How to reduce mangrove forest extraction in the mekong river delta of vietnam a common pool resource experiment

PROGRAM IN DEVELOPMENT ECONOMICS HOW TO REDUCE MANGROVE FOREST EXTRACTION IN THE MEKONG RIVER DELTA OF VIETNAM?. UNIVERSITY OF ECONOMIC INSTITUTE OF SOCIAL STUDIES VIETNAM – NETHERLAN

UNIVERSITY OF ECONOMICS, HO CHI MINH CITY VIET NAM – NETHERLANDS PROJECT FOR M.A PROGRAM IN

DEVELOPMENT ECONOMICS

HOW TO REDUCE MANGROVE FOREST EXTRACTION

IN THE MEKONG RIVER DELTA OF VIETNAM?

A COMMON POOL RESOURCE EXPERIMENT

-o0o -

A thesis submitted in partial fulfillment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

By

VU QUANG VINH

Academic Supervisor:

TRUONG DANG THUY

HO CHI MINH CITY, DECEMBER 2015

UNIVERSITY OF ECONOMIC INSTITUTE OF SOCIAL STUDIES

VIETNAM – NETHERLANDS

PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

HOW TO REDUCE MANGROVE FOREST EXTRACTION

IN THE MEKONG RIVER DELTA OF VIETNAM?

A COMMON POOL RESOURCE EXPERIMENT

A thesis submitted in partial fulfillment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

By

VU QUANG VINH

Academic Supervisor:

TRUONG DANG THUY

HO CHI MINH CITY, DECEMBER 2015

DECLARATION

“This is to certify that this thesis entitled “How to reduce mangrove forest extraction in the Mekong River Delta of Vietnam? A Common Pool Resource Experiment”, which is submitted by me in fulfillment of the requirements for the degree of Master of Art in Development Economics to the Vietnam – The Netherlands Programme (VNP)

The thesis constitutes only my original work and due supervision and acknowledgement have been made in the text to all materials used.”

Vu Quang Vinh

ACKNOWLEDGEMENTS

This thesis would not have been able to finish without the support and guidance from many people Firstly, I wish to express my deepest gratitude to my supervisor Truong Dang Thuy for the continuous support of my thesis, for his generosity and dedication in sharing his wisdom guiding

me to rethink and to deconstruct my thesis topic His excellent guidance helped me in all the time

of research and writing of this thesis He spent a lot of time to support me about the reference materials as well as give me advices about what should to do for my thesis Moreover, he took time to diligently review all my thesis drafts and help me correct errors and clear my thinking as well Moreover, the most wonderful thing is that he provided me an opportunity to join his team

as intern, let me experience the research of mangrove forest in the field and gave access to the experimental data and research facilities

Besides my supervisor, I would also like to thank all my friends here at VNP Together we have struggled through this whole treasured journey of learning and shared memorable moments

I thank my fellows for their insightful comments and the stimulating discussions, they were always willing to help and give me best suggestions Then, I express my gratitude to VNP officers as well

as VNP librarian for their support of comfort lab room and study materials and I want to say thanks

to EEPSEA for funding the data collection

Finally, I would like to thank my family for supporting me spiritually throughout writing this thesis and to my girl friend who always stood by me through the good times and difficult

ABBREVIATIONS

ASEAN Association of Southeast Asian Nations

CICES The Common International Classification of Ecosystem ServicesCPR Common Pool Resource

FAO Agriculture Organization of the United Nations

GIZ Deutsche Gesellschaft für Internationale Zusammenarbeit

ICZN Integrated Coastal Zone Management

ICZN/CCCEP Integrated Coastal Zone Management/ Climate Change and Coastal Ecosystems in Mekong Delta Vietnam Program

MARD Ministry of Agriculture and Rural Development

MONRE Ministry of Natural Resources and Environment

OLS Ordinary Least Square

PES Payments for Environmental Services

UNCED The United Nations Conference on Environment and Development

WCED World Council of Environmental and Development

ABSTRACT

Using data from series of experiment in six selected provinces in Southwestern Vietnam

in March and April 2015, this thesis aims to analyze the impact of mangrove forest policies and programs on people behavior towards exploring mangrove forest Common pool resource experiment is applied to examine the behavior of people in making decision To control for heterogeneity of individuals, I utilize the past behavior proposed by Frey and Meier (2004) There

is robust evidence that people will quickly change their behavior if their profitability is varied, however, the local authorities must control the unexpected influence in using financial reward In addition, it suggested that government should combine many measures to get more successful in inducing people to protect mangrove forest

Keywords: Mangrove forest, Common pool resource, Game theory, Experiment

TABLE OF CONTENTS

Chapter 1: Introduction 1

1.1 Problem statement 1

1.2 Research question 3

1.3 Scope OF STUDY 3

1.4 Methodology and data 4

1.5 The structure of the study 5

Chapter 2: Literature review 6

2.1 Mangrove’s ecological services 6

2.2 Mangrove forest and market failures 7

2.3 Literature review on the common pool resources experiments 9

2.4 Main hypotheses 15

Chapter 3: Methodology and data 19

3.1 Methodology 19

3.2.1 Sampling strategy 19

3.2.2 The interview 19

The structure of questionnaire 19

Procedures of interviewing 21

3.2.3 The experiment 21

Experimental design 21

Payoffs 23

Additional information 25

Procedures of conducting CPR game 25

3.2 Data 26

Chapter 4: Empirical results 29

4.1 Mangrove forest in vietnam 29

4.1.1 The current situation of mangrove forests in the Mekong River Delta 29

4.1.2 The situation of mangroves management in the Mekong River Delta 31

4.1.3 The current situation of other common pool resources in Vietnam 33

Fishing 33

Water 34

Forest 34

4.2 Non-parametric analysis 35

4.2.1 Descriptive statistics 35

4.2.2 Treatment 1: information 39

4.2.3 Treatment 2 and treatment 3: social comparison and profitability 39

4.3 Regression results 43

4.4 Discussion 49

4.4.1 Hypotheses 49

4.4.2 Other discussions and limitations 51

Chapter 5: Conclusions and policy implications 53

5.1 Conclusions 53

5.2 Policy implications 53

References 55

Appendix 60

LIST OF FIGURES

Figure 4-1: Chart of change in mangroves area of Vietnam from 1943 to 2013 29

Figure 4-2: ASEAN mangrove forest distribution map (NASA/USGS) 30

Figure 4-3: Chart of change in mangroves area of the South Vietnam from 1943 to 2008 31

Figure 4-4: The average extraction in 20 rounds 36

Figure 4-5: The average extraction with treatments distinction 37

Figure 4-6: Frequencies of extraction area in two stages by treatment groups 38

Figure 4-7: The average extraction in 20 rounds with treatment of information 39

Figure 4-8: The average extraction in 20 rounds with treatment of social comparison 40

Figure 4-9: The average extraction in 20 rounds with treatment of profitability 41

