lable at ScienceDirect

Ocean & Coastal Management 98 (2014) 70e78

Contents lists avai

Ocean & Coastal Management

journal homepage: www.elsevier .com/locate/ocecoaman

Determinants of participation of mangrove-dependent communitiesin mangrove conservation practices

Anjan Kumer Dev Roy a, b, *

a School of Commerce, Faculty of Business, Education, Law and Arts, University of Southern Queensland, Toowoomba, QLD 4350, Australiab Ministry of Planning, Government of Bangladesh, Sher-E-Bangla Nagar, Dhaka 1207, Bangladesh

a r t i c l e i n f o

Article history:Available online

* Ministry of Planning, Government of Bangladesh1207, Bangladesh. Tel.: þ88 (0) 1715630899, þ61 (9127813.

E-mail address: anjan_devroy@yahoo.com.

http://dx.doi.org/10.1016/j.ocecoaman.2014.06.0010964-5691/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

This article examines the factors that influence conservation practices in the world's largest mangroveforest, the Sundarbans in Bangladesh. It explores the relationship between mangrove density reductionsand past and present management practices. It is a cross-sectional study based on a close-endedstructured questionnaire (n ¼ 412), measuring variables of interest from theory of property rights.Randomly selected cross-section of households in mangrove-dependent communities (MDCs) wereinterviewed. Results show that MDCs who engage in resource harvesting under the current propertyrights regime have less participation in, and lower motivation for, the conservation of mangrove re-sources. Strict management by the Forest Department results in lower commitment by the community toconservation. Conservation of resources could be achieved by the allocation of management roles toMDCs through expanded property rights. With the granting of these rights they would be very willing toengage in conservation management. The results suggest that constructing effective strategies to pro-mote mangrove conservation require these two factors to be addressed if sustainable mangrove man-agement is to be achieved.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Mangroves are unique ecosystems located betweendry land, andshallow marine and brackish water. This natural characteristic in-troduces complexities to the management of resources due tocompeting and overlapping interests. It is estimated that more than50 000 square kilometres ofmangrove foresteor nearly one-quarterof its original, global cover e has been lost due to human activity(Roy et al., 2013a). Traditional uses of mangroves are often conflic-tual. Very often, ambiguities in property rights affect informal un-derstandings and customary rules concerning access and use ofmangrove resources by dependent communities (Walters et al.,2008). This customary use by different stakeholders causes con-flict and undermines incentives for, and use of, local knowledge forsustainable resource management. Especially in developing coun-tries, researchers have studied local people's perceptions regardingconservation with the assumption that sustainability, more re-sponsibility, and long-termmanagement of forest resources depend

, Sher-E-Bangla Nagar, Dhaka0) 422296010; fax: þ880 2

on local residents' support (Triguero-Mas et al., 2010). For successfulmangrove conservation, the issue of the allocation of property rightsto mangrove-dependent communities (MDCs) is considered vital toallow them to participate in decisions as a part of a shared re-sponsibility for the present and future state of the resources(Kathiresan and Qasim, 2005). Hence, there is a need for better un-derstanding of the factors influencing conservation choices, eco-nomic activities and existing conflicts in MDCs.

The world's largest Sundarbans mangrove forest (SMF) inBangladesh is losing its forest cover rapidly due to degradation. Onestudy cited 50% tree cover reduction of this forest in past 20 years(Kabir and Hossain, 2008). This is now facing high anthropogenicpressure as it provides livelihood support to 3.5 million localmangrove-dependent people living in its surrounding villages(Kabir and Hossain, 2008). The mangrove forest is managed by theBangladesh Forest Department (BFD) under conservation policies.They allow MDCs to harvest certain resources only under strictlyprescribed rules by issuing permits to MDCs. Although harvestingrights are restricted, MDCs regularly collect various resources, suchas timber, fuel-wood, stretching materials and fodder, legally orillegally. Consequently, MDCs use de facto rights with illegalentrance as well as their de jure rights established through permits.Although the BFD exercises exclusive rights tomonitor andmanage

A.K.D. Roy / Ocean & Coastal Management 98 (2014) 70e78 71

resource harvesting, they fail to ensure sustainable harvesting ofmangrove resources through merging de facto and de jure rights inthe case of the SMF.

The central policy goals are to keep intact the biodiversity of theSMF through integrated resource management (GOB, 1994). The pol-icy, thus, advocated a common property rights regime for “the activepartnership of the local people” (GOB, 1994) to attain sustained con-servation.However, this policy is violated by thenon-implementationof an active partnership with MDCs through allocating appropriatelevels of property rights. MDCs have made their livelihoods for gen-erations from the SMF with or without permits from the BFD. Nogovernment, presentor past, has allocated relevant property rights fortheir participation in management or policy making.

Attitudes of forest department staff and perceptions of conser-vation practices affect coastal community attitudes towards sus-tainability. For instance, MDCs' conflicts with the forest departmentstaff on resource harvesting allowances, strict rules on access, andrude behaviour or harassment by the forest department staff gen-erates negative attitudes towards conservation. The lack of MDCinvolvement in decision-making and management are vital de-terminants of negative attitudes towards conservation (Badolaet al., 2012; Datta et al., 2012). Thus, it is hypothesised that lackof involvement of local people and their customary knowledge inmanagement has a significant negative impact on mangroveresource sustainability. Although there is a large body of researchanalysing local people's attitudes towards conservation, an analysisin the context of the SMF is absent. To address this gap, this studyaims to assess MDCs' perceptions about property rights regimesand conservation practices. This will enable greater understandingof the link betweenMDCs' participation inmanagement and policy-making and sustainable use of the mangrove resources. Mangroveconservation policy andmanagement need an understanding of theimpact of factors influencing conservation choices, economic ac-tivities and existing conflicts between MDCs' actions and conser-vation practices. These are important as a means of transitiontowards a green economy through improving human well beingand simultaneously reducing environmental risks and ecologicalscarcities (UNEP, 2011). This paper initially focuses on an assess-ment of factors responsible for the existing level of MDCs' partici-pation in mangrove conservation. In spite of strict control by theBFD, a more inclusive approach needs to be implemented if sus-tainable management is to be achieved. To achieve conservationthrough co-management, the second aim of the study is to identifyfactors that could influence the conservation practices of MDCs.Thus, this study addresses the following research question: How dothe communities' perceptions of property rights regime impact thecurrent conservation regime?

