Printer Friendly

A choice experiment approach for assessing preferences to forest law configuration and compliance: the case of NTFP traders in Cameroon/Choix d'une approche experimentale pour evaluer les preferences de configuration et d'adherence aux lois forestieres: cas des commercants en produits forestiers autres que le bois au Cameroun.

Uso de un experimento de eleccion para evaluar las preferencias de configuracion y cumplimiento de la legislacion forestal: el caso de los comerciantes de PFNM en Camerun.

INTRODUCTION

Law compliance is an integral part of any governance system and refers to the extent to which subjects or actors respect prescribed behaviours (Blaser 2010, Ramcilovic-Suominen and Epstein 2012, Young 2005). In many countries, forests products including Non Timber Forest products (NTFP) are harvested, transported and traded in violation of national laws governing such activities (Assembe Mvondo 2009, Blaser 2010, Cerutti and Tacconi 2006, Tieguhong et al. 2010). The negative impacts of such illegal forest activities include among others, degradation of forests and related goods and services and loss of government and private revenues (Blaser 2010, Contreras-Hermosilla 2002, Kaimowitz 2003). The World Bank for example reports that illegal logging alone is responsible for annual losses in global market value of more than 10 million USD and in government revenue of as much as five billion USD. The magnitude of the problem has prompted national and international NGOs to put forest law compliance on the agenda of international forums (Blaser 2010, Tieguhong et al. 2010). In addition, researchers are continuously seeking solutions to advance scholarly investigations in order to assist countries to address the issue (Kaine et al. 2010, Ramcilovic-Suominen and Epstein 2012). Based on the outcomes of five of such international forums organised by the Food and Agricultural Organisation of the United Nations (FAO) and the International Tropical Timber Organisation (ITTO), the following points were identified as critical for a strategic approach to increase forest law compliance: ensuring broad and inclusive stakeholder participation; addressing the underlying causes of illegality; prioritising remedial actions and assessing the economic feasibility and social acceptability of proposed reforms (Blaser 2010). This study will pay attention to two of the above factors: (i) stakeholder participation and (ii) economic feasibility and social acceptability of proposed reforms.

In the forestry sector, the participation of stakeholders in the policy and law making process is highly recommended (Laird et al. 2009, Mayers and Bass 2004). Stakeholder participation should ensure that policies and laws reflect real needs and priorities and that they will be widely accepted (Kaine et al. 2010, Laird et al. 2009, Mayers and Bass 2004, Sutinen and Kuperan 1999). Stakeholder participation in the policy and law formulation process is in line with more recent approaches to understand compliance with the law, which are based on a normative perspective (Kaine et al. 2010, Sutinen and Kuperan 1999). This incorporates elements of personal morality, fairness and legitimacy. In this regard, individuals are expected to comply with the law to the extent that they perceive such a law as appropriate and consistent with internalised norms (Ramcilovic-Suominen and Epstein 2012, Sutinen and Kuperan 1999). In this case key variables determining compliance are individual perception of fairness and appropriateness of the law. From an economic perspective, compliance is based on a rational choice model of human behaviour, which incorporates calculated costs and benefits of compliance (Ramcilovic-Suominen and Epstein 2012).

Most top-down approaches and command and control-style policies and regulations are often considered to be unfair and ineffective because they do not take into consideration the perceptions of those targeted by such institutions (Horne 2006, Mayers and Bass 2004, Sutinen and Kuperan 1999). Consequently, when bottom-up approaches are used it is important to properly understand the extent to which individuals are motivated to change their behaviours to comply with proposed regulations (Kaine et al. 2010).

While considerable efforts are being deployed to reduce illegal forestry activities in the Congo basin forests, the literature suggests that the efforts made thus far seem to be concentrated on illegal timber logging and wildlife poaching, neglecting other illegal forest activities such as Non Timber Forests Products (NTFP) harvesting and trade (Tieguhong et al. 2010). The objective of this study is to contribute in filling this gap by investigating traders' preferences for several policy options that are hypothesised to reduce transaction costs (TCs) and other barriers in the process of complying with the permit system governing trade of NTFP in Cameroon. Suitable policies could reduce the inefficiencies in the NTFP value chains and by increasing respect for the law, the credibility of state institutions, which are prerequisites for economic performance will be enhanced (North 1990). A method often used to elicit preferences or to evaluate policy options taking insights from target beneficiaries (stakeholders' participation) into consideration is choice experiments (Alpizar et al. 2001, Hope et al. 2005, Horne 2006, Speelman et al. 2010, Sur et al. 2006). This method will be used in this study.

Effectiveness and ineffectiveness of policies and regulations may be theory based (related to bad laws) and process based (1) (bad implementation of even good laws). While these issues have been raised in the preceding paragraphs, analyses in this study are limited to traders' acceptability of existing and proposed policies and regulations. In this regard, the actual effectiveness of the proposed policies and regulations will not be tested.

The rest of the paper is organised as follows: first, background information on the process of obtaining NTFP permits in Cameroon is given. Secondly, the choice experiment method that will be used to asses traders' preferences for policy options will be explained. In describing the choice experiment method, the study will further elaborate on the permit system because the method requires that the set of policy options that are retained are properly described. The analytical and econometric foundation for analysing choice experiment data will come next. Finally the results of the study are presented followed by a discussion and conclusions.

CASE STUDY BACKGROUND INFORMATION

The right to sell NTFP in Cameroon is sanctioned by a permit system which deals with "special forests products" as prescribed by the 1994 Forestry Law (Government of Cameroon 1994). However it is not clear how the Government of Cameroon (GOC) defines "special forests products". Often it is understood that it consists of species with economic and environmental values (Ingram 2012). The unclear definition of special products actually makes it ambiguous which products are regulated by the permit system. Nevertheless, the day to day practice is that for most NTFP of high economic significance, traders are requested to show permits at road check points (Betti 2007, Foundjem-Tita et al. 2012b, Laird et al. 2010, Ndoye and Awono 2010). The majority of the traders, if not all, dealing in NTFP do not have their own permits, instead some of them rent permits. This practice is against the law as a result their activities are illegal (Awono et al. 2012, Betti 2007, Djeukam 2006, Ingram et al. 2012, Ingram 2012). Their decisions to operate illegally may not be an intentional act to overlook existing legislation, instead it may be related to the high TCs involved in the process of meeting the excessive legal requirements (Blaser 2010, Wells et al. 2007).Thus efforts are needed to reduce these high TCs and address elements that may be perceived as unfair.

Studies assessing transaction costs and options related to agri-envrionmental policies and regulations have been reported in Europe, the US (McCann et al. 2005, Mettepenningen et al. 2011, Stavins 1995) and in the forestry sector in some Latin American countries like Honduras and Nicaragua (Wells et al. 2007). Unfortunately the topic has not received much attention from researchers in Cameroon particularly not with regards to NTFP. Nonetheless, a few organisations like FAO and its partners have formulated proposals to the GOC aimed at reducing complications or in other words TCs related to the process of obtaining permits in the NTFP sector (FAO et al. 2010, Fokou Sakam 2010). The recommendations of FAO and its partners stem from stakeholders workshops. Literature on policy formulation processes however suggests that after the identification of such qualitative information by key informants through stakeholders workshops, policies should be subjected to a wider group for quantification and weighting of issues (Mayers and Bass 2004). This provides the rationale for our study.

METHODOLOGY

Why a choice experiment

A choice experiment is a form of choice modelling that has its roots in the field of marketing and transportation where it was originally used to study the tradeoffs between product characteristics or the characteristics of different transport services offered. Increasingly the method is being used in the valuation of non-market goods (Alpizar et al. 2001, Birol et al. 2006). In the policy evaluation sphere, the method can be used to provide information on target beneficiaries' preferences which can be interesting for the purpose of developing policies and laws (Hope et al. 2005, Horne 2006, Sur et al. 2006).

