Ben Tre where all fishers sell their
fishes to middlemen.

7.5. Technical efficiency of trawl fishery (Activity 5)

7.5.1. Introduction

Marine capture fisheries in Vietnam are open-access and predominantly small-scale

(Directorate of Fisheries, 2013; Long et al., 2008; Pomeroy et al., 2009; Raakjær et al., 2007).
Over the past few years, the number of fishing boats has increased dramatically, raising concern
about the sustainability of the fisheries (Directorate of Fisheries, 2013). Furthermore, the focus
of fishing activity has largely been coastal water areas, leading to depleted coastal marine
resources and over-capacity (Pham, De Laender, et al., 2014; Pomeroy et al., 2009). Recently,
actions have been taken to reduce the fishing pressure on coastal fisheries. These measures

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 148
have been aimed at reducing the number of fishing vessels operating in coastal areas and
encouraging the development of larger fishing vessels to fish in offshore areas (Directorate of
Fisheries, 2013; The Prime Minister, 2014). Small-scale trawl vessels in particular have been
prioritised for capacity reduction. The Master plan approved by the Central government on 1
March 2013 set a target for the total number of trawlers to be less than 15% of total fishing
vessels in Vietnam by 2015 (The Prime Minister, 2013a). However, in 2015, these boats
accounted for 17% of total fishing boats in Vietnam – 2% higher than the target (Department
of Capture Fisheries and Resources Protection, 2015). In addition, coastal trawl vessels are
considered the most harmful and unfriendly-environmental fishing method to coastal
ecosystems (Pham, De Laender, et al., 2014).

However, previous studies have demonstrated that the success of capacity reduction programs
is affected by variations in efficiency between boats (Pascoe, Cannard, et al., 2015; Pascoe &
Coglan, 2000). Given this variation, the choice of which vessel to remove has an impact on the
efficacy of the management measure, and may be more important than how many fishing
vessels are removed (Herrero & Pascoe, 2003). This is because removing inefficient fishing
boats will have proportionally less of an impact on the overall fishing capacity of the fleet
(Herrero & Pascoe, 2003; Pascoe & Coglan, 2000; Pascoe, Hassaszahed, et al., 2003). As a
result, the development of an effective fishing capacity reduction scheme needs information on
technical efficiency and its variation between fishing boats. Information on the level of
technical efficiency and distribution of efficiency and its drivers at fishing boat and/or fleet
levels are also useful to formulate appropriate regulations and policies (Sharma & Leung, 1998;
Tingley et al., 2005). For instance, fishers can improve their performance and earnings by
improving their efficiency while policymakers could use this information to formulate
appropriate and effective regulations, interventions and measures for sustainable fisheries
management.
Removal of inefficient vessels also assumes that these fishers have an alternative livelihood to
which they can move. However, persistence of inefficient vessels in a fishery may reflect the
low opportunity cost of them staying, which in turn is affected by the regional economic
conditions. While non-malleability of vessel capital in fisheries is well established (C. W. Clark
et al., 1979), if human capital is also non-malleable, for example due to a lack of alternative
employment opportunities, then fishers have an added incentive to remain in the fishery, even
if it is inefficient (Bjørndal & Munro, 2012; Nøstbakken, Thebaud, & Sørensen, 2011). Hence,
the opportunity cost of labour may be a further factor affecting the persistence of inefficient

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 149
vessels, and policy makers may best be able to improve fishery performance by providing
suitable alternative employment opportunities.

The objectives of this study are (i) estimate the level of technical efficiency of trawl fishing
vessels in Vietnam; (ii) identify factors affecting the level of technical efficiency of the trawl
vessel owners; and (iii) propose initiatives that fishery managers can implement to manage
capacity in the trawl fishery.

7.5.2. Methods

In this study, stochastic production frontier approach was applied to estimate technical
efficiency and its determinants. A translog production frontier was estimated, and the adoption
of three-step procedure to ensure theoretical consistency was applied (Henningsen & Henning,
2009). The three-step approach has been applied in a number of fisheries efficiency studies.
These studies include efficiency study of banana prawn fishery in Australia (Pascoe, Hutton,
et al., 2017); research on efficiency of tiger prawn fishery in Australia (Pascoe et al., 2012);
efficiency study of aquaculture in Sri Lanka (Kularatne et al., 2018); and technical efficiency
evaluation of semi-industrial fleet in Mexico (Quijano et al., 2018). An explicit inefficiency
model was also used (Battese & Coelli, 1995) to identify potential factors affecting
inefficiency. Further information about the stochastic production frontier and three-step
procedure approach used in this study were presented in Chapter 3.

7.5.3. Data

Detailed information on data collection was also previously presented in section 7.3.2. The
variables used for the efficiency analysis are given in Table 7.9. Definition of variables in the
analysis All input and output variables (other than the dummy variables) in the production
frontier and inefficiency model were logged and normalised such that their mean was equal to
zero (i.e. ln(𝑥) = ln(𝑦) = 0) (Pascoe, Innes, et al., 2017).

The inputs used in the production function were engine power, total fishing day and crew size
representing capital invested in the fishing vessels (e.g., large vessel requires bigger engine
power then requires higher capital invested), capital utilisation and labour, respectively. A
dummy variable representing the province was also included to reflect possible differences in
relative stock abundance and prices which will also affect catches and subsequent revenue.
Given the wide variety of species caught by the non-selective fishing gear and multispecies

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 150
nature of the fishery, annual revenue was used as the main output measure. This output measure
has been widely applied in a number of multispecies fisheries efficiency studies (e.g. Fousekis
& Klonaris, 2003; Ngoc et al., 2009; Pascoe, Innes, et al., 2017; Vinuya, 2010).

The inefficiency model included the key characteristics of the fishers and vessel characteristics
not captured in the main production function. Geographical location was also included to again
capture regional specific factors affecting inefficiency. Skipper characteristics included their
age and years of experience as a vessel owner. The impact of age and experience of skipper on
technical efficiency has been investigated in a number of studies (see e.g., Coglan & Pascoe,
2007; Esmaeili, 2006; Fousekis & Klonaris, 2003; Sesabo & Tol, 2007; Tingley et al., 2005).
The number of family members participating as crew members on-board was also included into
the model. This variable was selected to check whether there is a difference in efficiency
between vessel owners who use hired crew member and those who use a family member. Age
of vessel was also included in the model. Newer vessels were more likely to involve more up-
to-date technologies, more efficient vessel design, construction and material (Pascoe,
Hassaszahed, et al., 2003). Previous studies showed that a vessel’s age has affected its
efficiency (e.g., newer vessels tend to have higher efficiency than older ones) (e.g., Fousekis
& Klonaris, 2003; Greenville et al., 2006; Pascoe et al., 2001; Pascoe & Coglan, 2002; Pascoe
et al., 2012).

The distance from the fishing harbour to the main fishing ground was also included in the
analysis to capture the effects of local variations in marine resource abundance accessible to
the fisher. The origin of vessel’s engine (Japanese or otherwise) was also included to account
for potential difference between fishing vessels due to their different types of engines (other
than their rated power).

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 151
Table 7.9. Definition of variables in the analysis
Variable Definition Nature
Revenue Total annual revenue (Million VND)28 Continuous
Engine Engine power (CV) Continuous
Fishing day Total fishing days (days) Continuous
Crew size Number of people including skipper (persons) Continuous
Province 1 if data collected in Quang Ninh, and 0 otherwise Dichotomous
Age Age of skipper/vessel owners (years) Continuous
Experience Vessel ownership experience (years) Continuous
Family crew Family members involved in crew size (persons) Continuous
Vessel age Age of vessel (years) Continuous
Distance Distance from home port to fishing ground (Nm) Continuous
Engine type 1 for vessel with Japanese engine, and 0 for otherwise Dichotomous
Province 1 for data collected in Quang Ninh, and 0 for otherwise Dichotomous

28
1 USD = 21,890 VND (Exchange rate on 31 December 2015)
Source: https://www.vietinbank.vn/web/home/vn/ty-gia?theDate=31%2F12%2F2015
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 152
Table 7.10. Summary of the data variables used in the analysis
All observations (n=165) Min 1st Qu Median Mean 3rd Qu Max
Revenue (Million VND) 79 270 379.8 411.7 495 126.4
Engine (CV) 20 40 50 59.6 75 300
Fishing day (day) 95 122 146 148.6 180 220
Crew size (person) 1 2 2 2.4 3 5
Vessel age (year) 1 4 8 7.9 10 20
Family crew (person) 1 1 2 1.6 2 3
Age (year) 21 32 38 38.8 45 61
Experience (year) 1 6 10 11.3 15 31
Distance (Nm) 3 7 10 13.2 15 45
Quang Ninh (n=89) Min 1st Qu Median Mean 3rd Qu Max
Revenue (Million VND) 243 360 479.5 506.7 569.7 1264
Engine (CV) 20 50 69 73.7 84 300
Fishing day (day) 135 162 170 172.6 180 220
Crew size (person) 2 2 3 2.7 3 5
Vessel age (year) 1 3 8 8.0 11 20
Family crew (person) 1 1 2 1.8 2 3
Age (year) 22 33 39 39.0 45 61
Experience (year) 1 6 10 11.8 15 30
Distance (Nm) 6 10 15 17.8 20 45
Ben Tre (n=76) Min 1st Qu Median Mean 3rd Qu Max
Revenue (Million VND) 79 210.9 278 300.5 381.2 630
Engine (CV) 20 40 40 43.1 46.5 85
Fishing day (day) 95 110 120 120.4 128.5 150
Crew size (person) 1 2 2 2.2 2 3
Vessel age (year) 1 4 8 7.8 10 20
Family crew (person) 1 1 1 1.3 1.25 3
Age (year) 21 31.8 37 38.6 45 61
Experience (year) 1 6 10 10.8 15 31
Distance (Nm) 3 6 7 7.8 10 20

Summary statistics of the data are provided in Table 7.10. The average annual revenue of all
vessels was 411.7 million VND, ranging from 79 million VND to 1,264 million VND.
Similarly, the average of engine power varies from 20 CV to 300 CV with the mean of 60 CV.
The annual revenue and engine power of trawlers in Quang Ninh were greater than those in
Ben Tre, although the former’s sample also included larger vessels (≥90 CV) that were able to
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 153
fish offshore. Fewer fishing days in Ben Tre is also partly due to unfavourable weather in the
Northeast monsoon season (Thanh Long, 2015). The average crew size was approximately 2.5
persons, ranging from 1 to 5 people, with the crew size in Quang Ninh higher on average than
Ben Tre. This correlated with the size of the vessel as fishing vessels in Quang Ninh were
bigger on average than those in Ben Tre (i.e., 73.7 CV and 43.1 CV for Quang Ninh and Ben
Tre, respectively). In relation to variables in the inefficiency model, only distance travelled was
significantly different on average between the two regions.

The correlation between the logged variables used in the model is shown in Figure 7.7. The
correlation between input variables used in the production function was generally low, although
the correlation between engine power and crew size was 0.64. The correlation among variables
in the inefficiency model also appears a relatively low correlation although experience and age
of skipper was correlated, as might be expected.

Figure 7.7. Correlation among variables in the model

7.5.4. Results

7.5.4.1. Econometric analysis of technical efficiency

The results of stochastic model are presented in Table 7.11. Given the significance of the cross-
product term (i.e., fishing day square), a Cobb-Douglas production function was not tested, and
the translog production function was accepted as the more appropriate functional form (Pascoe,
Innes, et al., 2017). Using the translog model, the model was tested for the existence of a

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 154
stochastic frontier, and the result implies that a frontier rather than production function is a
more appropriate specification29(Pascoe, Hutton, et al., 2017; Pascoe, Innes, et al., 2017).

Three inputs in the production function, namely engine power, fishing day and crew size, have
significant positive effects on the production at 0.1% or less. Given that ln(𝑥) = 0, the
coefficients of the inputs also represent their production elasticities at the mean level of all
inputs (Pascoe et al., 2012). Hence, the results indicate that increasing these inputs would result
in a less than proportional increase in the output (i.e., elasticities of all inputs in the models are
less than 1).

The initial model (step 1) did not fully fulfil the monotonicity and quasi-concavity criteria,
especially in relation to the number of fishing days and crew size. Hence, the second and third
stages previously mentioned were carried out. The sign and magnitude of estimated production
frontier parameter values in final step (step 3) are generally similar to the initial estimates.
Further, all inputs in the production function fully satisfied the requirements of monotonicity
and quasi-concavity. As expected, the intercept term in the final step was not significantly
different from zero, and the scaling coefficient was not significantly different from 1, indicating
that no bias was introduced through the three-step procedure approach (Henningsen &
Henning, 2009).

The value of gamma (γ)30 in the model was about 0.8 and was statistically significant at the
0.1% level. This suggests that the variation in fishing production not accounted for by
differences in inputs is primarily driven by differences in technical efficiency rather than by
random variation, consistent with other efficiency studies on trawl fishery (see e.g. Pascoe et
al., 2012; Pascoe, Innes, et al., 2017; Viswanathan et al., 2001). The average value of technical
efficiency of the surveyed boats was 79%, ranging from about 36% to 98% (Table 7.122). This

29
The test for existence of a stochastic frontier was conducted with null hypothesis being specified as γ = 0, where
γ = 𝜎𝑢2 /(𝜎𝑢2 + 𝜎𝑣2 ) and the parameter, γ, lies between zero and one. If the null hypothesis (γ = 0) is not rejected,
this indicates that 𝜎𝑢2 equals zero and then the term 𝑢𝑖 should be removed from the model. The stochastic
production frontier then becomes the production function in which parameters can be consistently estimated using
ordinary least squares (Coelli, 1996). In this study, the null hypothesis (γ = 0) was rejected, indicating that a
production frontier was appropriate.
30
A value of gamma (γ) indicates the deviation from the frontier. The value of γ lies between zero and one. If
γ=1 indicates that all the deviations from the frontier are attributable to the technical inefficiency while if γ=0
means all deviations from the frontier are due to noise. If 0< γ <1 implies the deviation from the frontier is both
due to noise and technical inefficiency (Coelli et al., 2005).

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 155
result suggests that, on average, the surveyed fishers were only catching 79% of what was
possible (in value terms) given the level of inputs and technology used. Thus, there exists a
potential for output increasing without additional fishing inputs in the short term. In the long
term, however, this potential increase may not exist because of negative impact of additional
fishing effort on already overfished open-access nearshore fisheries in Vietnam.

The technical efficiency scores in Ben Tre province range widely from 0.36 to 0.93, with a
mean score of 0.7, whereas the technical efficiency of trawlers in Quang Ninh varies from 0.58
to 0.98, with a mean of 0.87 (Table 7.122). Notably, the trawl fleet in Quang Ninh province is
more homogenous in terms of efficiency level than the trawl fleet in Ben Tre province.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 156
Table 7.11. Initial and final model parameter estimates
Step 1 Step 2: Step 3
Initial Std. Min. Coef. Std. Derived

Pr(>|z|)

Pr(>|z|)
est. Error distance Error Final est
coef.

Constant 0.3908 0.0591 *** 0.3845 0.0002 0.0512 0.3850

lcfitted 1.0010 0.0594 ***
Ln(Engine) 0.4206 0.0562 *** 0.4177 0.4182
Ln(Day) 0.5967 0.1161 *** 0.6146 0.6153
Ln(Crew size) 0.4411 0.0729 *** 0.4547 0.4551
ln(Engine)2 0.2388 0.1784 0.1958 0.1960
Ln(Engine)*Ln(Day) 0.1920 0.2328 0.1854 0.1856
Ln(Engine)*Ln(Crew size) -0.1763 0.2501 -0.1283 -0.1285
2
Ln(Day) -1.9450 0.7260 ** -1.5308 -1.5324
Ln(Day)*Ln(Crew size) 0.3430 0.3526 0.2064 0.2066
2
Ln(Crew size) -0.3936 0.6037 -0.3114 -0.3117
Province -0.1918 0.0706 ** -0.2033 0.0622 **
Inefficiency model
Constant 0.4301 0.1188 *** 0.4270 0.1095 ***
Ln(Vessel age) 0.1380 0.0520 ** 0.1339 0.0445 **
Ln(Family crew) 0.1574 0.0759 * 0.1547 0.0771 *
Ln(Experience) -0.1099 0.0506 * -0.1068 0.0568 .
Ln(Age) 0.3977 0.1512 ** 0.4011 0.1506 **
Ln(Distance) -0.1863 0.1078 . -0.1866 0.0820 *
Engine origin -0.1491 0.0781 . -0.1463 0.0675 *
Province -0.3311 0.1246 ** -0.3335 0.1227 **
Sigma square 0.0407 0.0087 *** 0.0397 0.0086 ***
Gama 0.7992 0.1076 *** 0.7730 0.1012 ***
Log Likelihood 72.81 72.52
Monotonicity (%)
Ln(Engine) 100 100
Ln(Day) 93.9 100
Ln(Crew size) 99.4 100
Quasi-concavity (%) 96.4 100
Mean efficiency 0.79 0.79

Note: ***, **,* and ∙ the coefficients are statistically significant at 0.1%, 1%, 5%, and 10%,
respectively.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 157
Table 7.12. Summary of technical efficiency
Efficiency Min. 1st Median Mean 3rd Max.
All (n=165) 0.356 0.670 0.837 0.792 0.915 0.976
Quang Ninh (n=89) 0.577 0.815 0.908 0.866 0.940 0.976
Ben Tre (n=76) 0.356 0.596 0.678 0.704 0.851 0.933

The surveyed sample was divided into three groups of vessel based on engine power size and
survey area: coastal vessels in Quang Ninh and Ben Tre (i.e., with engine power <90 CV) and
offshore vessels in Quang Ninh (engine power ≥90 CV) (Table 7.13 and Figure 7.8). From
Table 7.13, coastal trawl vessels in Quang Ninh were more efficient than those of Ben Tre and
this difference is statically significant (p<0.01). Within Quang Ninh, the mean technical
efficiency of offshore vessel groups was also statistically significantly higher than that of
coastal vessel groups, averaging 0.914 and 0.857, respectively. From the results, the wider
distribution and lower technical efficiency of coastal vessel groups in Ben Tre and Quang Ninh
indicate more inefficient vessels existed in this vessel group while the tighter distribution and
higher technical efficiency of offshore trawl vessels means that most fishing boats are more
homogenous in terms of technical efficiency.

Table 7.13. Comparison of mean technical efficiency between different vessel groups
Quang
Quang Ninh Ben Tre
Ninh
t-value t-value
Engine<90 Engine ≥ 90 Engine <90 Engine <90
(n=75) (n=14) (n=75) (n=76)
Mean TE 0.857 0.914 -2.111** 0.857 0.704 7.698***
Min 0.577 0.622 0.577 0.356
Max 0.968 0.976 0.968 0.933
S.td 0.094 0.092 0.094 0.145
Note: ** and *** indicate p<0.05 and p<0.01, respectively.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 158
 Figure 7.8. Distribution of technical efficiency in each vessel group
Note: BT and QN refers to Ben Tre and Quang Ninh province, respectively.

7. 5.4.2. Determinants of technical efficiency

The factors affecting technical efficiency can be analysed by the algebraic sign, the magnitude,
and significance of the estimated coefficients in the inefficiency model (Jeon et al., 2006;
Squires et al., 2003). Results from the initial and final steps provide consistent outcomes
regarding the sign and magnitude of the estimated coefficients, as shown in the Table 7.11.
Note that in the inefficiency model, the dependent variable is the technical inefficiency value,
so a positive sign on the parameters in the inefficiency model shows a negative effect on
technical efficiency, and vice versa.

