Igor Vinci, Carol Djeddah, May Hani *

Introduction

As developing countries cope with limited financial and organisational resources, extensive Social Protection (SP) for all rural poor is far from being achieved. Accordingly, governments must put in place innovative institutional arrangements and partnerships with all relevant stakeholders to compensate for these gaps.

SP, intended as a combination of social assistance, insurance, services and mechanisms of social equity, is increasingly considered by decision-makers as a strategy to relieve rural families from extreme poverty. Rural Organisations (ROs) represent their members in decisionmaking processes and provide various services to them with good coverage and social outreach where public institutions struggle to intervene. For these reasons, there is an argument for acknowledging ROs as relevant stakeholders in the formulation and implementation of SP strategies. For the purposes of this study, Rural Organisations are intended as formal and informal groups (e.g. village associations, producers" organisations, agricultural cooperatives, etc.), based on the participation of members acting together to attain agreed socio-economic objectives. Members can be vulnerable smallholders or rural workers whose livelihood mostly depends on productive activities in the sectors of agriculture, livestock, fisheries and forestry.

Firstly, we will present a rationale for considering the role of ROs in SP based on the current weaknesses of national SP systems. Subsequently, we use practical examples and two case studies to reveal the two dimensions where ROs actively contribute to giving protection to the rural poor: 1. ROs operate as partners in national SP programs through the establishment of participatory mechanism; 2. ROs conduct collective practices of social assistance, insurance, services and mechanisms of social equity, increasing the level of protection of their members.

On these grounds, it is suggested that governments look at the role of ROs more thoroughly, acknowledging and increasing their contribution, collecting evidence on their impact and drawbacks, and supporting their flourishing.

A rationale for including ROs

Formal institutions of developing countries are still unable to fully satisfy the need of Social Protection (SP) of the rural poor in terms of social assistance, insurance, financial and mechanisms of social equity. In particular, they struggle to ensure extensive coverage for all regions affected by poverty and inclusiveness to reach the poorest citizens. Among rural poor, many operate within informal and often "invisible" economies that are difficult to incorporate. Particularly in rural areas, there are high percentages of informal employment and low rates of registration in civil registries, making access to national social security systems more difficult (e.g. job insurance, pensions, health).

Secondly, governments struggle to get comprehensive information on the socio-economic status of potential beneficiaries. Discrepancies emerge in terms of "who" needs protection and "what is needed". Accordingly, broad participation of those who will be affected by the policymaking process is of utmost importance in order for SP services to be relevant and adequate to the needs of beneficiaries.

Finally, governments struggle to put in place effective and efficient delivery mechanisms of SP. For instance, targeting mechanisms are sometimes unfit to identify the poorest in efficient and equitable ways, and programs fail to distribute limited financial resources with a real impact on target groups. Corruption and political distortions hamper the accountability of certain local institutions and the transparent administration of delivery mechanisms, increasing the misuse or dispersion of SP resources. (HLPE, 2012, Basset et al, 2012).

The constraints and gaps of SP systems are frequently ascribed to the lack of financial resources and infrastructures. However, weaknesses in the national and local governance of SP can be equally determining. At national level, scattered programs are not sufficiently connected and "scaled-up" to cover extensively all SP needs. SP systems lack sufficient recognition and enactment across governmental structures and sectors (agriculture, education, health, etc.), although governments are increasingly establishing inter-departmental commissions in the effort to coordinate SP across programs and sectors. Central governments also find difficulties in linking with local authorities, and fail to provide the right incentives for them to deliver programs efficiently. Finally, at local level, the frequent lack of formalisation of participatory mechanisms limits the contribution of civil society and its representative instances. (Devereux, 2014, Roelen, 2014).

The governments of some developing countries (e.g. Brazil, Mexico, Indonesia, to name a few) have already take action to respond to these challenges. There is an emerging tendency towards decentralisation processes and the devolution of more authority to local governments, increasing their responsibility and accountability. In such a context, the International Labour Organisation (ILO) has encouraged governments to pursue innovative governance systems in SP with the participation of relevant stakeholders affected by SP policies and programs. ILO"s Conventions No. 102 and 128 state that relevant stakeholders and representatives of SP beneficiaries (including ROs) should participate in the design, implementation and monitoring of SP beyond mere consultation.

