This paper should be cited as: Saxby, J. 2004. COEP - Comitê de Entidades no Combate à Fome e pela Vida. Mobilising against hunger and for life: An analysis of capacity and change in a Brazilian network. (ECDPM Discussion Paper 57C). Maastricht: ECDPM. http://www.ecdpm.org/dp57C

A case study prepared for the project ‘Capacity, Change and Performance’

Study of Capacity, Change and Performance:
Notes on the methodology

The lack of capacity in low-income countries is one of the main constraints to achieving the Millennium Development Goals. Even practitioners confess to having only a limited understanding of how capacity actually develops. In 2002, the chair of Govnet, the Network on Governance and Capacity Development of the OECD, asked the European Centre for Development Policy Management (ECDPM) in Maastricht, the Netherlands to undertake a study of how organisations and systems, mainly in developing countries, have succeeded in building their capacity and improving performance. The resulting study focuses on the endogenous process of capacity development – the process of change from the perspective of those undergoing the change. The study examines the factors that encourage it, how it differs from one context to another, and why efforts to develop capacity have been more successful in some contexts than in others.

The study consists of about 20 field cases carried out according to a methodological framework with seven components, as follows:

• Capacity: What are the components of capacity, i.e. the capabilities, in a given context?
• Endogenous change and adaptation: How do processes of change take place within an organisation or system?
• Performance: What has the organisation or system accomplished or is it now able to deliver? The focus here is on assessing the effectiveness of the process of capacity development rather than on impact, which will be apparent only in the long term.
• External context: How has the external context – the historical, cultural, political and institutional environment, and the constraints and opportunities they create – influenced the capacity and performance of the organisation or system?
• Stakeholders: What has been the influence of stakeholders such as beneficiaries, suppliers and supporters, and their different interests, expectations, modes of behaviour, resources, interrelationships and intensity of involvement?
• External interventions: How have outsiders influenced the process of change?
• Internal features and key resources: What are the patterns of internal features such as formal and informal roles, structures, resources, culture, strategies and values, and what influence have they had at both the organisational and multi-organisational levels?

The outputs of the study will include about 20 case study reports, an annotated review of the literature, a set of assessment tools, and various thematic papers to stimulate new thinking and practices about capacity development. The synthesis report summarising the results of the case studies will be published in 2005.

The results of the study, interim reports and an elaborated methodology can be consulted at www.capacity.org or www.ecdpm.org. For further information, please contact Ms Heather Baser (hb@ecdpm.org).


Contents

Introduction
Ten Years in the life of a development network
Performance: What has COEP accomplished?
Capabilities and capacity: What makes COEP work?
Looking ahead
Annex I: Notes on method
Annex II: Financing COEP
Annex III: List of signatories to COEP, 1993

Preface

This study rests on the hard work, ideas and support of a group of people spread across several continents. I wish to thank first those who set the table for this report. Heather Baser and her colleagues at the European Centre for Development Policy Management in Maastricht, The Netherlands, took an active interest in, and supported an analysis of COEP's experience in Brazil. The funders of ECDPM's multi-country study of capacity, notably the UK Department for International Development (DFID), the Canadian International Development Agency (CIDA), and the Ministry of Foreign Affairs of The Netherlands provided the core budget for this research. Peter Morgan, research coordinator of the ECDPM study, set out the broader conceptual framework and offered helpful comments on the drafts of this paper.

I am especially grateful to André Spitz and Gleyse Peiter of COEP, for inviting me to work with them on this project. COEP has funded a major part of this research, co-financed my expenses in Brazil and engaged Patricia Baldarelli as research officer for the project. Patricia's contribution has been invaluable. She has translated documents, conducted interviews and prepared summaries, and assembled secondary data, all with enthusiasm and good humour. Leila Vogel dos Santos has patiently handled the innumerable administrative details of my work with COEP over the last two years. Renato Cabral has been a generous interlocutor, introducing me to people in different parts of COEP and to Brazil more generally.

Finally, my thanks go to everyone in COEP who took part in this research, for their interest, thoughtfulness and good sense. Many people have put a lot of their lives into COEP over the last decade, and it is very dear to them. Their evident commitment to the network and its values, the people it serves, and to Brazil as a whole, has not compromised their readiness to step back and take a careful look at the organisation. I hope that the commentary here will be useful to them as they move COEP forward to the next phase of its remarkable life.

Acronyms

COEP Comitê de Entidades no Combate à Fome e pela Vida
(Committee of Entities in the Struggle against Hunger and for a Full Life)
CONSEA Council on Food Security
ECDPM European Centre for Development Policy Management
IPEA Institute for Research on Applied Economics
PRONINC National Programme of Popular Cooperative Incubators

Summary

This paper examines a Brazilian social solidarity network, COEP - o Comitê de Entidades no Combate à Fome e pela Vida (the Committee of Entities in the Struggle against Hunger and for a Full Life) - through the lens of organisational and social capacity and change. COEP is committed to building a just and inclusive society for all Brazilians, one without hunger and poverty. Its members include government agencies, parastatals, and organisations from the private sector and civil society. COEP is in fact a network of networks, active federally, in all of Brazil's 27 states, and now also at the municipal level. Its strategies include encouraging its members to support and participate in development projects to combat poverty, organising campaigns to mobilise public and institutional resources to end poverty, and promoting cooperation among its affiliates in their development work and campaigns.

The analysis of COEP is one of several case studies within a multi-country study, Capacity, Change and Performance, organised by the European Centre for Development Policy Management (ECDPM). The study seeks to understand better what 'capacity' is, what strategies are effective in developing capacity, and what 'performance' looks like from the perspective of capacity.