Figure 4-10: The average payoff and average extraction across social comparison treatment over 20 rounds 52

LIST OF TABLES

Table 2-1: Comparison between ecological wetland and ecological mangrove functions 6

Table 3-1: Description of some variables 27

Table 4-1: Summary about forest-holders and forest types 33

Table 4-2: The changes in forest area from 1945 to 2002 (unit: 1,000,000 hectares) 35

Table 4-3: The average extraction through treatments 38

Table 4-4: Summary of main characteristics 42

Table 4-5: Regressions for individual extracted area 43

Table 4-6: Regressions with three treatments 46

Table 4-7: The average payoff across treatments and stages (units: VND) 48

LIST OF APPENDIX Appendix 1: Introduction to game rules 60

Appendix 2: The payoff table 64

Appendix 3: Information sheet 69

Appendix 4: Consent form 72

Appendix 5: Questionnaire table 73

Appendix 6: Three types of mangrove forest and their main functions 117

Appendix 7: The situation of mangroves in some provinces in Mekong Delta region in 2012 118 Appendix 8: Global mangrove forest distribution map (NASA/USGS) 118

Appendix 9: The correlation matrix 119

CHAPTER 1: INTRODUCTION

1.1 PROBLEM STATEMENT

The concept of “sustainable development” appeared in the report “Our Common Future”

by the World Council of Environmental and Development (WCED) of United Nations in 1987

It is defined as “development that meets the needs of the present without compromising the ability of future generations to meet their own need.” The Earth Summit on Environment and Development (UNCED) in Rio de Janeiro in 1992 and the World Summit on Sustainable Development in Johannesburg in 2002 identified a “sustainable development” should satisfy economic development, social development and environmental protection Sustainable development is the inevitable trend in the evolution of human society In there, environmental protection is one of the imperative needs in the context of status of global climate changes Vietnam is one of countries that are greatly influenced by climate change; furthermore, the economic-society development still depends heavily on the exploitation of natural resources Therefore, protecting natural resources in general and common pool resources in particular are the priority issues have to be resolved thoroughly

Mangrove forest, which is one of common pool resources, provides material for healthy, living and manufacturing, energy source, protect people from storm, tsunami as well as contribute to mitigate the impacts of climate changes such as global warming In Vietnam, because of important role of mangrove forests in protecting seashore the government conducted many programs and policies to improve the situation of mangroves However, the obtained results are not satisfactory: during the early period (2000 – 2006), there is a significant increase

in area but in recent years, the mangrove forest area tends to decline (2008 – 2013) This result raises several questions about the efficiency as well as the rationality of related measures that government researched and deployed Moreover, the individuals and households who are subjects of these programs and policies are also considered as a factor in interpreting the results

In the mangrove forest area in Vietnam, the living conditions of most people are so difficult, and they often use backward methods to do for a living In addition, there is unreasonable in making practical and effective use of mangrove forest, the local communities extract firewood,

wood and seafood under the mangroves canopy but do not pay attention to restore the mangrove forest (Que and Hai 2012) Due to these reasons, the individuals and households are the weak links of a chain in executing mangroves conservation projects

When introducing new policies and regulations, according to conventional economic theories, policy makers assume that individuals will be motivated by their own interest without regard for others However, Cardenas, Stranlund and Willis (2000) found that people do not always comport themselves with this way but could between self and group interests They explained that the external regulations that aimed to control the local environment quality can

“crowd out other-regarding behavior” and hence people will become more self-interested in making decisions It means that the benefits from performing new regulations are unlike the initial expectations And the local regulations may be ineffective if “the cost of achieving acceptable levels of compliance” exceeds that benefits However, individuals respond not only the cost of performing new regulations but also other factors, i.e their wealth (Cardenas 2003), personal characteristics (Cárdenas and Ostrom 2004) or social context of the community

(Janssen et al 2013) In 2005, Cardenas conducted another experiment in Colombia with

villagers and students, in which the individuals were asked about their preferences toward the application of external regulations through voting mechanism Cardenas (2005) pointed out that there is a significant difference in behavior between group of villagers and group of student when an external regulation is imposed For this reason, the author suggested unless we replicate the experiment in the field after doing it in the laboratory, we would miss some related information Base on these suggestions and other relevant paper, the simultaneous interview is also needed Although there is some mangrove forest research in Vietnam such as Werthmann, Weingart and Kirk (2010), Thu and Populus (2007), many of them concentrated on survey data and used statistical methods to analyze (Tuyen 2014; McNally, McEwin and Holland 2011) Most of studies by Vietnamese also focused on the issues of mangrove restoration (GIZ Kien Giang Team 2012; Que and Hai 2012), mangrove forest management (Dung and Tinh 2006;

Que and Hai 2012), or the role of mangroves (Ngot et al 2014) However, the reality showed

that the human factors, especially those living in areas with mangrove forests, also play an important role in protecting mangrove forest Although other studies have also referred to the

human factor, they are not properly emphasized or just barely mentioned (Mai and Thang 2011) This study focuses on the human behavior by combining both descriptive statistics and regression model with the specific treatments Thereby, we propose specific measures to encourage people to participate in forest protection more effective

1.2 RESEARCH QUESTION

The objective of this study is to find out how farmers’ decision making is influenced by the economic benefit and other behavior, under different hypothesized situation For reducing mangrove forest extraction in the Mekong River Delta of Vietnam, we will deal with these following questions:

- If farmers know others’ benefits could be harmed as they do extraction, do they reduce mangrove forest extraction?

- If farmers know their extraction is higher than that of others, do they reduce mangrove forest extraction?

- If profit of farmers from extracting mangrove area change, do they reduce mangrove forest extraction?

We use regression analysis to examine the effect of these treatments in converting mangrove forest to other uses In the regressions, we also examine the influences of personal characteristics and social context on one’s making decision

1.3 SCOPE OF STUDY

The experiments and interviews will be conducted in some selected communes of six provinces in the South Vietnam Those six provinces are Tra Vinh, Soc Trang, Kien Giang, Ca Mau, Ben Tre and Bac Lieu where there is a significant share of mangrove forest area in the South In which, Soc Trang, Tra Vinh and Ben Tre are the province archived the highest rate of successful reforestation, 91.7%, 87.8% and 80.2% respectively while Bac Lieu, Kien Giang and

Ca Mau get the lower one with 43.6%, 57.3% and 79.5% respectively according to statistical data of Southern Forest Planning and Surveying Sub-Institute in 2010

The Mekong River valley is great potential of fishery, total production of fisheries in

