Transparency in benefit sharing and the influence of community expectations on participation in REDD+ Projects: an example from Mount Cameroon National Park

ABSTRACT Reducing Emissions from Deforestation and forest Degradation (REDD+) projects have the potential to conserve forest, generate income and attain development targets; but the level of transparency in benefit sharing and the community expected benefits are relatively unknown. Using cluster multi-stage random sampling, data were collected from 259 respondents living around Mount Cameroon National Park REDD+ project site. Results, analysed with SPSS and NVivo, show that community engagement in project interventions is significantly influenced by the perceptions that REDD+ will promote local community development and generate individual income. Although most community members do not know how forest revenues are presently distributed (no transparency) they are expecting more developmental projects and employment. It is belief that, there will be greater transparency in future REDD+ benefits because present benefits earned from the Ministry of Forestry and Fauna for participating in the present conservation initiative are unclear and negligible. Benefit-sharing mechanism and safeguards should be transparent and clearly designed to meet community expectations to enable sustainable development of communities and prevent early failure of REDD+ projects.


Introduction
Reducing Emissions from Deforestation and forest Degradation (REDD+) is an important component of the Paris Agreement.It emerged out of the increasing recognition of land-use-related emissions and focuses on the maintenance and enhancement of carbon stock that may result in the sustainable use of ecosystems services with improved livelihoods and social benefits (Karsenty 2011).REDD+ initiatives have become increasingly popular (Minang et al. 2012;Hoang et al. 2013) with local communities viewing REDD+ as an opportunity to earn additional income (Awono et al. 2014a).During the United Nations Framework Convention on Climate Change (UNFCCC) Conference of Parties (COP) from COP-13 in Bali (2007) through COP 21 in Paris (2015), there has been a focus on the implementation of REDD+.One of the ongoing debates within REDD+ circles has been around what safeguarding policies will protect local communities where property rights are ambiguous and governance are weak with the potential for actions to threaten local livelihoods (Pham et al. 2013).It is argued that unlike previous initiatives, REDD+ has a potential to provide socio-economic and ecological benefits (Seymour and Angelsen 2010).REDD+ has sparked a renewed hope that win-win scenarios can be delivered by conserving ecosystems and improving livelihoods and the well-being of the rural poor.The widespread uptake of voluntary certification standards, like Plan Vivo, proves that forest carbon projects can deliver socio-economic co-benefits (Diaz et al. 2011;Peters-Stanley et al. 2012), yet the empirical support on the environment-poverty win-win scenario is not adequately established in forest policy (Pham et al. 2013).Larson et al. (2010) stress that REDD+ is a climate change strategy rather than a poverty alleviation strategy, but Somorin et al. (2014) argue that the multiple benefits like biodiversity conservation, poverty alleviation and socio-economic developments are vital to legitimize REDD+.According to Luttrell et al. (2012), what matters is the net benefit that can be divided into three main different types: direct financial payments, better provision of ecosystem services and improved governance, capacity, technology transfer and infrastructure.REDD+ policies and legislations could create two forms of benefit-sharing mechanism, compensating for forgone opportunities and providing incentives that enable appropriate behaviour which can all be paid up-front or dispersed over time.Benefit sharing may occur from the state to local communities as well as across local stakeholders (Lindhjem et al. 2010), regional and governmental levels.Forest projects such as the Integrated Conservation and Development Projects and Community Forest Management all inspired great community expectations, yet none of them met those expectations and resulted largely in failure to stop deforestation and degradation and/or improve livelihoods (Fletcher et al. 2016;Lund et al. 2017).There are fears that REDD+ might follow the same path if local expectations and social safeguards are not met.
In January 2013, the Forest Carbon Partnership Facility (FCPF) approved $3.6 million to fund Cameroon's Readiness Preparation Proposal (FCPF 2013).Though funds, legal and technical resources were channelled to support the design and implementation of REDD+ policy, REDD+ is still facing challenges of inadequate clarity on benefit sharing and re-distribution of forest revenues despite high expectations from communities.According to Oyono et al. (2006), the lack of transparency by both the state and local delegates in Cameroon has been experienced with other forest management projects.Kronenberg et al. (2015) state that social cobenefit will not likely occur unless good institutions are in place to protect the rights of forest-dependent local communities.Good constitutional and collective institutions enhance local performance in incentivebased conservation initiatives; and top-down policy making plays a vital role in supporting communities to establish these institutions (Collen et al. 2016).Therefore, top-down strategies should be complimented with bottom-up institutional development to ensure REDD+ success.REDD+ has the potential to provide better incentive for the community to engage in forest management, so adjacent local communities to the park must be invited and sensitized to engage in REDD+ activities (Munishi 2013).But, the uncertainty in REDD+ policy and the market for forest carbon have made forest managers to delay sharing information, consulting communities and avoiding to use the term REDD+ to prevent raising expectations of local community's concerns on potential carbon payments and co-benefit potentials (Sunderlin and Sills 2012;Awono et al. 2014a).
The Cameroon Readiness Preparatory Plan recognizes the importance of engaging local communities in REDD+ activities, but there is no detail as to which level they can engage in revenue management or benefit sharing, with no clear definition of communities' roles.REDD + national benefit-sharing mechanisms should be shaped by legal and regulatory framework with clear objectives and targeted beneficiaries (Luttrell et al. 2012).However, Cameroon does not yet have any legislation defining carbon rights or benefit-sharing mechanisms at the national or sub-national levels.Cameroon is further challenged with corruption and management of REDD+ finances, and analysis has shown that previous finances for forestry programmes have not been effectively managed (Pham et al. 2013).Difficulties in obtaining Forest Environmental Programme Funds and ignorance of procedures of disbursement have been identified within the Ministry of Forestry and Wildlife in Cameroon (Dkamela 2011).Also, uniformity in benefit sharing tends to neglect local transaction differences and opportunity cost, leading to inequitable benefit sharing among local stakeholders (Costenbader 2010).Somorin et al. (2014) argue that REDD+ benefit-sharing mechanisms could be designed upon lessons learned from community forest and forest taxation schemes.However, the dominant form of tenure in Cameroon is forest concessions where fees and royalties are paid to the state before granting concession to exploit forest resources which are then redistributed across national, sub-national and community levels.According to the Cameroon1994 Forestry Law, 50% of all forest revenue goes to the state, 40% to rural council and only 10% plus village tax go to local communities adjacent to forest concessions (50:40:10) (Morrison et al. 2009).This ratio of 50:40:10 keeps local communities very far from adequate and equal benefits.Also Cameroon 1994 Forestry Law, section 50 and 61(3 and 4) requires proponents to establish developmental projects that provide social amenities to local communities such as roads, schools and hospitals.Scaling-up the benefit-sharing mechanism of forest revenues may be efficient, but unclear payment schemes and high level of under-payment are evident in Cameroon.Pham et al. (2013) stress the need to build on existing benefit-sharing framework to reduce cost of establishing and operating new REDD+ benefit-sharing institutions that could gain national political support.Lessons learned from benefit-sharing schemes in previous forest projects highlight the importance of clarifying and securing tenure rights for equitable and effective benefit sharing of REDD+ forest revenue (Hite and Barnett 2011).This paper, therefore, assesses the pre-existing benefit sharing within the national park to draw recommendations on what should be improved when designing and implementing the Mount Cameroon National Park (MCNP) REDD+ benefit-sharing scheme.It investigates how community expectations influence community participation in Mount Cameroon National Park (MCNP) conservation projects; and the level of transparency in present community benefits within MCNP conservation projects.Knowledge of community expectations and socio-economic outcomes of present conservation projects is vital in supporting policy debates towards the design and implementation of sustainable REDD+ projects.