As expected, all independent variables in the inefficiency model were statistically significant
at the 10% level or less. The vessel’s age was negatively statistically significant at the 5% level
in both the initial estimation and the final estimation. The result indicates that newer vessels
generally have a higher level of efficiency. The vintage of the boat may result in differences in
efficiency due to newer technologies and improved design being incorporated into the more
recent vessel (Pascoe, Hassaszahed, et al., 2003). Further, newer boats would function well at
sea and the likelihood of breakdowns is generally lower. Other studies have also confirmed the
link between technical efficiency and vessel’s age, with newer vessel generally being more
efficient than older vessels (e.g., Fousekis & Klonaris, 2003; Greenville et al., 2006; Pascoe et
al., 2001; Pascoe & Coglan, 2002; Pascoe et al., 2012).

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 159
Characteristics of fishers including age and experience are important factors in how a vessel
performs. The results suggest that the age of fishers was significantly negatively related to
technical efficiency at the 1% level of significance, whereas vessel ownership experience of
fishers had a positively significant impact on technical efficiency. The results imply that
younger fishers tend to be more efficient than older fishers with an equivalent level of
experience. In addition, fishers with more experience in vessel ownership (i.e., being vessel
owner) were found to be more efficient (for a given age). Fishers with more vessel ownership
experience presumably have more fishing experience, and as a result, have better knowledge
of fish location, weather patterns, current and tides, and bottom condition (Squires et al., 2003).
Further, vessel owners with more ownership experience are generally considered to have more
experience in arranging suitable crew size, decision and managerial skills and managing fishing
activities properly compared to new vessel owners.

The finding also suggests that technical efficiency decreased when more family members
(family crew variable) were involved in the crew. One possible interpretation is that the
majority of fishing boats are small-scale and family-owned fishing boats. The vessel owners’
family members (i.e., spouse and/or children) are generally unpaid and potentially
underemployed on the boat, contributing proportionally less to production than a similar
number of employed crew, who would only be employed if their contribution to production
exceeded their payment (i.e., their marginal value product was greater than their wage rate or
share). Most vessels used a combination of family and employed crew members, but those that
used more family members were also those which were less efficient.

Distance from fishing harbour to fishing ground was also a significant factor affecting technical
efficiency at the 5% level of significance. As expected, the negative sign of this variable implies
that the further vessels can venture from home port, the more fish they would catch. Marine
fish resources in offshore water areas are assumed to be still abundant and under-exploited
(Duy et al., 2015; FAO, 2005b), while fish stocks in near-shore areas are already overexploited
(Nguyen Long, 2003; Pham, De Laender, et al., 2014). Therefore, fishing in near shore areas
would result in lower efficiency. This finding supports the policies that have been applied to
encourage the development of offshore fleets by the Vietnamese government over the past
years (The Prime Minister, 2010). The aims of offshore fisheries development policies are: (i)
to lessen fishing pressure on near-shore areas by reducing the number of small-scale boats
operating in coastal areas, (ii) and encourage fishers to upgrade and/or build larger vessels to
fish in offshore areas. However, an offshore vessel requires high capital investment and high

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 160
operational expenses while most small-scale fishers are considered to be poor (Pomeroy et al.,
2009), and unable to afford offshore vessels without appropriate supporting policies for these
groups.

Engine type was also significant at the 5% level. Specifically, fishing vessels with a Japanese
engine were more efficient than those with engines from elsewhere. From discussion with
skippers, engines made in Japan were believed to be more stable, durable and more fuel-
efficient than the others.

Finally, the negative sign relating to the province suggests that trawlers in Quang Ninh province
were more technically efficient than their counterparts in Ben Tre province. The northern
region, as previously mentioned, is a developed industrial province, and offers a wider range
of economic activities than the southern region. As a result, the opportunity for alternative
employment is substantially greater for both crew and skippers, creating a higher opportunity
cost of fishing. In contrast, alternative activities are less available in the southern region, so
less efficient fishers do not have the option to undertake alternative employment opportunities.
As a result, they persist in the fishery, resulting in a lower level of average technical efficiency,
all other things being equal.

7.5.5. Discussion and conclusions

The principal aim of this study was to estimate technical efficiency and identify its determinant
for trawl fishing vessels in Vietnam. Using individual vessel data collected through face-to-
face surveys from trawlers in Vietnam, the efficiency and its drivers of the surveyed fishers
were analysed by applying a translog stochastic production frontier.

The results of the study indicate that there exists substantial variation in efficiency among the
trawlers in the sample data, and there is a considerable room to increase technical efficiency of
trawlers in the short run given the current level of inputs and technology used. However, as
marine capture fisheries in Vietnam are open-access, increasing technical efficiency without
also reducing fleet size could lead to the depletion of resource stock in the long term (Jeon et
al., 2006; Susilowati et al., 2005). Thus, while measures and management strategies to increase
technical efficiency are important for fishers, these should be incorporated with other actions
to address the problem of open-access regime that would otherwise lead to the problem of “too
many boats chasing too few fish”.

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 161
The analysis suggests that vessel’s age, vessel ownership experience, the number of family
members involved in crew size, fisher’s age, and engine type has a statistically significant
impact on technical efficiency. A key finding, however, was that fishers in Quang Ninh
province were more technically efficient than those in Ben Tre province, illustrating the impact
of opportunity cost of labour on efficiency of the fleet. This is an indirect effect, as the higher
opportunity cost of labour in Quang Ninh provides an opportunity for the less efficient
operators to find alternative livelihoods, increasing the average efficiency of the remaining
fleet.

These results have direct implications for capacity management. Marine capture fisheries in
Vietnam are predominantly small-scale and open-access, and the sector has played an
important role in Vietnam’s economy and socio-economic development. To a large extent,
small scale fishing has been seen as an employer of last resort in Vietnam (Van Phuong & Duc
Phu, 2013). Many management controls to limit fishing have only been considered reluctantly
to ensure ongoing livelihoods are available, hence the persistence of an open-access policy.
However, these fishing activities are heavily depended on coastal inshore waters, causing
heavy pressure on depleted inshore areas. Reducing the fishing pressure on overexploited
coastal areas by small-scale trawl vessels in particular has been a key priority for management.

Previously, buyback programs had been implemented in some regions in Vietnam to reduce
capacity (Pomeroy et al., 2009). However, the effectiveness of these programs was limited
because these schemes ignored the non-fishing livelihood options, socio-economic status of
fishers, limited investment capacity of the small-scale fishers, and the limited ability to adapt
to new livelihoods. For example, in 2008, a number of fishing boats operating in the Tam Giang
lagoon in Central Vietnam were purchased by the government, who also provided agricultural
land for displaced fishers to undertake new livelihood activities. However, in less than a year,
these fishers sold their land and went back to fishing for their livelihood because they lacked
the necessary skills to adapt to these new livelihood options even when they were provided
with land for agricultural production (Pomeroy et al., 2009; Quynh et al., 2018).

The apparent failure of the buyback program largely reflects the limited alternative livelihoods
available to fishers in many regions of Vietnam. The results of this study suggest that
improving sustainability of the fishery – both ecological and economic – may best be achieved
through regional economic development (Pomeroy et al., 2009). Providing viable livelihood
alternatives will encourage exit from the fishery of the less efficient producers, as well as
Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 162
providing alternative options for the family members of the more efficient producers, resulting
in lower levels of fishing effort and a higher average efficiency of the remainder of the fleet.
This is especially important in rural and poor regions like Ben Tre where such alternatives are
limited.

To some extent, this has been recognised in more recent management which is aiming at
providing alternatives to coastal fishing through promoting offshore fishing, with fisheries
resources in offshore waters being considered under-exploited (Duy et al., 2015; FAO, 2005b).
These include subsidies for fishing vessel construction, fuel, and insurance (for both crew and
vessel) for larger vessels (The Prime Minister, 2014). Policies to promote and develop offshore
fishing fleets in Vietnam should, over time, reduce fishing pressure on near-shore fisheries, as
well as improve the average efficiency of the vessels fishing inshore. However, a precautionary
approach for the development of offshore fisheries needs to be applied to avoid the same
mistakes that have happened to coastal fisheries (FAO, 2005b).

Under open-access regime, overcapacity and over-fishing problems are inevitable, as seen in
the near-shore fisheries in which the small-scale fisheries in Vietnam operate. As raised
elsewhere in the literature, managing overcapacity in small-scale fisheries in developing
countries is complex due to a number of factors that may include its characteristics, regulation
enforcements, high reliance of fishers and fishing communities on fisheries resources,
conflicting policies, data limitation, and others (Pomeroy et al., 2009; Salayo et al., 2008). The
relatively low level of technical efficiency in Ben Tre province reflects the interactions between
fishery performance and local economic opportunities in particular. Thus, an integrated-
coordinated and holistic management approach to reducing fishing capacity needs to also
consider livelihood-based fisheries management and regional economic development along
with ecosystem-based management and community-based management in order to achieve a
sustainable fisheries sector (Pham, Gert, et al., 2014; Pomeroy et al., 2009).

Chapter 7: Social focused case study fishery: An open-access trawl fishery in Vietnam 163
 PART 3:
DISCUSSION AND CONCLUSION

Part 3: Discussion and conclusion 164

Chapter 8. Discussion and conclusion

8.1. Introduction

Understanding the link between the quantity of inputs used in fishing and amount of fish caught
is important information for effective fisheries management (Pascoe et al., 2001). Information
on efficiency level and its drivers are helpful to provide directions for designing proper
fisheries management strategies, regulations and policies. This information has important
implications for both fishers and fisheries policy makers (Sharma & Leung, 1998; Tingley et
al., 2005). Over the past 20+ years, research on evaluating efficiency in fisheries and studies
investigating the impact of fisheries management changes on efficiency have received
increasing attention worldwide. Nevertheless, knowledge on the potential impact of fisheries
management objective on technical efficiency has not been previously examined empirically.

It is widely acknowledged that fisheries management is characterised by multiple objectives.

These objectives generally include biological, economic, social and political objectives. The
relative importance of each of these objectives differs from fishery to fishery and even from
country to country.

To achieve different objectives fisheries managers afford different levels of quasi-property or

use rights to the fishers. Fisheries managed with a strong economic objective use different
levels of quasi-property or use rights to the fishers to deliver strong economic incentives that
result in fleet structure defined by lower heterogeneity in fishing vessels technical efficiency
and a high average level of technical efficiency. For example, fishers who are operating in
fisheries managed under ITQ systems, which can be seen as a proxy of economic objective in
fisheries management, have economic incentives to exit the fisheries when they are less
efficient operators. The adoption of management systems such as ITQs also provides incentives
for fishers who are more efficient to buy out less efficient fishers.

In contrast, it would be expected that fisheries managed to achieve a social objective, through
open access, have a fleet structure defined by greater heterogeneity and a lower overall level
of technical efficiency. Thus, the social objective is achieved through a greater level of overall
employment, ensuring subsistence income levels etc.

Chapter 8: Discussion and conclusion 165

The primary objective of this dissertation is to understand the potential impact of management
objectives on technical efficiency in fisheries. Understanding this relationship is important, as
fisheries managed with strong social objectives, such as maintaining livelihoods, may, in fact,
be trapping individuals in inefficient, low-income production systems. To our knowledge, there
is no study examining the impact of fisheries management objective on technical efficiency.
This thesis is the first one to address this gap, and seek to add to the existing literature on
fisheries management regarding fisheries management objective and efficiency. Better
understanding of the impact of fisheries management objective on technical efficiency might
provide implications for fisheries management in terms of the trade-off of technical efficiency
in achieving opposing objectives in fisheries management. This is helpful information for
policy implications for fisheries management regarding pursuing different fisheries
management objectives, especially social objectives (e.g., open-access fishery is a proxy) and
economic objectives (e.g., MEY reference point).

8.2. Hypothesis, objective and research aims

Four studies were undertaken in this thesis with the primary objective to provide an
understanding of the potential impact of fisheries management objective on technical
efficiency. Each study addresses specific research questions.

A meta-analysis in fisheries efficiency was conducted to identify factors, especially fisheries

management tools (including ITQ, open-access regime and input controls), and study specific
characteristics that might affect the level of technical efficiency in fisheries (Chapter 4). The
sensitivity of technical efficiency to the choice of input measures and model specifications in
fisheries context was presented in Chapter 5. The effect of an economic and social objective
on technical efficiency of the fleet was examined in Chapter 6 and Chapter 7, respectively. In
Chapter 6, a fishery managed to achieve MEY objective was used as an economic focused
fishery. The primary objective of this study is to evaluate the extent of efficiency and
productivity change over the banana prawn fishing season, and the implication of these changes
on the setting of the MEY target catch rate. In Chapter 7 technical efficiency and its
determinants were derived for an open-access fishery, where fisheries managers had an explicit
social focus objective. The final chapter (Chapter 8) presents a summary of research work
undertaken in this thesis, focusing on the main findings from the four studies; key contributions

Chapter 8: Discussion and conclusion 166

of the thesis are presented followed by a discussion about the limitation of the thesis and
possible directions for future research.

In the following sections, a summary of the key results of the four studies are presented.
Limitations and possible directions for further research are discussed, with the conclusions
presented in the final section.

8.3. Key research results

As indicated earlier, this thesis consists of four studies (corresponding to Chapter 4 to Chapter
7). The four studies address the primary objective of the thesis. Each of these studies provides
useful information for fisheries management, in particular for the Northern Prawn Fishery in
Australia and open-access trawl fishery in Vietnam, as well as related issues around efficiency
estimation in fisheries. The two empirical studies (i) an open-access trawl fishery in Vietnam
and (ii) a MEY banana prawn fishery in Australia, provide valuable information for future
management and policy decision making of these fisheries. The primary findings of each study
are presented as follows.

8.3.1. Meta-analysis fisheries efficiency

The first study undertaken in this thesis was a meta-analysis of fisheries efficiency (Chapter 4).
To our knowledge, this study is the first meta-analysis conducted in the field of technical
efficiency in fisheries. A primary aim of this study is to provide an understanding of the impact
of fisheries management objective on technical efficiency from the existing literature.

Results from the meta-regression model indicate that the estimated coefficient for ITQs
variable was positive and significantly different from other management tools including input
controls, and open access, as expected. By contrast, despite being non-significant, the negative
sign of the parameter on the open-access dummy variable may imply that fisheries managed
under open-access regime yield lower mean TE than fisheries managed under input controls,
as would be expected given that input controls afford greater (but imperfect) rights than open
access. Given the relationship between management objectives and management instruments,
these results provide prima facie empirical support for the broad hypothesis that a greater focus
on economic objectives results in the overall fleet becoming more technical efficiency. The
causal relationship between these is not necessarily direct, as the objectives define how the

Chapter 8: Discussion and conclusion 167

management system is implemented which in turn affects the level of efficiency, but the end
result is the same.

The results from this study also show that study specific characteristics including efficiency
estimation methods (SFA, DEA and SDF), the number of inputs and outputs used in efficiency
model, data year used, and journal ranking were factors that affect technical efficiency results.
Primary studies that used SDF yield lower technical efficiency scores than those papers that
used SFA and DEA approaches. However, the difference between efficiency scores obtained
from models used SFA and DEA is not significant. The meta-regression model also implies
that efficiency models with higher number of inputs and outputs get higher efficiency levels
than models used fewer numbers of inputs and outputs. Finally, primary studies published on
higher-ranking journals tend to have lower efficiency scores than papers published on lower
ranking journals. This last result is most likely an artefact of the quality of the analysis rather
than implying that the journal affects efficiency.

The results of the meta-analysis suggest that the estimated mean efficiency scores and their
distributions are not independent of the approach used and other study characteristics, so simple
comparisons between studies are not appropriate. As efficiency scores are relative measures;
there will always be at least some vessels that are more efficient than others within a fishery
and it is these vessels that define the frontier. In this regard, the distribution of technical
efficiency scores would provide the most information about the effectiveness of management.
A long tail suggests the persistence of inefficient vessels, most likely reflecting the strength of
any quasi-property or use rights in the fishery. Such a tail also reduces the mean efficiency
score, so this is an indicator also of the effectiveness of fisheries management in achieving
potential economic and social objectives.

8.3.2. Sensitivity of the technical efficiency to the choice of input variables

Existing literature show that misspecification of the production function, through the use of
inappropriate input measures, may result in a bias in the measures of efficiency (Pascoe,
Hassaszahed, et al., 2003). Consequently, management decisions based on biased measures of
efficiency could potentially lead to mismanagement and unsustainable fisheries. The aim of
this study is to examine the sensitivity of technical efficiency indexes to the choice of input
measures used in efficiency model and to model specifications (see Chapter 5). Results from
this study would provide valuable understanding regarding the choice of input measures as well

Chapter 8: Discussion and conclusion 168

as model specifications in fisheries efficiency research. Examination of the sensitivity of
technical efficiency to the choice of input measures addresses a gap in the literature.

For the purpose of the study, a data set from the Northern Prawn Fishery for the 2010-2015
period was used for efficiency analysis. Two measures of fixed inputs (i.e., either engine size
or hull length) and three measures of variable inputs (either hour trawled or hour fished or day
fished), and fish stock index were included into stochastic production frontier models.
Efficiency values were estimated for different combinations of input variables in the production
function, as well as different model specifications.

The results from this study showed that the choice of input variables have an impact on the
level of technical efficiency. More specifically, models that used vessel length produced lower
technical efficiency scores than models that incorporated engine power. Furthermore,
regarding fishing effort measures (fishing time in this study) the research findings indicate that
when fishing time measure get less precise (i.e., ranging from hour trawled to hour fished to
day fished), technical efficiency levels are lower.

It is also noteworthy that model specifications such as the functional form, inefficiency
distribution and time-variant efficiency assumptions also have an impact on mean technical
efficiency but the magnitude of these impacts were found to be small. In particular, Cobb-
Douglas models yielded lower technical efficiencies than those of translog models; and time-
invariant efficiency models produced lower technical efficiency scores compared with time-
variant models. Finally, models in which technical efficiency is assumed to have a half-normal
distribution exhibit lower efficiency indexes than stochastic frontier models where the technical
efficiency follows a truncated normal distribution.

8.3.3. Economic objective focused fishery case study: Northern Prawn Fishery

This study focuses on banana prawn fishery, a sub-fishery of the Northern Prawn Fishery in
Australia, which has had an objective of achieving maximum economic yield (MEY) since
2007 (Dichmont et al., 2010) (see Chapter 6). The banana prawn component is managed
through a catch rate trigger and seasonal closures to achieve MEY objective. The MEY-based
trigger catch rates for the banana prawn season is estimated largely using average catch, price
and fishing cost information. This effectively assumes that the average cost is equal to the
marginal cost, which in turn implies that structure and fishing behaviour of the fleet does not
change over the season. In other words, the estimation of MEY-based trigger is based on the

Chapter 8: Discussion and conclusion 169

assumption of relative homogeneity of the fleet in terms of efficiency, productivity and cost
structure. Thus, insufficient understanding about how technical efficiency and productivity
changes over the seasons means it is unclear on how well the trigger catch rates are achieving
MEY in the fishery.

A primary objective of study was, thus, to assess the extent of efficiency and productivity
change over the banana prawn fishing season, and the implications of these changes, if any, on
the setting of the MEY trigger/target catch rate. To this end, we estimated the technical
efficiency of the boats operating in the banana prawn season, and examined how average
efficiency changes over the season.

In this study, we found that, even strongly focusing on economic objective in fisheries
management, there is a wide distribution of technical efficiency levels but the model efficiency
scores are high. The results also show that technical efficiency varied between vessels while
average efficiency was relatively constant over the fishing season. However, while average
efficiency did not change substantially in most years, productivity of the remaining vessels at
the end of the season was generally greater than the fleet in the earlier weeks as smaller vessels
(in terms of engine power) exited the fishery earlier. The remaining vessels at the end of the
fishing season were generally larger in terms of engine power, and trawled more hours per day.
Higher engine power and hours trawled would likely increase the average cost of fishing
towards the end of the season, with implications for setting the target catch rate.