2. Analytical framework

This study uses the theory of property rights to assess bundles ofproperty rights and their association with achieving coastal man-agement. Allocation of property rights is very important to ensureMDC participation in resource management under a co-management structure. This participation also involves social as-pects of participation in networks, shared values, understandingand norms (Plummer and FitzGibbon, 2006). Thus, it is hypoth-esised that conservation can be achieved by maintaining goodmangrove conditions with inclusive and participatory coastalconservation policy and management with defined property rights.Coastal conservation policies fail in many countries under the solebureaucratic management of forest departments (Badola et al.,2012; Stone et al., 2008) which could not allocate property rightsto MDCs at an appropriate level (Schlager and Ostrom, 1992). Forsustainable conservation to be achieved, these policies and

practices need to focus on the understanding of attitudes, needsand aspirations of local people (Mehta and Heinen, 2001).

‘Classical’ forms of property rights theory pay great attention tohistorical and institutional contexts (Coase, 1937, 1959, 1960).However, the ‘classical’ theory fails to explain situations whereinefficient outcomes persist. For instance, it is unable to link theroles of MDCs to mangrove conservation efforts because of ill-defined property rights (Barbier, 2006; Nugent, 2003). Whiletransaction costs and agency theories of organisation focus onoptimistic perspective of equilibrium, classical property rightstheory is better placed to handle shared ownership (Kim andMahoney, 2005).

One of themost important issues regarding resource governanceis to explore the relationship between equality of power amongstakeholders and governance outcomes (Brockington, 2002). Fromthis perspective, previous studies have applied the theory of prop-erty rights to examining sustainable outcomes for resources at thelocal level (Hayes and Persha, 2010; Irimie and Essmann, 2009). Thisconcern of inequality-related factors shaping outcomes of common-pool resources is addressed in Schlager and Ostrom's (1992) theoryof property rights. This theory can show how different mangrovedependent user groups can contribute to sustainable mangroveoutcomes without being full owners of mangroves. The theorysuggests the allocation of property rights bundles at lower butappropriate levels, such as: access, withdrawal for operational levelrights andmanagement, exclusion and alienation for collective levelrights (Schlager and Ostrom, 1992) can achieve this same outcome.Schlager andOstrom (1992) suggest that these rights, irrespective oftheir de jure or de facto nature, affect the incentive structure,participation of MDCs and outcomes observed in natural resourcemanagement situations.

Theoretical analysis of collective action was originated by Olson(1965) and Hardin (1968) who supported the roles of both gov-ernment and private property, both of which were necessary con-ditions to avoid overharvesting. This theory was not challengeduntil Ostrom (1990) proposed that common property rights withappropriate rights bundles can make self governance likely at thecommunity level, which would avoid the otherwise inevitabledeterioration into Hardin's tragedy of commons. In fact, Ostromproposed that property rights theory needed to link with otherresource-based concepts such as co-management and social capitalto enable sustainable management to be achieved (Ostrom andAhn, 2003; Schlager and Ostrom, 1992).

Central to the co-management is the distribution of propertyrights and responsibilities related to a particular resource to severalparties all of whom will gain from exhibiting reciprocal altruism(Plummer and Fennell, 2007; Plummer and Fitzgibbon, 2004).Thus, the co-management approach supports the application ofcommunity customary knowledge and joint monitoring to capillegal harvesting by providing property right bundles to local usergroups at an appropriate level (Jentoft, 1989; Matose, 2006). Co-management involves the process of sharing management de-cisions between centralised government agencies and local usergroups (Beem, 2007). In fact, co-management structures involve arange of relationships by conferring a power sharing, decision-making authority, with a people-centred governance approach forproblem solving (Berkes, 2009). However, co-management hasfailed to improve resource condition in cases where limitedattention to the nestedness of community-based organisation wasmade (Brown et al., 2007). Gough et al. (2008) identify improperattention to enhancing community social capital as the underlyingreason for such failures. Spellerberg (2001) defines social capital asrelationships among actors of individuals, groups and/or organi-sations that drive the development of mutual benefit or commonpurpose. In the absence of social capital, cooperative behaviours are

Fig. 1. A framework of factors influencing mangrove condition.

A.K.D. Roy / Ocean & Coastal Management 98 (2014) 70e7872

not sufficiently developed to solve problems requiring collectivegovernance (Adhikari and Goldey, 2010). Consequently, the theoryof property rights fails to recognise shared ownership as a validincentive structure between contracting parties, in the absence ofappropriate social capital.

Fig. 2. Map of the Sundarbans Mangrove

2.1. Framework for sustainable mangrove management

The above theory is applied in developing a framework forsustainable management of the SMF. Fig. 1 presents a framework offactors influencing mangrove condition.

Investigating MDCs' attitudes, taking into account their needs,and respecting their opinions should become a priority to enablecommunity participation in conservation of the SMF. FollowingAgarwal (2009), mangrove density is used as a proxy for mangroveconservation. This approach replicates the existing top-downconservation approaches of the BFD. Consistent with Schlager andOstrom (1992), this study translated permits as access and with-drawal, termed as operation-level rights, and management andexclusion as collective level rights. MDCs use collective level rightsto gain livelihood benefits from mangrove ecosystems whichinvariably require collective choice governance approaches tofunction as a collective endeavour. Consequently, they are morelikely to use customary knowledge and conflict resolution mecha-nisms to reduce illegal harvesting through joint monitoring in a co-management regime. Moreover, community-based conservationenhances sustained conservation through effective partnershipsand secure joint benefits by sharing government-owned restricted

Forest, Bangladesh and study area.

Table 1Household characteristics (N ¼ 412).