Accordingly, governments are searching out new partnerships with non-State actors to share the financial and administrative burden of SP; and they are increasingly adopting participatory mechanisms with civil society organisations for defining eligibility criteria, methods of delivery, mechanism for monitoring social services and accountable institutions, etc. ROs can find their place in this evolving institutional scenario characterised by decentralisation, partnerships and participatory policy-making. ROs play two main roles that partly compensate for the weaknesses of SP systems.

ROs can represent the rural poor, and advocate and negotiate for their needs with public and private actors. In the position of intermediaries, ROs have the potential to inform the design of SP policies and to take up certain tasks in the implementation of SP programs, e.g. targeting, registering, monitoring. ROs, therefore, operate as external service providers to publiclymanaged SP programs.

ROs conduct collective practices of SP in favour of their members through risk-sharing arrangements and the creation of economies of scale. Such practices support members to overcome their financial constraints, reduce transaction costs and explore better economic opportunities, thus managing risk more efficiently. In this case, ROs are internal service providers for their members, developing an important protection function in informal economies and societies.

Participatory mechanisms in SP systems

Improving the institutional framework of SP means recognising the important role and functions that can be conducted by non-State actors such as international organisations, NGOs, the private sector, civil society and Rural Organisations. These institutions, with their experience, information, resources and networks, contribute to filling the gaps shown by governmental institutions, improve the level of technical expertise, and increase the interaction between providers and users of SP services. Governments are establishing mechanisms for the participation of ROs in the attempt to improve the design and implementation of SP programs (e.g. targeting of eligible citizens, distribution of benefits, transparency and accountability of SP programs among others).

We have classified the participation of ROs in governmental social protection mechanisms in different forms according to higher degrees of autonomy transferred to ROs.

Consultation occurs when central governments seek the advice of local institutions. These can take the form of social councils/commissions/committees with the participation, among others, of ROs" representatives. In the program Oportunidades in Mexico, social councils are responsible for watching over inclusion and exclusion errors and informing about priority SP needs. Social councils in Brazil (program Fome Zero) are also called upon to inform State and central governments on the monitoring of supporting infrastructures and the performance of oversight activities. Furthermore, they inform and facilitate access of beneficiaries to social services. Particularly in South America, consultation mechanisms have represented the foremost effort to establish a dialogue with civil society around the development of social protection and other strategies for reducing poverty.

Collaboration occurs when governments decide to assign some tasks to ROs, according to their capabilities and potential to improve the delivery of SP services to beneficiaries. For instance, ROs collaborate in programs of local and institutional procurement (the cases of the Brazilia Sp programs PAA and PNAE are illustrated below) that operate on two different levels of protection: 1) they guarantee income generation for poor smallholders by securing access local markets; and 2) they provide food assistance to pupils in schools, encouraging their attendance and improving their nutrition security. Community targeting operates with the participation of community associations to improve the transparency and accountability of targeting processes (examples are found in the Vision Umurenge Program in Rwanda, the Gratuitous Relief in Bangladesh, or the Food Security Program in Ethiopia).

This approach has sometimes contributed to avoid problems of social cohesion within communities, caused, in other cases, by the top-down appointment of beneficiaries (e.g. Oportunidades in Mexico). In other cases, social tensions were caused by local power structures taking over the decision-making process and by the stigma against certain groups (e.g. women, disabled people, indigenous groups). Community targeting can be intrusive as families undergo public exposure of their most intimate conditions, thus turning into a selftargeting mechanism where only the more destitute and desperate are ready to accept the stigma of selection.