Background to the mount cameroon REDD+ projects
Mount Cameroon supports humid forest that experienced a total forest loss of 46.2% between 1987 and 2010 in Cameroon; has clear geographical boundaries for project intervention; and has been a target for implementing REDD+ (Sunderlin et al. 2010).A study carried out by the Cameroon Mountain Forest Conservation Foundation in 2001 to evaluate Cameroon Mountain potential for carbon programme showed that carbon sequestration is possible through re-growth and forest conservation on 4300 ha of forest area and the MCNP TREMA database is made up of geo-referenced forest inventory with 20,000 data points from 300 samples with built in functions to derived indices of 'bio-quality' which can be modified for carbon management purposes (EcoSecurities 2002).

History of MCNP and previous management schemes
The first Mount Cameroon conservation project started from 1998 to 2003 and was managed by the German International Cooperation (GIZ), Minister of Environment and Forestry, and the World Wide Fund for nature (WWF) (Awono et al. 2014a).In 2006, Mount Cameroon came under the Program for Sustainable Management of Natural Resources in the Southwest Region (PSMNR-SW) which supported conservation and livelihoods initiatives within this region.The Mount Cameroon National Park (MCNP) REDD+ project was latter launched in 2010 by the Cameroon government and is managed together with the GIZ under the PSMNR-SW.The project is aimed at conserving biodiversity, reducing deforestation and forest degradation, improving community livelihoods through employment, income generation and the provision of development projects.These community projects include farm-tomarket roads, pipe-borne water, schools, hospitals, crop preservation facilities and a range of income generating projects like nursery for improved palm and cocoa seedlings, domestication of Non-Timber Forest Products (NTFP) and processing of palm oil and cassava (Moki 2010).Before the establishment of MCNP park, participatory forest management activities were carried out by communities such as protection of specific plant species, inventory of special forest products, enactment of forest by-laws with clear use right for specific products among others; but the establishment of this park changed this community scheme to a centralized forest governing REDD+ scheme where local participation is limited to manual labour, member of committee and reporting of illegal activities within the park; thus resulting to conflicts and rendering local stakeholders powerless, as communities are reluctantly losing their land rights and access to forest and forest resources (Awung and Marchant 2016).

