«Dissertations Forestales 149 Forest law compliance in the High-Forest Zone of Ghana: an analysis of forest farmers’ livelihoods, their forest ...»
A literature review of forest communities’ livelihoods and livelihood assessment methods was conducted. Data were collected using a structured questionnaire (Appendix II). Closedand open-ended questions were used to identify (i) the livelihood-related issues covered in the VPA, (ii) the social groups whose livelihoods are most likely to be affected by FLEGT VPA implementation, and (iii) the potential impacts on the communities’ livelihoods. In addition, using an open-ended question, respondents were asked to provide general and more descriptive information about the issues, including (i) the concept of livelihoods and poverty alleviation within the FLEGT VPA in Ghana; (ii) the negotiation of the FLEGT VPA in Ghana (e.g., stakeholders’ views, policy objectives); and (iii) the final FLEGT VPA between the EU and Ghana (e.g., main elements, benefits, social issues and objectives).
Qualitative data were analysed using manual coding (Creswell 2007) and manual content analysis (Silverman 2006), whereas for quantitative data, arithmetic means were calculated.
4.3.2 Materials and methods in Articles II and III
Articles II and III are similar in terms of their methodological approach, data collection and data analysis. Data for both articles were collected using semi-structured face-to-face interviews (Appendix IV). Both articles are based on closed- and open-ended questions from the interviews and make use of descriptive statistics (frequencies and percentage) and non-parametric statistical tests in the data analysis phases.
Article II assesses the relative importance that farmers ascribe to certain forest values and the potential associations between their forest values and compliance with the treefelling rule (i.e., the ban on harvesting timber trees on farmers’ lands). Data on forest values were collected in two subsequent exercises: (i) the identification of all forest values and (ii) the ranking of the importance of twelve predefined categories of forest values. In the first exercise, the respondents were asked to name all of the things they value about the forest; in the second exercise, they were asked to rank the importance of twelve predefined forest values (Appendix I). Next, using an open-ended question, the respondents were asked to give reasons for ranking a certain forest value as the most important and another as the least important. Respondents could give more than one reason, allowing for a multiple response option. Farmers’ compliance with the tree-felling rule was assessed using the following question: “Would you fell timber tree/trees without a permit?”, with the following answer options: yes, only in difficult situations, and no. Subsequently, they were asked to provide reasons for their reported compliance behaviour. Non-parametric Friedman tests and multiple pairwise comparisons of subsets of values, with Bonferroni adjusted p-values, were conducted to establish the order of importance of forest values. Multivariate binary logistic regression (Hosmer and Lemeshow 2000, Hancock and Mueller 2010) was used to explore the potential relationships between compliance with the tree-felling rule (dependent variable) and forest values (explanatory variables). Finally, to understand why the respondents perceive certain forest values as the most and others as the least important, the given reasons were analysed and the percentages of responses were calculated.
Article III, on the other hand, assessed farmers’ compliance with formal forest rules and the reasons and motivation for their behaviour. Data collection on compliance with forest rules included (i) farmers’ own compliance behaviour; (ii) farmers’ perceptions about the compliance behaviour of their peers; and (iii) farmers’ approval for non-compliance with the studied forest rules. Concerning the factors that affect compliance, farmers were asked to respond to a variety of pre-defined statements, established with reference to general compliance theory (see section 2). Additionally, farmers were asked to give reasons for their reported compliance/non-compliance with specific forest rules. Data about forest law compliance and the factors influencing compliance were analysed using basic descriptive statistics on the numbers and percentages of respondents.
4.3.3 Materials and methods in Article IV
The last article is a response to the observed complexity surrounding forestry law, compliance behaviour, and the observed lack of theoretical, empirical and analytical insights into rule and law compliance in the forest sector. The article adopts an inductive approach that draws upon the available literature on sources of non-compliance in the forest sector and the interdisciplinary theoretical literature on rule compliance. More specifically, the article reviews the literature on compliance in forestry to identify a comprehensive list of the most common sources of non-compliance in the forest sector. It then continues by reviewing different theoretical perspectives on rule compliance and emerges with three dominant models that collectively highlight a variety of individual motivations for compliance that generally consist of (i) instrumental benefits and costs, (ii) social and personal norms, and (iii) legitimacy. Finally, the article integrates the empirical and theoretical reviews to present an analytical framework for compliance in the forest sector that embraces multiple theoretical models of human behaviour.
The studies on compliance in forestry reviewed in the article adopt a global perspective but focus mostly on countries where, for different reasons, high rates of illegal forest activities exist: the Amazon, Central Africa, Mesoamerica, South-East Asia and West Africa, with some consideration of European context (Contreras-Hermosilla 2002, Brack 2003, Hirakuri 2003, Tacconi et al. 2003, Contreras-Hermosilla and Peter 2005, World Bank 2006, Kishor and Damania 2007, Tacconi 2007, Tacconi 2007a, Blaser 2010).
The results of this research are summarised in four sections; each presenting the main results of one specific PhD article. As previously mentioned, the Article I discusses the concept of livelihood, as outlined in the VPA and the potential impacts of law enforcement on the forest communities’ livelihoods within the VPA in Ghana. The three subsequent articles explore different dimensions of compliance with forest rules, focusing in particular on: farmers’ forest values (Article II), factors that impede/facilitate compliance/noncompliance with rules (Article III), and theoretical and analytical developments in the study of forest law compliance.