Governments opted for the devolution of the autonomous management of community grants in the cases of Indonesia (community empowerment program PNPM) and Tanzania (social action fund TASAF). Benefits are transferred by central government to community associations, and accompanied by advisory support and general supervision. In both cases, the driving principle is to enhance features of collective action and local governance to develop community projects for the benefit of the most vulnerable. Communities receive adequate training to establish and run local associations that are responsible for the selection of beneficiaries and deliberate on the investment of block grants. These experiences have shown that little central bureaucracy is needed if appropriate training is given and if the program design remains simple. It is also crucial to provide the right incentives for communities to invest cash transfers in favour of the most vulnerable thus fulfilling the original purposes of SP.

Brazil: Programa de Aquisição de Alimentos & Programa Nacional de Alimentação Escolar (Da Silva, 2010, Del Grossi, 2010, De França, 2010).

The PAA and the PNAE are local procurement programs established by the government of Brazil. Both programs were initiated in 2003 within the framework of the broad initiative Zero Hunger, aimed at increasing the level of food security among the Brazilian poor. Food purchases are operated through ROs (producers" organisations and cooperatives) and distributed to local public institutions (in the case of PAA) and schools (PNAE).

The objectives of Zero Hunger declared by the Brazilian government are multiple and include relevant aspects of SP: 1) stimulating production for consumption to improve the food security of disadvantaged communities; 2) generating surpluses for sale in institutional markets to guarantee an income floor to poor smallholders; 3) developing local markets for the sustainable inclusion of smallholders in the economy; 4) supplying stock of quality food to be distributed in food programs, respecting local traditions and habits (e.g. Cesta Basica, the Brazilian program for the unconditional food transfer to destitute citizens); and 5) strengthening family farming to foster social inclusion, solidarity economy and good agroecological practices.

Local procurement of food is governed both at national and at local level through a complex, but highly interactive institutional framework, where ROs are given a pivotal role. Programs are coordinated by the Ministry of Social Development (MDS) and count on the collaboration of other ministries through the establishment of a cross-ministerial Managing Group. This is linked to local municipalities that are responsible coordinating operations with ROs.

The government considers ROs true partners and intermediaries for the smooth delivery of the programs, particularly because of their ability to connect with rural families and their sensitivity to the issues faced by vulnerable families. ROs are responsible for preparing offers flagging local demand and potential for supplying it to the municipality. They agree on product qualities, prices and quantities and, accordingly, carry out the selection of farmers. Once the RO has started supplying its products, the municipality proceeds with the payments in the organisation's private account. ROs have raised awareness about the programs and its requirements among farming families facilitating broad inclusion in the programs.

The Brazilian government manages to fulfil different SP functions putting in place specific mechanisms within PAA and PNAE. The direct purchasing for simultaneous donation allows supplying schools and other local institutions (day-care centres, shelters, hospitals, subsidized restaurants and community kitchens) ensuring the food security of pupils and other targeted poor. Direct purchasing is also valuable for stabilising the incomes of smallholders providing an income floor and, according to the demand, a certain regularity of purchases. ROs also stock products in their localities through governmental infrastructures(stocks for family farming). Farmers are able to sell in more rewarding periods and secure better incomes, while stocks ensure food availability to purchasing institutions. The MDS also offers the opportunity for ROs to buy their product in advance (advanced procurement) allowing farmers to afford seeds and other inputs so that they can initiate production. This mechanism is particularly useful for the poorest farmers who have no access to other forms of microcredit and have started to see this modality as the best way to get funded without any banking red-tape.

PAA and PNAE were particularly successful thanks to the government's commitment to allocate resources¹ and build a dialogue with farmers" organisations and local municipalities. The acknowledgement of the role of farmers" groups by the Brazilian government has proven to empower smallholders and smooth the programs" delivery operations. The Managing Group has established a successful collaboration with local municipalities (although it is desirable to expand coverage to all local municipalities). The government was also successful in providing supporting programs (such as training and access to credit) and to delegate the democratic selection and organisation of these to localROs.