2008 is 3.9 million tons in which fishing contributed 1.9 million tons and shrimp farming

contributed 2.0 million tons In addition, total economic value of fishery is estimated about 3.9 – 7.0 million USD (Mekong River Commission (MRC) 2010) In Vietnam, the communities in Mekong River valley could be divided into two groups: island group and coastal group For coastal group, fishing only dominate about labor force as well as generating income for people

in a few communities while agricultural activities play a major role in island group However, offshore fishing is only suitable for some wealthier household while inshore fishing is not encouraged by government Besides that, farming is coming up against difficulties about water deficiency Although shrimp farming is just ranked number two following agricultural activities, the community recognized that this is a potential and sustainable livelihood (Mekong River Commission (MRC) 2010) These activities are also influenced directly if there are any changes

in mangroves situation such as storm and flood, saltwater intrusion and the changes in volume

of aquatic species Therefore, in this study the subjects include who do at least one of three types

of livelihoods following: shrimp farming, fishing and farming

1.4 METHODOLOGY AND DATA

As we mentioned in the research question section, there is consider consistency in the causal relationships between the assumptions and the consequences In these cases, the experiments are often used to predict phenomena and explained the causation It means that experimental research is able to manipulate the predictor variable and subjects to identify the relationship between causes and effects (cause-and-effect relationship) Therefore, the experiment method is appropriated for this study

In this study, we will conduct a series of experiments in six selected provinces in the Southern Viet Nam As noted in the introduction, we designed the experiments to emulate the actual incentives that households face where their benefit from converting mangrove forestland

to agricultural land for production purposes with a social dilemma concerning environmental quality The individual’s benefits usually increase for one more unit of converted area, but the aggregate extraction will reduce the common benefits from keeping mangrove forest, such as flood and storm protection and desalinization Therefore, payoffs from playing game would be generated from a function level of converted area that individual choose and the total converted area in a group The mangrove forest policies and programs are also introduced into the game

as the treatments To focus the impact of these on individual extracting – behavior as well as eliminate other influences such as free rider, the non-cooperative game is recommended The interview is conducted simultaneously to control other personal characteristics Our experiments were undertaken during March and April 2015 in Tra Vinh, Soc Trang, Ben Tre, Kien Giang, Bac Lieu and Ca Mau The specific communes were chosen because they meet fully requirement

of sampling with significant mangrove forest

For analyzing, we just use the graph and simple OLS method with dependent variable is the converted area of mangroves in each round The round is treated as time series data and the identification code of each player is the cross-section data The data will be collected from the experiments and from the interviews in paper, and then we enter data into the computer After doing data cleaning, we will use them for descriptive statistics and doing regressions

1.5 THE STRUCTURE OF THE STUDY

Organization of the paper is as follows In chapter 1, we provide an overview about relationship between sustainable development and mangrove forest conservation, the necessity

of this study including research questions and brief methodology Chapter 2 will summary about the ecological services of mangroves and the findings of other relevant research with suggested control variables and the hypotheses After that, we will present details of the methodology and data description in chapter 3 The first section of Chapter 4 focuses the current situation of mangrove forest in the Mekong River Delta as well as the other common pool resources in Vietnam The next section of Chapter 4 is about the experiment results in non-parametric analysis and regression interpretation and we will discuss about the hypotheses and other conflict results in the last section We conclude the thesis with policy implications in the final chapter

CHAPTER 2: LITERATURE REVIEW

2.1 MANGROVE’S ECOLOGICAL SERVICES

Mangrove forest is a group of trees and shrubs that live in coastal intertidal zone including shorelines and estuaries in the tropics and sub-tropics In these places, the water is

100 times saltier than elsewhere, so only mangroves can grow normally All over the world, there is about 137.760 square kilometer of mangrove forest (2010), which was announced by NASA/USGS after analyzing data from The Landsat Program In which the largest areas are Indonesia, Brazil and Australia account for 21%, 9% and 7% of total mangrove forest area respectively

Mangrove forests provide a wide range of vital services as well as play an important role

in the ecological system In general, economists divide the economic value of mangroves into three parts: direct use values, indirect use value and non-use values Base on this classification

and modifications from previous papers, Brander et al (2006) summarized ten ecological

functions of wetland In Salem and Mercer (2012), the authors adjusted these functions for mangroves with seven functions including (i) Flood and flow control, (ii) Storm buffering/sediment retention, (iii) Water quality maintenance/nutrient retention, (iv) Habitat and nursery for plant and animal species, (v) Biodiversity, (vi) Carbon sequestration and (vii) Natural environment, as shown in Table 2-1 Besides, there are other divisions based on ecosystem service of mangrove, e.g CICES with Delphi technique in Mukherjee et al (2014) and Lee et al (2014)

Table 2-1: Comparison between ecological wetland and ecological mangrove functions

Flood and flow control Flood and flow control Flood protection* Indirect use

buffering/sediment retention

Storm protection* Indirect use

recharge/discharge

Improved water* quality Indirect use Waste disposal* Direct use

Habitat and nursery

for plant and animal

species

Habitat and nursery for plant and animal species

Commercial fishing and hunting*

Direct use

Energy resources* Direct use

Biological diversity Biodiversity Appreciation of species

existence*

Non-use

Micro-climate

stabilization

Climate stabilization Indirect use

Carbon sequestration Carbon sequestration Reduced global warming* Indirect use

Recreational activities* Direct use Appreciation of uniqueness

to culture/heritage

Non-use

Source: Adapted from Brander et al (2006), Cabrera et al (1998) and Salem and Mercer (2012) *Denotes the economic goods and services are presented in Salem and Mercer (2012); in function “Natural environment”, Salem and Mercer modified its structure with two components: “Recreation, tourism” (instead of “Recreational activities”) and “Existence, bequest, option value” (instead of “Appreciation of uniqueness to culture/heritage)

2.2 MANGROVE FOREST AND MARKET FAILURES

Every day, there are many people extract common pool resources or gain from them; even we do not use or consume them Moreover, because of these externalities, many of these common resources are market failure

Assume that we have free a free market where buyers and sellers do transactions freely without any government control (or just a little bit), there are two persons, one extracts common pool resources, and other one does not The person who does extraction will use the material

from common pool resources to manufacture goods and services, and then provides them to remain one In the free market, the price will be set based on a mutual agreement between buyer and seller As we mentioned above, the common pool resources benefit us by somehow and it

is also a rival goods, consequently the extraction of this one will reduce the happiness or fortunes

of that one The seller should take these values of common pool resources into account as a discount when he or she builds the price It is similar to someone have to pay an access fee to the common pool resources and it will be subsidized to others who do not extract that resource Ambec and Sebi (2011) showed that because of extraction costs differ for different ones, there are someone loses while others win The lack of information in this case causes the market failure

In an article about introduction to market failure of Professor Edward Morey, University

of Colorado Boulder, the author indicated another reason that makes market fails That is, in case of resource free access no one bear in mind “the opportunity cost to society” of doing extraction because people do not have to pay any costs Due to some particular characteristics,

it is very hard to exclude anyone from exploiting common pool resources The common pool resources are almost free to access and therefore, people often put a zero price on common pool resource