Variable Description Number %

Age 18e29 26 630e39 145 3540e49 134 3350e59 49 1260þ 58 14

Education level No school 133 32Primary 174 42Secondary 102 25Higher secondary/graduate 3 1

A.K.D. Roy / Ocean & Coastal Management 98 (2014) 70e78 73

and protected forestlands. Clear economic benefits encouragecommunity-based plantation to increase mangrove condition. Thisshould enhance conservation practices through building socialcapital among MDCs. Social capital promotes vertical and hori-zontal relationships between MDCs and BFD staff. Thus, percep-tions regarding the existing property rights regime andconservation of mangrove resources are assessed to understand theexisting level of conservation practices. However, it can bementioned that although natural disasters affect mangrove condi-tion, these variables are outside the scope of this study because oftheir irrelevance to the framework.

Gender Male 400 97Female 12 3

Primary sources of income Fishing and crab harvesting 349 85Wood and firewood harvesting 36 8Honey harvesting 3 1Gol leaves harvesting 22 5Other 2 1

Monthly income <Tk. 2000 16 4Tk. 2001e4000 216 52Tk. 4001e6000 107 26Tk. 6001e8000 38 9Tk. 8001e10 000 31 8>Tk. 10 000 4 1

3. Methods

3.1. Study area

The SMF covers an area of 601 700 ha, which is 3% of globalmangrove areas (Khan, 2009). This area is equivalent to 4.16% of thetotal area of Bangladesh and 40% of the total forest areamanaged bythe BFD (Fig. 2). This mangrove forest is diverse with 334 species oftrees, shrubs and plants and 269 species of wildlife including theRoyal Bengal Tiger (Roy, 2004). The United Nations Education,Scientific and Cultural Organisation (UNESCO) included it in theWorld Heritage list in 1997, and accordingly, the Government ofBangladesh declared this a World Heritage site in 1999 (Roy et al.,2012). This mangrove wetland is also included as a Ramsar Site inthe Convention of Wetlands of International Importance.

The SMF has a long history of conservation under the ReserveForest Policy since 1878 (Kabir and Hossain, 2008). Despite strictconservation policies being in place with restriction on entry, thisforest is being degraded rapidly. Tree density had decreased due tocontinuous logging (legal and illegal) and encroachment as a resultof anthropogenic pressure and natural disasters such as storms andcyclones (Iftekhar and Saenger, 2008). Shrimp farming andanthropogenic influences have destroyed 45% of mangrove wet-lands in Bangladesh (Islam and Gnauck, 2008). Barbier and Cox(2004) also found increased shrimp farming accounts for 30e70%of mangrove loss in the world. Loucks et al. (2010) predicted thatthe SMF will experience a 28 cm sea level increase by 2070, whichwill cause severe habitat loss. The main economic agents of theMDCs are: Bawalis (wood collectors), Mawalis (honey collectors),Golpata Sangrahakari (gol leaves collectors), Jele (fishers) andChunery (oyster collectors) (Kabir and Hossain, 2008). Householdsocio-economic and demographic characteristics of the re-spondents are presented in Table 1. The domination of male headedhouseholds is due to the issuance of permits to males only. The BFDdo not allow females to enter the SMF because of security problemsand social norms (Kabir and Hossain, 2008).

3.2. Sample frame and data collection

Primary data were collected using a random sampling tech-nique. The research aimed to study the populations' perceptions inthe districts (Khulna, Satkhira and Bagerhat) directly neighbouringthe protected mangrove. Out of three districts, Khulna wasrandomly selected. Key government departments are located here.Khulna district has 14 Upazilas (sub-districts) of which 6 Upazilas

Prðcommunity members agree with mangrove conservation in term

border the forest. Of these 6 Upazilas, Koyrawas selected randomly.Likewise, 2 Unions were randomly selected out of 5 adjacentUnions of this Upazila. These 2 Unions have 12 villages of which 6were again selected randomly. Lastly, out of 1705 households, 412households were randomly selected from these six villages. A lot-tery method was used for all of these random selections. The headof the family was selected as the respondent. It took between 22and 54 minutes for the respondents to complete the survey. Thesample size is more than the optimal size suggested for a quanti-tative research (Perry, 2008). Existing voter lists were used tofinalise the sampling frame. The survey was conducted betweenNovember 2010 and February 2011 using experienced interviewersand a pre-tested questionnaire. The interviews were conducted inthe local Bengali language. Data were analysed using SPSS softwareversion 19.

3.3. Statistical model

To identify factors influencing the perceptions of MDCs towardssustainable conservation practices, a discrete variable logit analysiswas adopted. Participants' views as to whether they believed thepossibilities of sustainable mangrove conservation practices underthe existing coastal management system were framed as a binary-choice model using species density changes over the last ten years.The model assumed a choice between two alternatives e an in-crease or a decrease in species density.

Let Ti represents a dichotomous variable. It is 1 if the partici-pant believes species density has decreased and 0 otherwise. So,the probability of a participant ‘agrees’, Pr (Ti ¼ 1), is a cumulativedensity function F evaluated at Xib, where Xi is a vector ofexplanatory variables and b is a vector of unknown parameters. Alogistic probability function is used to model this cumulativedensity function. The form of the logistic probability function is asfollows:

s of densityÞ ¼ PrðTi ¼ 1Þ ¼ expðXibÞ1þ expðXibÞ

(1)

A.K.D. Roy / Ocean & Coastal Management 98 (2014) 70e7874

This study builds an empirical relationship between the prob-abilities of dichotomous options [Xi(1,0)] and a set of explanatoryvariables relating to current conservation policy. The logistictransformation estimation form of this probability of communitymembers' opinions in favour of sustainable conservation practicesPr(Ti ¼ 1) is expressed in the following form:

LnPr Ti ¼ 1ð Þ1� Pr Tið Þ

� �¼ b0 þ b1X1 þ b2X2 þ…………þ bnXn (2)

Where, b0 is the “intercept” and b1, b2, b3, ………, bn are the“regression coefficients” of X1, X2, X3, ………, Xn respectively.

To estimate the parameters of the variables, a maximum likeli-hood method is used (Chatterjee and Hadi, 2006; Gujarati, 2003).Analyses of the results are based on the Wald statistic where acoefficient is 0 and has a c2 distribution corresponding at the 1%, 5%and 10% level of significance to determine the acceptance orrejection of the null hypothesis.