Institutional set-up, process and main measures within MCNP
The co-management approach of MCNP incorporates 41 peripheral villages; these are divided into four geographical clusters (Bomboko, Buea, Muyuka and West-Coast) based on natural boundaries and livelihoods differences (Figure 1).Agriculture is the main source of livelihood in all four clusters, but the second source of livelihood in each cluster differs -Bomboko (timber exploitation), Buea (salaried workers), Muyuka (hunting) and West-Coast (fishing).A cluster platform is then established to coordinate information exchange between the clusters; that is, the information hub between the National Park managers and the 41 park villages.The benefit-sharing schemes, the planned/proposed village development projects and the planning, implementation and evaluation of the park activities are all discussed at the cluster platform level.One facilitator from each cluster liaises between the park managers, the villages and the cluster platform.Together with the park managers, they organized the cluster platform meetings.The facilitators discuss the agenda with the park managers and the village suggestions that are collected during regular village meetings.Three delegates (at least one woman) are elected by the Village Forest Management Council from each village to represent at the cluster platform meeting.Negotiation and signing of the Conservation Development Agreement (CDA) is carried out by park authorities and the three village delegates from each community.Awung and Marchant (2017) show that community delegates within MCNP are granted access to decision making forum (cluster platform), but their opinions do not count nor influence the planning, implementation and management of the REDD+ project; therefore, their inclusion is just a means to legitimize REDD+ rather than an empowering process to address issues of social justice.This topdown strategy deters voluntary community engagement which now depends mostly on the benefits or economic dimension of the project.The success of REDD+ depends on full and effective participation of local communities.Designing or adapting benefit-sharing systems to suit community expectation will serve as an opportunity to address the gaps in attaining social co-benefits and success of projects.The CDA defines roles, responsibilities and states incentives or benefits for community collaboration in park activities.The CDA is also used to settle disputes and get villages 'on board' for comanagement.The park managers, together with the conservation partners, proposed a six-month work plan for each cluster.This work plan, which is called the Collaborative Management Activities (CMA) is discussed, complimented and adopted by the members of the four clusters.These activities are performance-based rewarded and include park boundary tracing, reporting of illegal park activities and encroachment, tree planting and sustainable harvesting of NTFPs especially Prunus africanacommonly called pygeum.Pygeum is an Afromountain rough bark hardwood tree use for the treatment of prostate cancer and is a major source of income for forest dwellers and local enterprises (Awung and Marchant 2016).About 90% of exploitable prunus is found within MCNP with five management cluster for a five-year rotatory harvesting plan which is fully integrated into the MCNP management plan and through sustainable harvesting, the resources are able to regenerate and increase in quality and quantity (Eben 2011).
The benefits expected by the 41 communities involve in the co-management of MCNP initiatives include Conservation Credit (CC); Conservation Bonuses (CB); income through direct employment and community development projects.All these are implemented in partnership with the conservation partners such as the German International Cooperation (GIZ); World Wide Fund (WWF); Wildlife Conservation Society (WCS); Ministry of Environment, Nature Protection and Sustainable Development (MINEPDED); and the Ministry of Forestry and Fauna (MINFOF).A community earns CC and CB through participation in the CMA of the MCNP projects such as forest conservation, tree planting, boundary demarcation, forest guards, reporting illegal activities, sustainable harvesting of Prunus, amongst others.The CC is 20% of the income earned by members of each community during its CMA.This amount is meant to contribute to a 10% compulsory village contribution towards their development projects such as roads, schools, markets, pipe-borne water, hospitals and toilets.Conservation-bonus is a virtual voucher earned from participation in CMA and local communities' commitment to co-management of the National Park and is only use for village development projects that falls in line with the conservation objectives stipulated by the conservation managers.