Additionally, the inequality could be considered as a cause of market failure The gap between income between each other leads to the fact that someone will gain more from common resources and someone else receive less For example, a fisher who has a larger fishing vessel will catch and store more fish than a fisher has a smaller one Because the first one can earn more by selling more fishes, he can buy more vessels or upgrade his vessel to be larger, and then he will earn much more Because fishes are “rival”, if the first one catches more and more fish, the second one may be having nothing

And the common reason that free market fails with common pool resources is not to assign property right adequately There are two cases, no assignation and assign to wrong person (Mankiw 2012) said that if the property right were not well established, market would fail in allocating resource efficiently In case a part of common pool resource does not have an identified property right, mean that there is no one control that part legally, someone will make

corrupt use of his or her power to violate the resource The Vedan company pollutes Thi Vai River is a good example For many years, the inhabitants who live along the Thi Vai River realized the worsening quality of water due to infectious sewage from manufacturing activities

of Vedan However, no one has the right to ask Vedan to be responsible for their actions until the authorities involved In the second case when the legal rights are put into wrong ones, it is the same as we give someone some advantages over others but unofficial or something like that For example, a household who live near the forest is distributed a smaller part of forest for management whereas a commercial man can “buy” a larger one with unclear purposes Base on

suggestion of Lant et al (2008) in similar issues, property law can make natural capital, i.e

common pool resources, become marketable goods and services and, therefore, make market fails

2.3 LITERATURE REVIEW ON THE COMMON POOL RESOURCES EXPERIMENTS

Mangrove forest is one of common pool resources, so that, understand the research about other common pool resources also help us in studying this resource Firstly, we will to learn how people behave toward common pool resource and what would happen in that case Hardin (1968) described the scenario of grazing cattle as “the tragedy of commons”, this theory states that “individuals acting rationally and independently” according to their own self-interest will use up the supply of common pool resource, even if it is conflicting with the best interest of the group In story of Hardin (1968), the individual farmers brought as many cattle as possible to the free grazing area, because it was beneficial for them or in other words, they were “acting rationally and independently” for themselves However, the pasture became unusable quickly when each individual acted in this way In addition, everyone in-group would suffer this loss as

a whole as the shared resource was depleted The tragedy of the commons is often brought up when people discuss environmental issues In this study, we examine the case of deforesting mangrove forest for production purpose such as farming and shrimp farming While converting mangrove forestland to agricultural land may directly benefit those who own and use the land, the cost of losing that mangrove forestland is more widely distributed As an individual, it may

be beneficial to earn money by selling fish and shrimp from ponds that is converted from

mangrove forestland Each individual would try to convert as many mangroves area as possible

to earn more money even if all the others were doing the same thing Extracting mangrove forest leads to increase salinization, erosion, pollution and environmental degradation However, because the mangrove forest is 'common' and shared by everyone, the overall public has to deal with these problems each person contributes As this behavior continues, the mangrove forest (the common pool resource) is eventually depleted and all people over the world (the group as

a whole) suffer

Secondly, we inspect the behavior of people in extracting common pool resource in detail to determine what factors affect that behavior In field experiments of Cardenas, Stranlund and Willis (2000) the crowding-out behavior of group-oriented is affected by external interventions People will change their activities if they feel something outside affects their benefits “…Crowded out group- regarding behavior in favor of greater self- interest” (Cardenas, Stranlund and Willis 2000) is the main reason, which makes the negative effect of external control policy It means that, when publishing any policies to control the improving environmental quality, the government should recognize who will be affected by that policy and who confront that policy completely Cardenas (2004) indicated that the effect of regulatory is different in different level of penalty In additional, people respond more factors than just the expected cost of regulation in reality Besides, there are many motivations in using a common resource, so that policies also satisfy those motivations In reaching the targets of improving environmental quality, the manipulation of information can make the results are not good as we expect Cardenas (2004) suggested that the effect of state regulators can “often be negative in terms of inducing a self-governed solution to the commons dilemma, particularly where regulation by the state is costly and difficulty” Titmuss, Oakley and Ashton (1997) concluded that people are willing to do something if they volunteer to do that rather than give them financial rewards Similarly, the authors also suggested that we should not try to correct them

by enforce a stronger method, because it will make all thing worse

Zero distribution hypotheses said that the government should impose a set of regulations

to help the users of common pool resource reaching their own long-term interests However, Ostrom (2014) showed a different view in evaluating social norms Her study represented,

although it needs more works to demonstrate conclusively, the users of common pool resource will manage local resource better if they can “devise and enforce” their regulations and no need government enforcement Even the effort of national government is one of threats to sustained collective action according to this research

Altruism, reciprocity and conformity can be used as an interpretation in many cases of cooperating to use common pool resources Cox (2004) confirmed there exist “trust” and

“reciprocity” in investment game in his research Berg, Dickhaut and McCabe (1995) developed the investment game; in there all subjects are divided into two groups and be arranged in two separate rooms, called the room A group and the room B group At the beginning, each subject will be received ten bills of $1 Each individual subject in room A has the right to transfer any integer number of their own bills (from 0 to 10) to a paired person who they do not know in room B and he or she can keep the remainder The experimenter had multiplied this amount by three times before delivered them to that anonymous person To the room B subjects, they could return part, all, or none of that tripled amount, which they received from the anonymous paired person in room A Based on this game, James C.Cox conducted an experiment and through setting up one of three treatments is investment game, he identified trusting behavior and reciprocating behavior in one’s preferences Velez, Stranlund and Murphy (2009) combined all altruism, reciprocity and conformity with self-interest in multiple model of behavior to test that statement Moreover, these authors showed that the conformity is more sensible than reciprocity

in explaining conditional cooperation; even the conformity could generate the reciprocal behavior of subjects They also suggested that this finding is potential implication for managing resources or publishing relative regulations Cox (2004) also emphasized in conclusion “Beliefs about other altruistic and reciprocity behavior should be incorporated in the model.” Furthermore, the living context of each subject has some certain influences to their decisions in experiments In research of Castillo and Saysel, they suggested modelers to use modeling as a tool to “determine the relevant factors in decision making in common pool resource management” And in this research, they found that the cooperation attitudes of an individual in real life correlate with the decision-making process in the experiments of that one

Another theory is also used to explain the one’s behavior is theory of “Conditional cooperation” This theory says the contribution of people will increase on average if they know that others also contribute and the effect will vary depending on their behavior in the past Frey and Meier (2004) performed a survey at the University of Zurich to test this theory in their research Each student will be informed about two charitable funds, one is used to support foreign students and one offered to students who faced the financial difficulty the low-interest loans They had the opportunity to contribute to one fund or both of two funds with a different compulsory tuition fee for each fund, or even did not give any contribution After nine semester periods, the authors found the relationship between one’s expectation about others’ doing and his behavior Moreover, the results from this research confirmed the presence of fairness preferences of people such as reciprocity in their behavior Ferraro, Miranda and Price (2011) conducted an experiment in Cobb County, a suburban county of Georgia, U.S, about the saving water of residents This study wants to learn about the long-term patterns in using water of residents through some social treatments in attempt to reduce “water use” The most important finding is the treatment of comparison still impacts on water use significantly for some periods