3.3.1. Model variables

3.3.1.1. The dependent variable. The dependent variable is sustain-able conservation practices in terms of species density changes.Thus, causal influences explain the interaction between existingproperty rights, BFD management and the level of social capitalevidenced.

3.3.1.2. Independent variables. It is hypothesised that the propertyrights categories of Schlager and Ostrom (1992), communityparticipation and social capital, influence participants' perceptionstowards sustainable conservation practices. Four rights categoriesare drawn from the Schlager and Ostrom's (1992) model: access,withdrawal, management and exclusion; and used as variables.However, as a reserve forest, the SMF is owned by the governmentfor which alienation right is not considered possible. Other factors,listed in Table 1, are extracted from the literature (Behera, 2009;Ostrom and Ahn, 2003; Salam et al., 2005) and from the pre-testpilot of the survey instrument. All the factors are selected fromthe theoretical model outlined in Fig. 1 and translated into quan-tifiable variables and described in Table 2.

Table 2Independent variables.

Variablecode

Variable title Variable description

AR Access right ‘Access’ rights to enter the SMF under presentconservation practices allocating permit licencesto the MDCs

WRD Withdrawal Right Harvest with permits as per MDC demandsWRT Withdrawal Right Completion of harvesting time as per permitsMR Management

rightMDCs' decisions about internal use patterns andtransform harvested resources by makingimprovements

ER Exclusion right MDCs' decisions about which community membercan use the SMF

CM Benefit-sharingpartnership

MDCs' intention to be involved in management ofthe SMF

CHD Customaryknowledge

Consultation between BFD and MDCs regardingharvesting and distribution

PMCI Conflict resolution Conflicts between present management andcommunity interests

SSP Plantation MDCs' satisfaction in species plantationCBO Institutional

capacity buildingMDCs' satisfaction with involvement incommunity-based organisations

DCP Corruption anddiscrimination

Discrimination or corruption in allocating permits

3.3.2. Model specificationSocio-economic and demographic characteristics of the com-

munities are important in assessing their relationships withexisting conservation practices (Infield and Namara, 2001; Macuraet al., 2011). Hence, a multiple regression model was used toexamine the relationships between socio-demographic attributesof participants and their attitudes towards mangrove conservation.The following model was estimated using the socio-economic at-tributes of the participants such as age, education, income,household size, sex, dwelling size and marital status.

Ln

�PrðTi ¼ 1Þ1� PrðTiÞ

�¼ b0 þ b1ðageÞ þ b2ðeducationÞ þ b3ðincomeÞ

þ b4ðhhsizeÞ þ b5ðsexÞ þ b6ðdwellingÞþ b6ðmaterialÞ þ2

(3)

In the following specification, all variables from Table 1 areincluded. Given the above-hypothesised factors in favour of sus-tainable mangrove conservation, the model to be estimated is:

Ln

�PrðTi ¼ 1Þ1� PrðTiÞ

�¼ b0 þ b1ðARÞ þ b2ðWRDÞ þ b3ðWRT Þ þ b4ðMRÞ

þ b5ðERÞ þ b6ðCHDÞ þ b7ðSSPÞ þ b8ðPMC1Þþ b9ðDCPÞ þ b10ðCBOÞ þ b11ðCMÞ þ2

(4)

4. Results

Respondents in 412 households were interviewed. Thefollowing sections explain the results of the study.

4.1. Analysing the socio-demographic attributes

Research findings show that MDCs' socio-economic character-istics such as: age, gender, schooling, affluence, household size,types of dwelling, marital status, property rights, co-managementand social capital can partially determine attitudes towards con-servation (Heinen and Shrivastava, 2009; Infield and Namara, 2001;Macura et al., 2011; Mehta and Heinen, 2001; Shibia, 2010).

Socio-demographic attributes (Eqn. (3)) accounted for 15% ofthe variation in the attitudes towards conservation (see Table 3).Although the goodness of fit of the model appears to be very low,similar range of values are found for other cross-section studies(Atmis et al., 2007). Overall, the regression analysis provides manyimportant insights into the socio-demographic variables that affectMDCs' perceptions towards sustained mangrove conservation.

Table 3Results of the logistic regression analysis with socio-demographic attributes(N ¼ 412).

Dependent variable: MDCs' attitudes against sustained conservation

Independent variables Co-efficient Standard error Odds-ratio

Sex �0.739 0.898 0.478Age �0.006 0.014 0.994Education �0.087* 0.046 0.917Income 0.000** 0.000 1.000Marital status 0.809 0.520 2.246Dwelling �0.522*** 0.147 0.593Household size 0.225*** 0.093 1.252Constant 1.249Model c2 38.397***Nagelkerke R2 0.150

Note: ***p < 0.01; and **p < 0.05; *p < 0.10.

Table 4Results of the logistic regression analysis (N ¼ 412).

Dependent variable: MDCs' attitudes against sustained conservation

Independent variables Co-efficient

Standarderror

Odds-ratio

Whether believe entrance into the mangroves asper their will (AR)

0.975** 0.471 2.651

Whether harvest with permits as per theirdemands (WRD)

�1.062** 0.381 0.346

Whether harvesting time completes as per permits(WRT)

1.283*** 0.374 3.608

Whether able to take decisions about internal usepatterns and transform harvested resources bymaking improvements (MR)

0.961*** 0.400 2.614

Whether able to make decisions about whichcommunity member can use the mangroves(ER)

0.950*** 0.396 2.585

Whether intend to be involved in co-management(CM)

�1.211** 0.556 0.298

Whether the BFD consults MDCs for harvestingand distribution (CHD)

0.976*** 0.414 2.654

Whether present management conflicts withcommunity interests (PMCI)

�0.873** 0.392 0.418

Whether satisfied with species plantation (SSP) 1.313* 0.745 3.717Whether satisfied with the involvement in various

community-based organisations (CBO)1.282* 0.694 3.603

Whether there is any discrimination or corruptionin giving permits (DCP)

�0.898*** 0.382 0.407

Constant �0.795Model c2 82.792***Nagelkerke R2 0.357

Note: ***p < 0.01; and **p < 0.05; *p < 0.10.