The decision to use a choice experiment in this study is guided by the opportunity it provides to identify compatible policy options that may reduce TCs and other setbacks in the NTFP permit system in Cameroon. The technique entails presenting to target beneficiaries a number of combinations of characteristics (attributes) at different levels that describe a good. In this study the attributes are related to regulations governing access to permits. For each choice scenario they are asked to pick the combination they prefer most or that gives them the highest utility. When analysed, results can determine the relative importance of each attribute in determining preferences (Alpizar et al. 2001, Ryan et al. 2001a, Sur et al. 2006). In choice experiments that study the value of goods and services, respondents' willingness to pay is often indirectly evaluated by including the price of the good as one of the attributes (Hanley et al. 2001, Speelman et al. 2010, Street et al. 2005). In this study the respondents' willingness to pay for shifts from one policy option to another was evaluated by including a forest regeneration tax as one of the attributes.

There are several advantages of using choice experiments over other stated preference methods such as contingent ranking, contingent rating and paired comparisons. One reason is that, choice experiments are suited for scenarios such as ours, where changes in the attributes are multidimensional and tradeoffs between them are of particular interest to the respondents. In fact the technique has the ability to identify the value of individual attributes. Although contingent valuation techniques can be used as well, they are considered expensive and cumbersome as a series of contingent valuation scenarios would need to be included in the questionnaire (Hanley et al. 2001). Another advantage of the method used here is that, respondents are given the opportunity to express their preferences on varied but significant unknown future events (Hanley et al. 2001, Hope et al. 2005). In addition, compared to ranking methods, choice experiments are most preferred because in choice experiment the decisions to be made are more similar to those which respondents make in their everyday life (Ryan et al. 2001b).

When designing choice experiments, besides having a good definition of the problem at hand, there is a need to identify relevant attributes and attribute levels (Alpizar et al. 2001, Speelman et al. 2010). These are discussed below.

The choice of attributes and attribute levels

Attributes selected should be those that affect respondents choices and that are relevant to policy (Alpizar et al. 2001, Speelman et al. 2010). In this study, the selection of attributes was guided by factors that are assumed to contribute to entice traders to comply with the permit system. Literature reviews, focus group discussions and stakeholder meetings are important in selecting attributes and their levels (Alpizar et al. 2001, Horne 2006). Based on this premise, a first set of attributes selected for this study was derived from a literature review amongst which a proposition made by FAO and partners (2) (FAO et al. 2010) to the Ministry of Forestry and Wild Life (MINFOF) for consideration in subsequent review of the 1994 Forestry Law. The proposition of FAO and partners to MINFOF was the result of two different stakeholders' workshops to identify weaknesses in current legislation governing the NTFP sector in Cameroon. Above all the workshops were organised in order to generate inputs to amend law No. 94/01 of 20th January 1994 governing the forestry, wildlife and fisheries regimes and its text of application (Fokou Sakam 2010). These workshops were organised in Bamenda in the Northwest region of Cameroon and in Ebolowa in the South region by FAO and its partners. Both workshops brought together a total of 173 actors representing 12 different categories of actors (3) in the NTFP sector representing 9 out of the 10 administrative regions in the country.

Three main attributes of the process of obtaining permits to sell NTFP, relevant for this study, were identified at this stage: the process of obtaining an approval, the process of obtaining an exploitation permit and the payment of the forest regeneration tax. These are not the only requirements to obtain the legal status to sell NTFP in Cameroon, but they are considered the most difficult. Details about the process of obtaining legal recognition are found in Djeukam (2007) and Beti (2007). The selected three attributes were further subjected to detailed focus group discussions involving traders and also individual discussions with 9 different experts working in the field of NTFP amongst whom: staff of the Netherland Development Organisation (SNV), FAO, ICRAF, CIFOR and MINFOF. Traders involved in the focus groups were a broad representation of the sampled group. Three different focus groups were organised with the traders. The first two was made up of 7 and 8 different traders and the third was a consolidation of the results of the first two. Ten traders from the previous two sessions attended the third focus group. The objective of the focus groups and the discussions with experts was first to gain insights into the three most important attributes identified in the stakeholders workshops and secondly to define the levels for each attribute.

Using more than 5 attributes in choice experiments may be detrimental to the quality of the information gathered because it increases the complexity of the choice task (Mazotta and Opaluch 1995). A too high task complexity may cause fatigue and boredom in responses leading to careless responses. The same may occur if the number of choice sets presented to each individual is too high or if there is correlation between attributes (Alpizar et al. 2001). To reduce complexity only the most essential attributes were retained in this study. Based on the focus group discussions and the interviews with experts, the three initially selected attributes were further broken down into five attributes. These five attributes are described below and summarised in Table 1.

The approval

This is the first legal document to be obtained by a trader. It gives the holder the right to exploit any forest resource in Cameroon. This means that only when this document is obtained traders may proceed to process other legal papers. The problem related to obtaining the approval is that the process is centralised in Yaounde, the capital of Cameroon. In addition, the process of following-up the files is considered expensive with a lot of transaction costs involved (Laird et al. 2010). NTFP traders with comparatively limited capital are expected to file in the same documents as timber exploiters because the same approval applies to both categories of actors (FAO et al. 2010, Fokou Sakam 2010, Government of Cameroon 1994, 1995). This is considered illegitimate by NGOs and NTFP traders.

According to Kaine et al. (2010) one strategy to promote compliance among individuals with unfavourable attitude towards an intervention is to include behaviours advocated by these individuals provided policy objectives are met. With regards to an approval it is advocated that a special approval should be issued for NTFP traders and the process should be made easier (FAO et al. 2010, Fokou Sakam 2010). Some traders and experts consulted respectively through focus group and expert meetings even argued that this document should be completely cancelled for NTFP. Based on the argument that selected attributes need to be policy relevant and meet the expectation of traders without jeopardising government objectives of conserving the natural resource, the research team concluded that the government may not abide to the option of entirely deleting the approval. It was thus imperative to search ways to reduce TC involved in the process rather than thinking of abolishing it.

The next question therefore was how to simplify the process? Two proposals came up during the focus group and expert discussions: (i) the number of documents to be submitted may be reduced and (ii) the process should be decentralised to the regional level. To make the choice experiment more easy these two proposals were grouped as one option and then the status quo of obtaining the same document in Yaounde was maintained. The main assumption here is that by instituting a special and simplified procedure to obtain an NTFP approval, obtainable at regional level, TCs involved in the process may be reduced and traders would be encouraged to apply for it.

Joint or individual approval

Another issue that came up during the expert and focus group discussions was whether traders would be encouraged to apply for permits if they were given the opportunity to work as a group. Actually the 1994 Forestry Law states that any individual or moral persons (group) can apply for an approval. The simplest form of organising farmers and other small scale rural enterprises in Cameroon is through common initiative groups (CIGs), associations or cooperatives (Government of Cameroon 1992). Currently, these types of groups/associations are not allowed to apply for permits and they are required to transform into limited liability companies to do so (Ingram 2012, Nguenang et al. undated).

Some traders have struggled to organise themselves into CIGs but have not been successful in the process (Ingram 2012). Moreover the imperative to transform CIGs and associations into limited liability companies may have a negative effect on a traders' decision because they believe that such companies/enterprises are very big structures that have to pay a lot of taxes. Facilitating joint ownership of approvals through CIGs may allow a group of traders to pull their resources together and in this way reducing costs. It is even argued that when individuals operate in groups, peer pressure will compel group members to abide with the law (Sutinen and Kuperan 1999). It is also known that some individuals would also want to work individually in order to avoid negative effects of group activities on their business. Thus, two options were proposed to the traders: individual approvals or joint application through CIGs.