Methods
A pilot study was conducted in one park village in each cluster (five months before main field survey) to identify gaps, field challenges and to enhance acquaintance and acceptance of the researchers by the community members.The main survey was conducted between October and December 2013.The basic sampling unit was household and park villages with a population between 100 and 1000 inhabitants.Cluster multi-stage random sampling was used because it permits subsequent sampling and each cluster's characteristics can be estimated (Daniel 2011).From each cluster, three villages without common boundaries were randomly chosen -Buea (Likombe, Bwassa and Bova 2), West-Coast (Bibunde, Njonji and Bankingili), Muyuka (Mundame, Liola-Buea and Bavenga) and Bomboko (Afolofo, Kotto 1 and Bomana) (Figure 1).In each cluster, at least 17% of households, 34% of each gender and 10% of each age-group participated (Table 1) in the face-to-face questionnaire with multiple-responses, ranking, closeended and open-ended.
All together 42 face-to-face interviews and 17 consultations were recorded with key proponents from the international -GIZ; national -MINEPDED and MINFOF; sub-national -South West Region delegation of MINFOF (MINFOF-SWR); local initiative group -Mount Cameroon Prunus Common Initiative Group (MOCAP); and the local community (LC) members.Familiarization with members of communities were accomplished by participating in village activities such as farming, fishing, singing, dancing and story-telling at the village squares.These socialization and field observations enable collection of data without affecting the feelings, attitudes and behaviour of members of community (Kvale and Brinkmann 2009).
Data were analysed through a mixed-methods approach which integrates findings from both quantitative and qualitative analysis to provide a better understanding of results (Bazeley 2012).Validated data from questionnaires were coded and imported into SPSS for quantitative analysis (t-test, Mann-Whitney U test, linear regression model as well as independent-samples Kruskal-Wallis test and Jonckheere-Terpstra trends test).The resulting quantitative results show differences between clusters; and between local participants and non-participants (horizontal analysis).Interview transcripts were qualitatively and thematically analysed using (NVivo) and relevant information were incorporated into quantitative results.Qualitative results provide different views within different levels of governance involved in MCNP initiative, from international, national, sub-national, local groups and local communities (vertical analysis).Relationship was examined, relationship investigated and models established.
Expectation of income generation influences participation in MCNP conservation projects.Result shows that community engagement is significantly influenced by expectation that MCNP projects will generate personal income (U = 9,742, p < 0.001, r = 0.33) with an effect size of 33% in all  Respondents in Bomboko (timber exploiters) and West-Coast (fisher men) clusters agreed that the MCNP will promote development (Figure 2) and generate income (Figure 3).While in Muyuka (hunters), respondents slightly agreed that local development will be promoted (Figure 2) and those in Buea (salaried  workers) slightly disagreed that the projects will generate income to community members (Figure 3).This may be due to differences in secondary livelihoodsthey are more salaried workers in Buea cluster who may consider any income from MCNP as negligible.
Therefore, a 'one-size-fit-all' approach should be avoided because of these differences in perception/ expectations.However, those with greater expectations of income generation and local development are more engaged in MCNP activities.
Significant linear regression models presenting how expectation of local development (A) and income generation (B) influence participation (P) is explained by Equation (1):

Transparency in benefits sharing from MCNP forest revenue
The present benefit-sharing scheme within the MCNP which may be utilized for REDD+ scheme is not transparent because only 15% of respondents are aware of local developmental projects carried out from forest revenue (Figure 4(a)) and those who did not know, blamed the absence of developmental projects to lack of funds (56%) or unknown reasons (44%).Again, up to 53% of respondents did not know how forest revenues are distributed (Figure 5 (a)).These development projects range from provision of zinc to roof local market places, pipe-borne water, to construction of public toilet, community hall and schools/classroom (Figure 4(a)), but only 2% of respondents could state how much was given for the project(s) (Figure 4(b)).While 7% of respondents in Buea know that 100,000-250,000 CFA has been given for community hall/chair and/or construction of public toilet, only 3% of respondents in Bomboko know that less than 100,000 CFA has been given for construction of toilets and/or pipe-borne water; and respondents in Muyuka and West-Coast are not aware of any projects (Figure 4(b)).

Local community expectations
Despite the unawareness and/or lack of transparency in benefit sharing, 97% of respondents are seeking for community developmental projects (46%) and employment to generate income (36%) as the major ways to ensure equitable revenue distribution, with employment expectation highest in Bomboko (Figure 5(b)).Participation could be impaired if these developmental and financial (employment) expectations are not met.The absence of basic necessities like electricity, tap water, health centres, market and roads in some villages; and the lack of formal/informal credit institutions to generate capital for investment (Figure 6(a)) have also deprived local communities of improved livelihoods.
The most preferable community expectation from MCNP is employment (for income generation) except in Muyuka where the community members' preferred electricity supply (Figure 6(b)).This is because there is no electricity installation in Muyuka (Figure 6(a)).To enable better livelihood, communities are expecting that MCNP-REDD+ projects will provide the following 10 top benefits (choice made through ranking): (1) Employment through which income will be generated; (2) Finance for establishing small businesses; (3) Provide them with pipe-borne water;