The results of this study also indicate that, given the changes in fleet productivity over the
season, and the observed changes in catch rates as a result, equating the MEY target and trigger
catch rates is a reasonable approach for pragmatic fisheries management. However, the
likelihood of higher fishing costs at the end of the season suggest that the current MEY trigger
estimates are too low. As the estimated target levels in recent years have been below the lower
precautionary buffer, they have not affected the actual implemented target and trigger. The
study also implies that while the MEY trigger catch rate provides a proxy for implementing
MEY in the fishery, it may not fully achieve MEY. Other factors may also affect MEY in the
fishery, particularly the interaction with the tiger prawn fishery (see e.g. Pascoe, Sharp, et al.,
2015). Under this circumstance, given that vessels operate across both fisheries, MEY in each
fishery should be assessed accordingly taking into consideration conditions in the other.

Chapter 8: Discussion and conclusion 170

8.3.4. Social objective focused fishery study: An open-access trawl fishery in Vietnam

This study focuses on an open-access trawl fishery in Vietnam which can serve as a good means
of achieving social objectives (see Chapter 7). This study aims at two main objectives. The first
objective (Activity 4) was to develop a socio-economic status of trawlers in Vietnam based on
primary data collected from fishers in the industry. The second objective (Activity 5) was to
estimate the level of technical efficiency and its determinants of the trawlers. For these
purposes, a field fisheries survey was conducted in two provinces (Quang Ninh and Ben Tre)
where trawl fishery is common in these areas.

The major results from Activity 4 provide a detailed and broader picture of socio-economic
status of trawlers in the surveyed areas. These results may provide useful information for
fisheries managers to regulate and design fisheries management strategies, fishing effort
reduction programs (e.g., reducing the number of small-scale boats operating in overexploited
nearshore areas) in particular. Results from the survey indicate that the level of education of
fishers is relatively low and most fishers rely solely on fishing activities for their livelihoods.
Furthermore, a majority of fishers are heavily dependent on middlemen to sell their catch,
especially in Ben Tre province where all fishers sell their fish through middlemen. It is also
notable that fishing is traditional (i.e., more than two generations involved in fishing) for a
large number of the surveyed fishing households. The results found in this case fishery study
show that the livelihood options of fishers are limited while in order to reduce fishing capacity
by removing the number of fishing vessels, the difference in regional socio-economic
conditions and creating non-fishing livelihoods options for fishers needs to be taken into
account.

The results from the study (Activity 5) suggest that there is potential room to increase technical
efficiency in the short run given inputs and technology used. However, in the long term, given
the open-access regime, increasing technical efficiency of the fishery may be impossible
without effective management changes. Further, increasing efficiency in the open-access
fishery may be counterproductive if it results in lower stock levels and subsequently lower
incomes.

A number of factors such as the age of fishing vessel, fisher’s age, experience, the number of
family members involved in crew size, engine type, distance from fishing ground, and regional
conditions are shown to be important determinants of variation in technical efficiency. The

Chapter 8: Discussion and conclusion 171

differences in technical efficiency between vessel groups in different regions can be partly
explained by differences in environmental conditions, socio-economics, demographic
characteristics and regional economic conditions. Specifically, the efficiency score of trawlers
in Quang Ninh, where regional socio-economics and geo-economics development is
favourable, is higher than that of Ben Tre where unfavourable socio-economics and natural
conditions exist. This might imply the difference of opportunity cost of fishing in two different
areas. This finding implies that reducing the number of small-scale fishing vessels operating in
nearshore areas by removing inefficient fishing boats needs also to take into account the
differences and constraints of regional economic conditions in each region. This is because
removal of inefficient vessels assumes that these fishers have an alternative livelihood to which
they can move. The finding in this study also supports the policies and strategies that are in
place to encourage the development of offshore fisheries in Vietnam. As shown in Chapter 7,
offshore vessels get higher technical efficiency than that of coastal-vessel counterparts as the
further fishing boats can venture the higher their technical efficiency.

The combined results found from study 1 (Chapter 4), study 3 (Chapter 6) and study 4 (Chapter
7) undertaken in this dissertation generally support the hypothesis that:

Fisheries management that has an economic focus results in a more

technically efficient fleet than management that has a less economic focus.

Specifically, findings in study 1 show that technical efficiency of ITQ fisheries (a management
instrument used largely to achieve economic objectives) would result in higher efficiency than
that of open-access (often justified through social objectives such as maintaining livelihoods
or maintaining regional fishing communities) and input-controlled (a proxy measure of halfway
between social focused and economic focused fisheries). However, other factors such as
opportunity cost of fishing, efficiency estimation approaches used efficiency model
specifications, input measures, and others that might have an impact on the technical efficiency
score.

8.4. Limitations and possible directions for further research

Efficiency scores are relative measures, and hence a comparison of these across fisheries is
confounded by a range of other factors. Thus, the distribution of technical efficiency scores
would provide the most information about the effectiveness of management. A long tail

Chapter 8: Discussion and conclusion 172

suggests the persistence of inefficient vessels, most likely reflecting the strength of any quasi-
property or use rights in the fishery. Relatively few studies included in the meta-analysis
reported distributions of the technical efficiency scores. Focusing on the distribution of
technical efficiency rather than the mean technical efficiency to examine the effectiveness of
fishing management in achieving management objectives may provide further insights,
although this would require practitioners to report these statistics in their papers.

Regarding research on technical efficiency of banana prawn fishery (Chapter 6), due to limited
data, factors that may affect technical efficiency were not included into the efficiency model.
This could be a potential scope for further research. Additionally, the results revealed that the
possible higher fishing costs at the end of the season suggest that the current MEY trigger
estimates are too low. Therefore, it is ideal to estimate a fishery specific cost function to refine
these costs and improve the estimate of the trigger and target.

The work done in Chapter 7 estimated the technical efficiency of trawl vessel owners in
Vietnam. A lack of data on input and output price did not allow us to evaluate cost and revenue
efficiency analyses. Knowledge on cost and revenue efficiencies would be valuable
information for sustainable fisheries management and future economic viability of this fishery.
These topics would be potential for future studies. Furthermore, more information about cost
and earning, as well as socio-economics of trawl fishery in Vietnam would be valuable to
provide comprehensive knowledge and helpful information for future management strategies
of the trawl fishery.

8.5. Conclusion

Fisheries management is largely characterised by multiple objectives that primarily include

biological, economic, social and political objectives. Different management tools and systems
applied to achieve different objectives would afford a different level of quasi-property or use
rights for fishers. This, in turn, affects the behaviour of fishers operating in the fishery, and
thereby the structure of fishing fleet in terms of economic performance. Although there is a
range of research that examines the impact of management changes on efficiency of fishers,
the impact of fisheries management objectives on technical efficiency has, until now, not been
empirically tested. The main purpose of this dissertation is therefore to enhance the
understanding of the impact of fisheries management objective on technical efficiency.

Chapter 8: Discussion and conclusion 173

This relationship is potentially important. Social objectives aimed at reducing poverty or
maintaining livelihoods in coastal communities may be counterproductive if they effectively
trap fishers into a low efficiency industry. The importance of opportunity cost in facilitating
fisheries adjustment, even under social objectives, suggests that solutions to low incomes in
fisheries may require non-fisheries related solutions.

The studies undertaken in this thesis generally support the hypothesis that there is a potential
relationship between fisheries management objective on the resultant technical efficiency of
the fleet. Specifically, fisheries management systems applied to achieve economic management
objectives have a positive impact on technical efficiency. In this thesis, we concluded that ITQ
fisheries result in higher technical efficiency than open-access fisheries and input-controlled
fisheries. However, other factors such as efficiency estimated methods used, model
specifications, input measures, and others are also found to have an impact on technical
efficiency levels. Therefore, comparing the level of technical efficiency estimates between
fisheries should be interpretable with cautions. A wide distribution of technical efficiency
suggests the persistence of inefficient vessels, and most likely reflects the strength of any quasi-
property or use rights in the fishery. A wide distribution of efficiency scores also reduces the
mean efficiency score, so this is perhaps a better indicator of the effectiveness of fisheries
management in achieving potential economic and social objectives. Such information is rarely
reported in empirical studies of technical efficiency, but perhaps should be in future.

The individual fishery studies also provide implications for management and policy decision
making in these fisheries. For the open-access fishery in Vietnam, a key recommendation is
that creating non-fishing livelihoods for inefficient fishers and promoting regional economic
development would be important measures to reduce fishing pressure on coastal areas and
increase technical efficiency of the fishing fleet as inefficient trawlers have options to leave the
fishery. Otherwise, their remaining in the fishery for living and livelihoods is understandable.
Promoting the development of offshore fisheries is also a potential short-term solution to
reduce fishing pressure on overexploited nearshore areas, but without adequate management
controls may result in these fisheries performing poorly in the future.

For the MEY banana prawn fishery in Australia, important recommendations included that
given the changes in fleet productivity over the season, and the observed changes in catch rates
as a result, equating the MEY target and trigger catch rates is a reasonable approach for
pragmatic fisheries management. Nevertheless, the possibility of higher fishing costs at the end
Chapter 8: Discussion and conclusion 174
of the season suggest that the current MEY trigger estimates are low. As the estimated target
levels in recent years have been below the lower precautionary buffer, they have not affected
the actual implemented target and trigger. Ideally, a fishery specific cost function should be
estimated to refine these costs and improve the estimate of the trigger and target.

Chapter 8: Discussion and conclusion 175

References

ABARES. (2016a). Australian fisheries and aquaculture statistics 2015. Australian Bureau of
Agricultural and Resource Economics and Sciences. Canberra, Australia
ABARES. (2016b). Australian Fisheries Economic Indicators Report 2015: Financial and
Economic Performance of the Northern Prawn Fishery. Australian Bureau of
Agricultural and Resource Economics and Sciences. Canberra, Australia
AFMA. (2009). Background information on the Northern Prawn Fishery. Australian Fisheries
Management Authority. Canberra, Australia
AFMA. (2014). Northern Prawn Fishery-Bycatch and discarding workplan. Australian
Fisheries Management Authority. Canberra, Australia
AFMA. (2016a). Annual report 2015-2016. Australian Fisheries Management Authority.
Canberra, Australia
AFMA. (2016b). Northern Prawn Fishery Directions and Closures. Australian Fisheries
Manegement Authority. Canberra, Australia.
AFMA. (2016c). Nothern Prawn Fishery Directions and Closures. Australian Fisheries
Manegement Authority. Canberra, Australia
AFMA. (2017). Northern Prawn Fishery Directions and Closures. Australian Fisheries
Management Authority. Canberra, Australia
Ahmad, M., & Bravo-Ureta, B. E. (1996). Technical efficiency measures for dairy farms using
panel data: A comparison of alternative model specifications. Journal of Productivity
Analysis, 7(4), 399-415.
Aiello, F., & Bonanno, G. (2016). Efficiency in banking: a meta-regression analysis.
International Review of Applied Economics, 30(1), 112-149.
Aiello, F., & Bonanno, G. (2018). On the sources of heterogeneity in banking efficiency
literature. Journal of economic surveys, 32(1), 194-225.
Aigner, D., Lovell, C. A. K., & Schmidt, P. (1977). Formulation and estimation of stochastic
frontier production function models. Journal of Econometrics, 6(1), 21-37.
Aldon, M. E. T., Fermin, A. C., & Agbayani, R. F. (2011). Socio-cultural context of fishers’
participation in coastal resources management in Anini-y, Antique in west central
Philippines. Fisheries Research, 107(1–3), 112-121.
Allison, E. H., & Ellis, F. (2001). The livelihoods approach and management of small-scale
fisheries. Marine Policy, 25(5), 377-388.
Andalecio, M. N. (2011). Including coastal resource users in fisheries management evaluation
of San Miguel Bay, Philippines. Ocean & Coastal Management, 54(10), 760-770.
doi:https://doi.org/10.1016/j.ocecoaman.2011.08.004
Andersen, J. L. (2002). Using different inputs and outputs to estimate technical efficiency in
fisheries: An application to Danish Seiners in the North Sea and Skagerrak. Working
paper. University of Copenhagen, Denmark
Andersen, J. L. (2005). The Inclusion of Stocks in Multi-species Fisheries: The Case of Danish
Seiners. Marine Resource Economics, 20(2), 163-184.

References 176
Anderson, C. M., Krigbaum, M. J., Arostegui, M. C., Feddern, M. L., Koehn, J. Z., Kuriyama,
P. T., . . . Sanders, J. (2019). How commercial fishing effort is managed. Fish and
Fisheries, 20(2), 268-285. doi:doi:10.1111/faf.12339
Andrew, N. L., Béné, C., Hall, S. J., Allison, E. H., Heck, S., & Ratner, B. D. (2007). Diagnosis
and management of small‐scale fisheries in developing countries. Fish and Fisheries,
8(3), 227-240.
Asche, F., Garlock, T. M., Anderson, J. L., Bush, S. R., Smith, M. D., Anderson, C. M., . . .
Vannuccini, S. (2018). Three pillars of sustainability in fisheries. Proceedings of the
National Academy of Sciences, 115(44), 11221-11225. doi:10.1073/pnas.1807677115
Barrio, M., & Loureiro, M. L. (2010). A meta-analysis of contingent valuation forest studies.
Ecological Economics, 69(5), 1023-1030.
Battese, G. E., & Broca, S. S. (1997). Functional Forms of Stochastic Frontier Production
Functions and Models for Technical Inefficiency Effects: A Comparative Study for
Wheat Farmers in Pakistan. Journal of Productivity Analysis, 8(4), 395-414.
Battese, G. E., & Coelli, T. J. (1988). Prediction of firm-level technical efficiencies with a
generalized frontier production function and panel data. Journal of Econometrics,
38(3), 387-399.
Battese, G. E., & Coelli, T. J. (1992). Frontier Production Functions, Technical Efficiency and
Panel Data: With Application to Paddy Farmers in India. Journal of Productivity
Analysis, 3(1/2), 153-169. doi:10.1007/BF00158774
Battese, G. E., & Coelli, T. J. (1995). A model for technical inefficiency effects in a stochastic
frontier production function for panel data. Empirical Economics, 20(2), 325-332.
Beddington, J. R., Agnew, D. J., & Clark, C. W. (2007). Current Problems in the Management
of Marine Fisheries. Science, 316(5832), 1713-1716.
Belhabib, D., Sumaila, U. R., & Pauly, D. (2015). Feeding the poor: Contribution of West
African fisheries to employment and food security. Ocean & Coastal Management,
111, 72-81. doi:https://doi.org/10.1016/j.ocecoaman.2015.04.010
Ben Tre Statistics Office. (2016). Ben Tre Statistical Yearbook 2015. Statistical publishing
house. Hanoi, Vietnam.
Béné, C. (2009). Are Fishers Poor or Vulnerable? Assessing Economic Vulnerability in Small-
Scale Fishing Communities. The Journal of Development Studies, 45(6), 911-933.
Béné, C., Arthur, R., Norbury, H., Allison, E. H., Beveridge, M., Bush, S., . . . Williams, M.
(2016). Contribution of Fisheries and Aquaculture to Food Security and Poverty
Reduction: Assessing the Current Evidence. World development, 79, 177-196.
doi:https://doi.org/10.1016/j.worlddev.2015.11.007
Béné, C., Macfadyen, G., & Allision, A. H. (2007). Increasing the contribution of small-scale
fisheries to poverty alleviation and food security. FAO Fisheries Technical Paper. No.
481. Rome. Italy: Food and Agriculture Organization of the United Nations.
Betcherman, G., & Marschke, M. (2016). Coastal livelihoods in transition: How are
Vietnamese households responding to changes in the fisheries and in the economy?
Journal of Rural Studies, 45, 24-33.
Bjørndal, T., & Munro, G. (2012). The economics and management of world fisheries: Oxford:
Oxford University Press.

References 177
Boström, M. (2012). A missing pillar? Challenges in theorizing and practicing social
sustainability: introduction to the special issue. Sustainability: Science, Practice and
Policy, 8(1), 3-14. doi:10.1080/15487733.2012.11908080
Bravo-Ureta, B. E., Solís, D., López, V. H. M., Maripani, J. F., Thiam, A., & Rivas, T. (2007).
Technical efficiency in farming: a meta-regression analysis. Journal of Productivity
Analysis, 27(1), 57-72.
Brewer, D., Heales, D., Milton, D., Dell, Q., Fry, G., Venables, B., & Jones, P. (2006). The
impact of turtle excluder devices and bycatch reduction devices on diverse tropical
marine communities in Australia's northern prawn trawl fishery. Fisheries Research,
81(2-3), 176-188.
Broemme, K., & Stolpe, H. (2007). Developing environmental concepts for Vietnamese coal
mines. Paper presented at the International Workshop Geoecology and Environmental
Technology, 25-27 October 2007, Hanoi, Vietnam.
Bromley, D. W. (2009). Abdicating Responsibility: The Deceits of Fisheries Policy. Fisheries,
34(6), 280-290.
Brons, M., Nijkamp, P., Pels, E., & Rietveld, P. (2005). Efficiency of urban public transit: A
meta analysis. Transportation, 32(1), 1-21.
Brons, M., Nijkamp, P., Pels, E., & Rietveld, P. (2008). A meta-analysis of the price elasticity
of gasoline demand. A SUR approach. Energy Economics, 30(5), 2105-2122.
Brooks, K. (2010). Sustainable development: Social outcomes of structural adjustments in a
South Australian fishery. Marine Policy, 34(3), 671-678.
Brooks, K., Schirmer, J., Pascoe, S., Triantafillos, L., Jebreen, E., Cannard, T., & Dichmont,
C. M. (2015). Selecting and assessing social objectives for Australian fisheries
management. Marine Policy, 53(0), 111-122.
Brown, C. J. (2016). Social, economic and environmental effects of closing commercial
fisheries to enhance recreational fishing. Marine Policy, 73, 204-209.
Buckworth, R. C., Ellis, N., Zhou, S., Pascoe, S., Deng, R., Hill, F., & O’Brien, M. (2013).
Comparison of TAC and current management for the White Banana Prawn fishery of
the Northern Prawn Fishery. Final Report to the Australian Fisheries Management
Authority. Brisbane, Australia: CSIRO Marine & Atmospheric Research.
Buckworth, R. C., Venables, W. N., Lawrence, E., Kompas, T., Pascoe, S., Chu, L., . . .
Rothlisberg, P. C. (2014). Incorporation of predictive models of banana prawn catch
for MEY-based harvest strategy development for the Northern Prawn Fishery. Final
Report to the Fisheries Research and Development Corporation, Project 2011/239.
Brisbane, Australia: CSIRO Marine & Atmospheric Research.
Carvalho, P., Marques, R. C., & Berg, S. (2012). A meta-regression analysis of benchmarking
studies on water utilities market structure. Utilities Policy, 21, 40-49.
Caves, D. W., Christensen, L. R., & Diewert, W. E. (1982). The economic theory of index
numbers and the measurement of input, output, and productivity. Econometrica:
Journal of the Econometric Society, 1393-1414.
CCAFS-SEA. (2016). Assessment Report: The drought and salinity intrusion in the Mekong
River Delta of Vietnam. Research Program on Climate Change, Agriculture and Food
Security- Southeast Asia (CCAFS-SEA). Hanoi, Vietnam: CGIAR Research Centers in
Southeast Asia.