It means that the method of comparing the activities of customers to that of their neighbors is efficient However, the decreasing over time also shows that there exist some restrictions

Through learning about the behavior of resource users in Thailand and Colombia, Janssen et al

(2013) also found out the expectations of people about their neighbors is a major reason for the different in making decision of each one The outsiders also have the similar impacts The experiences in managing resources as well as the regulations just play a secondary role Agree with the work of Velez, Stranlund and Murphy (2009), one attention point more is we should put the context’s residents like cultural village as a fundamental factor to interpret the results of experiment The establishment of rules bases on norms is also used as a reason in explaining the behavior of residents It can make people cooperate with others to improve welfare or do against the social benefit

Determinants of behavior toward common resources

As we discussed, there are many factors affect to the process of decision-making of each player in experiments Some above research recommended several variables following:

Age and education: Velez, Stranlund and Murphy (2009) examined the effects of age and

education on individual behavior in fishing management for each region Age were not statistical significant, but we should include it in our interview We expect that the older are often more thoughtful and careful than the young people, so their behavior will be more stable and vice versa Education has positive and significant effect in that research and this is possible result Higher educated person understands the rules better than lower educated one and then they choose a good extraction level to get more payoffs Lower educated ones, on the contrary, even do not know what level they should choose Cardenas, Stranlund and Willis (2000) also included age and schooling in the research and found significance of them by analyzing each group “Age” variable and “Education” variable will be measured by the years of age and the schooling years respectively

Gender and time living in region: even though gender variable and time living in region is

collected in research of Velez, Stanlund and Murphy (2009), the authors did not include them

in regressing because of lacking variability However, gender may be an important variable and

it is suggested in research of Cardenas (1999), Cárdenas, and Ostrom (2004) In our expectation, male subjects are more likely to take risks Additionally, in selected provinces that we should conduct series of experiments, male subjects could have higher education level than female on average Gender is a qualitative variable, so that number “1” denotes male and number “0” denotes female For time living in region variable, we measure by the number of years that person lives in current region and if they were born and grown in there we will note by number

“99”

Income level: besides age and schooling, Cardenas, Stranlund and Willis (2000) added

income level to determine whether the effect of income on making decision of participants Cardenas (1999) also found that the impact of actual wealth and occupation in terms of explaining cooperation in his experiments Income of each one come from their primary occupation and may in fact be from mangrove forest related activities It is hard to collect the information about exploiting mangrove forest of households, especially the activities relate to money We still expect that there is only a part of community has income from extracting mangrove forest higher than from their main doing

Type of livelihood: due to the type of common pool resource and the object of research, each

study considers many different kinds of livelihood of communities In 2000, Cardenas and his colleagues conducted a series of experiments near the area of local tropical cloud forests; so in that place the usual work are forestry activities Velez, Stanlund and Murphy (2009) only

focused on three regions with the primary work are fishery In the research of Janssen et al

(2013), the author crafted three kinds of game depend on resident’s context including irrigation game, fishery game and forestry game Ferraro, Miranda and Price (2011) even did not care much about customer’s career; the impact of treatments is the most interested issue Therefore, choosing what kind of livelihood relies on where we carry out the experiments In selected Southwestern provinces, fishery, aquaculture, and growing are three primary activities of the local people In contrast to above previous research, we use type of livelihood as a control variable to explain the behavior of each player base on their work We expect that if their work

is more dependent on mangrove forest, they may be more cautions in extracting It also shows their experience toward managing mangrove forest

The past behavior and behavior of others: Frey and Meier (2004) have shown that the

behavior of one person in contributing correlates express expectation This expectation comes from two sources: from themselves and from others In 2004, Bruno S Frey and Stephan Meier found that if one person contributed in the past, he would tend to contribute in the present In additional, when they know the information about others’ contribution, they will change their contribution slightly depending on the level of contribution of others The players will base on their past contribution as well as the total contribution for each round to make decision In our experiment, this information will be notified to all players and we can collect them easily

Level of communication: due to discussions of Cardenas, Stranlund and Willis (2000),

Cárdenas and Ostrom (2004), Velez, Stanlund and Murphy (2009), Ferraro, Miranda and Price

(2011) and Janssen et al (2013) about context’s residents, we see that a person who

communicate with the community tend to correct their behavior conformably To identify the level of communication of each one, we ask them some questions about the projects of conserving mangrove forests in their residence:

- Do you realize any changes of mangrove forest in some recent years (in area)?

- What are the reasons for these changes?

- How many conservation mangrove forests projects in your residence that you know?

- Did you enjoy those projects?

Each answer will be marked by “Yes” or “No” and then to be handled by computer The more “Yes” answers are collected, the more level of communication of that one

The quantity of mangrove forest exploited in the area: this is an explanatory variable in

our regression Through a game with 20 rounds, we will calculate the quantity of mangrove forest that player extracted in area on average Using the information of other characteristics as well as treatments in game, we will find out what determinants impact on exploiting behavior

of player and explain it base on related theories

Moreover, these variables also affect the effects of treatments on behavior Information treatment provides more information to arouse people’s compassion, so we expect that the female and the older reduce the level of extraction more than the male and the younger subjects Those who often participate in community activities (higher level of communication) could have positive response to this treatment, and it leads them to reduce self-extraction In social comparison treatment, the information is given specifically and, therefore, people may strongly react Apart from age, gender and level of communication, type of livelihood are included Because famer and shrimp farmer work on mainland, they have more chance to communicate with each other and hence they should be better in comparing their work with the work of others

It means that farmers and shrimp farmers follow the treatment of social comparison by far and their extraction level will be closer to the specific information than that of fishers Due to the adventurousness, the fishers also extract more than the farmers and shrimp farmer under the treatment of profitability Similarly, the individual with lower income is willing to take risks to expect higher return and they will try to maximize their profit by increase their extraction level

2.4 MAIN HYPOTHESES

Based on three hypotheses in developing a theory of social comparison processes of

Festinger (1954), four suggested conjectures of Janssen et al (2013) in the research of

harvesting forestry, the results of Cox (2004) in identifying trust and reciprocity, the hypotheses

in making decision from framework of Cárdenas and Ostrom (2004) and the work of Cardenas, Stranlund and Willis (2000), Cardenas (2004) in designing experiment, we will test four following hypotheses:

Hypothesis 1: People will exploit the common pool resources and exhaust its stock as much

as possible

In case of no communication, the first ten rounds, all players will try to convert the mangroves area as much as they have It can be explained that in thinking of all players, the benefit in case of converting is greater than that of leaving it on the ground The more the unit

of converted area, the higher expected benefit Due to they are not allowed to communicate with each other and, therefore, do not know how much others convert, they believe that others will

do like them Moreover, in their mind, if they do not extract common pool resources, others will

do We also expect that; they do not care this lead to reduce the social welfare of all people because they are not bound up with any rules whatsoever in baseline treatment stage