Exploitation permit

An exploitation permit is issued to traders after having obtained the approval. The exploitation permit gives the holder the right to exploit and sell a particular NTFP. The permit specifies the quantity and zone of action where the trader is supposed to collect the produce. The permit is issued for one year and a prerequisite for this is the payment of a regeneration tax. Literature reviews (Betti 2007, Djeukam 2007) and focus group discussions highlighted two weaknesses of the permit system: its transferability and its duration.

a. Transferability of permits

Whether NTFP permits are transferable from one person to another or not is not clear in the law. Confusion arises from article 42 (1) of the 1994 Forestry Law which states that holders of permits can transfer or subcontract some of their rights. Contradictorily section 42(2) stipulates that rights are personal and non transferable. Detailed discussions with forestry officials and NGOs advocating for farmers' and traders' rights reveal that transferability is currently illegal.

Some NGOs and traders are against transferability with the argument that it favours non-NTFP professionals (mostly powerful business men) to use their financial powers to obtain permits (Betti 2007, Kaimowitz 2003). In line with Kaimowitz (2003), these financially viable permit holders are accused of encouraging corruption and of increasing the transaction costs in the permit system. That is, it is alleged that they use their financial powers to corrupt Government officials to tighten conditions so that they can have the possibility to rent out their permits to the small scale traders who are unable to secure permits (Tieguhong et al. 2010).

Other traders argue that the process of obtaining a permit is too expensive for simple traders and limiting transferability will kick them out of business. Other studies show that because of the complications to obtain permits traders prefer to rent permits rather than to apply for the permits themselves (Awono et al. 2012, Betti 2007). Other stakeholders propose that transferability should be allowed but should be regulated by the government. That is if a permit is transferred it should be officially recognized by the government. This contrasts the current situation where business operators merely photocopy their letter heads and give it to traders who show them at road check points. In this way they pretend that they are working for the permit holder. The choice experiment assesses which of these three options is most preferred by traders: no transferability, transferability with authorisation from the state or transferability based on mutual agreement between the permit holder and the borrower.

b. Duration of exploitation permit

FAO and its partners report that the short duration of a permit limits investment (Djeukam 2006, Fokou Sakam 2010). This is true because money is involved in the process and thus a permit can be considered as an investment in the NTFP sector. Since NTFP exploitation permits are only valid for one production season, traders complain that the delays involved in the application process prevent them from exploiting the quotas that are attributed to them. For example if it takes till October before the exploitation permit for one ton of Irvingia gabonensis is issued, and the trader pays the required regeneration tax to exploit one ton but only succeeds to exploit 50% of it by the end of December; s/he loses the remaining 50%. In this regard by increasing the duration of the permit, traders would be able to carry over their quotas into the following year. This is expected to more easily allow costs recovery. However in order to avoid problems concerning the sustainability of NTFP exploitation, the validity of the permits should be limited to two or three years. To simplify the choice experiment, it was agreed that the status quo (duration of one year) and an improvement to three years should be chosen.

Regeneration tax

One of the reasons why governments sometimes devises policies and policy instruments such as the regeneration tax is to raise income to regenerate depleted resources (Betti 2007). The regeneration tax has been described to be inconsistent and inefficient in controlling trade, in promoting regeneration and to be too high for traders (Ingram 2012). Thus, it can be considered as one of the reasons that discourage traders to apply for permits. Yet other actors in the chain think that the amount charged is low and that it does not take into consideration the true value of products (Betti 2007, FAO et al. 2010, Fokou Sakam 2010). During the focus group discussions, some traders stated that the current charge of 10 FCFA/kg is not an issue for them and that they can pay more. This is corroborated by the fact that they actually currently pay more for bribe (Foundjem-Tita et al. 2012a). Hence, a good question to answer at this stage is, what is traders' willingness to pay (WTP) in term of regeneration tax? In this regard, during traders focus groups and the expert discussions, it was decided that three values should be selected as attribute levels for the regeneration tax; the current amount of 10FCFA, a lower amount of 5FCFA and a higher amount i.e. 20 FCFA. A summary of all the selected attributes and their levels is shown in Table 1.

DESIGNING THE CHOICE SETS

In choice experiments, experimental design aims at creating choice sets in an efficient way. This involves combining the levels of the attributes into alternative profiles (options) to be presented to respondents in choice sets (Alpizar et al. 2001, Hanley et al. 2001). Based on the number of attributes and their levels (three attributes with two levels and two attributes with three levels) a total of 72 different combinations ([2.sup.3] * [3.sup.2]) would be a complete factorial design. It is obvious that 72 different combinations would be too many for an individual to choose from. It was thus important to reduce the 72 combinations into a manageable number from which traders could make their choices.

The approach most commonly used is the so-called orthogonal design which results in uncorrelated alternatives of the attributes (Alpizar et al. 2001). The orthogonal design was developed using the ortho-plan function in SPSS version 17. This resulted into 16 different combinations. These 16 combinations were still considered too much and difficult to evaluate at a time (Hanley et al. 2001). For this reason it was decided that each respondent is given a maximum of 4 options at a time from which to choose. A procedure reported by Street et al. (2005) was used to construct the four different options. The procedure allows for independent estimation of all effects and leads to optimal designs for main effects and near optimal design for main effects with two interactions (Street et al. 2005). The basic principle of the procedure is to use the orthogonal output from SPSS as a starting point generally as the first option for each choice set and then make systematic level changes using a generator and modular arithmetic to get option two, and another to obtain option three and so on. The end result is that as many pairs of profiles as possible have different levels for each attribute. Based on this methodology 16 choice sets made up of 4 different options were obtained.

In order to use all the 16 choice cards, they were grouped in to 4 blocks. The essence of the blocking is to reduce respondent fatigue which potentially could result into low consistency (Adamowicz et al. 1998, Speelman et al. 2010). Each block contained 4 choice sets (cards) and each choice set contained 4 options from which the respondents had to select one. The 4 blocks were later divided into subsamples of our sample. Each respondent was randomly assigned to a block containing 4 choice cards. Since it was expected that some of the traders would not be able to read properly, some attribute levels were transformed into pictograms (Speelman 2008). These pictograms were explained to the respondents at the start of the experiment. An example of a choice set is shown in Table 2.

ANALYTICAL FRAMEWORK

The theoretical foundation for analysing data from a stated preference choice experiment is based on Lancaster's (1966) model of consumer choice and the econometric analysis is based on McFadden's (1974) conditional logit model grounded in the theory of random utility (Birol et al. 2006, Hanley et al. 1998, Horne 2006).

The consumer choice model

Based on Lancaster, individuals derive utility (satisfaction) from the attributes or the characteristics of a good rather than from the good itself. For an illustration of the model, consider a trader's choice for a given option, and consider that his utility (related to the level of transaction costs) depends on the choices made from a choice set C, which contains a number of all possible policy alternatives (four in our case). The trader is assumed to have a utility function of the form:

[U.sub.ij] = V([Z.sub.j], [S.sub.i]) + e([Z.sub.j], [S.sub.i]), (1)

Where, [U.sub.ij] represents the utility of trader i choosing a given option j from the set C. The utility derived from any of the policy options depends on the attributes Z of the policy option and the socio economic characteristics of the trader S which may include his experience in the business, capital, age, education, availability of alternative sources of income etc.