References 178
Ceyhan, V., & Gene, H. (2014). Productive efficiency of commercial fishing: evidence from
the Samsun Province of Black Sea, Turkey. Turkish Journal of Fisheries and Aquatic
Sciences, 14, 309-320.
Chakraborty, K., Biswas, B., & Lewis, W. C. (2001). Measurement of Technical Efficiency in
Public Education: A Stochastic and Nonstochastic Production Function Approach.
Southern Economic Journal, 67(4), 889-905. doi:10.2307/1061576
Charles, A. (1989). Bio-Socio-Economic Fishery Models: Labour Dynamics and Multi-
Objective Management. Canadian Journal of Fisheries and Aquatic Sciences, 46(8),
1313-1322.
Charles, A. (1992). Fishery conflicts: a unified framework. Marine Policy, 16(5), 379-393.
Charnes, A., Cooper, W. W., & Rhodes, E. (1978). Measuring the efficiency of decision
making units. European Journal of Operational Research, 2(6), 429-444.
Chávez, E., Quiroga, M., & Dresdner, J. (2017). The Effect of Collective Rights-Based
Management on Technical Efficiency: The Case of Chile’s Common Sardine and
Anchovy Fishery. Marine Resource Economics, 33(1), 87-112. doi:10.1086/696130
Chen-Lei, Gupta, R., Mukherjee, Z., & Wanke, P. (2016). Technical efficiency of Connecticut
Long Island Sound lobster fishery: a nonparametric approach to aggregate frontier
analysis. Natural Hazards, 81(3), 1533-1548. doi:10.1007/s11069-015-2144-5
Cheung, W. W. L., & Sumaila, U. R. (2008). Trade-offs between conservation and socio-
economic objectives in managing a tropical marine ecosystem. Ecological Economics,
66(1), 193-210.
Christensen, V. (2010). MEY = MSY. Fish and Fisheries, 11(1), 105-110.
Chu, C. (2009). Thirty years later: the global growth of ITQs and their influence on stock status
in marine fisheries. Fish and Fisheries, 10(2), 217-230.
Clark, C. W. (1973). The Economics of Overexploitation. Science, 181(4100), 630-634.
Clark, C. W., Clarke, F. H., & Gordon, R. M. (1979). The Optimal Exploitation of Renewable
Resource Stocks: Problems of Irreversible Investment. Econometrica, 47(1), 25-47.
Clark, M. R., Althaus, F., Schlacher, T. A., Williams, A., Bowden, D. A., & Rowden, A. A.
(2015). The impacts of deep-sea fisheries on benthic communities: a review. ICES
Journal of Marine Science, 73(suppl_1), i51-i69.
Cochrane, K. L. (2000). Reconciling sustainability, economic efficiency and equity in fisheries:
the one that got away? Fish and Fisheries, 1(1), 3-21.
Coelli, T. J. (1996). A guide to FRONTIER version 4.1: a computer program for stochastic
frontier production and cost function estimation. CEPA Working papers 96/07. Centre
for Efficiency and Productivity Analysis. University of New England, Armidale, NSW
2351, Australia.
Coelli, T. J., Battese, G. E., Donnell, C. J., & Rao, D. S. P. (2005). An Introduction to Efficiency
and Productivity Analysis. New York, USA: Springer Science + Business Media.
Coelli, T. J., Rao, D. S. P., & Battese, G. E. (1998). An introduction to efficiency and
productivity analysis. Boston: Kluwer Academic Publishers.
Coglan, L., & Pascoe, S. (1999). Separating Resource Rents from Intra-marginal Rents in
Fisheries' Economic Survey Data. Agricultural and Resource Economics Review,
28(2), 219-228.
References 179
Coglan, L., & Pascoe, S. (2007). Implications of human capital enhancement in fisheries.
Aquatic Living Resources, 20(3), 231-239.
Colegrave, A. D., & Giles, M. J. (2008). School cost functions: A meta-regression analysis.
Economics of Education Review, 27(6), 688-696.
Cornwell, C., Schmidt, P., & Sickles, R. C. (1990). Production frontiers with cross-sectional
and time-series variation in efficiency levels. Journal of Econometrics, 46(1-2), 185-
200.
Costello, C., & Deacon, R. (2007). The Efficiency Gains from Fully Delineating Rights in an
ITQ Fishery. Marine Resource Economics, 22(4), 347-361.
Coulthard, S., Johnson, D., & McGregor, J. A. (2011). Poverty, sustainability and human
wellbeing: A social wellbeing approach to the global fisheries crisis. Global
Environmental Change, 21(2), 453-463.
doi:https://doi.org/10.1016/j.gloenvcha.2011.01.003
CSIRO. (2018). Northern Prawn Fishery: World class management. Commonwealth Scientific
and Industrial Research (CSIRO). Retrieved from
https://www.csiro.au/en/Research/OandA/Areas/Marine-resources-and-
industries/Sustaining-Australian-fisheries/Northern-Prawn-Fishery (31 May 2018).
DAFF. (2007). Commonwealth fisheries harvest strategy: policy and guidelines. Department
of Agriculture Fisheries and Forestry (DAFF). Canberra, Australia.
Department of Agriculture and Rural Development of Quang Ninh. (2016). A report on the
fishing year 2015 and plan for fishing year 2016. Department of Agriculture and Rural
Development of Quang Ninh. Quang Ninh, Vietnam.
Department of Capture Fisheries and Resources Protection. (2015). Vessel statistics for
trawling vessels in Vietnam in 2015. Department of Capture Fisheries and Resources
Protection, Directorate of Fisheries, Ministry of Agricultural and Rural Development.
Hanoi, Vietnam.
Department of Capture Fisheries and Resources Protection of Ben Tre. (2016). Summary report
of trawl fisheries in Ben Tre. Department of Rural and Agricultural Development of
Ben Tre. Ben Tre, Vietnam.
Dichmont, C. M., Deng, R., Punt, A. E., Venables, W. N., & Haddon, M. (2006). Management
strategies for short-lived species: The case of Australia's Northern Prawn Fishery. 1.
Accounting for multiple species, spatial structure and implementation uncertainty when
evaluating risk. Fisheries Research, 82(1-3), 204-220.
Dichmont, C. M., Jarrett, A., Hill, F., & Brown, M. (2014). Harvest Strategy for the Northern
Prawn Fishery under Input Controls. Australian Fisheries Management Authority and
CSIRO. Australia
Dichmont, C. M., Loneragan, N. R., Brewer, D. T., & Poiner, I. R. (2007). Partnerships
Towards Sustainable use of Australia's Northern Prawn Fishery. In T. R. McClanahan
& J. C. Castilla (Eds.), Fisheries Management: Progress Towards Sustainability,
Chapter 10 (pp. 205-230): Blackwell Publishing Ltd.
Dichmont, C. M., Pascoe, S., Kompas, T., Punt, A. E., & Deng, R. (2010). On implementing
maximum economic yield in commercial fisheries. Proceedings of the National
Academy of Sciences of the United States of America, 107(1), 16-21.

References 180
Dichmont, C. M., Punt, A. E., Deng, R., Dell, Q., & Venables, W. N. (2003). Application of a
weekly delay-difference model to commercial catch and effort data for tiger prawns in
Australia’s Northern Prawn Fishery. Fisheries Research, 65(1–3), 335-350.
Die, D. J., & Ellis, N. (1999). Aggregation dynamics in penaeid fisheries: banana prawns
(Penaeus merguiensis) in the Australian Northern Prawn Fishery. Marine and
freshwater research, 50(7), 667-675.
Diop, B., Sanz, N., Duplan, Y. J. J., Guene, E. H. M., Blanchard, F., Pereau, J.-C., & Doyen,
L. (2018). Maximum Economic Yield Fishery Management in the Face of Global
Warming. Ecological Economics, 154, 52-61.
doi:https://doi.org/10.1016/j.ecolecon.2018.07.027
Directorate of Fisheries. (2012). The master plan on fisheries development through 2020 with
a vision toward 2030. Directorate of Fisheries, Ministry of Agriculture and Rural
Development. Hanoi, Vietnam
Directorate of Fisheries. (2013). The National Master Plan on fishing capacity management.
Directorate of Fisheries, Ministry of Agriculture and Rural Development. Hanoi,
Vietnam
Directorate of Fisheries. (2018). Summary report on status of fishing, preserving, processing
and marketing of seafood product. Directorate of Fisheries, Ministry of Agriculture and
Rural Development. Hanoi, Vietnam
Djokoto, J. G., & Gidiglo, K. F. (2016). Technical Efficiency in Agribusiness: A Meta‐Analysis
on Ghana. Agribusiness, 32(3), 397-415.
Dresdner, J., Campos, N., & Chávez, C. (2010). The impact of individual quotas on technical
efficiency: does quality matter? Environment and Development Economics, 15(5), 585-
607.
Dupont, D. P. (2000). Individual transferable vessel quotas and efficient restructuring of the
primary harvesting sector. Annals of Operations Research, 94(1), 275-294.
Duy, N. N. (2016). The Economics of Open-Access Fisheries: Subsidies and Performance of
Vietnamese Fisheries. (PhD thesis), Norwegian College of Fishery Science - University
of Tromso (UiT) - The Arctic University of Norway, Norway.
Duy, N. N., & Flaaten, O. (2016). Efficiency analysis of fisheries using stock proxies. Fisheries
Research, 181, 102-113.
Duy, N. N., Flaaten, O., Anh, N. T. K., & Ngoc, Q. T. K. (2012). Open-access fishing rent and
efficiency -The case of gillnet vessels in Nha Trang, Vietnam. Fisheries Research, 127–
128(0), 98-108.
Duy, N. N., Flaaten, O., & Long, L. K. (2015). Government support and profitability effects –
Vietnamese offshore fisheries. Marine Policy, 61, 77-86.
Efendic, A., Pugh, G., & Adnett, N. (2011). Institutions and economic performance: A meta-
regression analysis. European Journal of Political Economy, 27(3), 586-599.
Eggert, H., & Tveterås, R. (2007). Potential rent and overcapacity in the Swedish Baltic Sea
trawl fishery for cod (Gadus morhua). ICES Journal of Marine Science, 64(3), 439-
445.
Esmaeili, A. (2006). Technical efficiency analysis for the Iranian fishery in the Persian Gulf.
ICES Journal of Marine Science, 63(9), 1759-1764.

References 181
Espey, M. (1998). Gasoline demand revisited: an international meta-analysis of elasticities.
Energy Economics, 20(3), 273-295.
Fall, F., Akim, A.-m., & Wassongma, H. (2018). DEA and SFA research on the efficiency of
microfinance institutions: A meta-analysis. World development, 107, 176-188.
FAO. (1995). Code of conduct for responsible fisheries. Rome, Italy: Food and Agriculture
Organization of the United Nations (FAO).
FAO. (2002). A fishery manager's guidebook. Management measures and their application.
FAO Fisheries Technical Paper No. 424. Rome, Italy: Food and Agriculture
Organization of the United Nations (FAO).
FAO. (2005a). Fishery and Aquaculture Country Profiles. Country Profile Fact Sheets. Rome,
Italy: FAO Fisheries and Aquaculture Department, Food and Agricultural Organization
of the United Nations.
FAO. (2005b). Report of the conference on the national strategy for marine fisheries
management and development in Viet Nam. Hanoi, Viet Nam, 26-27 April 2005.
FAO/FishCode Review. No. 16. 2005. Rome, Italy: Food and Agriculture Organization
of the United Nations (FAO).
FAO. (2012). The state of World Fisheires and Aquaculture 2012. Rome, Italy: Food and
Agriculture Organization of the United Nations.
FAO. (2014). The state of World Fisheries and Aquaculture 2014. Rome, Italy: Food and
Agriculture Organization of the United Nations.
FAO. (2016). The state of World Fisheries and Aquaculture 2016. Contributing to food security
and nutrition for all. Rome, Italy: Food and Agriculture Organization of the United
Nations.
Färe, R., & Grosskopf, S. (2000). Theory and Application of Directional Distance Functions.
Journal of Productivity Analysis, 13(2), 93-103. doi:10.1023/A:1007844628920
Färe, R., Grosskopf, S., & Walden, J. B. (2015). Productivity change and fleet restructuring
after transition to individual transferable quota management. Marine Policy, 62, 318-
325. doi:http://dx.doi.org/10.1016/j.marpol.2015.05.015
Färe, R., Kirkley, J., & Walden, J. B. (2006). Adjusting technical efficiency to reflect
discarding: The case of the U.S. Georges Bank multi-species otter trawl fishery.
Fisheries Research, 78(2–3), 257-265.
doi:http://dx.doi.org/10.1016/j.fishres.2005.12.014
Färe, R., Kirkley, J., & Walden, J. B. (2011). Measuring Fishing Capacity When Some Outputs
Are Undesirable. Eastern Economic Journal, 37(4), 553-570. doi:10.1057/eej.2010.7
Färe, R., & Lovell, C. A. K. (1978). Measuring the technical efficiency of production. Journal
of Economic Theory, 19(1), 150-162. doi:http://dx.doi.org/10.1016/0022-
0531(78)90060-1
Farrell, M. J. (1957). The Measurement of Productive Efficiency. Journal of the Royal
Statistical Society.Series A (General), 120(3), 253-290.
Fatima, H., & Yasmin, B. (2016). Efficiency and productivity analysis of Pakistan's farm
sector: A meta-analysis. Pakistan Journal of Agricultural Research, 29(3).

References 182
Felthoven, R. G., Hiatt, T., & Terry, J. M. (2002). Measuring fishing capacity and utilization
with commonly available data: an application to Alaska fisheries. Marine Fisheries
Review, 64(4), 29-39.
Fissel, B. E., Felthoven, R. G., Kasperski, S., & O'Donnell, C. (2015). Decomposing
productivity and efficiency changes in the Alaska head and gut factory trawl fleet.
Marine Policy, 62, 337-346.
Fousekis, P. (2002). Distance vs. ray functions: an application to the inshore fishery of Greece.
Marine Resource Economics, 17(4), 251-267.
Fousekis, P., & Klonaris, S. (2003). Technical efficiency determinants for fisheries: a study of
trammel netters in Greece. Fisheries Research, 63(1), 85-95.
Fox, K. J., Grafton, R. Q., Kompas, T., & Che, T. N. (2006). Capacity reduction, quota trading
and productivity: the case of a fishery. Australian Journal of Agricultural and Resource
Economics, 50(2), 189-206. doi:10.1111/j.1467-8489.2006.00331.x
Fuss, M., McFadden, D., & Mundlak, Y. (1978). A survey of functional forms in the economic
analysis of production. In M. Fuss & D. McFadden (Eds.), Production Economics: A
Dual Approach to Theory and Applications (Vol. 1, pp. 219-268): McMaster University
Archive for the History of Economic Thought.
Gallet, C. A., & Doucouliagos, H. (2014). The income elasticity of air travel: A meta-analysis.
Annals of Tourism Research, 49, 141-155.
Gallup, J. L. (1995). The Economic Value of Children in Vietnam. Paper presented at the A
Seminar on Aging, Development, and Population. Institutes of Economics and
Sociology, Hanoi, Vietnam.
Gänsbauer, A., Bechtold, U., & Wilfing, H. (2016). SoFISHticated policy – social perspectives
on the fish conflict in the Northeast Atlantic. Marine Policy, 66, 93-103.
García del Hoyo, J. J., Castilla Espino, D., & Jiménez Toribio, R. (2004). Determination of
technical efficiency of fisheries by stochastic frontier models: a case on the Gulf of
Cádiz (Spain). ICES Journal of Marine Science, 61(3), 416-421.
Gedara, K. M., Wilson, C., Pascoe, S., & Robinson, T. (2012). Factors Affecting Technical
Efficiency of Rice Farmers in Village Reservoir Irrigation Systems of Sri Lanka.
Journal of Agricultural Economics, 63(3), 627-638. doi:10.1111/j.1477-
9552.2012.00343.x
General Statistics Office of Vietnam. (2016a). The 2014 Viet Nam Population and Housing
Survey: Population sex-age structure and related socio-economic issuses in Viet Nam.
Thong Tan Publishing House. Hanoi, Vietnam.
General Statistics Office of Vietnam. (2016b). Result of the Vietnam household living
standards survey 2014. Statistical Publishing House. Hanoi, Vietnam.
General Statistics Office of Vietnam. (2018a). Percentage of employed workers at 15 years of
age and above among population by province. Retrieved from
https://www.gso.gov.vn/default_en.aspx?tabid=774 (20 May 2018).
General Statistics Office of Vietnam. (2018b). Percentage of literate population at 15 years of
age and above by province. Retrieved from
https://www.gso.gov.vn/default_en.aspx?tabid=774 (19 May 2018).