Hypothesis 2: People will reduce the level of exploitation if they know that their exploiting has a negative impact on others

There are many evidences to suggest that people will choose to do things that harm for others in trying to get benefits, unless those things are also wrong for themselves However, Cox (2004) identified trust and reciprocity in an investment game; it is also consistent with the conclusion of Frey and Meier (2004) about the existence of “some level of fairness preference such as reciprocity” It means that when a person realizes or fells that his or her interests are not harmed or just a little due to doing something right for others, he or she will help those people

In the case of this game, the players will reduce their extraction to some extent if they know that their earning from converting mangroves land to agricultural land will harm others’ earning They think it is reasonable because they “trust” others also make concessions like them

Hypothesis 3: People will reduce the level of exploitation if they know that many others do as well

In the contribution experiment of Frey and Meier (2004), the authors found that the individual tends to contribute more, i.e for the charitable funds, if we let them know that others

also do contribute Therefore, we assume that in an extraction experiment if people know their extraction level is higher than the average, they will reduce it close to that of others It can also

be explained by a theory of social comparison, which is developed by Festinger based on a previous theory about the process in social group According to Festinger (1954), through the comparison, people decide to do something depends on other doing, and they just choose who

is similar to themselves In 2009, a company called OPOWER run series of energy conservation programs by sending the letters to residential utility customers to inform them about their electricity use and that of their neighbors Allcott (2011) conducted a research to evaluate this project and the author succeeds in using social comparison theory to explain the impact non-price intervention on consumer behavior In our experiment, when we provide the information about the extraction of other groups on average is lower than the average extraction of current group, we expect that all players in that group will reduce their level of extraction

Hypothesis 4: People tend to exploit more common pool resources if their profit changes, whether it increases or decreases

In our common pool resource game, no one is allowed to discuss with each other, and also cannot see the changes in others’ profit Therefore, when we issue them new payoff table, they only know how their profit changes and make decision base on that Cardenas (2004) found that although the penalties can success in influencing behavior of participants, there is a little difference between two levels, i.e high and low, of the treatment It means that people tend to behave the same when they face the problems that are similar in the essence, even if there is a difference about level In our game, players do not know how different in profit change of others but they can sure that there is a difference For player with lower payoff table, their profit will

be decreased so they try to convert as much as they can to increase the profit For player with higher payoff table, they believe that their profit is still lower than other and, therefore, they also increase the profit by doing more extraction For these reasons, we expect that whether the profit

of players increases or decrease, all players will convert more mangroves land

Beside four main hypotheses stated above, there is some trends should be considered When conducting the experiments, we will choose some communes in selected provinces and gather the groups of people living together and sharing responsibilities Even though in each

game no one is allowed to communicate with each other, it is hard to guarantee that they have not met yet in reality So when they play game with these individuals, their decision will be influenced partly

In the absence of both a physical and a social comparison, the behavior in extracting resource of one is unstable In normal people often compare with others who close to their

ability or opinion to evaluate their ability or opinion According to this action, they know their capability in doing something However, in the case of lack of information, people do not know who is close to them and what ability or opinion should be compared They should learn by process in some rounds of game Try-and-error method may be applied, and it leads to unstable results

Relevant resource management experience affects the behavior in the experiments In

the field, each household faces different problem in terms of exploiting mangrove forest Government pays someone a flat rate for planting and protecting mangrove forest, but someone

is not delivered any piece of ground Therefore, they have different experience and it causes they make a different decision when playing the game

CHAPTER 3: METHODOLOGY AND DATA

This chapter will provide a simple review of the problems that should be considered when preparing a study of common pool resource like mangroves, the approaches that is used

in conducting the study, a description of how to collect data as well as descriptive statistics of these data In addition, some explanations and suggestions from previous researches are included

3.1 METHODOLOGY

3.2.1 Sampling strategy

Six selected provinces in Southern include Ben Tre, Bac Lieu, Tra Vinh, Soc Trang, Kien Giang and Ca Mau In each province, we will choose a commune, which has a part of mangrove forest Then we go to each household, explain to them understand about the purpose

of this research and ask for their permission to interview them After completing the interview, the interviewee will be invited to a pre-determined location where we conduct a series of experiments Each eight of interviewee is gathered in to a group and play all 20 rounds for a game session When all players finish the game, they will receive the amount of money corresponding to their efforts privately

All participants have to be guaranteed that they are voluntary and they have the right to stop the interview or quit the game if they want Moreover, the local authorities supervise these activities Collecting data will be taken place from March 1st to mid-April

3.2.2 The interview

The structure of questionnaire

The questionnaire consists of three main parts

- Part 1: in this part, interviewee will be asked about his/her background:

 Gender, age, schooling years and primary occupation of each family: basic personal information of interviewee

 The number of years they live in current place: it aims to define the level of strong attachment to the current community

 The order of the current occupations in comparison to its in the past If there have any changes, interviewee will explain about the reasons for these changes

- Part 2: the awareness of residents about mangrove forests:

 The change of area of mangrove forest in five recent years in the opinion of interviewee: it presents the attention of everyone about the change of mangrove forest in particular Therefore, someone is aware of this change and someone is not In case the area of mangrove forest decreases, the interviewee will answer more about the causes of this decline

 A question is about the order of the disaster in causing damage to the interviewee and the local general Three disasters are listed in this question involve the reducing of mangrove forest

 The number of projects of conserving mangrove forests that interviewee knows: there are many projects to protect mangrove forests but not everyone knows about them If the interviewee is not well informed, some questions will be ignored

 In the opinion of interviewee, the mangrove forest is actually important or not and which benefits from it (the benefits will be listed in a table) The respondent just answers yes or no for each one

 Similarly, we also put some questions about the hindrances/difficulties that mangrove forest brings to them

 The area of mangrove forest that government allocates to interviewee’s family and the annual paid Government pays them a flat rate for managing the forest and people have the right to convert a part of this area to productive land

 There are some questions about living activities For example, what material they often use to fire (for cooking) and where they take it

- Part 3: this part will be different for each interviewee depends on their occupation

In this research, only fishing, aquaculture, and farming are obtained in the questionnaire The object of this part is calculating the profit of each family in relation to the above occupations Because of the difference in each type of occupation, the questions will be designed bases on the features of each one The

diversity in culture of each region is also considered, especially unit measurement, name of produce… It is learned before conducting interview

These questions are used to identify control variables that we mentioned above Each one of group has to answer them before or after playing game By connecting the final result of player and his characteristic in comparison to others, we will examine what factors in fact influence to his decisions