The random utility theory describes the utility U of each alternative in C as a sum of two components: a deterministic component (V) and an error component (e). The deterministic component is a vector of the alternative specific attributes and the characteristics of the individual respondent (trader). The error component (e) is independent of the deterministic part and represents errors that arise or influences that are unobservable by the researcher. It is generally assumed that inclusion of the error term makes the choice random (Holmes and Adamowicz 2002, Horne 2006). In other words the error component implies that predictions cannot be made with certainty (Birol et al. 2006).

Based on the random utility theory, an individual i will choose a given option from the choice set C if the utility j derived from that particular option is greater than that obtainable from any other choice k (equation 2). This means that,

[U.sub.ij]> [U.sub.ik] [right arrow] [V.sub.ij] + [e.sub.ij] > [V.sub.ik] + [e.sub.ik] for all j [not equal to] k, j, k [euro] C (2)

Choices made between different alternatives are a function of the probability that the utility associated with a particular choice are higher than that related to another alternative. This means that the probability that an individual i chooses alternative j is the same as the probability that the utility of alternative j is greater than any other utility k in the choice set C as stated in equation 3.

P (j) = P ([V.sub.ij] + [e.sub.ij] > [V.sub.ik] + [e.sub.ik]) for all j [not equal to] k, j, k [euro] C (3)

The econometric model

An assumption often used is that the error terms [e.sub.ij] are independently and identically distributed with an extreme value (Weibul) distribution (Hanley et al. 2001). This assumption will lead to the conditional logit model, which is most widely used in analysing multi attribute choices (Horne 2006). The conditional logit model (Greene 1997, McFadden 1974) for individual i, choosing option j over the others can be given by equation 4:

P (j) = [exp.sup.Vij]/[SIGMA][exp.sup.Vik] (4)

Generally the conditional indirect utility function estimated is:

[V.sub.ij] = [beta] + [[beta].sub.1][Z.sub.1] + [[beta].sub.2][Z.sub.2] + ......... [[beta].sub.n][Z.sub.n] + [[lambda].sub.1][S.sub.1] + [[lambda].sub.2][S.sub.2] + ... [[lambda].sub.m][S.sub.m] (5)

where [beta] is a constant that captures the utility of any attribute that is not specified in the choice set, n is the number of attributes and m the number of socio economic variables describing the trader. The coefficients [[beta].sub.1] to [[beta].sub.n] and [[lambda].sub.1] to [[lambda].sub.m] respectively capture the influence of the vector of attributes studied and the vector of socio-economic characteristics describing each trader. The socio economic variables do not vary across the different options selected by each trader. In this regard they are only included in the model as interaction terms with the choice specific attributes (Birol et al. 2006, Hanley et al. 2001). The coefficients obtained from the interaction terms can be used to interpret the effect of the socioeconomic characteristics in determining choices (Speelman et al. 2010).

Once the parameter estimates have been obtained, willingness to pay can be calculated for each attribute (Alpizar et al. 2001, Hanley et al. 2001). For a linear utility function as that expressed in equation 1, willingness to pay can be simply articulated as in equation 6:

WTP = - [b.sub.c]/[b.sub.y] (6)

Where [b.sub.c] is the coefficient of any of the attributes and by is the coefficient of the monetary value in equation one, which in this study refers to the regeneration tax. Equation 6 is also referred to as the path worth (implicit price) formula and represents the marginal rate of substitution between the monetary value (regeneration tax) and the other attributes in the equation.

DATA COLLECTION AND ENTRY

The study population was made up of traders specialised in selling four different NTFP in Cameroon. Data were limited to traders because most producers sell at the door step and are not familiar with the regulations (Schreckenberg et al. 2006). The four selected NTFP include Cola spp, Ricinodendron heudelotii, Gnetum spp and Irvingia gabonensis. The four studied species are among the most widely traded in Cameroon (Ndoye et al. 1998) and are products which traders most frequently cite as attracting attention from police and forestry officials requesting permits (Betti 2007, Ingram et al. 2012, Ndoye and Awono 2010). Traders were identified in eight different markets which were either major local assembly, urban or export markets for the respective products. Since traders are often very mobile, each of the markets was visited at least two times on two different days within a week in order to maximise the chance of meeting a greater number of different traders.

Before each interview, the objective of the exercise was explained to the respondent. Specifically, a brief explanation was made on how the exercise could contribute to improved policies related to obtaining permits. Traders were also explicitly informed that participation was voluntary. The symbols that relate to each attribute and their levels were carefully explained to each respondent in a step by step process and a test was made to ensure that they had clearly understood the process. Those who failed the test were excluded from the interview. The questionnaire also included questions related to the socio economic characteristics of the respondents. A description of the socio economic variables is found in Table 3. Each survey took between 50 minutes and 90 minutes depending on the ability of traders to grasp the concept of the choice experiment and the other questions. The questionnaire was pretested and administered by the lead author. This was to make sure the quality of information collected was guaranteed.

In all, 280 choices were elicited from 70 traders who accepted to take part in the survey and passed the choice experiment test. Data collected were entered in SPSS version 17 and later saved in the statistical package Stata 11.0. The conditional logit models were analysed using Stata 11.0. In performing the conditional logit models, all the quantitative variables such as duration of permits and regeneration tax were directly entered. The coefficients in the model therefore show the effect of the magnitude of these variables. The qualitative attributes were effect-coded. This means that the base level of a given attribute is assigned a value of -1 while the other levels may take values of 0 or 1. Holmes and Adamowicz (2002) showed that this way of coding reduces problems of collinearity with the intercept, which might occur if the attribute levels were dummy coded. Interested readers are referred to their book chapter for the technicalities of this (Holmes and Adamowicz 2002). The coefficients related to the qualitative attributes therefore measure the impact of a change in level relative to a base level. The base levels for the three qualitative variables used are as follows: for type of documents, the base level was "complicated-centralised procedures"; for joint or individual exploitation agreement, the base level was "individual" and for transferability of permits, the base level was "no transferability".

RESULTS

A description of the socio-economic characteristics of the traders interviewed is given in Table 3. Results of the conditional logit models are shown in Table 4. Model 1 is the basic model and illustrates the effect of different attributes in determining traders' decision to apply and obtain NTFP trade permits. The attribute specific constant (ASC) included in this model represents the utility gains linked to not choosing any of the options provided to the traders. Even though the ASC was not significant, the negative sign suggests that choosing the opt-out alternative decreases implicit utility. This means that all things being equal, traders would prefer to see some changes in the current procedures based on the alternatives provided to them.

Based on model 1, all the attributes except transferability with authorisation from the government had a positive and significant effect in determining traders' choices. In other words, there is a preference for a decentralised and simplified process compared to a centralised and complicated one; for joint permits through CIGs and not through limited liability companies. Also, transfer of permits between traders and an increased duration of a permit improved the chance of an option being selected. On the other hand and in line with basic economic theory, an increase in forestry regeneration tax had a negative significant effect on the probability for a particular option to be chosen. This therefore means that all things being equal traders would prefer lower regeneration taxes.

In order to target policies that satisfy traders with different socio-economic characteristics it may be useful to know how certain socio-economic variables influence traders' choices. To meet this objective, only those socio-economic factors that were anticipated to influence preferences for specific attributes were chosen. For example it was anticipated that, traders with relatively less capital, and those who are less educated would prefer joint exploitation agreements compared to traders who have more capital. However, because of incidences of co-linearity that arise when many socioeconomic variables are included in a model (Holmes and Adamowicz 2002), several interactions were tested following Speelman et al. (2010) by running different models and those that were consistent were finally retained.

Results of different interactions revealed that most socioeconomic variables shown in Table 3 were not significant in determining choices. However, membership in traders association, sex of the trader and experience in business were found to significantly influence the choice of respondents.