References 183
Géraldine, H., Henningsen, A., & Jensen, U. (2015). A Monte Carlo study on multiple output
stochastic frontiers: a comparison of two approaches. Journal of Productivity Analysis,
44(3), 309-320. doi:10.1007/s11123-014-0416-9
Ghosh, B., & Kar, T. K. (2013). Maximum sustainable yield and species extinction in a prey–
predator system: some new results. Journal of Biological Physics, 39(3), 453-467.
Giannakas, K., Tran, K. C., & Tzouvelekas, V. (2003). On the choice of functional form in
stochastic frontier modeling. Empirical Economics, 28(1), 75-100.
Gilman, E., Passfield, K., & Nakamura, K. (2014). Performance of regional fisheries
management organizations: ecosystem-based governance of bycatch and discards. Fish
and Fisheries, 15(2), 327-351. doi:doi:10.1111/faf.12021
Giron-Nava, A., Johnson, A. F., Cisneros-Montemayor, A. M., & Aburto-Oropeza, O. (2018).
Managing at Maximum Sustainable Yield does not ensure economic well-being for
artisanal fishers. Fish and Fisheries, 0(0), 1-10. doi:doi:10.1111/faf.12332
Glass, G. V. (1976). Primary, secondary, and meta-analysis of research. Educational
researcher, 5(10), 3-8.
Glass, G. V., McGaw, B., & Smith, M. L. (1981). Meta-analysis in social research. Beverly
Hills: Sage Publications.
Gong, B.-H., & Sickles, R. C. (1992). Finite sample evidence on the performance of stochastic
frontiers and data envelopment analysis using panel data. Journal of Econometrics,
51(1), 259-284. doi:https://doi.org/10.1016/0304-4076(92)90038-S
Gordon, H. S. (1953). An economic approach to the optimum utilization of fishery resources.
Journal of the Fisheries Board of Canada, 10. doi:10.1139/f53-026
Gordon, H. S. (1954). The Economic Theory of a Common-Property Resource: The Fishery.
Journal of Political Economy, 62(2), 124-142.
Grafton, R. Q. (1996). Individual transferable quotas: Theory and practice. Reviews in Fish
Biology and Fisheries, 6(1), 5-20.
Grafton, R. Q., & Kompas, T. (2014). Three Pillars of Fisheries Policy. Asia & the Pacific
Policy Studies, 1(3), 609-614.
Grafton, R. Q., Kompas, T., Chu, L., & Che, N. (2010). Maximum economic yield. Australian
Journal of Agricultural and Resource Economics, 54(3), 273-280.
Grafton, R. Q., Squires, D., & Fox, K. (2000). Private Property and Economic Efficiency: A
Study of a Common‐Pool Resource. Journal of Law and Economics, 43(2), 679-714.
Gréboval, D. (2002). Report and documentation of the International Workshop on Factors
Contributing to Unsustainability and Overexploitation in Fisheries. Bangkok,
Thailand, 4-8 February 2002. FAO Fisheries Report. No. 672. Rome, Italy: Food and
Agriculture Organization of the United Nations.
Greene, W. H. (1990). A Gamma-distributed stochastic frontier model. Journal of
Econometrics, 46(1), 141-163.
Greene, W. H. (2008a). Econometric analysis (6th ed.). Upper Saddle River, N.J:
Pearson/Prentice Hall.
Greene, W. H. (2008b). The econometric approach to efficiency analysis. In H. O. Fried, C. A.
K. Lovell, & S. S. Schmidt (Eds.), The measurement of productive efficiency and
productivity change. Oxford: Oxford University Press.
References 184
Greenville, J., Hartmann, J., & MacAulay, T. G. (2006). Technical Efficiency in Input-
Controlled Fisheries: The NSW Ocean Prawn Trawl Fishery. Marine Resource
Economics, 21(2), 159-179.
Griffin, R. C., Montgomery, J. M., & Rister, M. E. (1987). Selecting functional form in
production function analysis. Western Journal of Agricultural Economics, 216-227.
Groves, T., & Squires, D. (2007). Lessons from Fisheries Buybacks. In R. Curtis & D. Squires
(Eds.), Fisheries Buybacks. USA: Blackwell Publishing.
Guillen, J., Macher, C., Merzéréaud, M., Bertignac, M., Fifas, S., & Guyader, O. (2013).
Estimating MSY and MEY in multi-species and multi-fleet fisheries, consequences and
limits: an application to the Bay of Biscay mixed fishery. Marine Policy, 40, 64-74.
Gulland, J. A. (1956). On the fishing effort in English demersal fisheries. London: Her
Majesty's Stationery Office.
Gurevitch, J., Koricheva, J., Nakagawa, S., & Stewart, G. (2018). Meta-analysis and the science
of research synthesis. Nature, 555, 175.
Hannesson, R. (1983). Bioeconomic Production Function in Fisheries: Theoretical and
Empirical Analysis. Canadian Journal of Fisheries and Aquatic Sciences, 40(7), 968-
982. doi:10.1139/f83-123
Hannesson, R. (2015). World Fisheries in Crisis? Marine Resource Economics, 30(3), 251-
260.
Harper, S., Grubb, C., Stiles, M., & Sumaila, U. R. (2017). Contributions by Women to
Fisheries Economies: Insights from Five Maritime Countries. Coastal Management,
45(2), 91-106.
Harper, S., Zeller, D., Hauzer, M., Pauly, D., & Sumaila, U. R. (2013). Women and fisheries:
Contribution to food security and local economies. Marine Policy, 39(1), 56-63.
doi:10.1016/j.marpol.2012.10.018
Hart, P. J., & Reynolds, J. D. (2008). Handbook of Fish Biology and Fisheries. USA: John
Wiley & Sons.
Hauge, K. H., Cleeland, B., & Wilson, D. C. (2009). Fisheries depletion and collapse. Risk
governance deficits: an analysis and illustration of the most common deficits in risk
governance. International Risk Governance Council. Geneva, Switzerland
Heen, E. E., Heen, K., & Leung, P. (2014). Conflicting goals in fisheries management - A study
of the Norwegian cod fisheries. Marine Policy, 49, 73-80.
Henningsen, A., & Henning, H. C. A. C. (2009). Imposing regional monotonicity on translog
stochastic production frontiers with a simple three-step procedure. Journal of
Productivity Analysis, 32(3), 217-229.
Hentrich, S., & Salomon, M. (2006). Flexible management of fishing rights and a sustainable
fisheries industry in Europe. Marine Policy, 30(6), 712-720.
Herrero, I. (2005). Different approaches to efficiency analysis. An application to the Spanish
Trawl fleet operating in Moroccan waters. European Journal of Operational Research,
167(1), 257-271.
Herrero, I., & Pascoe, S. (2003). Value versus Volume in the Catch of the Spanish South-
Atlantic Trawl Fishery. Journal of Agricultural Economics, 54(2), 325-341.

References 185
Herrero, I., Pascoe, S., & Mardle, S. (2006). Mix Efficiency in a Multi-species Fishery. Journal
of Productivity Analysis, 25(3), 231-241.
Hilborn, R. (2007a). Defining success in fisheries and conflicts in objectives. Marine Policy,
31(2), 153-158.
Hilborn, R. (2007b). Managing fisheries is managing people: what has been learned? Fish and
Fisheries, 8(4), 285-296.
Hilborn, R. (2007c). Moving to Sustainability by Learning from Successful Fisheries. AMBIO:
A Journal of the Human Environment, 36(4), 296-303.
Hilborn, R., Stewart, I. J., Branch, T. A., & Jensen, O. P. (2012). Defining Trade-Offs among
Conservation, Profitability, and Food Security in the California Current Bottom-Trawl
Fishery. Conservation Biology, 26(2), 257-268. doi:doi:10.1111/j.1523-
1739.2011.01800.x
Hoff, A. (2006). Bootstrapping Malmquist indices for Danish seiners in the North Sea and
Skagerrak. Journal of Applied Statistics, 33(9), 891-907.
doi:10.1080/02664760600742151
Holland, D., Gudmundsson, E., & Gates, J. (1999). Do fishing vessel buyback programs work:
A survey of the evidence. Marine Policy, 23(1), 47-69.
Holma, M., Lindroos, M., Romakkaniemi, A., & Oinonen, S. (2018). Comparing economic
and biological management objectives in the commercial Baltic salmon fisheries.
Marine Policy. doi:https://doi.org/10.1016/j.marpol.2018.11.011
Hoshino, E., Pascoe, S., Hutton, T., Kompas, T., & Yamazaki, S. (2017). Estimating maximum
economic yield in multispecies fisheries: a review. Reviews in Fish Biology and
Fisheries. doi:10.1007/s11160-017-9508-8
Houghton, R. G. (1977). The fishing power of trawlers in the western English Channel between
1965 and 1968. ICES Journal of Marine Science, 37(2), 130-136.
doi:10.1093/icesjms/37.2.130
Huang, L., Ray, S., Segerson, K., & Walden, J. B. (2018). Impact of Collective Rights-based
Fisheries Management: Evidence from the New England Groundfish Fishery. Marine
Resource Economics, 33(2), 177-201. doi:10.1086/697478
Huang, L., & Smith, M. D. (2014). The dynamic efficiency costs of common-pool resource
exploitation. American Economic Review, 104(12), 4071-4103.
Iliyasu, A., Mohamed, Z. A., Ismail, M. M., & Abdullah, A. M. (2014). A Meta-Analysis of
Technical Efficiency in Aquaculture. Journal of Applied Aquaculture, 26(4), 329-339.
Innes, J., & Pascoe, S. (2008). Productivity Impacts of Veil Nets on UK Crangon Vessels.
Journal of Agricultural Economics, 59(3), 574-588.
Irsova, Z., & Havranek, T. (2013). A meta-analysis of bank efficiency measurement in
transitional countries. Transformations in Business and Economics, 12(2), 163-183.
Jamnia, A. R., Mazloumzadeh, S. M., & Keikha, A. A. (2015). Estimate the technical efficiency
of fishing vessels operating in Chabahar region, Southern Iran. Journal of the Saudi
Society of Agricultural Sciences, 14(1), 26-32.
Jennings, S., & Kaiser, M. J. (1998). The Effects of Fishing on Marine Ecosystems. In J. H. S.
Blaxter, A. J. Southward, & P. A. Tyler (Eds.), Advances in Marine Biology (Vol. 34,
pp. 201-352): Academic Press.

References 186
Jeon, Y., Ishak Haji, O., Kuperan, K., Squires, D., & Susilowati, I. (2006). Developing country
fisheries and technical efficiency: the Java Sea purse seine fishery. Applied Economics,
38(13), 13.
Jondrow, J., Knox Lovell, C. A., Materov, I. S., & Schmidt, P. (1982). On the estimation of
technical inefficiency in the stochastic frontier production function model. Journal of
Econometrics, 19(2), 233-238. doi:https://doi.org/10.1016/0304-4076(82)90004-5
Jones, J. (1992). Environmental impact of trawling on the seabed: a review. New Zealand
Journal of Marine and Freshwater Research, 26(1), 59-67.
Kearney, R. E. (2001). Fisheries property rights and recreational/commercial conflict:
implications of policy developments in Australia and New Zealand. Marine Policy,
25(1), 49-59. doi:https://doi.org/10.1016/S0308-597X(00)00035-X
Kearney, R. E. (2002). Co-management: the resolution of conflict between commercial and
recreational fishers in Victoria, Australia. Ocean & Coastal Management, 45(4), 201-
214. doi:https://doi.org/10.1016/S0964-5691(02)00055-8
Kempf, A., Mumford, J., Levontin, P., Leach, A., Hoff, A., Hamon, K. G., . . . Rindorf, A.
(2016). The MSY concept in a multi-objective fisheries environment – Lessons from
the North Sea. Marine Policy, 69, 146-158.
doi:http://dx.doi.org/10.1016/j.marpol.2016.04.012
Kennelly, S. J. (1995). The issue of bycatch in Australia's demersal trawl fisheries. Reviews in
Fish Biology and Fisheries, 5(2), 213-234. doi:10.1007/bf00179757
Khondker, M.-e.-J., Belton, B., & Viswanathan, K. K. (2014). Communication strategies for
managing coastal fisheries conflicts in Bangladesh. Ocean & Coastal Management, 92,
65-73. doi:https://doi.org/10.1016/j.ocecoaman.2014.01.003
Khondker, M.-e.-J., Salayo, N. D., & Kanagaratnam, U. (2009). Managing fisheries conflicts
through communication planning: Experience from inland fisheries of Bangladesh.
Fisheries Research, 99(2), 112-122. doi:https://doi.org/10.1016/j.fishres.2009.04.009
Kiadaliri, A. A., Jafari, M., & Gerdtham, U.-G. (2013). Frontier-based techniques in measuring
hospital efficiency in Iran: A systematic review and meta-regression analysis. BMC
Health Services Research, 13(1), 312-312.
Kim-Hyun, J. (2019). South Korea's use of force against Chinese illegal fishing in the course
of law enforcement in the Yellow Sea. Marine Policy, 99, 148-156.
Kim Anh, T. N., & Flaaten, O. (2011). Facilitating Change: A Mekong Vietnamese Small-
Scale Fishing Community. In (pp. 335-357). Dordrecht: Springer Netherlands.
Kim, D.-H., Seo, J. N., Kim, H. S., & Lee, K. (2012). Estimation of productivity growth,
technical progress, and efficiency changes in the Korean offshore fisheries. Fisheries
Science, 78(3), 743-751.
Kirkley, J., Morrison Paul, C. J., Cunningham, S., & Catanzano, J. (2004). Embodied and
disembodied technical change in fisheries: An analysis of the sète trawl fishery, 1985-
1999. Environmental and Resource Economics, 29(2), 191-217.
Kirkley, J., Squires, D., & Strand, I. E. (1995). Assessing technical efficiency in commercial
fisheries: the mid-Atlantic sea scallop fishery. American Journal of Agricultural
Economics, 77(3), 686-697.

References 187
Kirkley, J., Squires, D., & Strand, I. E. (1998). Characterizing Managerial Skill and Technical
Efficiency in a Fishery. Journal of Productivity Analysis, 9(2), 145-160.
Kompas, T. (2005). Fisheries Management: Economic Efficiency and the Concept of
'Maximum Economic Yield'. Australian Commodities: Forecasts and Issues, 12(1),
152-160.
Kompas, T., & Che, T. N. (2005). Efficiency Gains and Cost Reductions from Individual
Transferable Quotas: A Stochastic Cost Frontier for the Australian South East Fishery.
Journal of Productivity Analysis, 23(3), 285-307.
Kompas, T., Che, T. N., & Grafton, R. Q. (2004). Technical efficiency effects of input controls:
evidence from Australia's banana prawn fishery. Applied Economics, 36(15), 1631-
1641.
Kompas, T., Dichmont, C. M., Punt, A. E., Deng, R., Che, T. N., Bishop, J., . . . Zhou, S.
(2010). Maximizing profits and conserving stocks in the Australian Northern Prawn
Fishery. Australian Journal of Agricultural and Resource Economics, 54(3), 281-299.
Kompas, T., & Gooday, P. (2007). The failure of ‘command and control’approaches to fisheries
management: lessons from Australia. International journal of global environmental
issues, Vol. 7(Nos. 2/3).
Kopp, R. J., & Smith, V. K. (1980). Frontier Production Function Estimates for Steam Electric
Generation: A Comparative Analysis. Southern Economic Journal, 46(4), 1049.
Kularatne, M. G., Pascoe, S., Wilson, C., & Robinson, T. (2018). Efficiency of culture-based
fisheries production in village irrigation systems of Sri Lanka. Aquaculture Economics
& Management, 1-21. doi:10.1080/13657305.2018.1497104
Kumbhakar, S. C., & Lovell, C. A. K. (2000). Stochastic frontier analysis. Cambridge:
Cambridge University Press.
Laird, A. (2018). Northern Prawn Fishery Data Summary 2017. NPF Industry Pty Ltd,
Australia
Lampe, H. W., & Hilgers, D. (2015). Trajectories of efficiency measurement: A bibliometric
analysis of DEA and SFA. European Journal of Operational Research, 240(1), 1-21.
Lau, L. J. (1978). Testing and Imposing Monotonicity, Convexity, and Quasiconcavity. In M.
Fuss & D. McFadden (Eds.), Production economics: A dual approach to theory and
applications (Vol. 1: The theory of production). North-Holland, New York.
Le, T. T., Nguyen, K. A. T., & Jolly, C. M. (2016). Perception, preparedness and severity of
climate change triggered events in Ben Tre Province, Vietnam. International Journal
of Food and Agricultural Economics, 4(4), 29.
Lee, S.-G., & Rahimi Midani, A. (2015). Productivity change under the vessel buyback
program in Korean fisheries. Fisheries Science, 81(1), 21-28.
Legović, T., Klanjšček, J., & Geček, S. (2010). Maximum sustainable yield and species
extinction in ecosystems. Ecological Modelling, 221(12), 1569-1574.
Leung, P., Heen, K., & Bardarson, H. (2001). Regional economic impacts of fish resources
utilization from the Barents Sea: Trade-offs between economic rent, employment and
income. European Journal of Operational Research, 133(2), 432-446.

References 188
Leung, P., Muraoka, J., Nakamoto, S. T., & Pooley, S. (1998). Evaluating fisheries
management options in Hawaii using analytic hierarchy process (AHP). Fisheries
Research, 36(2), 171-183. doi:https://doi.org/10.1016/S0165-7836(98)00097-6
Lichter, A., Peichl, A., & Siegloch, S. (2015). The own-wage elasticity of labor demand: A
meta-regression analysis. European Economic Review, 80, 94-119.
Lindhjem, H. (2007). 20 years of stated preference valuation of non-timber benefits from
Fennoscandian forests: A meta-analysis. Journal of Forest Economics, 12(4), 251-277.
Liu, J. S., Lu, L. Y. Y., Lu, W.-M., & Lin, B. J. Y. (2013). A survey of DEA applications.
Omega, 41(5), 893-902.
Lokina, R. B. (2009). Technical efficiency and the role of skipper skill in artisanal Lake
Victoria fisheries. Environment and Development Economics, 14(4), 497-519.
Long, L. K., Flaaten, O., & Anh, N. T. K. (2008). Economic performance of open-access
offshore fisheries—The case of Vietnamese longliners in the South China Sea.
Fisheries Research, 93(3), 296-304.
Löthgren, M. (1997). Generalized stochastic frontier production models. Economics Letters,
57(3), 255-259. doi:http://dx.doi.org/10.1016/S0165-1765(97)00246-2
Mai, N. T. T., Rahtz, D. R., & Shultz, C. J. (2014). Tourism as catalyst for quality of life in
transitioning subsistence marketplaces: Perspectives from Ha Long, Vietnam. Journal
of Macromarketing, 34(1), 28-44.
Mardani, A., Zavadskas, E. K., Streimikiene, D., Jusoh, A., & Khoshnoudi, M. (2017). A
comprehensive review of data envelopment analysis (DEA) approach in energy
efficiency. Renewable and Sustainable Energy Reviews, 70, 1298-1322.
doi:https://doi.org/10.1016/j.rser.2016.12.030
Mardle, S., & Pascoe, S. (2002). Modelling the effects of trade-offs between long and short-
term objectives in fisheries management. Journal of environmental management, 65(1),
49-62.
Mardle, S., Pascoe, S., Boncoeur, J., Gallic, B. L., Garcı́a-Hoyo, J. J., Herrero, I., . . . Mathiesen,
C. (2002). Objectives of fisheries management: case studies from the UK, France, Spain
and Denmark. Marine Policy, 26(6), 415-428.
Mardle, S., Pascoe, S., & Herrero, I. (2004). Management objective importance in fisheries: an
evaluation using the analytic hierarchy process (AHP). Environmental Management,
33(1), 1-11.
Mareth, T., Thomé, A. M. T., Cyrino Oliveira, F. L., & Scavarda, L. F. (2016). Systematic
review and meta-regression analysis of technical efficiency in dairy farms.
International Journal of Productivity and Performance Management, 65(3), 279-301.
Martin, S. M., Lorenzen, K., & Bunnefeld, N. (2013). Fishing Farmers: Fishing, Livelihood
Diversification and Poverty in Rural Laos. Human Ecology, 41(5), 737-747.
doi:10.1007/s10745-013-9567-y
Mason, F. (2002). The Newfoundland Cod Stock Collapse: A Review and Analysis of Social
Factors. Electronic Green Journal, 1(17).
Meeusen, W., & Broeck, J. v. D. (1977). Efficiency Estimation from Cobb-Douglas Production
Functions with Composed Error. International Economic Review, 18(2), 435-444.