Procedures of interviewing

A group of interviewers would be trained about the research and its objectives, the interests and benefits that the interviewee will receive when participating in interview, and the interview techniques such as how to ask questions properly After all legal procedures had been done, the agent of authorities leads these interviewers to the households who conforming to standards At each household, the local authority introduces them to each other Then an interviewer introduces himself as earlier training, talks about the research and seeks interview from the householders The interviewer must present the objectives of this research and do not forget to mention that the interviewee can pause or stop the interview whenever they want If the householder or the proxy accepts after hearing about the regulations carefully, the interview will begin Another way is to invite them to play game first, then conduct the interview

An interview takes about 30 – 45 minutes including time to introduce about the research and explain about something, which confuse the interviewee This information would be checked several times and verified by the interviewee before inputting data

3.2.3 The experiment

Experimental design

Group size: In case the group size is small, it costs too much money and time to conduct the

experiments Especially in the group consist of two players, this one is easy to guest the other does what and some unexpected behaviors will appear In comparison between dyads and triads, the group of three persons is more stable than the group of two persons The third person acts

as a mediator as the relationship between two other becomes strained When the number of person in a group increases, each one will be more separated, and therefore they tend to be less

influenced by decision of others However, too many players in a group also make the results more variable According Cardenas (1999) and Cárdenas and Ostrom (2004), a group of eight players is recommended

Game round: The number of rounds in a game is neither too many nor too fewer If we

conduct a game with many rounds, time and money need to be reconsidered again Besides, as players participate in the game for as long, it makes them feel tired They do make decision unreasonably or even quit the game However, if the number of turns is too few, player does not have enough time to learn by process Then, they do not know how to make decision reasonably

In the series of experiments, each game has two stages and ten rounds per stage The first stage

is baseline treatment; in this stage, people do extraction as they want without any regulations And in the second stage, there are three treatments will be applied

The endowment: Because of the research object – mangrove forest, we use the area of

mangrove forest as the endowment In Vietnam, the government published many documents in regards to manage mangrove forest, including which percentage of mangrove forest is allowed

to convert into agricultural land This percentage depends on the allocated area and the highest percentage corresponds to the 10-hectare forest area or more Using this number helps local residents to feel less worried in making decisions In the beginning of each round, we assume that the government will allocate 10 hectares of mangrove forest to each player and the players are allowed to convert mangrove forest area to agriculture land for production purposes, the range of area that one can extract is from zero to ten hectare The concept of zero extraction is easy to explain, because the households also have their livelihood, their income does not depend

on forest Beside that, in the experiment game, we also strongly emphasize that the maximum number of mangrove forest area is just an assumption

The treatments: In our common pool resource game, the players will play 20 rounds equally

divided into two stages The first stage is baseline treatment stage and in this stage they decide which number of mangroves area to be extracted as they wish Before moving to the second stage, which is the treatment game session, the facilitator will introduce the treatment that is imposed in that stage There are three kinds of treatments that are used in this paper include information treatment, social comparison treatment and profitability treatment

In the treatment of information, all players knew that they would receive more payoffs

as they decide to extract more forest They are also informed that, however, these actions also reduce the earning of others This treatment is used to identify the influence of “trust” and

“reciprocity” in game of investment Cox (2004) concluded, “Beliefs about others’ altruistic and reciprocating behavior should be incorporated in the model.”The second treatment is social comparison In this treatment, subjects were informed about the average number of mangroves area that their “neighbors” – other groups – converted in previous games Through Frey and Meier (2004), we expect that the information about others’ doing influence the behavior of individual and, therefore, each one should compare their exaction level and follow their

“neighbors” in the same action And in the last treatment, different players receive the different payoff tables There are two types of Payoff Table, one is high rate of return and the other one

is low rate of return Because of change in profit, individual will behave differently One of results from Cárdenas and Ostrom (2004) also indicates that in different levels the effect of regulation is different

The operations of the mangrove forest policies and programs are concretized and imitated into three treatments The study of these treatments will help us understand how the current mangrove forest regulations work and how they influence on the behavior towards extracting mangrove forest in practice

Payoffs

Based on work of Cardenas, Stranlund and Willis (2000), Cardenas (2003), Cardenas (2004) and many previous experiment researches, the relations between the parts or elements of payoffs function are determined The problem of exploiting mangrove forest is instructed to them as a game and no penalties as well as not in the present Each parameter is presented here

At the beginning of each round, each player is given an endowment of area, call e Individual i can decide how many areas he/she wants to extract, call x i , of the total e In each

unit of area is kept in one year the government pays the individual an amount corresponding to

w It means that individual i will receive w(e – xi) from keeping mangrove forest However,

because the differences in mangroves management in the southern provinces, this amount is different in different provinces, even in different communes in the same province Therefore,

this amount will be treated as a constant value to obtain the consistent outcome from different

provinces, f(x) = C

In case individual i decide to extract forest, we suggest the form of private benefit which

individual i will receive when do convert g(x i ) = d i - x i ) 2 /2) This form follows the law of

Diminishing Marginal Utility to assume that individual i will receive the benefit from exploited area until reaching the threshold and if he or she exceeds this point, for each additional unit of exploited area his/her marginal utility will be declined It can also explain threatens from over-harvest; there is a probability of penalty as if the local authorities detect The parameter d is used

to transform this extraction into payoffs and it should be positive and  are positive numbers and they guarantee individual i always has benefit from extracting mangrove forest, or g(xi) >0 with xi ∈ [1,e]

Mangrove forest provides not only products that people sold for money (direct benefit) but also more than that (indirect benefit), such as storm protection, the living environment of many aquatic species, fresh air It would be insufficient if we did not take these elements into the function Assume that no one converts mangrove forest; everyone will receive the same

indirect benefit q o In other words, q o is the mangrove forest quality in the absence of extraction Because mangrove forest is common pool resource, so the extraction of anyone will cause harm

to the interests of the rest Therefore, the indirect benefit is a function of total extraction h(∑x i )

= i(q o - ∑x i ) 2 /2), where i is the parameter which transforms the indirect benefits into payoffs

andα is also a positive number in part of guarantee the reality: more extracting, less payoffs

In our experiment, qo is the total endowment of all participants The quadratic form of the aggregate amount of exploited area represents the rapid decline of resource when many people join the exploitation It could be seen as the high growth rate/low growth rate of the resource depending on each case For example, in case someone converts a small number of area units, mangrove forest can be renewal and people still gain from it However, when everyone does extraction in a large area, even all units available, it slows down the speed of recovery as well

as reduce the benefit from mangrove forest Because of the complex formula, we do not put the

influence of recovery into this function This quadratic form, so that, help us in part in this problem

Given these components, the utility function of an individual in terms of extracting the local mangrove forest is defined following:

Additional information

This is a non-cooperative game; so no players are allowed to communicate with anyone

as well as their decisions were kept private During the game, all players are always reminded about this Players can know their payoffs base on the calculator table that is given to them from the beginning but do not know other payoffs