A.K.D. Roy / Ocean & Coastal Management 98 (2014) 70e78 75

Except for sex, age and marital status, all other attributes werefound to be statistically significant. Participants' household size andincome have positive effects on sustained conservation, indicatingthat the heads of the households with more dependent familymembers are more engaged in resource harvesting then otherswith lesser numbers of dependants. Many households have highdependency ratios and they harvest more, causing species densitydegradation. This result supports earlier findings identifying in-dividuals with larger families as being more dependent on re-sources for household livelihoods (Coulibaly-Lingani et al., 2009).The significant relationship of income supports the earlier findingthat higher income is expected to be associated with higher prob-abilities of engaging in resource management activities (Joshi andArano, 2009).

It was found that 74% of households are primarily dependenton this forest for their livelihoods. Most participants wereengaging in Gol leave collection. In most cases, MDCs collectthese materials illegally at the same time as harvesting fish andcrabs. Consequently, dwelling type has a significant negativerelationship with mangrove conservation. Likewise, educationhas a significant but negative relationship with conservationmeaning that MDCs with some education are better informedabout the pros and cons of conservation. This presence of nega-tive correlation between education and environmental conser-vation is due mainly to lack of alternative livelihoodopportunities and inappropriate allocation of property rights toMDCs. Hence, higher education is expected to be associated withhigher probabilities of resource harvesting.

In other studies, attributes such as age, sex and marital statushave indicated positive or negative associations with resourceconservation (Coulibaly-Lingani et al., 2011; Joshi and Arano, 2009).However, regression analysis results provide no such associationbetween mangrove conservation and these attributes.

4.2. Analysing other model variables

Results of estimating the logistic regression model (Eqn. (4))outlined the overall perceptions of MDCs against sustained con-servation. 83% of the participants do not believe that sustainableconservation practices are being followed. Majority of the partici-pants (90%) do not want wilful entrance to be allowed and 58% aredissatisfied with the permit system. Only 35% expressed the viewthat they had sufficient time to complete their harvesting as pertheir permit. Many participants (66%) expressed that they could notmake decisions regarding internal use patterns and transformationof harvested resources. Over 80% of the participants did not feelthat they could make any decisions regarding mangrove use forcommunity members. Most participants (75%) disagree that theBFD consults MDCs regarding conservation activities. 85% of par-ticipants are dissatisfied with the plantation practices of BFD staffwhilst 81% think that existing conservation practices create severeconflicts with discrimination and corruption occurring in theissuing of permits being identified by 75%. However, 67% of par-ticipants are happy with their involvement with community-basedorganisations other than the BFD. 93% expressed their willingnessto participate in conservation practices in a co-managementstructure.

4.3. Parameter estimation of the logistic regression model

Table 4 presents the logistic regression model results with testsof its 11 independent variables. The results suggest that partici-pants' beliefs in entrance to the forest as per their will (AR) issignificantly different from zero at a 5% significance level with ahighly expected positive sign. The odds-ratio indicates that

participants who show satisfaction with their wilful entrance are2.7 times more likely to show an interest in sustained conservationthan others. Harvesting with permits as per community demands(WRD) is significantly different from zero at the 5% level with anunexpected negative sign. Alternatively, participants are highlysatisfied with the harvesting time completion (WRT) which issignificantly different from zero at a 1% level with an expectedpositive sign. The odds-ratio indicates that the participants who areassured of completing harvest in their allocated time as per theirpermits are 3.6 times more likely to believe in mangrove densityreduction than those who do not have sufficient legal time in theSMF. Community concurrence towards taking decisions about in-ternal use patterns and transform harvested resources (MR) issignificant at the 1% level with an expected positive sign. Evenconcurrence of the participants to make decisions about using theforest (ER) is significant at the 1% level with an expected positivesign. MDCs' desire to be involved in co-management (CM) is sig-nificant at the 5% level with an unexpected negative sign. The BFD'sconsultations with the MDCs are also significant at the 1% levelwith an expected positive sign meaning fruitful consultationsmight reduce mangrove density reduction. Existing conflicts due topresent conservation management (PMCI) is significant at the 5%level with an expected negative sign. Community satisfaction withspecies plantation (SSP) is significant at the 10% level with an ex-pected positive sign. The odds-ratio shows that participants whoare satisfied with SSP are 3.7 times more likely to indicate thatmangrove density has reduced. Involvement in community-basedorganisations (CBOs) is also significant at the 10% level with anexpected positive sign. The CBO odds-ratio indicates that partici-pants who are involved in various CBOs are more likely to agreethatmangrove density has been reduced. Existing discrimination orcorruption under present management system (DCP) is also foundto be highly significant at the 1% level with an expected negativesign.

A.K.D. Roy / Ocean & Coastal Management 98 (2014) 70e7876

5. Discussion

In this section, the results are discussed.Significant relationships of the variables - access (AR) and

withdrawal (WRD andWRT) rights e to conservation mean that thepresent conservation system imposes strict restrictions on access toand withdrawal of mangrove resources to permit holders. Researchfindings show that this significant relationship between harvestingof resources and mangrove conservation enhances an increasingrate of harvesting by the permit holders living in proximity. Treedensity degradation in the last 5 years was estimated at 20% (Royet al., 2013b). After the loss of livelihoods from agriculture andshrimp farming by Aila, poorer MDCs have increased their har-vesting from the mangrove forest. Existing uniformed de jure har-vesting rights allocated through permits cannot ensure an equaldistribution of benefits as local community elites gain much fromtheir corrupt arrangements with BFD staff. This top-down processof formalising communal harvesting rights leaves greater de jureharvesting rights insecure and has increased the incidence ofoverharvesting. This is supported by Duchelle et al. (2011) whofound a high incidence of theft in tropical forests because of thisproblem. Hence, this pressure has increased resource conflictswhere illegal harvesters use de facto open access harvesting prac-tices. Usually, fish and crab harvesting time starts from the date ofissuing permits, but MDCs frequently fail to cover longer springtides period when harvesting is most bountiful. They undertakeunauthorised harvesting by offering bribes to BFD staff.