References 189
Mekasha, T. J., & Tarp, F. (2013). Aid and Growth: What Meta-Analysis Reveals. The Journal
of Development Studies, 49(4), 564-583.
Michaelides, P. G., Vouldis, A. T., & Tsionas, E. G. (2010). Globally flexible functional forms:
The neural distance function. European Journal of Operational Research, 206(2), 456-
469. doi:https://doi.org/10.1016/j.ejor.2010.02.013
Ministry of Agriculture and Rural Development. (2012). The Plan to restructure of marine
capture fisheries in Vietnam. Ministry of Agriculture and Rural Development. Hanoi,
Vietnam.
Ministry of Agriculture and Rural Development. (2015). Circular No.9443/BNN-TCTS about
enhancing the capacity of marine capture fisheries management. Ministry of
Agriculture and Rural Development. Hanoi, Vietnam
Ministry of Labour - Invalids and Social Affairs. (2016). The Decision No. 1095/QĐ-LĐTBXH
on approval of results of the Vietnam poor households survey 2015. Hanoi, Vietnam
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred reporting items for
systematic reviews and meta-analyses: the PRISMA statement. Bmj, 339.
Moore, W., & Thomas, C. (2010). A meta‐analysis of the relationship between debt and
growth. International Journal of Development Issues, 9(3), 214-225.
Moriera Lopez, V., & Bravo-Ureta, B. (2009). A study of dairy farm technical efficiency using
meta-regression: An international perspective. Chilean Journal of Agricultural
Research, 69(2), 214-223.
Muawanah, U., Pomeroy, R. S., & Marlessy, C. (2012). Revisiting Fish Wars: Conflict and
Collaboration over Fisheries in Indonesia. Coastal Management, 40(3), 279-288.
doi:10.1080/08920753.2012.677633
Murillo‐Zamorano, L. R. (2004). Economic efficiency and frontier techniques. Journal of
economic surveys, 18(1), 33-77.
Nam, N. H. (2013). Vietnam seafood export: Opportunities and challenges. Retrieved from
https://www.ide.go.jp/library/English/Events/Sympo/pdf/130919_unido_04.pdf (30
June, 2017).
Nelson, J. P., & Kennedy, P. E. (2009). The use (and abuse) of meta-analysis in environmental
and natural resource economics: an assessment. Environmental and Resource
Economics, 42(3), 345-377.
Ngoc, Q. T. K., Flaaten, O., & Kim Anh, T. N. (2009). Efficiency of Fishing vessels affected
by a Marine Protected Area – The case of small-scale trawlers and The Marine
Protected Area in Nha Trang Bay, Vietnam. In Integrated Coastal Zone Management.
US: Wiley Blackwell.
Nguyen, K. A. T., Jolly, C. M., Bui, C. T., & Le, T. H. (2015). Climate change, rural household
food consumption and vulnerability: The case of Ben Tre province in Vietnam.
Agricultural Economics Review, 16(2), 95.
Nguyen Kim, H., & Coelli, T. (2009). Quantifying the effects of modelling choices on hospital
efficiency measures: a meta-regression analysis. Working Paper Series No.
WP07/2009. Centre for Efficiency and Productivity Analysis, School of Economics,
University of Queensland, Australia

References 190
Nguyen Long. (2003). A preliminary analysis on the socioeconomic situation of coastal fishing
communities in Vietnam. In G. Silvestre, L. Garces, I. Stobutzki, M. Ahmed, R. A.
Valmonte-Santos, C. Luna, L. Lachica-Alino, P. Munro, V. Christensen, & D. Pauly
(Eds.), Assessment, management and future directions for coastal fisheries in Asian
countries. WorldFish Center Conference Proceedings (Vol. 67, pp. 657-688). Penang,
Malaysia: WorldFish Center.
NOAA. (2017). Understanding Fisheries Management in the United States. National Oceanic
and Atmospheric Administration (NOAA), USA. Retrieved from
https://www.fisheries.noaa.gov/insight/understanding-fisheries-management-united-
states (09 October 2018).
Norman-López, A., & Pascoe, S. (2011). Net economic effects of achieving maximum
economic yield in fisheries. Marine Policy, 35(4), 489-495.
doi:10.1016/j.marpol.2010.12.001
Nøstbakken, L., Thebaud, O., & Sørensen, L.-C. (2011). Investment behaviour and capacity
adjustment in fisheries: a survey of the literature. Marine Resource Economics, 26(2),
95-117.
O’Donnell, C. J., & Coelli, T. J. (2005). A Bayesian approach to imposing curvature on
distance functions. Journal of Econometrics, 126(2), 493-523.
doi:10.1016/j.jeconom.2004.05.011
Odeck, J., & Bråthen, S. (2012). A meta-analysis of DEA and SFA studies of the technical
efficiency of seaports: A comparison of fixed and random-effects regression models.
Transportation Research Part A: Policy and Practice, 46(10), 1574-1585.
OECD. (2011). Fisheries Policy Reform: National Experiences. Paris: OECD Publishing.
Ogundari, K. (2014). The Paradigm of Agricultural Efficiency and its Implication on Food
Security in Africa: What Does Meta-analysis Reveal? World development, 64, 690-702.
Ogundari, K., Amos, T. T., & Okoruwa, V. O. (2012). A Review of Nigerian Agricultural
Efficiency Literature, 1999–2011: What Does One Learn from Frontier Studies?
African Development Review, 24(1), 93-106.
Ogundari, K., & Brümmer, B. (2011). Technical Efficiency of Nigerian Agriculture: A Meta-
Regression Analysis. Outlook on Agriculture, 40(2), 171-180.
Oliveira, M. M., Camanho, A. S., & Gaspar, M. B. (2010). Technical and economic efficiency
analysis of the Portuguese artisanal dredge fleet. ICES Journal of Marine Science,
67(8), 1811-1821.
Oliveira, M. M., Camanho, A. S., & Gaspar, M. B. (2013). The influence of catch quotas on
the productivity of the Portuguese bivalve dredge fleet. ICES Journal of Marine
Science, 70(7), 1378-1388. doi:10.1093/icesjms/fst098
Oliveira, M. M., Camanho, A. S., Walden, J. B., & Gaspar, M. B. (2016). Evaluating the
influence of skipper skills in the performance of Portuguese artisanal dredge vessels.
ICES Journal of Marine Science, 73(10), 2721-2728. doi:10.1093/icesjms/fsw103
Oliveira, M. M., Gaspar, M. B., Paixão, J. P., & Camanho, A. S. (2009). Productivity change
of the artisanal fishing fleet in Portugal: A Malmquist index analysis. Fisheries
Research, 95(2–3), 189-197. doi:http://dx.doi.org/10.1016/j.fishres.2008.08.020

References 191
Onyango, P. O. (2011). Occupation of Last Resort? Small-Scale Fishing in Lake Victoria,
Tanzania. In S. Jentoft & A. Eide (Eds.), Poverty Mosaics: Realities and Prospects in
Small-Scale Fisheries (pp. 97-124). Dordrecht: Springer Netherlands.
Orea, L., Alvarez, A., & Paul, C. J. M. (2005). Modeling and measuring production process
for a multi-species fishery: Alternative technical efficiency estimates for the Northern
Spain hake fishery. Natural Resource Modeling, 18(2), 183-213.
Pascoe, S., Andersen, J. L., & de Wilde, J. W. (2001). The Impact of Management Regulation
on the Technical Efficiency of Vessels in the Dutch Beam Trawl Fishery. European
Review of Agricultural Economics, 28(2), 187-206.
Pascoe, S., Brooks, K., Cannard, T., Dichmont, C. M., Jebreen, E., Schirmer, J., & Triantafillos,
L. (2014). Social objectives of fisheries management: What are managers' priorities?
Ocean & Coastal Management, 98(0), 1-10.
Pascoe, S., Cannard, T., Jebreen, E., Dichmont, C. M., & Schirmer, J. (2015). Satisfaction with
fishing and the desire to leave. Ambio, 44(5), 401-411.
Pascoe, S., & Coglan, L. (2000). Implications of differences in technical efficiency of fishing
boats for capacity measurement and reduction. Marine Policy, 24(4), 301-307.
Pascoe, S., & Coglan, L. (2002). The Contribution of Unmeasurable Inputs to Fisheries
Production: An Analysis of Technical Efficiency of Fishing Vessels in the English
Channel. American Journal of Agricultural Economics, 84(3), 585-597.
Pascoe, S., Coglan, L., Punt, A. E., & Dichmont, C. M. (2012). Impacts of Vessel Capacity
Reduction Programmes on Efficiency in Fisheries: The Case of Australia's Multispecies
Northern Prawn Fishery. Journal of Agricultural Economics, 63(2), 425-443.
Pascoe, S., Dichmont, C. M., Brooks, K., Pears, R., & Jebreen, E. (2013). Management
objectives of Queensland fisheries: Putting the horse before the cart. Marine Policy,
37(0), 115-122.
Pascoe, S., Dichmont, C. M., Vieira, S., Kompas, T., Buckworth, R. C., & Carter, D. (2013).
A Retrospective Evaluation of Sustainable Yields for Australia's Northern Prawn
Fishery: An Alternative View. Fisheries, 38(11), 502-508.
Pascoe, S., Hassaszahed, P., Anderson, J., & Korsbrekke, K. (2003). Economic versus physical
input measures in the analysis of technical efficiency in fisheries. Applied Economics,
35(15), 1699-1710.
Pascoe, S., & Herrero, I. (2004). Estimation of a composite fish stock index using data
envelopment analysis. Fisheries Research, 69(1), 91-105.
Pascoe, S., Hutton, T., Coglan, L., & Nguyen, V. Q. (2017). Implications of efficiency and
productivity change over the season for setting MEY‐based trigger targets. Australian
Journal of Agricultural and Resource Economics, 59, 1-18.
Pascoe, S., Hutton, T., van Putten, I., Dennis, D., Plaganyi-Lloyd, E., & Deng, R. (2013).
Implications of Quota Reallocation in the Torres Strait Tropical Rock Lobster Fishery:
Implications of quota reallocation. Canadian Journal of Agricultural Economics, 61(2),
335-352.
Pascoe, S., Innes, J., Courtney, A., & Kienzle, M. (2017). Impact of reducing investment
disincentives on the sustainability of the Moreton Bay prawn trawl fishery. Fisheries
Research, 186, 121-130.

References 192
Pascoe, S., Kirkley, J., Gréboval, D., & Morrison, P., C.J. (2003). Measuring and assessing
capacity in fisheries. 2. Issues and methods. FAO Fisheries Technical Paper. No. 433/2.
Rome, Italy: Food and Agriculture Organization of the United Nations.
Pascoe, S., & Mardle, S. (Eds.). (2003). Efficiency analysis in EU fisheries: stochastic
production frontiers and data envelopment analysis. CEMARE Report 60, CEMARE,
University of Portsmouth, UK.
Pascoe, S., Plagányi, É. E., & Dichmont, C. M. (2016). Modelling multiple management
objectives in fisheries: Australian experiences. ICES Journal of Marine Science, 74(2),
464-474.
Pascoe, S., Proctor, W., Wilcox, C., Innes, J., Rochester, W., & Dowling, N. (2009).
Stakeholder objective preferences in Australian Commonwealth managed fisheries.
Marine Policy, 33(5), 750-758.
Pascoe, S., Punt, A. E., & Dichmont, C. M. (2010). Targeting ability and output controls in
Australia's multi-species Northern Prawn Fishery. European Review of Agricultural
Economics, 37(3), 313-334.
Pascoe, S., & Robinson, C. (1998). Input controls, input substitution and profit maximisation
in the English Channel beam trawl fishery. Journal of Agricultural Economics, 49(1),
16-33.
Pascoe, S., Sharp, J., & Buckworth, R. C. (2015). Modelling effort levels in a sequential
fishery. ICES Journal of Marine Science, 73(2), 503-511.
Pascoe, S., & Tingley, D. (2007). Capacity and Technical Efficiency Estimation in Fisheries:
Parametric and Non-parametric Techniques. In A. Weintraub, C. Romero, T. Bjørndal,
R. Epstein, & J. Miranda (Eds.), Handbook Of Operations Research In Natural
Resources (pp. 273-294). Boston, MA: Springer US.
Pascoe, S., Vieira, S., Dichmont, C. M., & Punt, A. E. (2011). Optimal vessel size and output
in the Australian northern prawn fishery: a restricted profit function approach.
Australian Journal of Agricultural and Resource Economics, 55(1), 107-125.
doi:10.1111/j.1467-8489.2010.00526.x
Pascoe, S., Vieira, S., & Thebaud, O. (2015). Allocating repairs and maintenance costs to fixed
or variable costs in fisheries bioeconomic models. Applied Economics Letters, 22(2),
127-131. doi:10.1080/13504851.2014.929619
Pearse, P. H. (1992). From open access to private property: Recent innovations in fishing rights
as instruments of fisheries policy. Ocean Development & International Law, 23(1), 71-
83.
Perret, S. R., & Yuerlita. (2014). Adapting to declining fish resources: the differentiation of
livelihood systems and fishing strategies in Singkarak Lake’s fishing community, West
Sumatra. Regional Environmental Change, 14(3), 1203-1214. doi:10.1007/s10113-
013-0554-z
Pfeiffer, L., & Gratz, T. (2016). The effect of rights-based fisheries management on risk taking
and fishing safety. Proceedings of the National Academy of Sciences, 113(10), 2615.
doi:10.1073/pnas.1509456113
Pham, V. A., De Laender, F., Everaert, G., Chu, T. V., & Goethals, P. (2014). An integrated
food web model to test the impact of fisheries management scenarios on the coastal
ecosystem of Vietnam. Ocean & Coastal Management, 92(0), 74-86.

References 193
Pham, V. A., Gert, E., Chu, T. V., & Goethals, P. (2014). Need for integrated analysis and
management instruments to attain sustainable fisheries in Vietnam. Sustainability of
Water Quality and Ecology, 3–4, 151-154.
Pho Hoang Han, M. M. M. (2007). Fisheries development in Vietnam: A case study in the
exclusive economic zone. Ocean & Coastal Management, 50(9), 699-712.
Pomeroy, R. (2012). Managing overcapacity in small-scale fisheries in Southeast Asia. Marine
Policy, 36(2), 520-527.
Pomeroy, R., Kim Anh, T. N., & Thong, H. X. (2009). Small-scale marine fisheries policy in
Vietnam. Marine Policy, 33(2), 419-428.
Pomeroy, R., Parks, J., Mrakovcich, K. L., & LaMonica, C. (2016). Drivers and impacts of
fisheries scarcity, competition, and conflict on maritime security. Marine Policy, 67,
94-104. doi:https://doi.org/10.1016/j.marpol.2016.01.005
Pomeroy, R., Parks, J., Pollnac, R., Campson, T., Genio, E., Marlessy, C., . . . Thu Hue, N.
(2007). Fish wars: Conflict and collaboration in fisheries management in Southeast
Asia. Marine Policy, 31(6), 645-656.
Punt, A. E., Deng, R., Dichmont, C. M., Kompas, T., Venables, W. N., Zhou, S., . . . Kienzle,
M. (2010). Integrating size-structured assessment and bioeconomic management advice
in Australia's northern prawn fishery. ICES Journal of Marine Science, 67(8), 1785-
1801.
Punt, A. E., Deng, R., Pascoe, S., Dichmont, C. M., Zhou, S., Plagányi, É. E., . . . van der
Velde, T. (2011). Calculating optimal effort and catch trajectories for multiple species
modelled using a mix of size-structured, delay-difference and biomass dynamics
models. Fisheries Research, 109(1), 201-211.
Quang, N., & Leung, P. (2009). Do fishermen have different attitudes toward risk? An
application of prospect theory to the study of Vietnamese fishermen. Journal of
Agricultural and Resource Economics, 34(3), 518-538.
Quang Ninh Statistics Office. (2016). Quang Ninh Statistical Yearbook 2015. Statistical
publishing house. Hanoi, Vietnam.
Quijano, D., Salas, S., Monroy-García, C., & Velázquez-Abunader, I. (2018). Factors
contributing to technical efficiency in a mixed fishery: Implications in buyback
programs. Marine Policy, 94, 61-70. doi:https://doi.org/10.1016/j.marpol.2018.05.004
Quynh, C. N. T., Hailu, A., Schilizzi, S., & Iftekhar, S. (2018). Fisher participation in
monitoring: Does it help reduce excessive investment in fishing capacity? Fisheries
Research, 206, 138-149. doi:https://doi.org/10.1016/j.fishres.2018.04.024
Raakjær, J., Manh Son, D., Stæhr, K.-J., Hovgård, H., Dieu Thuy, N. T., Ellegaard, K., . . .
Giang Hai, P. (2007). Adaptive fisheries management in Vietnam: The use of indicators
and the introduction of a multi-disciplinary Marine Fisheries Specialist Team to support
implementation. Marine Policy, 31(2), 143-152.
Richard, B., Kevin, M., & Mihaela, Z. (2017). MRAG - Northern Prawn Fishery Final Report
and Determination. FL 33702, USA: MRAG Americas, Inc.
Rindorf, A., Dichmont, C. M., Thorson, J., Charles, A., Clausen, L. W., Degnbol, P., . . . Levin,
P. (2017). Inclusion of ecological, economic, social, and institutional considerations
when setting targets and limits for multispecies fisheries. ICES Journal of Marine
Science, 74(2), 453-463.

References 194
Rindorf, A., Mumford, J., Baranowski, P., Clausen, L. W., García, D., Hintzen, N. T., . . . Reid,
D. (2017). Moving beyond the MSY concept to reflect multidimensional fisheries
management objectives. Marine Policy, 85, 33-41.
doi:https://doi.org/10.1016/j.marpol.2017.08.012
Robson, D. S. (1966). Estimation of the relative fishing power of individual ships. Research
Bulletin of the International Commission for North-West Atlantic Fisheries, 3, 5-14.
Rust, S., Yamazaki, S., Jennings, S., Emery, T., & Gardner, C. (2017). Excess capacity and
efficiency in the quota managed Tasmanian Rock Lobster Fishery. Marine Policy, 76,
55-62.
Salayo, N., Garces, L., Pido, M., Viswanathan, K., Pomeroy, R., Ahmed, M., . . . Masae, A.
(2008). Managing excess capacity in small-scale fisheries: Perspectives from
stakeholders in three Southeast Asian countries. Marine Policy, 32(4), 692-700.
Sale, P. F. (2002). The Science We Need to Develop for More Effective Management. In P. F.
Sale (Ed.), Coral Reef Fishes (pp. 361-376). San Diego: Academic Press.
Sauer, J. (2006). Economic Theory and Econometric Practice: Parametric Efficiency Analysis.
Empirical Economics, 31(4), 1061-1087.
Sauer, J., Frohberg, K., & Hockmann, H. (2006). Stochastic efficiency measurement: The curse
of theoretical consistency. Journal of Applied Economics, 9(1), 139-165.
Schmidt, P. (1985). Frontier production functions. Econometric Reviews, 4(2), 289-328.
Schofield, C. H., Sumaila, R., & Cheung, W. (2016). Fishing, not oil, is at the heart of the South
China Sea dispute. (16 August) The Conversation 1-5.
Schrank, W. E., & Roy, N. (2013). The Newfoundland Fishery and Economy Twenty Years
after the Northern Cod Moratorium. Marine Resource Economics, 28(4), 397-413.
Sciberras, M., Hiddink, J. G., Jennings, S., Szostek, C. L., Hughes, K. M., Kneafsey, B., . . .
Kaiser, M. J. (2018). Response of benthic fauna to experimental bottom fishing: A
global meta-analysis. Fish and Fisheries, 19(4), 698-715. doi:doi:10.1111/faf.12283
Sesabo, J. K., & Tol, R. S. J. (2007). Technical efficiency of small-scale fishing households in
Tanzanian coastal villages: an empirical analysis. African Journal of Aquatic Science,
32(1), 51-61.
Sharma, K. R., & Leung, P. (1998). Technical Efficiency of the Longline Fishery in Hawaii:
An Application of a Stochastic Production Frontier. Marine Resource Economics,
13(4), 259-274.
Sharp, B. M. H., Castilla-Espino, D., & García del Hoyo, J. J. (2004). Efficiency in the New
Zealand rock lobster fishery: A production frontier analysis. New Zealand Economic
Papers, 38(2), 207-218.
Simar, L., & Wilson, P. W. (1998). Sensitivity analysis of efficiency scores: How to bootstrap
in nonparametric frontier models. Management Science, 44(1), 49-61.
Simar, L., & Wilson, P. W. (2000). Statistical Inference in Nonparametric Frontier Models:
The State of the Art. Journal of Productivity Analysis, 13(1), 49-78.
doi:10.1023/a:1007864806704
Soliman, A. (2014). Using individual transferable quotas (ITQs) to achieve social policy
objectives: A proposed intervention. Marine Policy, 45, 76-81.