Daily wages in these regions is about 120,000 - 150,000 dong (VND) during the period

of this research (March - April, 2015)

Procedures of conducting CPR game

This experiment adopted a case common pool resource (CPR) experiment approach with

a few changes to fit the context The eight householders or proxies who are interviewed or would

be in later were gathered into a group Each household only appoint one representative to join the game and this one will be identified by an identification code (ID)

In the beginning of each game, the monitor has to read game instructions to all eight players and they are played some draft rounds to get the rules clearly before playing the game

Of course, those draft rounds are not accounted into the final earnings The experiment instruction is presented in the appendix

Each game will be divided into two parts: in the first part, all players play the game without any information and play as they want; in the second part, the monitor will provide some treatments (these treatment is different for each group depends on group composition) and players play bases on new information The monitor will issue the player a record table, which record the converted area in each round as well as the corresponding payoff, and it is also the basis for the payment To start each round, each player will receive a small ballot paper with his

or her ID Then, when the monitor asks all players to fill out the number of area of mangrove forest they want to convert, they will write the number to that ballot Some assistants should go around ensuring that all players do not communicate to each other and make decision independently When all players write down their decision to the record, the monitor will collect all ballot paper After doing some calculations, the monitor informs to all players about the total converted area in this round Based on this number each player calculate the others’ converted area and find their payoff in the payoff table, it is also recorded The assistants will support the players who feel difficult in calculating or finding the payoff Next round will begin if everything is finished After the first ten rounds, the monitor will spend a few minutes to introduce about the regulations in next ten rounds before continuing the game to the end

Each game took approximately 1.5 hours In addition, on average each player will receive about 130,000 - 150,000 dong (VND) in expectation They get this earning by cash after the end of game immediately

We do the same for each region that we conduct experiment to examine whether the results were replicated across the regions Even there is some different between regions, for example geography features, method of fishing - farming - shrimping, weather we still expect that some common determinants will be used to explain our results

3.2 DATA

All data are obtained from the experiment including interview and CPR game A description for variables that are used in the regression is presented in the following table

Table 3-1: Description of some variables

measurement

Extraction The converted area which is recorded in each round for

each player, it must be an integer and its value is from 0

to 10

Hectare(s)

Age The number of year old, it is also calculated by taking

the current year (2015) minus the year of birth of that person

Year

Education The number of schooling years In case the interviewee

just remembers their grade, assuming that each person only spends one year in school for a class For example,

if a man answers that he was fourth grader, the number

of schooling years will be 4

Schooling year

Gender Male and female Number “1” denotes male one and

number “0” denote female one

to be a sign of farming, shrimp farming and fishing respectively

communication

It demonstrates the concern of individuals towards the issues that related to community and mangroves For example, if the interviewee knows the number of mangrove forest projects are deployed in local as well

as theirs purposes (protecting and/or planning), the communication level of this one will be higher than that

of other one who do not know The participation of individuals in such projects also contributes to increase their level of communication

Benefit from

mangroves

The benefits of mangroves that they receive or they think so It includes some sub-options (each option presents a function of mangrove forest) in our questionnaire table and the interviewee will choose what they want For example, if the interviewee thinks that mangrove forest could prevent the floods, he or she will answer “Yes” and otherwise choosing “No” “Yes”

Variables Definitions Unit of

measurement

answer indicates that the function of mangrove forest benefits people and community, “No” answer could be due the interviewee do not gain any benefits from that function in thinking

Hindrances from

mangroves

The difficulties or hindrances of mangroves that they face or they think so It includes some sub-options (each option presents a hindrance) and the interviewee will choose what they want For example, if the interviewee thinks that the mangrove forest creates difficulties for farming or shrimp farming, he or she will answer “Yes”

and otherwise choosing “No” “Yes” answer indicate that mangrove forest has negative impact on real living

of interviewee and “No” answer means that that one is not trouble with mangrove forest

Management

experience

The experience of household in managing mangrove forest Only those who were allocated an area of mangroves to manage could answer the relevant questions These question include the area of mangrove forest is allocated, the annual amount that interviewee is paid corresponding to the allocated area, the amount from pruning off mangrove branches and after fixed years

Past behavior Past behavior is considered as the proxy for individuals’

heterogeneity (Frey and Meier 2004) There are three alternative options: the individual extracted area in the previous round, the total extracted area in the previous round and the total extracted area of others in the previous round

CHAPTER 4: EMPIRICAL RESULTS

4.1 MANGROVE FOREST IN VIETNAM

4.1.1 The current situation of mangrove forests in the Mekong River Delta

Vietnam locates in the region is considered as “the world’s largest and best developed”

of mangroves in Giesen et al (2007) Nevertheless, the awareness about mangrove forest does

not correspond to its position Food and Agriculture Organization of the United Nations (FAO) announced that, from 1980 to 2105, the area of mangrove forest is lost about 20 – 35% and developing countries have the highest rate of mangrove forest loss Vietnam is also not an exception; the mangrove forests in 2013 are just left about 40% (in comparison with) of that in 1943.The chart below will provide an overview about the changes in mangrove area of Vietnam from 1943 to 2013

Source: the estimated area in 1943 from work of Maurand in 1943; Hong and San (1993) and Hong P N (2003)

collected data in 1953, 1982 and 1990; Ministry of Agriculture and Rural Development do statistics for 2000,2008 and 2013 The blue column denotes the total area of mangrove forest in Vietnam

Figure 4-1: Chart of change in mangroves area of Vietnam from 1943 to 2013

Based on the opinions of Phan Nguyen Hong in 1984 and then be restated in Hong and San (1993), there are four zones and 12 subzones of mangroves in Vietnam and the majority of

0 50.000

mangroves concentrated in the coast of Southern Vietnam In this zone, the river systems of Mekong have good conditions for growth and development of true mangroves as well as mangrove associates Most of mangroves in the South allocate in two areas, in Rung Sac – Ho Chi Minh City and in Ca Mau peninsula In Southwestern, the largest area of remaining mangroves is around Ca Mau Point in Ca Mau province And together with other seven provinces including Long An, Tien Giang, Ben Tre, Tra Vinh, Soc Trang, Bac Lieu and Kien Giang, it made up of a part of Mekong Delta region where has mangroves

Figure 4-2: ASEAN mangrove forest distribution map (NASA/USGS)

Because of the wars in Vietnam, especially the Vietnam War (1955-1975), the mangroves area of the South declined from 250,000 ha (1943) to 191,800 ha (1982), it means

we lost 1,492 ha per year on average However, in next period the human activities are the main factors that cause the mangrove area decreased rapidly from 191,800 ha (1982) to 108,579 ha (2000) or losing 4,623 ha per year on average, this decrease is three times higher than the previous period The government conducted many conservation projects, and in addition published several regulations to encourage people plant and protect mangroves Due to these appropriate actions, the mangroves area in the South as well as in the whole country is recovered gradually According to Ministry of Agriculture and Rural Development, the total area of