Qualitative results
Interviewees talked mostly about benefits, community, forest, conservation and money from where two themes were established: community development and/or benefits and park activities linked to village forest conservation credits.There is yet no clear benefit-sharing mechanism within MCNP.MOCAP respondents were concerned that although a management plan has been approved for Prunus harvesting, money from the block harvested in 2010 has not yet been re-distributed to local communities despite the strenuous harvesting process.
GIZ: 'At the moment there is no definite benefit sharing structure set-up….There has been a management plan approved for Prunus harvesting….
They have harvested just five blocks (2010)….But for now, nothing has been given to the community because we want to make this an integral part of the conservation development agreement.But the money is there, … about 7,000,000 CFA.' GIZ: '…MOCAP is representing only 33 out of 41 communities and we want it to represent all the 41 communities to avoid conflicts… The expected revenue is not as high as they expected in the beginning because they did an inventory and the prognosis fell below expectation… The harvesting is not fast enough as they thought and it is strenuous….They cannot harvest it as before because it was unsustainable, but now they have to plant the trees and harvest with a specific method which we are training them and only specific trees are harvested.' Ecotourism too is being re-structured and there is hope that money from tourism will also be redistributed to communities.GIZ: 'Mount Cameroon Equatorial Organisation … collect fee from visitors going up the mountain and a stakeholder fee which are supposed to be distributed to the community … .It is now dormant because the tourism sector is being restructured…' We are still working on a mechanism how all the benefits can be to the entire community.'This benefit-sharing mechanism is meant to enable the indigenous people to link benefits to the park's conservation initiatives.As reported by the GIZ Technical Assistant and MINFOF-SWR respondents, 50% of the bonus is based on village contribution to conservation efforts (illegal activities are regularly reportedcrosschecked with results from guard patrols); the second portion of the bonus (25%) will depend on the level of collaboration of the village (work spirit and efficiency to perform CMA); and payment of the last portion of 25% will depend on the overall cluster performance (regular cluster platform meeting attended by all members and coordinated conservation efforts).
Revenue distribution for local development is gained both directly on individual basis through employment to work in the park and earning an income (e.g.Prunus africana harvesters are paid 150 CFA/kg fresh barks) and indirectly through conservation credits and bonuses.Local respondents argued that, though some development measures and agricultural training are on the way to generate income like nursery, no tangible benefit could be traced because there is yet no market for the nursery.
GIZ: 'There are village development measures … to assist the communities in some specific development areas like small-scale water supplies and income generating activities like … multi-purpose nursery projects and cocoa improvement projects … (aimed at) improving the land and the output of the current land and not expansion of any agriculture.'LC: 'They trained us on how to nurse plants and we did produced many nurseries, but there is no market for the nurseries … we have planted the nurseries in our farms until there is no more space.'National: 'We are trying to see how local communities will be involved in the benefits.'As revealed by MOCAP and MINFOF-SWR only of revenue from Prunus harvesting are dedicated to the village development fund 7) and only 150 CFA/kg is paid for Prunus harvested which is lower than the 375 CFA which was being paid before the establishment of the park.With only 43% of the revenue set aside to pay harvesters, buy field equipment and medication (Figure 7), it is likely that only about 35% of the actual revenue are paid to harvesters.This low level of benefit has impeded engagement because 'payments are ridiculous'.
Sub-national: 'The benefit sharing mechanism decides how much goes to the harvesters, MOCAP and park management (Figure 7).This money is used as village contributions for different projects …, it should be around 40 million CFA … we make sure that the funds (16%) are used to the satisfaction of the villagers (about 7,000,000 CFA).' MOCAP: '…Some (forest revenue) are paid as salaries, some go to developmental projects that are designed by the community (Figure 7)….They (local community) plant in plantation.When the Prunus are mature; they harvest and sell to us every five years.'MOCAP: 'The 33 communities have just seven harvesters and they are paid just 150 CFA/kg when they carry a harvest from up the mountain right down.So people don't want to work as harvesters because payment is ridiculous.Everybody is complaining about the price.Before, they used to sell for more than 350 CFA/kg, but now the price of Prunus is determined by the park committee.' Though MOCAP respondent claimed to allocate money to local communities' projects, only one local respondent revealed that MOCAP has given them money to assist with building community hall and the amount seems negligible (250,000 CFA).
MOCAP: '…the money is spent on projects that are designed by the community.'LC: 'MOCAP has helped by giving us some money about 250,000 CFA to mould blocks, but the contractor started the project and did not complete it.' Communities were concerned that, though they have learned a lot about conservation, they are not employed to work in the park because employment is based on General Certificate of Education (G.C.E.) Ordinary Level certificate and blamed their dependence on forest resource to lack of jobs, school and other alternatives to improve livelihoods and well-being.the forest, we will do that well any other because we know that securing the will yield benefits which will help develop our community.'LC: 'My dad worked in the forest and harvested from the forest.I am doing the same thing now … if I had a different job I would not be doing same … maybe our children will also grow up to work in the forest and harvest from it.'LC: 'We ask the national park to build schools for those around the mountain areas, but they refused.'LC: 'We are suffering, no school, we don't have pipe born water and no health centre.We are suffering a lot.' LC: 'The youths have to work.If they employ the youths, they will work and feed themselves… But if we have no jobs and we are hungry, we will go directly to the forest to hunt or harvest food.' LC: 'They are employing based on G.C.E certificate, but that is wrong because common sense is more important than certificate … If I was working somewhere, I would not have time for hunting.If my kids have schools they would be occupied with school