5.1 Concept of livelihoods and the expected impacts of the FLEGT VPA on forest communities’ livelihoods in Ghana (Article I) 5.1.1 Concept of livelihoods in the EU-Ghana FLEGT VPA In addition to the five-livelihood assets given in the sustainable livelihood framework (SLF) (Carney 1999, Chambers and Conway 1992, DFID 2002), five supplementary elements of livelihoods were identified as relevant for the livelihood of forest communities in the context of FLEGT VPA. After Baumann (2000), these elements were termed as “policy and institutional livelihood assets” and include: (i) forest communities’ rights to forest resources, (ii) their access to resources, (iii) their participation in decision-making processes, (iv) equity in timber benefit sharing, and (v) land and tree tenure. The experts involved in the interview considered the policy and institutional livelihood assets as the most influential and the most relevant for the livelihood security of forest communities, within the VPA. The policy and institutional livelihood assets directly relate to the larger forest governance discourse (Cotula and Mayers 2009). More precisely these aspects define the bundle of rights and bundle of powers (Ribot 1998, Ribot and Peluso 2003), as well as deliberation processes in which these bundles are transferred from socially, politically, or financially stronger to the weaker groups (Agrawal and Ribot 1999, Ribot et al. 2006, Tacconi 2007b).
Natural (i.e. forest as natural resource stock) and social assets (social networks, relationships and norms), from the SLF, were also perceived as very relevant in the VPA process. The VPA implementation is expected to have an indirect impact on these assets.
For instance, the VPA implementation may improve forest management practices, which would lead to retention of forest resources, which in a long run, may strengthen forestbased livelihoods of communities. Social aspects are expected to improve, as a result of an expected improved access information, as well as formation of networks and consultation that took place during the VPA negotiation phase. Financial assets (savings, income), on the other hand, were regarded as relevant for the livelihood security, but largely overlooked in the VPA negotiation. The actual financial risks and uncertainties – including market prices and impact on communities – were considered higher than the assumptions made in the negotiation phase. Consequently, concerns about financial loss for both, VPA countries’ governments and the communities, were expressed. Similarly, human (skills, knowledge, labour) and physical assets (infrastructure, transport, energy, communications) were considered to be largely overlooked during the VPA negotiation process.
5.1.2 Potential impacts of the VPA implementation: who will be affected and how Groups within the forest communities whose livelihood is likely to be affected by the VPA implementation include: (i) small-scale and subsistence farmers; (ii) chainsaw operators;
(iii) youth; (iv) women; (v) middle class (e.g. cocoa farmers, land owners); and (vi) wealthier groups (e.g. chiefs and sub-chiefs).
It is expected that the VPA implementation may result in both, positive (e.g. justice in allocation of timber benefits, and better access to information, improved participation) and negative impacts (e.g. lost access to forest and forest resources, lost employment provided by the illegal forest activities) on the livelihoods of the impoverished, as well as the wealthier groups, within the forest communities (Table 2/Article I). The analysis of results imply that the actual impacts will largely depend on the extent to which the following issues are addressed in the VPA implementation: (i) reforms of the land and tree tenure, (ii) participation, (iii) transparency and accountability, and (iv) forest management practices.
Considering the legal plurality and the complex co-existence of statutory and customary tree and land tenure systems in Ghana (Amanor 1999, Larbi 2006, Boakye and AffumBaffoe 2008), a lot of hope is put on clarification and reform of ownership and tenure.
However, clarification and reforms of tenure and ownership rights will not necessarily benefit the forest communities and the vulnerable groups, unless their interests and concerns are taken into consideration in the reforming processes. Therefore, a direct participation and an honest consideration of the communities in the forest policy and legislation reforms ought to be strengthened. Increased transparency and accountability is expected to reduce the elite capturing of forest benefits, and potentially the existing corruption in the forest sector. Having said that, an increased transparency and accountability is expected to have potential negative impacts on the local elite; while at the same time, it is expected to benefit the vulnerable groups, through more equal and just sharing of timber benefits. Finally, certain expectations exist that the VPA will introduce better forest management practices, which will result in improved resource stocks, environmental services and non-wood forest products; which in a long run will have positive impacts on forest communities’ livelihoods.
5.2 Understanding the meaning and context of farmers’ forest values (Article II) 5.2.1 What farmers value about the forest Farmers identified over 100 forest value items (e.g., bushmeat, protein, air quality, farming land, soil fertility, shelter, timber, firewood, honey, wild fruits, inner peace). These items were grouped into 32 broader categories of forest values (Appendix V). As shown in the appendix, the identified categories of forest values were grouped in one of the following groups: subsistence, environmental, economic, learning, future, cultural and spiritual forest values.
Figure 2. List of the dominant forest values identified by farmers.
Percentage of respondents (N = 226) The most dominant (in terms of the percentage of farmers who identified the specific forest values) and the most diversified forest values (in terms of the numnumber of items of value identified) are subsistence values (e.g., food, wild crops and bushmeat), followed by environmental (e.g., water, rainfall) and economic (timber, income) values. Aesthetic and religion-related forest values were not identified by farm (Figure 2).
related farmers The rankings of the importance of different categories of forest values, classified into use and non-use forest values (Appendix I), were found to be statistically different use (Friedman test statistic for use forest values: 575.2, p=0.000, df df=5, n=225; Friedman test statistic for non-use forest values: 357.5, p=0.000, df=5, n=226). Environmental, use subsistence and economic values were ranked as the most important of the use forest values. They are followed by medicinal, learning, and finally aesaesthetic values as the least important of the use forest values.
Table 2. Pairwise multiple comparisons of subsets of values: order of importance of use and non-use forest values (1 – most important to 6 – least important)
Future forest value was ranked as the most important in the group of non-use values, followed by moral and cultural, then intrinsic and spiritual, and finally religion–related forest values as the least important in the respective group (Table 2). It should be noted that environmental, subsistence and economic forest values have the same order of importance, and so do moral and cultural and intrinsic and spiritual values. This means that in pairwise comparisons, there was no statistically significant difference in the rankings of importance of these subsets of forest values.