As was indicated in equation 6, by assuming a linear utility function, the monetary value of a marginal change from one attribute level to another (willingness to pay) is derived by dividing the implicit marginal utility of that attribute (coefficient) by the implicit marginal utility of the regeneration tax. Willingness to pay values (Table 6) derived from model 2, and subject to the socio-economic characteristics of traders can be viewed as the value of a marginal change from one attribute level to another. For example, a change from a centralized-complicated procedure to a decentralised-simplified procedure is valued at 22FCFA per kg (Table 5) while an increase in the duration of permits by one year is valued at 5.6 FCFA per kg. This implies that a decentralised and simplified process is four times as important as increasing the duration of permits from one to three years. The positive signs on the willingness to pay values suggest that traders are generally going to benefit from a change in any of the attributes.

DISCUSSION

This study used choice experiment as a method to incorporate stakeholders' views in the design and evaluation of policy options and to assess traders' perceptions about proposed regulations that are expected to entice them to better comply with the forestry law governing trade in NTFP in Cameroon.

Results of the study show that all the selected attributes contributed in influencing traders' choices of complying with the law. In other words, the traders would prefer to choose one of the proposed options rather than to opt not to choose any. Above all, these results suggest that all the attributes were relevant to the traders. This relevance may be related to the validity of the choice experiment which takes into consideration insights from the target beneficiaries through stakeholders' workshops and focus group discussions, the latter involving representatives of the sampled population (Alpizar et al. 2001, Horne 2006). This approach contributed in selecting those options that the traders' considered were fair or would contribute in reducing transaction costs in the process of complying with the permit system. The results therefore confirm two things (i) that participatory approaches are useful in selecting options relevant to formulating policies and legislations (Blaser 2010, Kaine et al. 2010, Mayers and Bass 2004, Ramcilovic-Suominen and Epstein 2012, Sutinen and Kuperan 1999) and (ii) that choice modelling is ideally suited to analyse the significance of the selected options (Birol et al. 2006, Hanley et al. 2001).

Based on the WTP measures, the option most valued by the sampled traders was decentralisation of the application process to the regions and reduction in the volume of paper work involved. For example, by decentralising and simplifying the process, traders would be made better off to the extent of 22 FCFA/kg. This is higher than the current regeneration tax of 10 FCFA/kg. This means that most traders may be willing to pay a higher regeneration tax in order for the process to be simplified. If this is done the benefit does not only accrue to traders because regeneration tax can be an important source of revenue for the Government (Betti 2007). The latter interpretation of traders willing to pay higher regeneration taxes is different from the negative sign obtained for the coefficient of the regeneration tax in Table 4. The sign for the coefficient of the cost component in the choice experiment is interpreted as a test for theoretical validity of the model (Alpizar et al. 2001, Speelman et al. 2010). The negative sign obtained in this study is thus in line with rational economic thinking that individuals would like to minimise cost which suggest that all things being equal, traders would prefer to pay lower regeneration taxes. However, they would prefer to pay higher regeneration taxes to see the process improved from complicated to simplified procedures.

Some socio-economic indicators such as turn-over, which was used as a proxy for capital, were hypothesised to influence most of the attributes and attribute levels. However, it did not turn out to be a significant factor. This is definitely because most traders interviewed are small scale traders which imply that variability amongst them was too low to capture an effect. The policy implication of this is that traders, at least of the sample population generally perceive problems related to permits in the same way and a common policy will be binding for all of them. However to have a bigger picture of the policy options to improve compliance in the forest sector in Cameroon, it would be important to extend this study to other sectors where illegal activities are common such as illegal timber exploitation and wildlife poaching.

Literature reports that marketing groups can play significant roles in reducing transaction costs related to the search of information, negotiation and monitoring of activities (Bienabe et al. 2004, Griffon 2001, Shepherd 2005, Tieguhong et al. 2012) and this has been empirically reported to be possible in the NTFP sector in Cameroon (Foundjem-Tita et al. 2011). As expected therefore, membership in traders association had significant positive effects on traders' decision to choose joint approvals. This is important as one of the political objectives of facilitating CIGs in Cameroon is for small scale actors to mobilise limited resources to meet costs, something that would not be possible if traders operated individually. Narrative discussions with some traders revealed that the desire to obtain official papers is one of the main objectives why such groups remained united. It is therefore obvious for traders who are in groups to derive higher utilities for joint exploitation agreements than those who operate individually. Although traders' focus in this case is on reducing transaction costs, the role of groups in facilitating compliance through the application of social sanctions to members has also been reported (Ostrom 1990, Ramcilovic-Suominen and Epstein 2012, Sutinen and Kuperan 1999). Thus if CIGs are also allowed to apply for permits members can put pressure on others to comply with the law.

Women are generally reported to be less successful in small businesses than men (Amine and Staub 2009, Chirwa 2008, Coleman and Robb 2009, Loscocco et al. 1991, Shelton 2006) but in this study women had significantly higher capital compared to men. Capital was not significant in determining choices, meaning that the nature of the business characterised by harassment from forestry and police officials requesting permits at check points may explain why more women compared to men would prefer to operate as a group. The choice of women may be related to the fact that they may not have equal resistance to confront the police and forestry officials, thus by joining common initiative groups they feel they can easily succeed in going through the process. The latter explanation is supported by Amine and Staub (2009) who report that women entrepreneurs in sub-Saharan Africa face daunting array of challenges which negatively affect their business. These negative challenges generally stem from the socio-cultural, economic, legal and political environment in which they live. These additional prejudices are expressed through differential attitudes towards women in general.

Compared to no transferability the introduction of a transferable permit system regulated by the government did not significantly increase the preference of the traders. However when the government regulated permit system was interacted with experience in business, the effect became negatively significant. The negative significance means that traders with experience in handling the species were less likely to be motivated by transfers of permits that had to pass through the government. This is probably related to negative experiences with bribery and corruption encountered in dealing with forests and police officials who enforce the law on permits (Cerutti and Tacconi 2006, Ndoye and Awono 2010). This negative experience might have resulted in a general lack of trust in government institutions by the experienced traders. The implication of this is that even if the Government of Cameroon allows transferable permits, they may not be readily accepted as traders will be distracted by high transaction costs involved in the transfer procedure if the government coordinates the transfers.

CONCLUSIONS

Given the difficulties stakeholders in the forestry sector face to comply with laws which sometimes are related to transaction costs resulting from excessive regulations or dissatisfaction as a result of perceived unfairness of laws and regulations; it is important for research to assist policy makers through the provision of advice about rational and legitimate preferences as perceived by targeted stakeholders and feasible on the part of the Government. The case study applied the choice experiment technique to elicit traders' evaluation and preferences for various policy options that are hypothesised may lure them to comply with the regulation on NTFP permits in Cameroon.

Results reveal that the following factors were significant in determining traders' choices to select an option and would thus encourage compliance with the permit system: (i) CIGs and not only limited liability companies should be recognised as organisational structures that are eligible to obtain permits; (ii) the process has to be simplified by reducing paper work and by decentralising procedures to the regions; (iii) transfers between permit holders should be allowed and, (iv) the duration of the permit should be increased by at least one year.

Because all the selected attributes were significant in determining traders' choices, the approach supports existing knowledge that stakeholders' participation is an important process in the design and formulation of policies and laws that can be widely accepted by the targeted group. Likewise, choice experiment can be an important method to enhance stakeholders' participation because it makes provision for a representative sample of the target group to take part in the selection of attributes and their levels. In this respect, aspects of economic feasibility (transaction costs reduction) and social acceptability (perceived fairness) of the proposed policies and regulations are assessed by the target group.