work … Do we have to die because we are not educated or what?If they go up to the mountain and an animal attacks them, is it the certificate that will save them?We are the ones they will call for help because we know how to tame these animals.'Benefits are also re-distributed through a performance-based approach by allocating conservation credits and bonuses to motivate local communities to engage in the initiative.But allocated amount is often negligible as compared to effort needed (Table 2).Each year, a village is supposed to be allocated a minimum of 200,000 CFA for Collaborative Management Activities, though, members of local communities know the amount to be 125,000 CFA.Conservation Credit (CC) is designed to facilitate the 10% compulsory contribution from each village towards their developmental measures.CC is additional credit generated which is actually 20% of money earned from Collaborative Management Activities.If a village earns 100,000 CFA, it will generate an additional 20 CC for the village (1000 CFA = 1 CC).With a minimum of 200,000 CFA allocated to each village, it can earn at least 40 CC per year.The average amount allocated to village projects is three million CFA.Therefore, for a village to carry out a project like this, it is obliged to contribute 10% which is 300,000 CFA or 300 CC.If it has about 200 CC accumulated say over five years, then it would be left with 100,000 CFA to contribute in cash, or in kind; or wait for two more years (total of seven years) to gain the required 300 CC.This time frame of seven years, delays development and development delayed is development denied!Sub-national: 'We have introduced the conservationcredits and conservation-bonuses in this collaboration management approach….Conservation-credit is a virtual voucher aimed at motivating the communities to contribute and respond to conservation needs of the MCNP management.It doesn't support illegal activ-ities….For conservation credits, we calculate 20% of what the park spends on any activity in the village where any member of the village is participating … If 300,000 CFA was spent on village "A" (income) … 20% of 300,000 CFA is registered for that village and the money is not given to them directly.We spend the money on water projects and other development projects for that community.The person in charge of these funds makes sure that the community contribute at least 10% of the cost of the project in kind or in cash.Say if a water project cost 1,000,000 CFA, the community will contribute 10% of that cost in cash or kind by digging the pipelines where the pipes have to pass.So if they have their conservation-credit say 100,000 CFA, they will not contribute anymore.' The full amount of the Conservation Bonuses are made available to park villages that report illegal activities regularly and carry out the Collaborative Management Activities properly; but a village community can lose its bonus (partially or completely) if the mentioned conditions are not fulfilled.The conservation bonuses rewarded can only be accessed through a Conservation Management Unit, where a village proposes a village development project in line with conservation objectives.The village development fund project is supervised by the park managers and not the community head, therefore, the community does not freely decide on how to use their bonusestheir choices are highly influence by the park managers.
Sub-national: 'The conservation-bonus goes to the whole village for participating in the conservation activ-ities….An assessment criterion determines what percentage of this conservation-bonus they would receive (see Table 3) … If an elephant is killed today and after one week another elephant is killed within that same area, you are automatically going to lose all … even at five O'clock; people call me to report that there are people with engine saw going to cut trees around.The use of this bonus must be for a project that is conservation friendly….We are not going to approve buying of an engine saw … We will approve buying of desks …, give the money to the president of the conservation committee of that village, and then come back after three weeks to inspect if the desk are there.So this is how it is designed to function and the strategy or approach is explained to the community.'Most local respondents are not happy with the benefit-mechanism because conservation credits bonuses are too small and be reduced to nothing in the case of illegal activities that are mostly carried out by 'outsiders'.In addition, they are neither employed as forest guards, nor given the authority to effectively stop encroachment.
LC: 'We are not motivated.They (park managers) also said if they want to give us money or benefits, they will come and check in the forest to see if people are harvesting there.Some outside people, sometimes, come and cut down trees and export them to other countries, but when they come and see that, they will say we are the ones who cut the trees.So it is better for them to give us authority because we know who is doing any illegal activity better than them and if we see someone not from this village who is cutting trees, we can interrogate the person to know if he has a license or not.' LC: 'When there is any illegal activity in the forest, they will reduce the 125,000 CFA that was supposed to help the village.But if some of the villagers were forest guards, they will make sure that no illegal activity is carried out in their forest so as to save that money for the benefit of the village.'LC: 'They want us not to harvest from the forest because they want to preserve it, but at the same time they are not supporting us to live.We can sell what we get from the forest and provide for our needs …, how are we going to get money to buy other provisions?So they really have to help us and the 125,000 CFA allocated for us is too small.'Apart from securing conservation credits for community development, local participants are also demanding individual payment for activities carried out especially in difficult terrain and under hard labour.
LC: 'They asked me to accompany someone to the for-est….On the way he could not walk anymore, I carried him on my back and carried his bag also until we reached the place we were going … When we came back from visiting the mountain, the man did not pay me, he said payment will be counted as part of the conservation-credit.The credit is for the whole village and not for me, what I did was an individual task.Moreover the credit is about 125,000 CFA a year and that is very small … They expect us to do jobs on the mountain on rocks with no insurance … if I die what will my family and I benefit?If I had killed antelopes and carried them down here, I would have sold them for maybe 160,000 CFA.But I used the energy to carry a man on my back for nothing … What I am saying is that the forest officials are paid far higher than what they give us just because we don't have certificates, but we are forced to do all the work.Unfair!'