References 195
Spijkers, J., Morrison, T. H., Blasiak, R., Cumming, G. S., Osborne, M., Watson, J., &
Österblom, H. (2018). Marine fisheries and future ocean conflict. Fish and Fisheries.
Squires, D. (1987). Fishing effort: Its testing, specification, and internal structure in fisheries
economics and management. Journal of Environmental Economics and Management,
14(3), 268-282. doi:https://doi.org/10.1016/0095-0696(87)90020-9
Squires, D. (1992). Productivity Measurement in Common Property Resource Industries: An
Application to the Pacific Coast Trawl Fishery. The RAND Journal of Economics,
23(2), 221-236. doi:10.2307/2555985
Squires, D. (2010). Fisheries buybacks: A review and guidelines. Fish and Fisheries, 11(4),
366-387.
Squires, D., Grafton, R. Q., Alam, M. F., & Omar, I. H. (2003). Technical efficiency in the
Malaysian gill net artisanal fishery. Environment and Development Economics, 8(03),
481-504.
Squires, D., & Kirkley, J. (1999). Skipper skill and panel data in fishing industries. Canadian
Journal of Fisheries and Aquatic Sciences, 56(11), 2011-2018.
Squires, D., Kirkley, J., & Tisdell, C. A. (1995). Individual transferable quotas as a fisheries
management tool. Reviews in Fisheries Science, 3(2), 141-169.
Squires, D., & Vestergaard, N. (2016). Putting Economics into Maximum Economic Yield.
Marine Resource Economics, 31(1), 101-116. doi:10.1086/683670
Stanley, T. D., & Doucouliagos, H. (2012). Meta-regression analysis in economics and
business. London: Routledge.
Stanley, T. D., Doucouliagos, H., Giles, M., Heckemeyer, J. H., Johnston, R. J., Laroche, P., .
. . Rost, K. (2013). Meta-analysis of economics research reporting guidelines. Journal
of economic surveys, 27(2), 390-394.
Stanley, T. D., & Jarrell, S. B. (1989). Meta‐Regression analysis: A quantitative method of
literature surveys. Journal of economic surveys, 3(2), 161-170.
Stephenson, R. L., Paul, S., Wiber, M., Angel, E., Benson, A. J., Charles, A., . . . Sumaila, U.
R. (2018). Evaluating and implementing social–ecological systems: A comprehensive
approach to sustainable fisheries. Fish and Fisheries, 19(5), 853-873.
doi:doi:10.1111/faf.12296
Stoeckl, N., Larson, S., Thomas, J., Hicks, C., Pascoe, S., & Marsh, H. (2017). Socioeconomic
Impacts of Changes to Marine Fisheries and Aquaculture that are Brought About
Through Climate Change. In Climate Change Impacts on Fisheries and Aquaculture
(pp. 925-958): John Wiley & Sons, Ltd.
Stouten, H., Heene, A., Gellynck, X., & Polet, H. (2011). Policy instruments to meet fisheries
management objectives in Belgian fisheries. Fisheries Research, 111(1–2), 8-23.
Sumaila, U. R., & Hannesson, R. (2010). Maximum economic yield in crisis? Fish and
Fisheries, 11(4), 461-465. doi:10.1111/j.1467-2979.2010.00381.x
Sumaila, U. R., & Munro, G. R. (2009). Fisheries Economics. In J. H. Steele (Ed.),
Encyclopedia of Ocean Sciences (Second Edition) (pp. 491-498). Oxford: Academic
Press.

References 196
Susilowati, I., Bartoo, N., Omar, I. H., Jeon, Y., Kuperan, K., Squires, D., & Vestergaard, N.
(2005). Productive efficiency, property rights, and sustainable renewable resource
development in the mini-purse seine fishery of the Java Sea. Environment and
Development Economics, 10(6), 837-859.
Symes, D., & Phillipson, J. (2009). Whatever became of social objectives in fisheries policy?
Fisheries Research, 95(1), 1-5. doi:http://dx.doi.org/10.1016/j.fishres.2008.08.001
Thanh Long, N. (2015). Analysis of financial and technical performance of offshore trawlers
(engine >90 CV) in Ben Tre Province. Can Tho University Journal of Science, 38(1),
88-94.
The European Parliament and The Council of The European Union. (2013). Regulation (EU)
No 1380/2013 of The European Parliament and The Council of The European Union
on The Common Fisheries Policy. Strasbourg, France
The Prime Minister. (2010). The Decree No. 33/2010/ND-CP dated on 31 March 2010 on
management of aquatic resource exploitation in Vietnam’s sea areas. Hanoi, Vietnam
The Prime Minister. (2013a). The Decision No. 375/QĐ-TTg dated on 01 March 2013 on the
Plan to restructure of marine capture fisheries in Vietnam. Hanoi, Vietnam
The Prime Minister. (2013b). The Decision No. 1445/QĐ-TTg dated on 16 August 2013 on the
master plan for fisheries development to 2020 and vision to 2030. Hanoi, Vietnam
The Prime Minister. (2014). The Decree No. 67/2014/NĐ/CP dated on 7 July 2014 by
Vietnam's Prime Minister on some fisheries developement policies. Hanoi, Vietnam
Thiam, A., Bravo-Ureta, B. E., & Rivas, T. E. (2001). Technical efficiency in developing
country agriculture: a meta-analysis. Agricultural Economics, 25(2-3), 235-243.
Thøgersen, T. T., & Pascoe, S. (2014). Combining performance measures to investigate
capacity changes in fisheries. Applied Economics, 46(1), 57-69.
Thompson, G. D. (1988). Choice of flexible functional forms: review and appraisal. Western
Journal of Agricultural Economics, 169-183.
Thong, N. B. (2013). A brief introduction to trawl fishery and management issues in Vietnam.
Regional expert workshop on tropical trawl fishery management. 30th September - 4th
October, 2013. Phuket, Thailand.
Thong, N. B. (2015). Use of trash and low value fish for aquaculture production in Vietnam:
issues linking wild harvest impacts and sustainable aquaculture production.
Sustainable Fisheries Partnership. Vietnam.
Thong, N. B. (2017). Study on trawl fishery socio-economics and supply chains in Kien Giang,
Viet Nam. In V. S. Siar, P. Suuronen, & R. Gregory (Eds.), Socio-economics of trawl
fisheries in Southeast Asia and Papua New Guinea. Rome, Italy: Food and Agricultural
Organization of the United Nations.
Thuy, P. T. T. (2013). Rent creation and distribution-Theory and the Vietnamese small scale
purse seine fishery. (PhD thesis), Norwegian College of Fishery Science - University
of Tromso, Norway.
Tidd, A. N., Reid, C., Pilling, G. M., & Harley, S. J. (2016). Estimating productivity, technical
and efficiency changes in the Western Pacific purse-seine fleets. ICES Journal of
Marine Science, 73(4), 1226-1234. doi:10.1093/icesjms/fsv262

References 197
Tingley, D., Pascoe, S., & Coglan, L. (2005). Factors affecting technical efficiency in fisheries:
stochastic production frontier versus data envelopment analysis approaches. Fisheries
Research, 73(3), 363-376.
Tingley, D., Pascoe, S., & Mardle, S. (2003). Estimating capacity utilisation in multi-purpose,
multi-métier fisheries. Fisheries Research, 63(1), 121-134.
Tinh, V. H., & Thanh, N. P. (2014). Research results on marine capture fisheries state - Ben
Tre province. Journal of Fisheries Science and Technology - Nha Trang University,
2014(2), 62-66.
Tomberlin, D., & Holloway, G. (2010). Bayesian hierarchical estimationof technical efficiency
in a fishery. Applied Economics Letters, 17(2), 201-204.
Torres-Irineo, E., Amandè, M. J., Gaertner, D., de Molina, A. D., Murua, H., Chavance, P., . .
. Lezama-Ochoa, N. (2014). Bycatch species composition over time by tuna purse-seine
fishery in the eastern tropical Atlantic Ocean. Biodiversity and Conservation, 23(5),
1157-1173. doi:10.1007/s10531-014-0655-0
Truong, N. X., Terje, V., Ngoc, Q. T. K., Kim Anh, T. N., & Thanh Thuy, P. T. (2011).
Technical Efficiency of Gillnet Fishery in Da Nang, Vietnam: Application of stochastic
production frontier. Fish for the people, 1(9), 26-39.
Tzanatos, E., Dimitriou, E., Papaharisis, L., Roussi, A., Somarakis, S., & Koutsikopoulos, C.
(2006). Principal socio-economic characteristics of the Greek small-scale coastal
fishermen. Ocean & Coastal Management, 49(7–8), 511-527.
Urquhart, J., Acott, T., & Zhao, M. (2013). Introduction: Social and cultural impacts of marine
fisheries. Marine Policy, 37(1), 1-2. doi:10.1016/j.marpol.2012.04.007
Valdmanis, V. (1992). Sensitivity analysis for DEA models: An empirical example using
public vs. NFP hospitals. Journal of Public Economics, 48(2), 185-205.
Van Nguyen, Q., Pascoe, S., & Coglan, L. (2019). Implications of regional economic
conditions on the distribution of technical efficiency: Examples from coastal trawl
vessels in Vietnam. Marine Policy, 102, 51-60.
Van Phuong, T., & Duc Phu, T. (2013). Managing overcapacity of small-scale fisheries in
Vietnam. In: Secretariat, Southeast Asian Fisheries Development Center.
Vance, D., Staples, D., & Kerr, J. (1985). Factors affecting year-to-year variation in the catch
of banana prawns (Penaeus merguiensis) in the Gulf of Carpentaria, Australia. ICES
Journal of Marine Science, 42(1), 83-97.
Varabyova, Y., & Müller, J.-M. (2016). The efficiency of health care production in OECD
countries: A systematic review and meta-analysis of cross-country comparisons. Health
Policy, 120(3), 252-263.
Venables, B., Hutton, T., Lawrence, E., Rothlisberg, P., Buckworth, R., Hartcher, M., &
Kenyon, R. (2011). Prediction of common banana prawn potential catch in Australia's
Northern Prawn Fishery. Prepared for Australian Fisheries Management Authority.
Brisbane, Australia: CSIRO.
Vestergaard, N., Squires, D., & Kirkley, J. (2003). Measuring capacity and capacity utilization
in fisheries: the case of the Danish Gill-net fleet. Fisheries Research, 60(2–3), 357-368.
Vinuya, F. D. (2010). Technical efficiency of shrimp fishery in South Carolina, USA. Applied
Economics Letters, 17(1), 1.

References 198
Viswanathan, K. K., Omar, I. H., Jeon, Y., Kirkley, J., Squires, D., & Susilowati, I. (2001).
Fishing Skill in Developing Country Fisheries: The Kedah, Malaysia Trawl Fishery.
Marine Resource Economics, 16(4), 293-314.
Walden, J. B. (2006). Estimating Vessel Efficiency Using a Bootstrapped Data Envelopment
Analysis Model. Marine Resource Economics, 21(2), 181-192.
Walden, J. B., Fissel, B., Squires, D., & Vestergaard, N. (2015). Productivity change in
commercial fisheries: An introduction to the special issue. Marine Policy, 62, 289-293.
Walden, J. B., Kirkley, J., Färe, R., & Logan, P. (2012). Productivity Change under an
Individual Transferable Quota Management System. American Journal of Agricultural
Economics, 94(4), 913-928.
Waldo, S., Jensen, F., Nielsen, M., Ellefsen, H., Hallgrimsson, J., Hammarlund, C., . . . Isaksen,
J. (2016). Regulating Multiple Externalities: The Case of Nordic Fisheries. Marine
Resource Economics, 31(2), 233-257. doi:10.1086/685286
Walker, N. D., Maxwell, D. L., Le Quesne, W. J., & Jennings, S. (2017). Estimating efficiency
of survey and commercial trawl gears from comparisons of catch-ratios. ICES Journal
of Marine Science, 74(5), 1448-1457.
Ward, J., Kirkley, J., Metzner, R., & Pascoe, S. (2004). Measuring and Assessing Capacity in
Fisheries: 1. Basic concepts and management options (9251051437). Retrieved from
Rome:
Wattage, P., Mardle, S., & Pascoe, S. (2005). Evaluation of the importance of fisheries
management objectives using choice-experiments. Ecological Economics, 55(1), 85-
95.
Weninger, Q., & McConnell, K. E. (2000). Buyback programs in commercial fisheries:
effciency versus transfers. Canadian Journal of Economics, 33(2), 394-412.
Wilson, D. C., & McCay, B. J. (2001). Fishery Management, Human Dimension. In J. H. Steele
(Ed.), Encyclopedia of Ocean Sciences (Second Edition) (pp. 522-527). Oxford:
Academic Press.
World Bank. (2018). Data for High income, Low income, Lower middle income, Upper middle
income. Washington DC. Retrieved from
https://data.worldbank.org/country/australia?view=chart (15 July 2018).
Wu, J., Wang, N., Hu, Z.-H., Hong, Z., & Wang, Y.-G. (2019). Incorporating Social Objectives
in Evaluating Sustainable Fisheries Harvest Strategy. Environmental Modeling &
Assessment. doi:10.1007/s10666-019-9651-9
Yane, S., & Berg, S. (2013). Sensitivity analysis of efficiency rankings to distributional
assumptions: applications to Japanese water utilities. Applied Economics, 45(17), 2337-
2348.
Yang, C., Lou, X., Matsui, T., & Zhang, J. (2016). Evaluating the technical efficiencies of
fishing vessels to achieve effective management of overexploited fisheries. Mitigation
and Adaptation Strategies for Global Change, 1-14.
Ye, Y., Cochrane, K., Bianchi, G., Willmann, R., Majkowski, J., Tandstad, M., & Carocci, F.
(2013). Rebuilding global fisheries: the World Summit Goal, costs and benefits. Fish
and Fisheries, 14(2), 174-185. doi:10.1111/j.1467-2979.2012.00460.x

References 199
Zhou, S., Buckworth, R. C., Ellis, N., Deng, R., & Pascoe, S. (2015). Getting all information
out of logbooks: estimating banana prawn fishable biomass, catchability, and fishing
power increase, with a focus on natural mortality. ICES Journal of Marine Science,
72(1), 54-61.

Zhou, S., Dichmont, C., Burridge, C. Y., Venables, W. N., Toscas, P. J., & Vance, D. (2007).
Is catchability density-dependent for schooling prawns? Fisheries Research, 85(1), 23-
36. doi:https://doi.org/10.1016/j.fishres.2006.11.034
Zhou, S., Pascoe, S., Dowling, N., Haddon, M., Klaer, N., Larcombe, J., . . . Wayte, S. (2010).
Quantitatively defining biological and economic reference points in data poor fisheries.
Final Report on FRDC Project 2010/044. CSIRO Marine and Atmospheric Research,
Canberra, Australia, 44.
Zhu, S., Ellinger, P. N., & Shumway, C. R. (1995). The choice of functional form and
estimation of banking inefficiency. Applied Economics Letters, 2(10), 375-379.
Zijlstra, J. J., & de Veen, J. F. (1963). On the Relative Fishing Power of Dutch Trawlers. ICES
Journal of Marine Science, 28(2), 241-245. doi:10.1093/icesjms/28.2.241

References 200
Appendices

Appendix 1: Human research Ethics approval certificate

Appendices 201
Appendix 2: Questionnaire for trawl fishery survey in Vietnam
Research title:
Technical efficiency in Vietnamese trawl fisheries
Individual information in this survey (i.e., respondent, vessels owner, and skipper/operator) will be treated as
strictly confidential. No individual or business will be able to be identified from the results in any reports.
This questionnaire is applied to collect information about fishing activities (e.g., catch, revenue,
and production cost and so on) and related information that took place in the year 2015 (From
1th January 2015 to 31th December 2015).

Questionnaire number (ID) ..........................................................................................................

Date of survey ..............................................................................................................................
Name of fishing port ...................................................................................................................
Date of inputting data...................................................................................................................
1. Vessel owner information

1.1 Age of vessel owner (years)?

1.2 What is the gender of vessel □ Male
owner? □ Female
1.3 What is your ethnicity? □ Kinh
□ Other (please specify)
___________________

1.4 Is vessel owner household □ Yes

head? □ No
1.5 Total years of experience in
fishing (years)?
1.6 Number of years as vessel
owner (years)?
1.7 Vessel owner’s educational □ None
level? □ Primary school
□ Secondary school
□ High school
□ Vocational educational training
□ Higher level (specify)
___________________________

Appendices 202
 1.8 Vessel owner’s father is/was fishermen?
□ Yes
□ No
□ Not known
Vessel owner’s grandfather/previous generation is/was fishermen?
□ Yes
□ No
□ Not known
Which generation are you (if known)?__________
1.9 Household size of vessel owner (people)? (Including vessel owner’s spouse, and
their dependent children)____________________________________
In which (specify the number of members based on gender and age as follows):

Age (years) Under 15 From 15 to 60 (Male) Above 60 (Male)

From 15 to 55 (Female) Above 55 (Female)
Male __________ ______________ _____________
Female __________ ______________ _____________
1.10 Number of dependent children □ In total……………………..
who are schooling (children)? In which the number of children are schooling at:
□ Primary ……………………..
□ Secondary …………………..
□ High school…………………
□ Other (University/college/vocational
training school)………………
1.11 Number of household members □ In total……………………….
involved in fishing (people)? □ Number of male ……………...
□ Number of female……………
1.12 How many fishing vessels do you have currently (vessels)? .................................
(If vessel owner has more than one fishing vessel, please specify as follows :)
□ The number of trawl fishing vessels:__________________________
□ The number of other fishing vessels (specify type of fishing vessels e.g.,
gillnet, longline):_______________

2. Vessel characteristics (Trawl fishing vessels only, and if vessel owner has more than one
trawl fishing vessel, separate questionnaire is applied for each trawl fishing vessel)

2.1. Hull’s length (m)? ................................................................................................................

2.2. Hull’s width (m)? .................................................................................................................
2.3. Hull’s material? (Please tick one box only)
□ Wooden hull
□ Steel hull
□ Composite hull
□ Other material

Appendices 203
2.4. Total engine power (HP or KW, please specify)? ................................................................
In which:
□ Main engine (HP or KW) .................................................................................................
□ Minor engine (HP or KW) ...............................................................................................
2.5. Gross registered tonne – GRT (tons) ....................................................................................
2.6. Vessel information?