Regarding collective-level management (MR) and exclusionary(ER) rights, statistical analyses identified some significant factorsthat might influence MDCs' perceptions of sustainable conserva-tion. 49% of participants are not satisfied with the exclusive rightsof the BFD to make decisions about community use of the SMF.However, management and exclusionary rights are significantfactors in achieving sustainability. This can be addressed byinvolving communities in management and decision-making pro-cess by allocating appropriate property rights. However, this doesnot occur as BFD officials make managerial and technical decisionsat every stage. Under the present management structure, thesefield officials have no opportunity to incorporate local indigenousand customary knowledge in designing conservation strategies.This is also a violation of the latest forest policy (GOB, 1994) whichsupports mangrove management and conservation through activepartnerships with the local people. Thus, present conservationpolicy has failed to develop a partnership approach to achievingmangrove sustainability through accommodating the multi-dimensional interests of the MDCs in the processes of decision-making and policy implementation.

Elsewhere, the introduction of community-based mangrovemanagement in Thailand and Indonesia has made economic sus-tainability of MDCs and ecological sustainability of the mangroves(Datta et al., 2012). However, current management by the BFD hasfailed to embrace this model in the SMF.

The results of the benefit-sharing partnership significantlysupport that the lack of community involvement in managementhas a negative impact on mangrove sustainability. This neithercurbs deforestation nor promotes investment in regeneration toenhance the contribution of the mangroves towards a greenereconomy.

The results also significantly support that the use of communitycustomary knowledge is positively related to sustained conserva-tion. Past research has found that mangrove management becomesmore successful in achieving sustainable coastal managementwhere forest departments work jointly with the local people (Dattaet al., 2012; Ellison, 2012). This success is due mainly to two rea-sons: ensuring mutual benefits by allowing MDCs' greater access to

mangrove resources and by reducing forest department costs topolice resource extraction (Behera, 2009; Stone et al., 2008). Theresults significantly support the hypothesis that consultations be-tween the BFD and MDCs regarding the use of communitycustomary knowledge for resource harvesting and distribution(CHD) will have a positive impact on the condition of mangroves.However, this is currently absent due to a lack of social capitalformation at the community level. Hence, social equity and envi-ronmental objectives are notmet in a truly green economywith thecurrent top down conservation practices (Victor and Jackson, 2012).

Conflicts between present management and MDC interests(PMCI) have a negative impact on achieving sustainable manage-ment. The present cultural, political and legal obstacles encoun-tered by these communities to be involved in co-managementarrangements deny their participation in decision-making.

MDCs agree with species plantation (SSP) which will have apositive impact on sustainability. The results implied that planta-tions could be a very significant factor in offsetting the current highdegradation trend. At present, plantation programmes are con-ducted under the sole authority of the BFD. Ministry of Environ-ment and Forests failed to ensure continued healthy growth ofplanted forests due mainly to a lack of appropriate planning,implementation and monitoring procedures (Roy, 2013). That is,there is no mechanism for maintenance of high conservation valuemangroves.

For better conservation, forest departments and local MDCsshould share responsibilities and control over forest managementto achieve sustainability. It implies contractual agreements speci-fying the distribution of authority, responsibilities, agreementtenure, and the sharing of benefits (Salam et al., 2005). Presentmanagement practices of the BFD do not allow such arrangements.No community-based organisation has any contact with the forestmanagers. In Andhra Pradesh, India, social capital, proxied by anumber of active community-based organisations, is found to havea positive effect on forest growth outcomes (Behera, 2009). It isindicated that communities having high social capital are morelikely to effectively manage their mangroves by enhancing theireconomic activities in general and the management of natural re-sources in particular (Sudtongkong and Webb, 2008). Past studiesalso confirm that social capital increases institutional capacitybuilding at the local community level which results in sustainedconservation of natural resources (Pretty, 2003).

The results reveal that participants see the existing corruptionand discrimination (DCP) of BFD staff in allocating permits has anegative impact on achieving sustained conservation. The findingsalso indicate that existing conservation practices lack transparencyand accountability to other organisations especially those run bycommunities or members of civil society. This gives ample oppor-tunity to BFD staff to engage in corruption.

Although increasing social capital results in reduced corruption(Peh and Drori, 2010), MDCs perceived that the BFD had not beensuccessful in implementing its conservation policies. Governmenthas elsewhere failed in its exclusive responsibilities like: lawenforcement and conservation practice and planning because ofnot achieving a degree of trust with MDCs. The BFD staff inter-viewed do not perceive that community level institutions canovercome this problem.

6. Conclusions and policy implications

In the SMF, the absence of a shared understanding about rules ofaccess, inappropriate government regulations, along with a lack ofeffective enforcement and dispute resolution through communityinstitutions, has resulted in reduced mangrove coverage and failedto achieve sustained conservation. Hence, this study examined the

A.K.D. Roy / Ocean & Coastal Management 98 (2014) 70e78 77

factors influencing sustained conservation of the SMF. It identifiedde facto and de jure property rights at different levels as factors fordriving community attitude change towards greater mangroveconservation. These factors may be conceived as the first realisticstep towards implementing community based sustained mangrovemanagement mechanism in the SMF.

Changing conservation practices and management can be aneffective strategy for attaining sustainable resource use in the SMF.A successful conservation strategy requires a partnership betweenMDCs and the BFD. Co-management with clearly defined propertyrights and local level community institutions are needed more to-ward sustainable use. Previous research suggests that this linkbetween assured benefits and sustainability is very strong inBangladesh (Salam et al., 2005). It is evident that having mutuallyenforceable, appropriate, secure and enduring bonds between theBFD and MDCs could help achieve mangrove conservation goals.Hence, Bangladeshi policy makers may adopt measures in line withthe determinants indicating positive signs and mitigate the effectsof the remainders indicating negative signs.

There is a need for the BFD to move toward allocating appro-priate level of property rights by acknowledging MDCs' knowledgeand skills, by developing local institutions to build social capital andstopping corruption. The participation of MDCs in managementand decision-makingmust be guaranteed by implementation of theexisting and supportive national policies. Therefore, from thepresence of high correlation between performance status of theMDCs and the condition of mangroves managed by them, it can beconcluded that the identified determinants can satisfactorilyperform the functionability assessment of an MDC. This research isexpected to build mangrove based determinants for the identifi-cation of environmental standards for mangroves and makes newinroads in achieving sustainability in mangrove managementthrough MDCs' participation in mangrove conservation practices.