Discussion
Though MCNP REDD+ conservation project is still at an early stage with no clear define REDD+ benefit-sharing scheme, the pre-existing benefit-sharing mechanism might not provide the benefits expected by communities as it negatively impacts community livelihoods by jeopardizing communities' resource rights and counteracting poverty alleviation objectives through low income, employment and negligible development initiatives; and these could hinder community engagement.knowledge and understanding of benefits, including obligations (buyers/sellers) and above all the willingness to pay.Therefore, REDD+ needs all stakeholders, especially local communities to better understand the benefit-sharing schemes and adopt adaptive management approach to address opportunity cost as well as build capacity and trust among stakeholders to ensure more transparency.

Financial and community development expectations
We show direct relationships between community participation (P) in MCNP activities and expectations that REDD+ will generate income (B) and enhance community development (A) [P = 0.615 + 0.353A + 0.118B].We join Munishi (2013) to state that community participation in REDD+ and forest management initiatives significantly correlate with increase in benefits and incentives for forest management.Therefore, unresolved challenges need to be address and the cracks between expectations and realities need to be renegotiated.Employment for income generation, finances for small businesses, pipe-borne water, training on animal husbandry, hospitals, schools and roads are the major benefits expected by local communities.The MCNP project has elaborated the village development plan and implemented few socio-economic infrastructures like pipe-borne water, community halls and public toilets.Some community members feel that benefitsharing unit should be household or individualized rather than community projects because the income generated from forest were used to solve unique family issues, so direct payment should be directed to those providing services to ensure sustainability of REDD+.It is hard to prove that everyone will benefit from the projects (exclusion of external users), and if adequate compensation will enable provision of alternatives livelihoods and feed the growing population.We join Krause and Loft (2013) to question the poverty alleviation attainment in incentivized conservation stewardship programs.Even if benefits are shared to customary rights owners, there are additional claims from migrants who are also members of these communities.
The lack of adequate benefits may also be due to unclear tenure.Delgado-Serrano (2017) shows that the lack of indigenous-based conservation strategy and legal recognition of customary land rights are some of the factors influencing the trade-off between conservation and development.The lack of access and restricted rights over land and trees hinder MCNP communities from influencing benefit sharing, thereby, removing them off main financial streams.We join Miah (2014) to state that trade-offs between forest conservation and local livelihoods must be well understood and negotiated before implementing REDD+.The provision of alternative livelihoods, employment, finances for small business and developmental projects may enhance participation and improve REDD+ success.We argue that, REDD+ has create an 'economy of expectation ' (Lund et al. 2017) which may result to a conservation fads (Fletcher et al. 2016), if the hopes and expectations of rural communities are not met.

MCNP benefit-sharing framework
There is still no benefit-sharing mechanism for REDD+ in Cameroon (Awono et al. 2014b), but members of local communities understand that they have to conserve the forest while expecting benefits through employment, conservation-credits and conservationbonuses.While waiting for REDD+ payment, park managers have negotiated CDAs; however, effective benefit-sharing mechanism depend on transparency, funds availability, accountability and capacity to manage finances at all levels (regional, national, subnational, community, etc.).The insecure tenure systems and national control rights have limited the scope of local barks) and this action has discouraged communities from engaging.
Incentives are offered to communities through a participatory-based approach as communities in deciding benefit sharing of REDD+ revenue with park services deciding on price of Prunus for farmers Prunus africana harvesters are paid 150 CFA/kg fresh compensation for their support and limited use of forest in the form of jobs and finances for community projects.Some community members are recruited for boundary demarcation, while some of them work as security guards trapping down defaulters and reporting forest activities that result in conflict within MCNP.Community heads report unlawful entrance or activity in the forest, creating conflict.the most disadvantaged members of the community are excluded because they neither get hired nor participate in community projects.Most community members are not being employed because of lack of G.C.E certificate (illiteracy) and that is an infringement on their rights.Mpoyi et al. (2013) show that, though some jobs are created, qualified positions are filling-up by external professionals because of lack of local capacity/skills relevant to forest management.There is a clear need for community education and training to build adequate skills and capacity to engage at all levels of interventions and also REDD+ implementation strategies should be adapted to incorporate local forest management practices.
The MCNP managers are using this benefitsharing scheme to manipulate, marginalize and force these vulnerable poor communities to comply, in expectation of negligible benefits and this is an infringement of community's rights.We argue that, the potential long-term REDD+ benefits of improved forest governance, transparency, accountability, community infrastructure, job creation, greater adaptive capacity of communities and development of social and human capital amongst others (Chhatre et al. 2012;Strassburg et al. 2012) are not yet evident in the MCNP initiative.Although there is limited understanding about the link between poverty determinant in local communities and deforestation, the sustainable management of natural resources and improved livelihoods of targeted communities should be considered so that predicted benefits could match the needs and community expectations.Therefore, a re-think of fair and transparent financial mechanisms to attract indigenous peoples and forest stewards in rural communities is urgent.It is also crucial to secure REDD+ benefits for community members if forest conservation, income generation and developmental potentials of REDD+ are to be realized.Above all, REDD+ should support forest stewardship activities of local communities providing benefits such as strengthening of community and resource rights, empowering local community institutions to engage in decision making, improving income through employment and engaging in community developmental projects.