Price (‘000 VND)

Curre
nt
marke Estimated
Estimat
Condition at t price current
Purchase ed
purchase Purchas for the price of
Fishing year lifespan
(please tick) ed price same the current
vessel (years)
purch vessel if
ased sold
vessel

□ New
□ Old
Engine □ New
□ Old
□ Japan
□ Others
Fishing gear __________ □ New
□ Old

3. Skipper and crew information

Skipper (skip these questions if vessel owner is skipper)

3.1 Vessel owner □ Yes
is skipper? □ No
3.2 Skipper □ None
educational □ Primary school
level? □ Secondary school
□ High school
□ Vocational educational training
□ Higher level (specify____________________

3.3 Age of skipper (years)? ____________________________

3.4 Total fishing years’ experience (years)? _________________
3.5 Total years as a skipper (years)?
____________________________________

Appendices 204
 3.6 Does skipper □ Yes
attend training □ No
courses for
skipper?
3.7 Other training □ No
courses or □ Yes (specify)
qualifications ____________________________________________
obtained
(specify)?
3.8 Skipper’s father is/was fishermen or skipper?
□ Yes
□ No
□ Not known
Skipper’s grandfather is/was fishermen or skipper?
□ Yes
□ No
□ Not known
Which generation are you (if known)?__________
3.9 What is □ Kinh
skipper’s □ Other (please specify)
ethnicity? ___________________
3.10 Is skipper a □ Yes
relative of □ No
vessel owner?
3.11 What type of □ Fixed-wage contract (‘000VND/month):
remuneration _____________________________________
does the □ Shared-revenue contract (%):
skipper _____________________________________
receive? □ Other, please specify
_____________________________________
Crew information (Asking only information for year 2015)
3.12 Crew size per □ On average_______________
trip (skipper □ Maximum________________
excluded)? □ Minimum_________________
How many crewmembers are vessel owner’s relative?
_____________________________
3.13 Average income for crew per month (‘000VND)?
________________________
3.14 Average annual income of crew (‘000VND)?
_________________________

Appendices 205
 3.15 What type of □ Fixed-wage contract (‘000VND/month):
remuneration _____________________________________
does crew □ Shared-revenue contract (%):
receive _____________________________________
(owner, □ Other, please specify
skipper and _____________________________________
mechanic
chief
excluded)?
3.16 Does crew □ Yes (if yes, move to the question 3.18)
have specialist □ No
mechanic?
3.17 What type of □ Fixed-wage contract (‘000VND/month):
remuneration _____________________________________
does mechanic □ Shared-revenue contract (%):
chief receive? _____________________________________
□ Other, please
specify_________________________________

4. Fishing trip and fishing ground (Asking information for year 2015)

4.1. Number of fishing trip per year on average (Trip)? ..............................................................

4.2. Total fishing months per year (Month)? ..............................................................................
4.3. Average days per trip? (Day) ................................................................................................
4.4. Number of days or hours approaching fishing ground and returning fishing port (specify
day or hour)? ................................................................................................................................
4.5. Operating distance from homeport on average (Nautical mile)? ..........................................
4.6. When is unfavorable weather for fishing? From month…………………….to month ........
(Specify period of time in year)
5. Information about total catch and revenue for year 2015 (Asking information for year
2015)
5.1. What is your total revenue from fishing activities for whole year 2015? (Only revenue from
the trawling vessel that has been asked about vessel characteristics above) (‘000 VND)
______________
5.2. Information about total catch per trip:
Information Total catch Total estimated Average price
(kg) percentage of catch (‘000 VND/kg)
(%)
Total prawn
Specie 1
Specie 2
Specie 3

Appendices 206
 Fish
Low value fish species
Squid
Crab
Other species
Total
5.3. Who do you mainly sell your catch to? (Please tick X and give estimated share where
appropriate)
□ Local market (%): ____________
□ Middlemen (%): ____________
□ Family consumption (%): ____________

5.4. Do you think your catch have increased, decreased or stayed the same in comparison with
4 or 5 years ago?(Please tick one box only, and give estimated percentage if possible)

[ ] Strongly decreased (%):_________ [ ] Strongly increased (%):_______

[ ] Decreased (%):__________________ [ ] Increased (%):________

[ ] Stay the same [ ] No idea

6: Information about average operating cost per trip (Asking information for year 2015)

Items
No Information Price Total expense
Quantity
(‘000 VND) (‘000 VND)

6.1 Diesel (litre)

6.2 Lubricant (litre)
6.3 Ice (kg)
Provisions, grocery expenses
6.4 and miscellaneous supplies for
crew (‘000 VND)
Total minor repair, running
cost and maintenance cost per
6.5
trip (gear fix, vessel repair,
cleaning etc.) (‘000 VND)
6.6 Other costs (‘000 VND)
Total operating cost
7. Information about onboard equipment (Please tick)

Has the equipment been used on

No Items
the fishing boat in year 2015?
Appendices 207
 □ No
7.1 Preservation equipment (i.e., ice
□ Yes
container, cool room)
□ No
7.2 Generators
□ Yes
□ No
7.3 Global Positioning System (GPS)
□ Yes
□ No
7.4 Echo sounder
□ Yes
□ No
7.5 Radio
□ Yes
□ No
7.6 Walkie-talkie
□ Yes
□ No
7.7 Safety gear
□ Yes
8. Annual insurance costs (Asking only information for year 2015)

No Item Expenses (‘000 VND)

8.1 Hull insurance
8.2 Engine insurance
8.3 Fishing gear insurance
8.4 Crew insurance
Total
9. Other annual costs (Asking only information for year 2015)

No Item Expenses (‘000 VND)

9.1 Annually registered fee, and harvesting
license fee
9.2 Annual interest payment
9.3 Major repair expenses (hull, engine and
gear repair)
9.4 Other administration costs
Total
10. Annual income of household
10.1. Share of income from fishing in comparison with total income in year 2015 (% or ‘000
VND)? ..........................................................................................................................................
Other non-fishing livelihood activities:
Agriculture activities ________ Trading, self-employment __________
Aquaculture ________ Others _____________
10. 2. In comparison with previous years (4 or 5 years ago), how does income from fishing
activities change (Please tick one box only, and give estimated percentage if possible)?

Appendices 208
 [ ] Strongly decreased (%):_______ [ ] Strongly increased (%):______

[ ] Decreased (%):_________ [ ] Increased (%):_______

[ ] Stay the same [ ] No idea

10.3. How satisfactions have you been with your income from fishing? (Please tick one box
only)
Very Neither Very
Somewhat Somewhat
satisfied or No idea
dissatisfied dissatisfied satisfied satisfied
dissatisfied
[ ] [ ] [ ] [ ] [ ] [ ]
10.4. What will you do with fishing activities in the future?

[ ] Keep the recent status [ ] Decrease

[ ] Expand fishing activities [ ] Stop fishing

(i.e., more investment) [ ] No idea

10.5. If you decide to stop fishing and quit fishing job (e.g., switching other job as construction
worker, agricultural production, and middlemen), what are the main reasons?
(Many choices if possible, please ticks X)
[ ] Decreased resource [ ] Lack of capital [ ] Lack of labour
[ ] Low income [ ] Lack of technology [ ] Others……………
[ ] Risky occupation (i.e.,
accident)
10.6. Any fishing related organizations you are membership (e.g., fishing cooperative
organization, agricultural and fisheries extension organization)?
[ ] Fishing cooperative [ ] Agricultural and fisheries extension organization

[ ] Farmer association [ ] None of them

10.7. Do you think these associations are beneficial for your fishing activities (Please tick one
box only)?

[ ] Strongly agree [ ] Acceptable [ ] Strongly disagree

[ ] Agree [ ] Disagree

10.8. How do you think the fishery should be managed? (Please rank these following options
based on the level of importance, 1: the most important objective, 2: the second most important,
3: the third most important and 4: the fourth most important)
[ ] Ensuring lots of jobs in fishery [ ] Maximizing incomes from fishing

Appendices 209
 [ ] Ensuring fish stock for fishery [ ] Other objectives (i.e., reducing conflicts
between fishing communities, border
security and sovereignty)
10.9. Any recommendations or suggestion to fisheries manager or policy makers about
fisheries management in Vietnam recently?
......................................................................................................................................................
......................................................................................................................................................
THANK YOU FOR YOUR COOPERATION

Research findings
(A summary report of research findings will be provided to fishers if they would like to have
it).
Would you like to receive a summary report of research findings through email address?
□ No □ Yes
Please give us email address? _______________________________

Appendices 210
Appendix 3: Fishing zone in Vietnam’s sea

Backward Coastal and inshore

areas

Forward
Offshore and high
sea areas

Source: The Decree No. 33/2010/ND-CP dated on 31 March 2010 on management of aquatic
resource exploitation in Vietnam’s sea areas (The Prime Minister, 2010).

Appendices 211
 Appendix 4: List of primary studies used in meta-analysis
No. First Published Country Fisheries/Fishing fleet Management Estimation Journal
papers author year tools approach

1 Kirkley 1995 USA Mid-Atlantic sea scallop fishery IP SFA American Journal of Agricultural Economics

2 Kirkley 1998 USA Mid-Atlantic sea scallop fishery IP SFA Journal of Productivity Analysis

3 Sharma 1998 USA Longline fishery in Hawaii IP SFA Marine resource economics

4 Eggert 2001 Sweden Swedish demersal trawlers IP SFA Working paper

5 Pascoe 2001 UK The United Kingdom vessel IP DEA ICES Journal of Marine Science

6 Viswanathan 2001 Malaysia Trawl fishery in Kedah, Malaysia OA SFA Marine resource economics

7 Andersen 2002a Denmark Danish Baltic Sea trawlers IP SFA Working paper

8 Andersen 2002b Denmark Danish seiner fleet IP SFA Working paper

9 Felthoven 2002 USA Bering sea and Aleutian Island Pollock IP DEA, SFA Marine resource economics
fishery
10 Fousekis 2002 Greece Inshore fishery in Greece IP SDF Marine resource economics

11 Vestergaard 2002 Denmark Danish trawl fisheries in the North Sea IP SFA Working paper

12 Fousekis 2003 Greece Trammel netters in Greece IP SFA Fisheries Research

13 Herrero 2003 Spain Spanish South-Atlantic trawl fishery IP SFA Journal of Agricultural Economics

14 Squires 2003 Malaysia Artisanal gillnet fishery in Malaysia OA SFA Environment and Development Economics

15 Weninger 2003 USA Northern Gulf of Mexico reef fish IP DEA Journal of Environmental Economics and
fishery Management
16 Baskaran 2004 USA The Atlantic herring fishery fishing IP SFA PhD thesis
fleets
17 García del 2004 Spain Purse-seine fishery IP SFA ICES Journal of Marine Science
Hoyo
18 Herrero 2004 Spain Sea bream and tuna fisheries in the IP DEA ICES Journal of Marine Science
Gibraltar
19 Kirkley 2004 USA Northwest Atlantic scallop fishery IP DEA, SFA Marine resource economics

Appendices 212
 No. First Published Country Fisheries/Fishing fleet Management Estimation Journal
papers author year tools approach
20 Kompas 2004 Australia Northern Prawn fishery IP SFA Applied Economics

21 Sharp 2004 New Rock lobster fishery ITQ SFA New Zealand Economic Papers
Zealand
22 Signorello 2004 Italy Fisheries in Southern Ionian Sea IP DEA Italian Review of Agricultural Economics

23 Andersen 2005 Denmark Danish Seiners IP DEA Marine resource economics

24 Espino 2005 Spain The Strait of Gibraltar red seabream IP DEA Marine resource economics
fishery
25 Herrero 2005 Spain The Spanish Trawl fleet IP DEA, European Journal of Operational Research
SFA, SDF
26 Orea 2005 Spain Northern Spain hake fishery IP SFA, SDF Natural Resource Modeling

27 Susilowati 2005 Indonesia The Java Sea mini-purse seine fishery OA SFA Environment and Development Economics

28 Tingley 2005 UK English Channel fishing boats IP DEA, SFA Fisheries Research

29 Kompas 2005 Australia Australian South East fishery ITQ SFA Journal of Productivity Analysis

30 Esmaeili 2006 Iran Iranian Persian Gulf fishing industry IP SFA ICES Journal of Marine Science

31 Herrero 2006 Spain The Spanish–Moroccan deep trawl IP DEA Journal of Productivity Analysis
fishery
32 Staffan 2006 Sweden Swedish Pelagic Fisheries IP DEA Working paper

33 Walden 2006 USA Sea scallops in the mid-Atlantic IP DEA Marine resource economics

34 Greenville 2006 Australia The NSW Ocean prawn trawl fishery IP SFA Marine resource economics

35 Jeon 2006 Indonesia The Java Sea purse seine fishery OA SFA Applied Economics

36 Brandt 2007 USA Mid-Atlantic surf clam fishery IP, ITQ SFA Journal of Economic Behavior &
Organization
37 Esmaeili 2007 Iran Fishery in Hamoon Lake IP DEA Journal of Applied Science

38 Lindebo 2007 Denmark Danish Norht Sea trawlers IP DEA European Journal of Operational Research

39 Pascoe 2007 UK Scottish fleet IP DEA, SFA Book chapter

Appendices 213
 No. First Published Country Fisheries/Fishing fleet Management Estimation Journal
papers author year tools approach
40 Pascoe 2007 UK The UK demersal trawl fleet IP DEA Applied Economics

41 Sesabo 2007 Tanzania Small-scale fishing households in OA SFA African Journal of Aquatic Science
coastal villages
42 Akanni 2008 Nigeria Artisanal fishery in Nigeria OA SFA Tropical Agriculture

43 Innes 2008 UK The North Sea brown fishery IP SFA Journal of Agricultural Economics

44 Maravelias 2008 Greece Purse-seine fleet in Greece IP DEA Fisheries research

45 Reddy 2008 Fiji Artisanal fisheries in Fiji OA SFA World Review of Entrepreneurship,
Management and Sustainable Development
46 Lokina 2009 Tanzania Artisanal Lake Victoria fisheries OA SFA Environment and Development Economics

47 Muldoon 2009 Australia The Great Barrier Reef and the Reef IP DEA PhD thesis
line fishery
48 Quach 2009 Vietnam Small-scale trawlers in the Central OA SFA Book chapter.
Vietnam
49 Idda 2009 Italy Small-scale fisheries the coastal waters IP DEA Marine policy

50 Akanni 2010 Nigeria Small-Scale Fishers in Lagos State OA SFA North American Journal of Fisheries
Management
51 D.H. Kim 2010 Korea Coastal composite fishery in Korea IP DEA, SFA The Journal of Fisheries Business
Administration
52 Dresdner 2010 Chile Pelagic fisheries in Central-Southern IP, ITQ SFA Environment and Development
Chile Economics
53 Holzer 2010 USA The New England Multispecies Fishery IP SDF PhD thesis

54 Chowdhury 2010 Banglades Industrial trawl fishery in Bangladesh IP SFA Economics Bulletin
h
55 Pascoe 2010 Australia Northern Prawn fishery IP SDF European review of Agricultural
Economics
56 Tomberlin 2010 USA The US hake fishery IP SFA Applied Economics Letters

57 Vinuya 2010 USA Shrimp fishery in South Carolina IP SFA Applied Economics Letters

58 Alencastro 2010 Ecuador The artisanal lobster fishing fleet in the IP SFA PhD thesis
Galapagos Marine Reserve

Appendices 214
 No. First Published Country Fisheries/Fishing fleet Management Estimation Journal
papers author year tools approach
59 Oliveira 2010 Portugal Artisanal dredge fleet IP DEA ICES Journal of Marine Science

60 Castillo 2011 Mexico Small-scale fisheries OA SFA PhD thesis

61 D.H. Kim 2011 Korea Sandfish coastal gillnet fishery IP SFA Fisheries Science

62 Truong 2011 Vietnam Gillnet fishery in Vietnam OA SFA Fish for the people

63 D. H. Kim 2012 Korea Korean offshore fisheries IP DEA Fisheries Science

64 Latif 2012 Malaysia Trawl fishing vessels in Penang OA DEA, SFA Journal of Applied Science

65 Khan 2012 Banglades Artisanal gillnet fisheries in the Bay of OA DEA PhD thesis
h Bengal
66 Latiff 2012 Malaysia Inshore Fisheries in Kuala Terengganu OA DEA International Journal of Agricultural
Management and Development
67 New 2012 Australia Eastern Tuna and Billfish fishery ITQ SFA Conference paper

68 Kumbhakar 2013 Norway Norwegian fishing trawlers ITQ SDF Journal of Productivity Analysis

69 Otumawu 2013 Australia The South Australian Rock Lobster IP, ITQ DEA PhD thesis
fishery
70 Pascoe 2013 Australia The Torres Strait tropical rock lobster IP SFA Canadian Journal of Agricultural
fishery Economics
71 Schnier 2013 USA Bering Sea and Aleutian Island crab IP SFA Land Economics
fisheries
72 Vázquez- 2013 USA Industrial purse seine fishing vessels IP DEA Marine Policy
Rowe
73 Perret 2014 Indonesia Small-scale fisheries in Singkarak OA DEA Regional Environmental Change
Lake, West Sumatra
74 Burns 2014 USA Mid-Atlantic surf clam fishery ITQ SFA PhD thesis

75 Castilla- 2014 Georgia The southeastern Black Sea anchovy IP DEA Journal of Marine Systems
Espino fishery
76 Ceyhan 2014 Turkey Commercial fishing from the Samsun IP DEA Turkish Journal of Fisheries
province and Aquatic Sciences
77 Collier 2014 USA West Coast Limited Entry Groudfish IP SFA Omega
Trawl Fishery

Appendices 215
 No. First Published Country Fisheries/Fishing fleet Management Estimation Journal
papers author year tools approach
78 Gbigbi 2014 Nigeria Artisanal fishery in the Niger-Delta OA SFA International Journal of Fisheries
and Aquatic Studies
79 Otoo 2014 Ghana Artisanal fishery in Ghana OA SFA Conference paper

80 Solis 2014 USA The Gulf of Mexico red snapper IP, ITQ SDF Food policy
commercial fleet
81 Thøgersen 2014 Denmark Danish North Sea demersal trawlers IP DEA, Applied Economics
SDF
82 Pham 2014 Vietnam Gillnet fishery in Vietnam OA DEA Marine policy

83 Chandrasekar 2015 India Ring Seiners in Kerala OA DEA Agricultural Economics Research Review

84 Fernando 2015 Sri Lanka Small-Scale Marine Fishing OA SFA International Journal of Economic Practices
Households and Theories
85 Solís 2015 USA The Gulf of Mexico red snapper IP, ITQ SDF Marine policy
fishery
86 Wategire 2015 Nigeria Small-scale shrimp fishers in Delta OA SFA Mediterranean Journal of Social Sciences
state
87 Fissel 2015 USA Alaska head and gut trawl fleet ITQ SFA Marine policy

88 Jamnia 2015 Iran Inshore and offshore fishing vessels IP SFA Journal of the Saudi Society of Agricultural
Sciences
89 Rahman 2016 Malaysia Inshore Fishery in Terengganu IP DEA Conference Proceedings

90 Duy 2016 Vietnam Offshore gillnet fishery in Vietnam OA SFA Fisheries Research

91 Gigentika 2016 Indonesia Tuna fisheries in Kupang OA DEA AACL Bioflux

92 Sowunmi 2016 Nigeria Artisanal fishermen in Lagos state OA SFA Environmental Monitoring and Assessment

93 Yang 2016 Japan Pacific saury fishery in Japan IP SFA Mitigation and Adaptation Strategies
for Global Change
94 Pinello 2016 Greece Small-scale and bottom trawler vessels IP DEA Sustainability
in Greece
95 Larryn 2017 Philippines Municipal fisherfolk in Philippine OA SFA Asian Fisheries Science

96 Mohammad 2017 Indonesia Mini purse seine fishing households OA SFA Habitat

Appendices 216
 No. First Published Country Fisheries/Fishing fleet Management Estimation Journal
papers author year tools approach
97 Pascoe 2017a Australia The Northern Prawn Fishery IP SFA Australian Journal of Agricultural and
Resource Economics
98 Pascoe 2017b Australia Moreton Bay prawn trawl fishery IP SFA Fisheries research

99 Rust 2017 Australia Tasmanian Rock lobster fishery ITQ DEA Marine policy

100 Quijano 2018 Mexico Semi-industrial fishing fleet of Yucatan IP SFA Marine Policy

101 Quynh 2018 Vietnam Small-scale fisheries in the Central IP DEA Fisheries research
Vietnam
102 Madau 2018 Italy The Sardinian small-scale fisheries IP DEA Marine Policy

103 Yamazaki 2018 Indonesia Small-Island Fisheries in Indonesia OA DEA Ecological economics

Appendices 217
 Appendix 5: Net economic return of the NPF from 2000/01 to 2014/15

Year 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15
Catch (tons)
Tiger Prawns 2 116 1 958 1 969 2 186 1 785 1 749 1 834 1 235 1 024 1 274 1 627 864 1 340 2 025 1 761
Banana Prawns 6 286 5 419 3 325 3 572 2 827 3 247 2 674 5 380 5 214 5 771 7 577 4 855 2 990 5 780 4 565
Endeavour Prawns 868 1 132 395 418 412 281 355 201 218 355 426 498 500 497 692
King Prawns 7 2 3 1 2 19 28 20 7 6 10 8 15 21 13
Other Prawns 6 7 6 15 9 8 1 1 2 1 0 2 2 5 14
Other Species 469 224 62 85 89 95 239 67 64 58 33 77 153 108 100
Total (tons) 9 752 8 742 5 761 6 277 5 124 5 400 5 131 6 903 6 529 7 465 9 673 6 304 4 999 8 436 7 145
a
GVP ($million) 239.0 190.0 113.1 99.0 84.9 92.2 78.4 88.5 85.3 100.1 103.7 69.1 74.2 117.2 106.8
NER ($million)a 83.1 43.2 28.6 11.7 -16.7 -10.9 -3.5 9.3 5.5 13.2 6.4 -4.0 5.1 12.2 12.3
Number of vessels 118 118 101 95 96 86 77 54 55 54 54 55 54 52 55
a
: 2014–15 dollars

Appendices 218