Acknowledgements

This research was funded by an Endeavour International PostGraduate Research Scholarship (E-IPRS) of the Australian Govern-ment awarded to the author. I gratefully thank the three anony-mous reviewers for their insightful comments. The University ofSouthern Queensland approved the research and provided ethicalclearance.

References

Adhikari, K.P., Goldey, P., 2010. Social capital and its “Downside”: the impact onsustainability of induced community-based organizations in Nepal. World Dev.38, 184e194.

Agarwal, B., 2009. Gender and forest conservation: the impact of women's partic-ipation in community forest governance. Ecol. Econ. 68, 2785e2799.

Atmis, E., Dasdemir, I., Lise, W., YildIran, €O., 2007. Factors affecting women'sparticipation in forestry in Turkey. Ecol. Econ. 60, 787e796.

Badola, R., Barthwal, S., Hussain, S.A., 2012. Attitudes of local communities towardsconservation of mangrove forests: a case study from the east coast of India.Estuar. Coast. Shelf Sci. 96, 188e196.

Barbier, E.B., 2006. Mangrove dependency and the livelihoods of coastal commu-nities in Thailand. In: Hoanh, C.T., Tuong, T.P., Gowing, J.W., Hardy, B. (Eds.),Environment and Livelihoods in Tropical Coastal Zones: Managing Agricultur-eeFisheryeAquaculture Conflicts. CAB International, Oxon.

Barbier, E.B., Cox, M., 2004. An economic analysis of shrimp farm expansion andmangrove conversion in Thailand. Land Econ. 80, 389e407.

Beem, B., 2007. Co-management from the top? The roles of policy entrepreneursand distributive conflict in developing co-management arrangements. Mar.Policy 31, 540e549.

Behera, B., 2009. Explaining the performance of state-community joint forestmanagement in India. Ecol. Econ. 69, 177e185.

Berkes, F., 2009. Evolution of co-management: role of knowledge generation,bridging organizations and social learning. J. Environ. Manag. 90, 1692e1702.

Brockington, D., 2002. Fortress Conservation: The Preservation of the MkomaziGame Reserve, Tanzania. James Currey, Oxford.

Brown, H.C.P., Lassoie, J.P., Wolf, S.A., 2007. An analytic approach to structuringcoemanagement of community forests in Cameroon. Prog. Dev. Stud. 7,135e154.

Chatterjee, S., Hadi, A.S., 2006. Regression Analysis by Example, fourth ed. WileyInterscience, New Jersey.

Coase, R.H., 1937. The nature of the firm. Economica 4, 386e405.Coase, R.H., 1959. The federal communications commission. J. Law Econ. 2, 1e40.Coase, R.H., 1960. The problem of social cost. J. Law Econ. 3, 1e44.Coulibaly-Lingani, P., Savadogo, P., Tigabu, M., Oden, P.C., 2011. Factors influencing

people's participation in the forest management program in Burkina Faso, WestAfrica. For. Policy Econ. 13, 292e302.

Coulibaly-Lingani, P., Tigabu, M., Savadogo, P., Oden, P.C., Ouadba, J.M., 2009. De-terminants of access to forest products in southern Burkina Faso. For. PolicyEcon. 11, 516e524.

Datta, D., Chattopadhyay, R.N., Guha, P., 2012. Community based mangrovemanagement: a review on status and sustainability. J. Environ. Manag. 107,84e95.

Duchelle, A.E., Cronkleton, P., Kainer, K.A., Guanacoma, G., Gezen, S., 2011. Resourcetheft in tropical forest communities: implications for non-timber management,livelihoods and conservation. Ecol. Soc. 16, 4 [online].

Ellison, J.C., 2012. Climate Change Vulnerability Assessment and Adaptation Plan-ning for Mangrove Systems. World Wildlife Fund, Washington, DC.

GOB, 1994. National Forest Policy 1994. Ministry of Environment and Forests,Government of Bangladesh, Dhaka.

Gough, A.D., Innes, J.L., Allen, S.D., 2008. Development of common indicators ofsustainable forest management. Ecol. Indic. 8, 425e430.

Gujarati, D.N., 2003. Basic Econometrics, fourth ed. McGraw-Hill Higher Education,Boston.

Hardin, G., 1968. The tragedy of the commons. Science 162, 1243e1248.Hayes, T., Persha, L., 2010. Nesting local forestry initiatives: revisiting community

forest management in a REDDþ world. For. Policy Econ. 12, 545e553.Heinen, J., Shrivastava, R., 2009. An analysis of conservation attitudes and aware-

ness around Kaziranga National Park, Assam, India: implications for conserva-tion and development. Popul. Environ. 30, 261e274.

Iftekhar, M.S., Saenger, P., 2008. Vegetation dynamics in the Bangladesh Sundarbansmangroves: a review of forest inventories. Wetl. Ecol. Manag. 16, 291e312.

Infield, M., Namara, A., 2001. Community attitudes and behaviour towards con-servation: an assessment of a community conservation programme aroundLake Mburo National Park, Uganda. Oryx 35, 48e60.

Irimie, D.L., Essmann, H.F., 2009. Forest property rights in the frame of publicpolicies and societal change. For. Policy Econ. 11, 95e101.

Islam, S., Gnauck, A., 2008. Mangrove wetland ecosystems in Ganges-Brahmaputradelta in Bangladesh. Front. Earth Sci. China 2, 439e448.

Jentoft, S., 1989. Fisheries co-management: delegating government responsibility tofishermen's organizations. Mar. Policy 13, 137e154.

Joshi, S., Arano, K.G., 2009. Determinants of private forest management decisions: astudy on West Virginia NIPF landowners. For. Policy Econ. 11, 118e125.

Kabir, D.M.H., Hossain, J., 2008. Resuscitating the Sundarbans: Customary Use ofBiodiversity & Traditional Cultural Practices in Bangladesh. Unnayan Onneshane The Innovators, Dhaka, pp. 1e98.