Local procurement programs in Brazil have shown good performance beyond stated objectives. According with their aims, PAA and PNAE have greatly increased the redistribution of economic resources to smallholders, promoting local agricultural production and better levels of economic stability for vulnerable rural families. Furthermore, a variety of mechanisms have facilitated farmers" access to credit and infrastructures. The pre-requisite to be a member of a PO has substantially encouraged participation in local organisations, favouring smallholders" access to the programs and to other services offered by POs (e.g. extension, collective purchases, access to local markets).

The Brazilian experience offers grounds for further improvements, however, both in terms of ensuring SP for beneficiaries and strengthening ROs" implementation capacity. (1) Oftentimes, ROs struggle to guarantee certain qualities and quantities of supplies, while purchasing institutions require a regularity of provisions. This might be due to a lack of technical ability, financial resources, information, or a combination of these. (2) At individual level, the poorest smallholders are often unable to meet program requirements, as they often lack resources, information, knowledge and time. (3) In many cases, farmers targeted for advanced procurement were reported to be previous beneficiaries of the agrarian reform (they have access to land) and credit defaulters (they had access to credit and refused to re-pay). Poor farmers that had never accessed credit did not get priority. This might be due to power capture by more influential farmers or lack of integration of the databases of different programs. (4) The presence of supporting infrastructure remains a crucial factor of success or failure. For instance, farmers" stocks improved their negotiation power and, thus, their incomes. On the other hand, the management of advanced procurement encounters difficulties due to the lack of local finance institutions to manage requests and finances. (5) Finally, local procurement programs are deemed by some policy analysts to create a culture of dependency in beneficiaries. More evidence on farmers" ability and willingness to access alternative markets would clarify the trade-offs between investing in local procurement and other strategies.

Manifold experiences on participatory mechanisms, such as the PNAE and the PAA in Brazil, point to a few key factors of success.

The existence of policies envisioning participatory mechanisms does not necessarily guarantee that they are operational. The effectiveness of these mechanisms depends on the governments" political commitment. Firstly, it is indispensable that participatory mechanisms at local level are formally institutionalised through national policies and enacted upon. Subsequently, government must be willing to allocate appropriate resources to SP programs and, in particular, to the effective coordination of participatory platforms with ROs. Success depends on the ability of central administrations to manage "horizontal" coordination across ministries and "vertical" coordination from national to local level, engaging local organisations and municipalities in decision-making, supporting their capacity development, and devolving to local governments the effective authority to convene relevant political and technical actors.

Broad representativeness is one of the cornerstones of participatory mechanisms. ROs" proactive and successful representativeness of the interests of the poor encourage: creating and strengthening strong bonds within ROs and other civil society organisations; making responsible decisions and re-distributing resources in favour of the poorest coherently with the SP aims of the programs; improving targeting mechanisms; linking groups with external actors and networks; and establishing a transparent dialogue among participating ROs and other actors avoiding corruption and nepotism. In Mexico, for instance, local ROs were excluded from the management of cash transfers due to their political use and co-optation by ROs leaders, and the lack of pro-active engagement of members in their agricultural cooperatives.

Local ownership is crucial for the internal functioning of ROs and within the broader institutional framework of SP programs. In Indonesia, village associations were successful in managing community grants, also thanks to the good inclusiveness and commitment. On the other hand, village associations were not able to influence the broader decision-making elites, who were not directly involved in the program, with their principles of transparency and good governance.

In participatory mechanisms, ROs perform a transformative function, where ""transformative" refers to the need to pursue policies that relate to power imbalances in society that encourage, create and sustain vulnerabilities". (Devereux, 2004, Sabates-Wheeler, 2004). ROs bring their members to the centre of decision-making in the design and implementation of SP, while improving their technical, organisational and negotiation skills. By taking part in decisionmaking processes, ROs are able to voice their members" needs and interests, make sure that the right benefits are delivered to the appropriate targets, and hold public institutions accountable for their action. This political inclusion constitutes the fundament of the broader economic and social inclusion of the poor.