To the best of our knowledge this is one of the few studies that used the choice experiment technique to assess policy options in the NTFP sector especially when applied to the Congo basin forests. As such, the relevance of the method needs to be tried in other empirical research especially in selecting the most important attributes and their levels that are relevant in the formulation of new policies and laws.

ACKNOWLEDGEMENTS

The authors would like to thank two anonymous referees for their valuable contribution in shaping the content and structure of the paper. Many thanks also go to all the traders and experts who respectively took part in focus group and expert discussions. The study was funded by the Belgian Development Cooperation through the project 'Agroforestry Tree Products for Africa', implemented in Cameroon and the Democratic Republic of Congo by the World Agroforestry Centre and partners. The authors would thus like to thank the donor organisation.

REFERENCES

ADAMOWICZ, W., LOUVIERE, J. and SWAIT, J. 1998. Introduction to attribute-based stated choice methods. Edmonton, Alberta-Canada, ADVANIS.

ALPIZAR, F., CARLSSON, F. and MARTINSSON, P. 2001. Using choice experiments for non market valuation. Working paper in economics no. 52. Working paper Department of Economics, Goteberg University.

AMINE, L.S. and STAUB, K.M. 2009. Women entrepreneurs in sub-saharan africa: An institutional theory analysis from a social marketing point of view. Entrepreneurship & Regional Development 21(2): 183-211.

ASSEMBE MVONDO, S. 2009. State failure and governance in vulnerable states: An assessment of forest law compliance and enforcement in cameroon. Africa Today 55(3): 85-102.

AWONO, A., INGRAM, V., SCHURE, J. and LEVANG, P. 2012. Guide for small and medium enterprises in the sustainable non-timber forest product trade in central africa Bogor, Indonesia, CIFOR.

BETTI, J.L. 2007. Perspectives d'une fiscalite appropriee promouvant le commerce et la gestion durable des produits forestiers non ligneux en afrique centrale. Gcp/raf/398/ ger renforcement de la securite alimentaire en afrique centrale a travers la gestion et l'utilisation durable des produits forestiers non ligneux., FAO: 59.

BIENABE, E., CORONEL, C., LE COQ, J.-F. and LIAGRE, L. 2004. Linking small holder farmers to markets: Lessons learned from literature and analytical review of selected projects. Study report. Washington D.C., World Bank.

BIROL, E., SMALE, M. and GYOVAI, A. 2006. Using a choice experiment to estimate farmers' valuation of agro-biodiversity on hungarian small farms. Environmental and Resource Economics 34: 439-469.

BLASER, J. 2010. Forest law compliance and governance in tropical countries: A region-by-region assessment of the status of forest law compliance and governance in the tropics, and recommendations for improvement. A. Sarre, FAO and ITTO: 27.

CERUTTI, P.O. and TACCONI, L. 2006. Forests, illegality, and livelihoods in cameroon. Working paper no. 35. Bogor, Indonesia, Center for International Forestry Research.

CHIRWA, E.W. 2008. Effects of gender on the performance of micro and small enterprises in malawi. Development Southern Africa 25(3): 347-362.

COLEMAN, S. and ROBB, A. 2009. A comparison of new firm financing by gender: Evidence from the kauffman firm survey data. Small Business Economics 33(4): 397-411.

CONTRERAS-HERMOSILLA, A. 2002. Law compliance in the forestry sector. An overview. Washington D.C., World Bank Institute.

DJEUKAM, R. 2006. Le cadre legislatif et reglementaire de l'utilisation des produits forestiers non ligneux au cameroun. Projet renforcement de la securite alimentaire en afrique centrale a travers la gestion et l'utilisation durable des produits forestiers non ligneux, gcp/raf/398/ger,fao comifac--gtz Yaounde, Cameroon, FAO: 47. www.fao.org/docrep/012/al029f/al029f00.pdf.

DJEUKAM, R. 2007. Le cadre legislatif et reglementaire de l'utilisation des produits forestiers non ligneux au cameroun. Projet renforcement de la securite alimentaire en afrique centrale a travers la gestion et l'utilisation durable des produits forestiers non ligneux, gcp/raf/398/ger,fao comifac--gtz Yaounde, Cameroon, FAO: 47. www.fao.org/docrep/012/al029f/al029f00.pdf.

FAO, EU, COMIFAC, SNV, ICRAF and CIFOR. 2010. Proposition d'amelioration du cadre politique et legale regissant la gestion des pfnl au cameroun. Sous groupe de travail thematique sue les produits forestiers non lignuex (pfnl). Relecture de la politique forestiere au Cameroun et des textes de la loi No. 94-01 du janvier 1994: 16 www.fao.org/forestry/enterprises/45716/en/.

FOKOU SAKAM, I. 2010. Rapport de la reunion de presentation des resultats des sous-groupes thematiques sur les pfnl, la gestion communautaire des ressources forestieres et la gouvernance forestiere. Salle de conference du minfof, le 03 novembre 2010. Www.Fao.Org/forestry/ enterprises/45716/en/. Yaounde., FAO-PFNL, SNV 59.

FOUNDJEM-TITA, D., D'HAESE, M., DEGRANDE, A., TCHOUNDJEU, Z. and VAN DAMME, P. 2011. Farmers' satisfaction with group market arrangements as a measure of group market performance: A transaction cost analysis of non timber forest products' producer groups in cameroon. Forest policy and Economics 13(7): 545-553.

FOUNDJEM-TITA, D., SPEELMAN, S., D'HAESE, M., A., D., VAN HUYLENBROECK, G., VAN DAMME, P. and TCHOUNDJEU, Z. 2012a. Trade permits dealys and costs in the non timber forests products value chain in cameroon Tree products value chains in Africa, sharing innovations that work for smallholders Yaounde, Cameroon

FOUNDJEM-TITA, D., TCHOUNDJEU, Z., SPEELMAN, S., D'HAESE, M., DEGRANDE, A., ASAAH, A., VAN HUYLENBROECK, G., VAN DAMME, P. and NDOYE, O. 2012b. Policy and legal frameworks governing trees: Incentives or disincentives for smallholder tree planting decisions in cameroon? Small Scale Forestry.

GOVERNMENT OF CAMEROON. 1992. Law no 92/006 of 14 august 1992 relating to co-operative societies and common initiative groups National Assembly of Cameroon.

GOVERNMENT OF CAMEROON. 1994. Loi no. 94/01 du 20 janvier 1994 portant regime des forets, de la faune et de la peche. Republic of Cameroon: 24.

GOVERNMENT OF CAMEROON. 1995. Decret no. 95/53/ pm du 23 aout 1995 fixant les modalites d'application du regime des forets Prime minister Head of Government. Yaounde.

GREENE, W.H. 1997. Econometric analysis New Jersey, Prentice Hall.

GRIFFON, M. 2001. Filieres agroalimentaires en afrique comment rendre le marche plus efficace. Paris, France, Directions Generales de la Cooperations Internationales. Ministere des affairs Etrangeres.

HANLEY, N., MOURATO, S. and WRIGHT, R.E. 2001. Choice modelling approaches: A superior alternative for environmental valuation Journal of Economics surveys 15(3): 435-462.

HANLEY, N., WRIGHT, R.E. and ADAMOWICZ, V. 1998. Using choice experiments to value the environment. Environmental and Resource Economics 11(3-4): 423-428.

HOLMES, T.P. and ADAMOWICZ, V.L. 2002. Attribute based methods A primer on non market valuation. K.J. Boyle, P.A. Champ and T.C. Brown. Boston Kluwer Academic publishers 588.

HOPE, R.A., BORGOYARY, M. and AGARWAL, C. 2005. Desinging a choice experiment to evaluate adoption of organic farming for improved catchment environmental services and poverty reduction Centre for land use and water resources research, University of Newwasle and Winrock International, India.