Conclusion
Participatory forest management schemes that was earlier practised within communities around MCNP has changed to a centralized forest governing scheme with low and unclear benefit opportunity for communities.The MCNP REDD+ initiative is expected to create employment and bring community development projects, but the existing size of benefits from previous forest projects is almost negligible and these could impair local participation in the conservation initiatives if expectations are not met.The effectiveness of REDD+ depends on how cost and benefits are shared and if incentives are sufficient enough to enable behavioural changes and policies.A structured national benefitsharing mechanisms are needed to ensure more sustainability of REDD+ projects, so there is need to improve financial transparency, information sharing, capacity and coordination amongst all stakeholders.Park managers should focus on social and livelihood aspects of REDD+ to provide strong incentives.
Community rights, social safeguards and adequate benefits from REDD+ incentives should enhance health, biodiversity and livelihoods of communities.Therefore, community forestry will be an essential part of sustainable REDD+ projects as a bottom-up approach will understand and consider the needs and concerns of local communities; and community members could manage it to directly generate income, develop communities and produce food for improve livelihood and well-being.We follow Fletcher et al. (2016) to call conservation managers to re-think and together with communities re-design a transparent benefit-sharing mechanism in place before implementing REDD+.Such re-thinking and re-designing should accommodate patterns and lessons learned from previous forest projects to support REDD+ in achieving its biodiversity conservation, income generation, employment, improved livelihoods and community development goals.

Figure 1 .
Figure 1.Map of Mount Cameroon National Park (MCNP) showing the cluster conservation zone and participatory villages involved in MCNP REDD+ project.FMU: Forest Management Unit; BFR: Bomboko Forest Reserve (Awung and Marchant 2016).

Figure 2 .
Figure 2. A Kruskal-Wallis plot showing variances in expectation that MCNP project will promote local development between clusters (a) and a t-test showing how this expectation influences participation in MCNP activities (b) and within each cluster (c).

Figure 3 .
Figure 3.A Kruskal-Wallis plot showing variances in expectation that MCNP project will generate income between clusters (a), and a t-test showing how this expectation influences participation in MCNP activities (b) and within each cluster (c).
(4) Training on animal husbandry and breeds; (5) Hospitals or health centres; (6) Schools to enhance education and technical skills; (7) Establishment of markets to sell surplus food and forest products; (8) Better access of motor-able roads; (9) Community halls for community gathering, discussion of community issues, socializing; and (10) Electricity supply to process and store food (Figure 6(b)).

Figure 4 .
Figure 4. Percentages of knowledge of projects carried out in MCNP clusters (a) and the amount given for projects (b).

Figure 5 .
Figure 5. Percentages of knowledge of present distribution of forest revenue (a) and expected ways to ensure equitable distribution of revenue (b).

Figure 6 .
Figure 6.Percentages of knowledge of present infrastructure within MCNP clusters (a) and expected developmental projects from MCNP projects (b).

Figure 7 .
Figure 7. Revenue distribution from the sale of Prunus africana (summary from interview with MOCAP and MINFOF-SWR respondents.Libation here refers to sacrifices offered to the ancestors of the communities).

Table 1 .
Demographic and statistical information from questionnaire survey.

Table 2 .
Mann-Whitney U test showing how participation is influenced by expectation of local development (a) and income (b) with their corresponding effect sizes.

Table 3 .
Summarized conditions for acquiring conservation-bonuses (CB) within MCNP conservation clusters as explained by GIZ and MINFOF-SWR respondents.+50 Respect minimum park patrols effort, keep records of illegal activities and deliver seized items to the ministry of Forestry and Fauna Report of class 'A' animal's poacher by patrol team −25 to −100 Village forest management unit should report all illegal activities Eco guards report class 'B' and 'C' animal poacher (commercial purposes) −10 to −50 Patrol team meet hunter with arms or killed animal