Collective practices of SP²

Collective practices of SP exist but are often disregarded by national SP policies and programs, and little or no support is provided to encourage and sustain these initiatives. Certain principles characterise ROs: they are driven by collective principles of solidarity, reciprocity and democracy, and, most importantly, count on the participation and ownership of members that are bound by common needs and interests. Their utmost importance is in covering areas often unreached by public programs. We propose a short classification of ROs and illustrate how they offer social assistance, insurance and agricultural, financial and social services, improving the level of protection of their members.

Several informal groups of rural poor are not recognised by national polices and laws, but entrenched in the social structures of local communities. They normally take the form of community savings and loan schemes, such as the Rotating Savings and Credit Associations (ROSCAs), the Accumulated Savings and Credit Associations (ASCAs), Village Savings and Loan Associations (VSLAs) and self-help groups. Other groups help needing members to access land, inputs and other indispensable resources to attain subsistence (e.g. water and natural resources user-groups) and basic social services (e.g. burial associations, shared tenancy groups).

Microinsurance organisations have the purpose of preventing rural people from losing/reducing their incomes and assets, saving livelihoods and lives in cases of severe shocks and destitution. Their services include compensation schemes covering the loss of personal assets, crop failures, health, death, loan protection, and shelter provision. (ICMIF, 2012).

Various types of member-based cooperatives, associations and mutuals provide financial support to needing members. Microfinance organisations, such as the Savings and Credit Cooperatives (SACCOs), were established to grant opportunities for accessing liquidity to vulnerable citizens who are normally excluded by the formal banking system.

Producers" organisations (POs), such as farmers" organisations and agricultural cooperatives, traders" and processors" organisations establish collective arrangements to access local procurement markets, loans (e.g. through rotating funds), and agricultural inputs (e.g. seed fairs). POs run grain reserves and warehouse receipt systems, allowing farmers to mitigate the risk of falling prices. Finally, POs can provide training for unskilled members and worksharing arrangements for accessing liquidity during halts of production or unemployment.

Malaysia: the Rotating Savings and Credit Associations (Satkunasingam, 2006, Shanmugam, 2006)

ROSCAs are lump sum funds composed by fixed contributions from members (normally 12-20) who receive the "pot" in turn, at fixed intervals and as a whole. ROSCAs are widespread in developing countries and in the last 15-20 years have been steadily increasing in Malaysia. They can be named differently according to the region: Tontines in Central Africa, Ekub in Ethiopia and Eritrea, or Chit Funds in India. Their importance lies in providing financial services to people, who are unlikely to have access to publicly managed programmes of private insurance and credit. Indeed, in remote rural areas insurance, savings and credit services are often a prerogative of a few government officials, armed forces, and employees of large companies. At the same time, the private sector is scared away by clients that are considered potentially insolvent. In Malaysia, ROSCAs compensate for the lack of financial inclusion in formal institutions by providing financial services to low-income and vulnerable rural citizens, particularly women.

ROSCAs are normally started by an "organiser" who gathers a group of people interested in savings and credit. The group then agrees on the composition of participants, the amount to be contributed and the frequency of contribution, and further mechanisms (e.g. what to do in case of emergencies). The "organiser" will be responsible for collecting and distributing the loans.

Malaysian communities normally practice three types of ROSCAs. 1) In the random ROSCA, the beneficiaries are appointed by lottery to determine the sequence of beneficiaries. Accordingly, these obtain and re-pay the loan in turn. Organisers are normally the first to obtain the "pot", which puts them in a risk-free position in case of defaulters. 2) The consumer durable ROSCA has identical rule of appointment, but different aims. The organizers settle a deal with wholesalers for consumer goods (electrical items, seeds, tools for agriculture) with a discount in view of future purchases for the following beneficiaries of the group. 3) Finally, in the bidding or auction ROSCA, beneficiaries bid for the "pot". The person who is willing to give up the highest percentage gets the loan for that month. This "surplus" is shared among members except the bidder.