HORNE, P. 2006. Forests owners' accpetance of incenetive based polocy instruments in forests bio-diversity conservation--a choice experiment based approach. Silva Fennica 40(169-178).

INGRAM, V., NDUMBE, L.N. and EWANE, M.E. 2012. Small scale, high value: Gnetum africanum and buchholzianum value chains in cameroon. Small Scale Forestry. INGRAM, V.J. 2012. Governance of non-timber forest products in the congo basin. ETFRN News (53(April)): 36-45.

KAIMOWITZ, D. 2003. Forest law enforcement and rural livelihoods. International Forestry Review 5(3): 199-210.

KAINE, G., MURDOCH, H., LOUREY, R. and BEWSELL, D. 2010. A framework for understanding individual response to regulation Food policy 35(2010): 531-537.

LAIRD, S.A., INGRAM, V., AWONO, A., NDOYE, O., SUNDERLAND, T.C.H., FOTABONG, E.L. and NKUINKEU, R. 2010. Integrating customary and statutory systems: The struggle to develop a legal and policy framework for ntfps in cameroon. Wild product governance: Finding policies that work for non-timber forest products: People and plants international conservation series. S.A. Laird, R.J. McLain and R.P. Wynberg. London Earthscan: 53-70.

LAIRD, S.A., WYNBERG, R. and MCLAIN, R.J. 2009. Wild product governance:Laws and policies for sustainable and equitable non-timber forest product use, A collaboration of: United Nations University, Centre for International Forestry Research, People and Plants International, Environmental Evaluation Unit, University of Cape Town, and the Institute for Culture and Ecology.

LANCASTER, K.J. 1966. A new approach to consumer theory. Journal of polictical economy 74(2): 132-157.

LOSCOCCO, K.A., ROBINSON, J., HALL, R.H. and ALLEN, J.K. 1991. Gender and small business success: An inquiry into women's relative disadvantage. Social Forces 70(1): 65-85.

MAYERS, J. and BASS, S. 2004. Policy that works for forestrs and people. Real prospects for governance and livelihoods. Sterling. Virginia, VA, Earthscan, Uk and USA.

MAZOTTA, M. and OPALUCH, J. 1995. Decision making when choices are complex: A test of heiner's hypothesis. Land Economics 71: 500-515.

MCCANN, L., COLBY, B., K.WILLIAM, E., KASTERINE, A. and K.V., K. 2005. Transaction cost measurement for evaluating environmental policies Ecological Economics 52: 527-542.

MCFADDEN, D. 1974. Conditional logit analysis of qualitative choice behaviour Frontiers in econometrics P. Zarembka. New York, Academic Press.

METTEPENNINGEN, E., BECKMANN, V. and EGGERS, J. 2011. Public transaction costs of agri-environmental schemes and their determinants--analysing stakeholders' involvement and perceptions. Ecological Economics 70(4): 641-650.

NDOYE, O. and AWONO, A. 2010. Policies for gnetum spp. Trade in cameroon: Overcoming constraints that reduce benefits and discourage sustainability Wild product governance: Finding policies that work for non timber forests products. People and plants international conservation series. S.A. Laird, R.J. McLain and R.P. Wynberg. London Earthscan.

NDOYE, O., RUIZ PEREZ, M. and EYEBE, A. 1998. The markets of non-timber forest products in the humid forest zone of cameroon, ODI, London, UK.

NGUENANG, G.M., FAPA, C., SCHINDLER, J. and MBOLO, J. undated. Le cout de la legalisation des organisations paysannes pour la collecte et la commercialisation des produits forestiers non ligneux au cameroun. Programme d'Appui au Programme Sectoriel Foret et Environnement (ProPSFE) Yaounde, Cameroon Deutsche Gesellschaft fur Internationale Zusammenarbeit (GIZ) GmbH.

NORTH, D.C., Ed. 1990. Institutions; institutional chnage and economic performance. The political economy of institutions and decisions. New York, Cambridge university press.

OSTROM, E. 1990. Governing the commons New York, Cambridge University Press.

RAMCILOVIC-SUOMINEN, S. and EPSTEIN. 2012. Towards and analytical framework for forest law complaince International Forestry Review 14(4).

RYAN, M., BATE, A., EASTMOND, C.J. and LUDBROOK, A. 2001a. Use of discrete choice experiments to elicit preferences. Quality in health care 10: 55-60.

RYAN, M., SCOTT, D.A., REEVES, C., BATE, A., VAN TEIJLINGEN, E.R. and RUSSELL, E.M. 2001b. Eliciting public preferences for healthcare: A systematic review of techniques. Aberdeen, UK, Health Technology Assessment.

SCHRECKENBERG, K., AWONO, A., DEGRANDE, A., MBOSSO, C., NDOYE, O. and TCHOUNDJEU, Z. 2006. Domesticating indiginous fruit trees as a contribution to poverty reduction Forests trees and Livelihoods 16: 35-51.

SHELTON, L.M. 2006. Female entrepreneurs, work-family conflict, and venture performance: New insights into the work-family interface*. Journal of Small Business Management 44(2): 285-297.

SHEPHERD, A.W. 2005. Association of market traders. Their roles, and potential for further development Agricultural Management, Marketing and Finance Service (AGSF) Occasional paper. Rome, FAO.

SPEELMAN, S. 2008. Water use efficiency and influence of management policies, analysis of the small scale irrigation sector in south africa, Ghent University Doctoral Thesis 191.

SPEELMAN, S., FAROLFI, S., FRIJA, A. and VAN HUYLENBROECK, G. 2010. Valuing improvements in the water rights system in south africa: A contingent ranking approach1. JAWRA Journal of the American Water Resources Association 46(6): 1133-1144. STAVINS, R.N. 1995. Transaction costs and tradeable permits Journal of Environmental Economics and Management 29: 133-148.

STREET, D.J., BURGESS, L. and LOUVIERE, J.J. 2005. Quick and easy choice sets: Constructing optimal and nearly optimal stated choice experiment International journal of research in marketing 22: 459-470.

SUR, D., COOK, J., CHATTERJEE, S., DEEN, J. and WHITTINGTON, D. 2006. Increasing the transparency of stated choice studies for policy analysis: Desiging experiments to produce raw response graphs Journal of Policy Analysis and Management 26(1): 189-199.

SUTINEN, J.O. and KUPERAN, K. 1999. A socio-economic theory of regulatory compliance. International Journal of Social Economics 26(1/2/3): 174-193.

TIEGUHONG, J.C., NDOYE, O., GROUWELS, S., ASSENG, Z.A., MASUCH, J. and SAKAM, I.F. 2010. Transparency and governance in non-wood forest product commodity chains, on the example of cameroon: A need to include nwfp in the flegt mechanism. COFO International Conference-Forum on Emerging Economic Mechanisms: Implications for forest related policies and sector governance. FAO Headquarters, Rome-Italy.

TIEGUHONG, J.C., NDOYE, O., GROUWELS, S., MALA, W.A. and BETTI, J.L. 2012. Rural enterprise development for poverty alleviation based on non-wood forest products in central africa. International Forestry Review 14(3): 363-379.

WELLS, A., DEL GATTO, F., RICHARDS, M., POMMIER, D. and CONTRERAS-HERMOSILLA, A. 2007. Rural livelihoods, forests law and the illegal timber trade in honduras and nicaragua. Illegal logging: "Law enforcement, livelihoods and the timber trade". L. Tacconi. London, Earthsacn.

YOUNG, O.R. 2005. Compliance and public authority: A theory with international applications. Baltimore, The John Hopkins University press.

(1) We thank an anonymous reviewer for this point.