Kathiresan, K., Qasim, S.Z., 2005. Biodiversity of Mangrove Ecosystems, first ed.Hindustan Publishing Corporation, New Delhi.

Khan, S.I., 2009. Protecting the protectors: lessons for adaptation strategies ofmangrove forests from Bangladesh. In: IOP Conference Series: Earth andEnvironmental Science, vol. 6, p. 382025.

Kim, J., Mahoney, J.T., 2005. Property rights theory, transaction costs theory, andagency theory: an organizational economics approach to strategic manage-ment. Manag. Decis. Econ. 26, 223e242.

Loucks, C., Barber-Meyer, S., Hossain, M.A.A., Barlow, A., Chowdhury, R.M., 2010. Sealevel rise and tigers: predicted impacts to Bangladesh’s Sundarbans mangrovese a letter. Clim. Chang. 98, 291e298.

Macura, B., Zorondo-Rodríguez, F., Grau-Satorras, M., Demps, K., Laval, M.,Garcia, C.A., Reyes-García, V., 2011. Local community attitudes toward forestsoutside protected areas in India. Impact of legal awareness, trust, and partici-pation. Ecol. Soc. 16 (3), 10 [online].

Matose, F., 2006. Co-management options for reserved forests in Zimbabwe andbeyond: policy implications of forest management strategies. For. Policy Econ.8, 363e374.

Mehta, J.N., Heinen, J.T., 2001. Does community-based conservation shape favorableattitudes among locals? an empirical study from Nepal. Environ. Manage. 28,165e177.

Nugent, S.S., 2003. Mangrove Alchemy: Turning Green into Gold, vol. 21. PERC e TheProperty and Environment Research Center, pp. 10e12.

Olson, M., 1965. The Logic of Collective Action: Public Goods and the Theory ofGroups. Harvard University Press, Cambridge.

Ostrom, E., 1990. Governing the Commons: the Evolution of Institutions for Col-lective Action. Cambridge University Press, Cambridge.

Ostrom, E., Ahn, T.K., 2003. Introduction. In: Ostrom, E., Ahn, T.K. (Eds.), Foundationsof Social Capital. Edward Elgar, Northampton, MA.

Peh, K., Drori, O., 2010. Fighting corruption to save the environment: cameroon'sexperience. AMBIO J. Hum. Environ. 39, 336e339.

Perry, C., 2008. A structured approach to presenting theses: notes for students andtheir supervisors. Australas. Mark. J. 6, 63e86.

Plummer, R., Fennell, D., 2007. Exploring co-management theory: prospects forsociobiology and reciprocal altruism. J. Environ. Manage. 85, 944e955.

A.K.D. Roy / Ocean & Coastal Management 98 (2014) 70e7878

Plummer, R., Fitzgibbon, J., 2004. Co-management of natural resources: a proposedframework. Environ. Manag. 33, 876e885.

Plummer, R., FitzGibbon, J., 2006. People matter: the importance of social capital inthe co-management of natural resources. Nat. Resour. Forum 30, 51e62.

Pretty, J., 2003. Social capital and the collective management of resources. Science302, 1912e1914.

Roy, A.K.D., 2013. An Investigation into the Adequacy of Existing and AlternativeProperty Rights Regimes to Achieve Sustainable Management of the Sundar-bans Mangrove Forest in Bangladesh. University of Southern Queensland,Toowoomba.

Roy, et al., 2012. A review of the role of property rights and forest policies in themanagement of the Sundarbans mangrove forest in Bangladesh. For. PolicyEcon. 15, 46e53.

Roy, et al., 2013a. Sustainability through an alternative property rights regime forBangladesh’s mangrove forest. Geogr. Rev. 103, 372e389.

Roy, et al., 2013b. Community perceptions of state forest ownership and manage-ment: a case study of the Sundarbans Mangrove Forest in Bangladesh.J. Environ. Manag. 117, 141e149.

Roy, M.K., 2004. Designing a Co-management Model for Protected Areas inBangladesh, International Seminar onProtectedAreaManagement. InternationalSeminar on Protected Area Management. University of Montana, Montana.

Salam, M.A., Noguchi, T., Koike, M., 2005. Factors influencing the sustained partic-ipation of farmers in participatory forestry: a case study in central Sal forests inBangladesh. J. Environ. Manag. 74, 43e51.

Schlager, E., Ostrom, E., 1992. Property rights regimes and natural resources: aconceptual analysis. Land Econ. 68, 249e262.

Shibia, M.G., 2010. Determinants of attitudes and perceptions on resource use andmanagement of Marsabit National Reserve, Kenya. J. Hum. Ecol. 30, 55e62.

Spellerberg, A., 2001. Framework for the Measurement of Social Capital in NewZealand. Statistics New Zealand, Wellington.

Stone, K., Bhat, M., Bhatta, R., Mathews, A., 2008. Factors influencing communityparticipation in mangroves restoration: a contingent valuation analysis. Ocean.Coast. Manag. 51, 476e484.

Sudtongkong, C., Webb, E.L., 2008. Outcome of state evs. Community-basedmangrove management in southern Thailand. Ecol. Soc. 13, 2 [online].

Triguero-Mas, M., Olomi-Sola, M., Jha, N., Zorondo-Rodriguez, F., Reyes-Garcia, V.,2010. Urban and rural perceptions of protected areas: a case study in DandeliWildlife Sanctuary, Western Ghats, India. Environ. Conserv. 36, 208e217.

UNEP, 2011. Modelling Global Green Investment Scenarios. Supporting the Transi-tion to a Global Green Economy. United Nations Environment Programme,viewed on 7/6/2012. http://www.unep.org/greeneconomy/Portals/88/documents/ger/GER_13_Modelling.pdf.

Victor, P.A., Jackson, T., 2012. A commentary on UNEP’s green economy scenarios.Ecol. Econ. 77, 11e15.

Walters, B.B., R€onnb€ack, P., Kovacs, J.M., Crona, B., Hussain, S.A., Badola, R.,Primavera, J.H., Barbier, E., Dahdouh-Guebas, F., 2008. Ethnobiology, socio-economics and management of mangrove forests: a review. Aquat. Bot. 89,220e236.