Through ROSCAs, members fulfil various functions to protect and promote their economic situation. ROSCAs are often used as a mechanism of forced savings by members who otherwise would lack the conditions or the commitment to put money aside. Malaysian women, for instance, declare that they join ROSCAs particularly to save money and cover the expenses for their children's health care and education; they use money to buy seeds and other inputs promoting the household's wellbeing; finally, their savings make the household safer to confront unexpected emergencies. Forced savings become then an empowering financial instrument for more vulnerable groups such as women and children (ROSCAs are often exclusively composed of women in Malaysia).

Secondly, members refer to ROSCAs to obtain funds to buy inputs locally and start small businesses, promoting the household's food security and subsistence, and the growth of the farming community.

Finally, ROSCAs can function as microinsurance. When a member faces an emergency, she might be granted access to the "pot" in sign of solidarity, for example by swapping position in the sequence or receiving the surplus funds from auction ROSCAS. Obviously, ROSCAs cannot work as insurance on income (which is a precondition to participate) or against covariant risk that affects the whole community (e.g. a drought which puts all farmers in the same situation).

As many other informal organisations, ROSCAs are based on cultural norms of reciprocity and mutual trust among members who share close social ties. These values smooth the governance of local informal institutions and ease the resolution of potential controversies among members. In small local groups, participants share similar socio-economic status, issues and risks, and are therefore more prone to cooperate and show solidarity to others.

Adverse selection is avoided thanks to mutual trust and mechanisms of peer-monitoring. ROSCAs will not cease to exist because participants are aware that they will be sanctioned and, possibly, ostracized by their community if they cheat on their fellow members. Malaysian ROSCAs have proven to be responsive to social and cultural needs. For example, needing participants are granted the loan when their costs occur at a specific time (school fees) or when unexpected (e.g. funerals). Mechanisms such as default percentages are implemented in auction ROSCAs to provide an alternative to interest rates, which are forbidden by the Muslim law.

Issues within ROSCAs are normally linked to the risk of defaults. It might happen that participants refuse to re-pay or, most likely, that they default before receiving the "pot" due to an unforeseen event. ROSCAs were made illegal by the Malaysian government to promote the use of formal financial institutions. Consequently, defaults can be particularly harmful as ROSCAs members are legally powerless against defaulters.

Formalising Malaysian ROSCAs means recognising the important financial and social functions that these informal institutions play for the rural poor. It would provide a legal instrument for participants to prosecute defaulters and would allow linking savings to public services. For instance, savings can be used to access health care and schooling through tailored programs. Formalised ROSCAs would also be able to guarantee a stronger status to money savers, by bridging with financial schemes offered by private banks. On the other hand, over-regulation might damage social relationships, since formal rules could be interpreted as a sign of mistrust, transforming cultural norms which people resort to so trustfully. In other countries such as Japan, over-regulation has led to the disappearance of ROSCAs. In India, ROSCAs have turned into big cooperatives losing their connectivity with the territory, and, most importantly, struggle to maintain the spirit of solidarity that underpins collective initiatives.

The case of Malaysian ROSCAs and a broader review of collective practices of SP bring to light important features of ROs, potential improvements and challenges. ROs show some strengths compared to central governments and the private sector. On one hand, ROs seem better suited to guarantee broad social reach at local level (particularly including the most vulnerable whom they represent and that are often unable to access formal SP services); high responsiveness to the needs of their members; and high sensitivity towards cultural dynamics and norms.

The contribution of these institutions is also valuable for sensitising the importance of riskcoping and for promoting the values of mutuality, responsibility and solidarity. Indeed, the strength of ROs lies in the solidity of cultural systems which people always refer to during and beyond times of despair, thus making collective arrangements resilient and affordable.

Successful ROs are normally initiated through bottom-up initiatives, are highly inclusive and transparent, and members are bound to the organisation by common interests, values and some sort of commitment (e.g. subscription fee). (Herbel, 2012, Crowley, 2012, Hadda, 2012, Lee, 2012).