(2) Partners refer to a consortium of international organisations among which the International Centre for forestry research (CIFOR); the World Agroforestry Centre (ICRAF) and the Netherlands Development Organisation (SNV Cameroon). These organisations including FAO as lead participated in the execution of the project titled Mobilisation and Capacity Building of Small and Medium size Enterprises in involved in NTFPs funded by the European Union.

(3) Actors invited to the stakeholders' workshop included: producers/collectors, traders, local NGOs, international NGOs, law enforcement officers including police and forest guards, Employees from MINFOF, the media, local councils etc.

D. FOUNDJEM-TITA [1,2,3], S. SPEELMAN [1], J.C. TIEGUHONG [4], M. D'HAESE [1], A. DEGRANDE [3], Z. TCHOUNDJEU [3], O. NDOYE [4], G. VAN HUYLENBROECK [1] and P. VAN DAMME [2,3]

[1] Ghent University, Faculty of bioscience engineering, Department of agricultural economics

[2] Ghent University, Faculty of bioscience engineering, Laboratory of tropical and subtropical agronomy and ethnobotany

[3] World Agroforestry Centre, West and Central African Region (ICRAF-WCA)

[4] Food and Agricultural Organisation of the United Nation, Non Wood Forest Product Project, Yaounde Cameroon

Email: dftita@yahoo.com and tita.divinefoundjem@ugent.be

TABLE 1 Summary of selected attributes and their levels

Description of the selected
attributes                          Attribute Levels

Type of document and where to       1.  Decentralised-simplified
apply for an approval (Type_EA)         procedures specific to NTFPs
Whether special approval should         obtained from the regions
be authorised for NTFPs different   2.  Centralised procedures
from what timber exploiters             obtained from Yaounde
do submit.                              (Statu quo)

Joint ownership of approval         1.  Approval can be jointly
(JI_EA)                                 owned by a group of traders
Whether or not a number of              through a CIG
traders can unite to apply for      2.  Approval should only be
an approval as a CIG                    given to a single trader

Transferability of Exploitation     1.  Can be transferred (status
permit (Tr_Ept)                         quo)
Whether a holder of an              2.  Cannot be transferred
exploitation permit should          3.  Can be transferred with
transfer his rights to another          authorisation from
trader provided the quantity            government
authorised in the permit has not
been exhausted

Duration of exploitation permit     1.  One year (Status quo)
(D_Ept)                             2.  Three years
Number of years an exploitation
permit should be valid

Regeneration Tax (R_Tax)            1.  5 FCA/kg
Up to how much can traders pay      2.  10 FCFA/Kg
as regeneration tax                     (Status quo)
                                    3.  20 FCFA/kg

Description of the selected         Effect being captured by the
attributes                          attribute:

Type of document and where to       Transactions costs
apply for an approval (Type_EA)     (time and costs to obtain the
Whether special approval should     legal) Effect of other costs e.g
be authorised for NTFPs different   Travel costs to Yaounde
from what timber exploiters
do submit.

Joint ownership of approval         Whether by jointly owning an
(JI_EA)                             approval traders reduce
Whether or not a number of          transaction costs and will
traders can unite to apply for      entice them to ask for permits
an approval as a CIG

Transferability of Exploitation     Whether by preferring to hire
permit (Tr_Ept)                     permits traders keep transaction
Whether a holder of an              costs low
exploitation permit should
transfer his rights to another
trader provided the quantity
authorised in the permit has not
been exhausted

Duration of exploitation permit     Cost of keeping the agreement
(D_Ept)                             for a short period i.e. if
Number of years an exploitation     traders go more for more than
permit should be valid              three years then the costs of
                                    obtaining the document just for
                                    one year is comparatively higher

Regeneration Tax (R_Tax)            To measure actual willingness to
Up to how much can traders pay      pay for regeneration tax
as regeneration tax

TABLE 3 Socio-economic characteristics of respondents

Variable        Description                       Percentage

Principal       If species is traders principal   No = 19
activity        source of revenue                 Yes = 81

Species                                           Gnetum = 30
                                                  Irvingia = 31
                                                  Kola = 26
                                                  Ricinodendron = 13

Membership      If respondent belong to a         No = 46
                traders association               Yes = 54

Sex             Sex of respondent                 Female = 40
                                                  Male = 60

Boundary of     Level of exportation              Low export = 37
exchnage                                          High export =63

Experience of   Number of years trader has
trader          been in the NTFP business

Value_FCFA      Trader's annual turnover in
                2011 in 1000 FCFA

Age             Age of respondent

Household       Household size

Education       Number of years of education

Variable          Mean (s.d)         Range

Principal
activity

Species

Membership

Sex

Boundary of
exchnage

Experience of     11.4 (9.16)        1-50
trader

Value_FCFA      24,768 (31,862)   250-150,000

Age                40 (8.30)         23-65

Household         7.9 (3.86)         1-18

Education         10.7 (3.20)        0-17

TABLE 4 Conditional logit model: determinants of choice to
obtain permits

                                                     Model 1

Attribute                                     Coefficient   SE

Attribute specific constant                   -14.73        47.331
Simplified decentralised procedures (base     1.06 ***      0.09
  level: complicated procedure)
Joined Ownership of an approval (base level   0.45 ***      0.08
  single ownership)
Traders based transferable permits            0 42 ***      0.11
  (non transferable permits)
Government based transfer (base level         0.05 ns       0.11
  non transferable permits)
Duration of exploitation permit               0.27 ***      0.08
Regeneration tax                              -0.06 ***     0.01
Membership in traders association
  * Joined ownership
Experience in business * Government based
  transferable permits
Sex * Join ownership
Model statistics
Log likelihood                                -362.15
Pseudo [R.sup.2]                              0.28
Chi square                                    0.00

                                                     Model 2

Attribute                                     Coefficient   SE

Attribute specific constant                   -15.30        512.97
Simplified decentralised procedures (base     1.10 ***      0.33
  level: complicated procedure)
Joined Ownership of an approval (base level   0.49 ***      0.18
  single ownership)
Traders based transferable permits            0.44 **       0.21
  (non transferable permits)
Government based transfer (base level         0.34 ***      0.15
  non transferable permits)
Duration of exploitation permit               0.28 **       0.13
Regeneration tax                              -0.05 ***     0.01
Membership in traders association             0.60 ***      0.17
  * Joined ownership
Experience in business * Government based     -0.027 **     0.011
  transferable permits
Sex * Join ownership                          -0.57 ***     0.17
Model statistics
Log likelihood                                -34.98
Pseudo [R.sup.2]                              0.31
Chi square                                    0.00

*** significant at 1%, ** siginifcant at 5%, ns = non significant

TABLE 5 Valuation of attribute changes

                                          Implicit willingness
Attribute change                            to pay FCFA /kg

Complicated to simplified                        22.00
Single to joined                                  9.80
No transfer--traders based transfer               8.80
No transfer--government based transfer            6.80
Duration of permits from 1 to 3 years             5.60
COPYRIGHT 2013 Commonwealth Forestry Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

 
Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Cameroon, Africa; non timber forest products
Author:Foundjem-Tita, D.; Speelman, S.; Tieguhong, J.C.; Dhaese, M.; Degrande, A.; Tchoundjeu, Z.; Ndoye, O
Publication:International Forestry Review
Article Type:Case study
Geographic Code:6CAME
Date:Jun 1, 2013
Words:11080
Previous Article:Exploring forest-related coping strategies for alleviating the HIV/AIDS burden on rural Malawian households/Exploration des strategies basees sur la...
Next Article:An economic approach to household collection of gum arabic from the wild/Une approche economique a la cueillette sauvage de la gomme...
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters