To recap so far, this paper argues that livelihood diversification is generally a good thing both for reducing rural poverty and for improving rural natural resources and environments. Diversification gives individuals and families wider experiences, it strengthens human capital, it generates cash resources that can be used for natural resource investments, it can take pressure off sensitive natural resources by providing alternative income streams, and so on. Yet diversification is inhibited by many branches of state policy. Mobility and spatially separated transactions are often inhibited, licensed and taxed. Transfers of people and goods across space are made difficult by formal and informal barriers to movement. Migration is often actively discouraged, and few states have facilitating regulatory frameworks for migration and remittance transfers whether these are internal within countries or across borders with other countries.
An important example of this discouraging policy environment is that of rural taxation, a topic that is relatively little researched in low income countries. Rural tax regimes are overlooked because their revenue yield at central government level is insignificant compared to other sources of government income. Some taxes are formal e.g. market dues and business licenses; others are informal and variable e.g. a basket of fish that must be given to the chief or village leader in order to be allowed to go fishing. Still others are bribes that must be paid to gatekeepers - fisheries, forestry and livestock officers, police, army, customs officials - in order to do things like move livestock, trade in fish, carry crops from one village or town to another, engage in cross border trade and so on.
The apparently small levels of these various payments give the false impression that they are unimportant to the economic vigour of rural areas, and of minor significance compared to other factors in explaining rising poverty and vulnerability. This is not so, as research in a number of African countries has begun to demonstrate (Fjeldstad, 2001a; 2001b; 2002; Ellis & Bahiigwa, 2003). The enormous range and variety of such payments adds to the oppressive hopelessness of individual and household level attempts to construct routes out of poverty. While small in tax revenue generation terms, market dues can represent up to 30 percent of the farm gate value of items being sold; likewise business licenses that are applied to even the most micro of start-up enterprises can represent a substantial proportion of net income generated. There is typically a strong anti-poor bias in tax and licensing regimes: market dues are proportionately higher on small quantities of produce (a basket or a heap) than on large ones (a sack or truckload); flat rate business licenses represent a larger proportion of turnover for smaller than larger enterprises operating in the same line of business.
Rural taxes tend to be multiple, complicated, bear no relation to service delivery, create numerous rent seeking opportunities, and accentuate relations of mistrust and subordination between ordinary citizens and those who possess revenue generating powers over them. They also inhibit market engagement and mobility since often the risk of incurring taxes, fees and fines increases markedly the more people engage in activities above the subsistence survival level or external to their communities of residence.
It might be thought that these types of barrier would tend to be tackled and moves made to dismantle them in Poverty Reduction Strategy Papers (PRSPs) that have poverty reduction as their overarching goal, and that are informed by widespread participation and consultation with stakeholder groups in society. However, PRSPs have proved disappointing as vehicles for adopting cross-sectoral policy thinking about poverty reduction. Despite the lip service they pay to contemporary development discourse (participation, empowerment, gender awareness etc.), the majority of PRSPs end up looking like elaborate sectoral expenditure plans, and indeed they are often supported in the background by strategic documents (agricultural development strategies etc) that are unashamedly based in an orthodox sector-based view of the world.
Migration is typical of the type of cross-cutting phenomenon that PRSPs seem poorly equipped to handle, and it therefore provides a suitable example of the mismatch between micro livelihood priorities and macro policy formulation that tends to occur a lot in practice. In a survey of 48 PRSPs undertaken in 2003, migration was found not to be mentioned at all in 21 of them. Most of the remaining 27 PRSPs referred to migration in negative or pejorative terms. Seventeen, for example, posed internal migration as a problem for development, eight cast migration as a cause of urban poverty, and others pointed to the negative effects of migration in spreading HIV/AIDS and contributing to crime. Eight PRSPs expressed the need for internal migration to be controlled by the state i.e. for rural-urban migration to be reduced (Black et al. 2003: pp.18-19).
The foregoing discussion suggests a number of areas in which action research could both improve our understanding of diversification-natural resource access linkages, and contribute actively to improving the policy environment for rural poverty reduction. These areas are listed here and amplified in subsequent discussion:
land tenure reform
rural taxation and licensing (linked to decentralization)
migration and remittances
diversification and CBNRMs
integration of cross-sector thinking in PRSPs
Almost no systematic research has hitherto been conducted on the links between land access, farm size, farm productivity and livelihood diversification; nor on the gender dimensions of these relationships. The dangers of superficial dismissal of customary tenure as an allocative mechanism for land have been emphasized by some researchers, and ensuing benefits of customary tenure identified (Shipton & Goheen, 1992; Platteau, 1996). The specific link between tenure and farm productivity was investigated for several SSA countries by Place & Hazell (1993) with the finding that customary tenure made no statistically significant difference to farm yields as compared to freehold in those case-studies. On the gender side, a number of observers have noted the erosion of matrilineal land inheritance systems in SSA societies where this has been a customary feature of land tenure.
With large numbers of countries revising their legislative frameworks governing land tenure over the past fifteen years (Toulmin & Quan, 2000), and decentralization proceeding apace in many countries (see also below), this is an appropriate juncture to carry out research at district and community levels into the more complex inter-relationships between land tenure and poverty reduction that are suggested by previous discussion. In particular, such research might seek to explore, document, and derive policy-relevant findings with respect to:
effectiveness of new legislation at achieving stated goals, when applied at district, sub-district and community levels;
institutional mechanisms that have arisen at local levels in response to national level changes; their effectiveness in providing frameworks for equitable land access and for solving disputes;
distributional benefits and harm so far realized, related to access and farm size aspects of land tenure;
detailed examination of livelihood diversification relationships to land tenure and access; the effects of temporary and long term absence by family members from different income and wealth groups; re-allocations that occur or are inhibited by customary and legislative frameworks;
investigation of workable mechanisms for introducing more flexibility into land access, while protecting the access rights of those least able to protect themselves. The emphasis here is on workable i.e. transparent, low cost, village level institutions that can help to stimulate and regulate a more vigorous rental market in land; and
the position of women in all matters related to livelihood diversification and land access would be a fundamental feature of all the foregoing dimensions of research.
The critical positioning of decentralization in contemporary donor thinking about poverty reduction has been noted earlier in this paper. There is a huge literature on this, mainly dating from the 1990s (Crook & Manor, 1998; Manor, 1999). The principal thrust has been towards so-called democratic decentralization, implying local elections of councillors to represent their constituents in district level councils or assemblies. In some countries, previous larger administrative entities (provinces or regions) have been broken up into smaller ones with a resulting proliferation of parallel local government functions. For example in Tanzania decentralization resulted in the creation or redesignation of 96 rural districts replacing former local government functions conducted in 20 regions (Ellis & Mdoe, 2003). The democratic procedures of decentralization vary widely. In some instances, there are direct elections by citizens of councillors according to area-based constituencies, in others a tiered electoral system operates whereby directly elected representatives at sub-district tier comprise the electorate that then vote for alternative candidates standing for district council positions.
As already noted a critical research issue in decentralization concerns local revenue generation. While district level government continue to be predominantly supported by central government transfers, with intended increasing autonomy in making spending decisions (fiscal decentralization), the ultimate intention is for them to rely more on generating their own resources thence achieving a desirable synergy between taxes paid and citizens expectations of improved public revenues.
Local revenue generation is important for the intersection between livelihood diversification and natural resource management in several ways:
local taxes are often levied on crops, livestock and other NR products;
such taxes can distort relative prices, adversely effect resource allocation decisions, and inhibit the expansion of the monetised economy in rural areas;
such taxes are often regressive in effect, even if not in intention, due to flat rate impositions that penalise small quantities compared to large quantities or things produced by the poor more than those produced by the better off;
business licenses are often utilized by local councils primarily as a revenue enhancement device rather than as a tool for establishing an enabling regulatory environment for small businesses;
land itself may be taxed under some local revenue systems, or the potential exists for such tax as a progressive alternative to regressive commodity taxes, or land taxes may be used to discourage non-utilization of land by absentee landowners; and
tax collection by itself often exhibits serious flaws manifested by substantial leakages between point of collection and delivery of receipts to local tax offices
Overall, substantive policy issues surround the impact of local taxation systems both with regard to the enabling versus disabling public sector environment for livelihood diversification, and on the utilization of land as a resource, and fertile area for action research could be developed in this area.
Migration plays crucial roles in diminishing vulnerability and lessening poverty in low income countries (de Haan, 1999; Skeldon, 2002). Migration may be seasonal (e.g. to participate in harvests), circular (involving periods away and periods at home), rural-urban (with a degree of permanence) or international (cross-border and distant migrations). Recent literature has emphasised the significance of remittances in international financial flows to developing countries (Nyberg et al., 2002); as well as the complex social as well as economic ties that bind migrants to the livelihood circumstances of those they leave behind (de Haan & Rogaly, 2002; Kothari, 2003).
Migration also involves important interactions with land, land tenure and the use made of land. Those left behind by migrants act as stewards and retainers of land that in the absence of secure ownership rights might otherwise be in jeopardy of being lost to the family. In the absence of a rental market, this land may be poorly utilized. Alternatively, migrant earnings can enable improved agricultural practices and raise yields as discussed earlier in this paper. A series of research questions are raised by these interactions and cumulative processes:
how is management of land affected when individual family members migrate?
what differences are made between male and female migration for land management?
what is the evidence concerning use of remittance income for agricultural improvement compared to other uses?
are those ‘left behind’ impoverished or their livelihoods improved when family members migrate?
how do land tenure systems affect migration decisions?
Decentralization and support for community-based management provide significant opportunities for rural people to gain increasing representation in decision-making that affects their access to and control over natural resources. Such opportunities hold out the promise, prominent in the rhetoric of both decentralization and CBNRM, of providing the poor with increased opportunity to defend their rights and interests. In practice, however, decentralization and CBNRM initiatives have been criticised from a range of perspectives.
An important strand of this criticism is essentially that such initiatives have been either, in the hands of politically naïve bureaucrats and technocrats or of political elites, deliberately designed to retain central control while decentralizing the burden of government (Ribot, 1999). There is a clear need for more political and historically contextualised analysis of CBNRM and its relationship to processes such as market liberalization and decentralization, in order to provide a challenge to blue-printed, instrumental versions of these processes that fail to provide any significant challenge to the conditions that allow poverty and inequality to prevail. Most donor-programmes, of necessity, strive for political neutrality on the surface, while strongly promoting current international neo-liberal orthodoxies in practice.
CBNRM initiatives are often conceived and implemented on sectoral lines. East African villages often have newly-developed fisheries management committees, forest-user groups, grazing committees, water-users’ associations and various farmer-based organizations. These are often supported by donors or NGOs and may co-exist with earlier state-created village environment committees and with the remnants of customary land tenure and resource access arrangements (themselves often created by colonial governments under indirect rule, rather than being, in some sense ‘traditional’ and ‘indigenous’; Ribot, 1999). Market-based institutions such as producer-organizations and trade co-operatives are also promoted under state and donor-funded programmes, alongside various civil-society organizations that may have direct or indirect impact on natural resource access. This complex institutional environment is seldom properly analysed prior to the design of interventions by donor programmes. It is not clear how conducive these institutional arrangements are to supporting livelihood diversification, nor how effective they are in blocking or facilitating people’s activities.
A frequently-cited reason for the failure of CBNRM institutions to represent the interests of the poor is the observation that the poor (and particularly poor women) are often so burdened by work that they do not have the time and resources to become involved in decision-making, conflict resolution, management planning, lobbying and other activities of village-based CBNRMs. This constraint is multiplied in households with diversified livelihoods in villages where there is a different CBO regulating access to every natural resource. Each of these institutions needs to be financially self-sustaining, and potentially develop a surplus for investment in community-goods (e.g. successful forest-user groups in Nepal). Financial sustainability often comes from membership contributions; to be a member of a beach village committee or a forest-user group or a water-users association requires subscription. The more diversified the CPR basis of a livelihood, the more subscriptions need to be paid. Obviously this may act as a barrier to diversification. Cross-sectoral analysis of local-level institutions, with emphasis on their comparative strengths and weaknesses, may help guard against unsustainable institutional proliferation and overlap.
Policy-level analysis of consistency between forestry, fisheries, communal grazing, water resources and land tenure programmes in PRSPs may be useful. Programmes to reform and improve the management of each of these resources are nearly always carried out separately, often with funding from different donor agencies, and often with little communication. There are so many common themes and policy directions, as well as practical overlaps at ground level that keeping such experience in separate compartments is counter-productive. While the theoretical literature on CBNRM is well integrated across the sectors, the empirical literature is invariably sector-focused; fisheries, forestry and irrigation experiences are seldom shared at the level of practitioners and policy makers at national level.
FAO has supported CBNRM in a variety of natural resource sectors (fisheries, forestry, pastoralism, irrigation), yet, to our knowledge, no comparative analysis of its project and programme support across these different sectors has taken place, despite their conceptual similarity and potential for lesson-learning. This would be relatively easy to achieve and would have the advantage, hopefully, of drawing together work carried out under a common organizational mandate, with similar expectations and standards of reporting. This would bypass one of the major constraints to comparative analysis of CBNRM experiences: the use of different measures of success (independent variables) and proliferation of possible causative factors (dependent variables), confounded by different levels and modes of reporting by different agencies (Agrawal, 2001). An action-research agenda to address these issues could focus on the following questions:
Is CBNRM conceived primarily in instrumental terms, or does it seek to challenge social inequities, in association with decentralization?
How can CBNRM institutions be designed to both sustain the natural resource base and challenge existing social inequities, to benefit the poor?
Do sector-based CBNRM institutions at local level hinder livelihood diversification?
Do diversified livelihood strategies undermine the effectiveness of sector-based CBNRM institutions?
What lessons for CBNRM programme design and process can be learnt by comparing FAO’s sector-based programmes under a common analytical framework?
As the overarching framework for achieving poverty reduction, PRSPs should hopefully take the lead in creating policy environments that work with, rather than against, people’s own efforts to build routes out of poverty. This would entail considering carefully observable successful livelihood strategies and seeking to reduce barriers that inhibit people’s ability to pursue such strategies. It would also require PRSPs to take on a much more active coordinating role across different branches of government than has so far been manifested in first round PRSPs.
A promising policy research agenda lies in deconstructing and reconstructing a selection of PRSPs from the foregoing perspective. As already observed, PRSPs have a tendency to maintain rather rigid sectoral demarcations and to develop pro-poor policies along traditional sector lines. These may, or may not, work at cross-purposes to each other. PRSPs are also formulated at a national, aggregate, level, and seldom are they critically examined with respect to consistency between national level strategic pronouncements and the actual discharge of policy at intermediate levels of government down to the district and village. The former observations about rural taxation typify situations where different layers of the public sector institutional context to people’s lives may pull in opposite directions to each other. PRSPs may champion enabling environments as national level poverty reduction goals, but local governments may simultaneously be putting in place increasing barriers and blockages to exchange and mobility in the local rural economy.