Inevitably, local systems of protection show some limitations. They are localised, unstable and weak in the face of covariant shocks, i.e. shocks that affect a whole village or farming community. Local organisations are limited to the community boundaries with little or no reach to external networks. They often lack resources, in terms of finances, infrastructures, capabilities and information, to cover for all urgencies and needs. In some cases, they risk perpetuating traditional forms of hegemony (local political elites, village leaders, chiefs) and discriminatory practices against weaker groups. Finally, organisations based on contributory participation might exclude the poorest who cannot afford an entry-fee (which, at the same time, guarantee the full commitment of members).

Some successful ROs were able to grow (clustering) to regional and national unions or federations. Examples are given by the MUSCCO cooperatives in Malawi, the cooperative movement in Kenya, the Nepal Federation of Savings and Credit Cooperative Unions, and many others. These organisations can reach more beneficiaries and networks, increase their negotiation power with the public and private sector, ensure coherence of vision and mission among participating groups, and constitute communal platforms for discussions and exchanges. On the other hand, clustering might decrease the sense of proximity among members and complicates the coordination between local and central administrations, reducing some advantages of these economies of scale.

Finally, the Malaysian case shows how the formalisation of local groups might help in acknowledging their important function, in linking them to other public services, and providing legal instruments to enable small money savers to protect themselves. Governments should consider building upon existing nets of protection, using and/or complementing their structures instead of merely substituting them. However, this process should avoid damaging cultural norms that are the very foundations of collective practices.

Conclusion

The main objective of this study was to describe the roles that ROs play in protecting the rural poor. We have identified two main dimensions. (1) ROs are recognized as service providers by  governments, and implement some tasks in the delivery of SP programs through participatory mechanisms (from consultation to the autonomous management of community transfers). (2) ROs are service providers for their members and contribute to their social protection by arranging collective practices for sharing and managing risk. Both dimensions are particularly meaningful in the face of the financial and governance weaknesses of public institutions, which limit an extensive and adequate coverage of vulnerable rural citizens.

A review of examples and two case studies have provided some elements of reflection regarding the factors enabling the participation and functioning of ROs. Complex institutional systems require coordination both among different ministries and sectors, and between central and local governments.

In this institutional scenario, the rural poor must be fully represented and pro-active in order to make their needs and wants taken into account. Such is the transformative role played by ROs in participatory mechanisms. ROs show strengths and limitations compared to national SP system. If, on one hand, they are more likely to understand and have concern for their members" issues, on the other, they are often constrained in scope and resources. Clustering could improve ROs" networks and reach, but this might imply losing connectivity with the representative base.

Considering this trade-off is of utmost importance. The institutionalisation of participatory mechanisms and the formalisation of certain collective practices (where relevant) are important steps to be taken to acknowledge their importance and facilitate their flourishing.

The evidence on ROs" contribution to SP suggests three main conclusions. Governments should consider more systematically the role of ROs to the design of SP policies to the implementation of SP programs, building upon existing mechanisms and resources, and complementing rather than substituting them.

Governments and development practitioners should collect in-depth and contextual evidence demonstrating added values, opportunities and risks implied by devolving more responsibilities to ROs in SP. This study struggled to find specific reports or analyses of SP programs with a focus on ROs" contributions.

Finally, governments should consider committing to formalising participatory mechanisms and allocating appropriate resources for their implementation. Additional resources can strengthen ROs" technical, financial and organisational performance aimed at improving their SP services to benefit most vulnerable members.

References

Bassett L., Giannozzi S., Pop L., Ringold D. (2012) "Rules, Roles and Controls - Governance in Social Protection with an Application to Social Assistance" Washington: World Bank.

Da Silva J., G., Del Grossi M., E., De França C., G. (2010) "The Fome Zero (Zero Hunger) Program: The Brazilian experience" Brasília: Ministério do Desenvolvimento Agrário.

Devereux S. & Roelen K. (2014) "Inclusive Social Protection and the Post-2015 Agenda" Brazilia: International Comparison Programme - United Nations Development Program.

Devereux S., Sabates-Wheeler R. (2004) "Transformative social protection" Brighton (UK): Institute of Development Studies.

Herbel D., Crowley E., Haddad N. O., Lee M. (2012) "Good practices in building innovative rural institutions to increase food security" Rome