COEP attracted the interest of the organisers of the wider study for several reasons.

• It is uniquely Brazilian, part of that country's renewed democracy and of the broad social movement against hunger. Created by Brazilian initiative, COEP has been sustained by Brazilian resources.
• It is an intriguing hybrid - a voluntary nationwide network that embodies many aspects of a civil society organisation, but which operates in the border area between the state, the parastatal sector, private business and civil society.
• It is engaged with the paramount development issues of the day, mobilising citizens and organisations to work to end poverty and for social justice. It commands legitimacy as a development actor, legitimacy that helps to keep these issues on the public agenda.

The case study examines three broad areas of COEP's organisational life: its origins, growth and change over a decade; its performance; and the factors that explain its capacity.

Growth and performance
COEP's performance has been remarkable. Since 30 national enterprises declared their affiliation in August 1993, the network has grown and diversified, and now counts more than 800 member organisations (including 46 at the national level) active in networks in all 27 states, and recently created its first municipal networks. Originally limited to public entities, COEP's membership has become much more diverse, including government departments, private firms, labour unions, NGOs and other civil society organisations. Because membership in COEP is voluntary, sustained growth of this kind is itself a sign of success - ever-growing numbers of people are voting with their feet and are participating in the organisation.

The network has drawn upon its considerable internal resources to manage this growth. It has adapted its governance and management structures to changing circumstances while preserving consistency with its original principles. Its leadership has nurtured COEP's legitimacy as an actor in public life by maintaining a politically nonpartisan stance while retaining its original values and purpose.

COEP's programme provides other indicators of success. Through its national and state-level networks, COEP has conducted vigorous campaigns to mobilise institutions and the public to support the fight against poverty and misery, and to encourage 'active citizenship' (cidadania). COEP participates in government campaigns such as Fome Zero (Zero Hunger), and carries out its own activities. The latter include a Week of Mobilisation against hunger, held each year to mark the death in 1977 of Herbert de Souza, 'Betinho', the visionary founder of COEP. By engaging the public entities in its campaigns, COEP has helped to change the public discourse, and to keep poverty on the national agenda. Since establishing its first state-level networks in 1995, COEP has encouraged its affiliates to support community development initiatives in all parts of the country. COEP's secretariat, Oficina Social, maintains a database of almost 850 such projects and programmes, many of which are documented in its publications (cadernos) and a series of videos. In its campaigns and development projects, COEP promotes cooperation and partnerships among its affiliates, and has persuaded them to commit substantial resources to social responsibility and action.

The broader results of COEP's work are less clear, both in its affiliates' development initiatives with communities, and in its public and institutional mobilisation campaigns. COEP and those it works with could benefit from systematic assessment of its work to address the question: what difference do we make? The network clearly has the capability to organise the research and reflection this question requires. Such an assessment of the impact of its work could prove invaluable to COEP in its next phase, as it extends its networks to municipalities, and moves to create closer links with communities and their organisations.

Capabilities and capacity: what makes COEP work?
The report seeks to explain why COEP has been successful. The critical forces are to be found within COEP's substantial internal resources, both intangible and material. There are three principal factors at work, which reinforce each other. First, COEP has benefited from creative, even inspired, leadership at national and state levels, which enjoys substantial legitimacy within the network and beyond. That legitimacy, moreover, has been recognised and carefully maintained. Second, the network has shown a considerable and sustained capability for strategic thinking and change, continuously renewing itself while maintaining its character and principles. None of this would be possible without the evident commitment of its institutional members, and even more the passion, ideas and energy that the people in COEP bring to their work.

Complementing these intangibles are two significant material factors. COEP has been sustained by major financial and in-kind contributions from its affiliates. Also, the network relies on an effective web of electronic communications, and is able to draw upon the substantial technical resources and national reach of its member entities.

Our research suggests another level of analysis that might extend this assessment. The network shows numerous paradoxes in its style, dynamics and structure. For example:

• COEP is not formally registered, and its non-hierarchical design is intended to make space for diversity, creativity and energy, and to encourage participation. In its daily and strategic workings the network relies heavily on informal power - the chemistry among its people, their knowledge and contacts. Yet it is not a loose or simple organisation. Its guiding principles are clear and its statutes detailed. It is tightly organised and increasingly complex. Its governance structures are consistent across networks at national, state and municipal levels, and as COEP has grown, it has elaborated its administrative and operational capabilities.
• Politically, COEP is consciously non-partisan, yet must be politically astute because it operates in a charged institutional milieu. It is engaged with the big development issues - poverty and social justice - which are inescapably political.
• COEP is both a network of organisations and a network of people. Its membership is institutional, yet the quality of the organisation's contributions depends very much on the individuals involved. It is the people who take part in COEP who make it work, with their commitment, passion and competence.
• COEP's activists feel strongly about the organisation and its work, and there are strong personalities within the network, yet it has largely been free of divisive conflict, and has not been captured by personal, political or commercial agendas.
• Intangibles like leadership, creativity, confidence and legitimacy give COEP its energy, and attract new participants, yet it can only do what it does because institutional members make sizeable financial contributions, and donate an even larger pool of in-kind resources, in particular the paid time of their employees. Planned or fortuitous, this symbiosis is powerful. The whole undertaking, moreover, works on a national scale in such a large country partly because of Brazil's reliable communications infrastructure.

These paradoxes, and others, could be focal points for further reflection within COEP. They appear to be sources of creative tension within the network, and as such suggest another set of social forces at work, still to be articulated and understood.

Looking ahead
Three additional strategic issues are likely to challenge COEP's capacity in the future.

• Organisational identity. A long-standing question remains relevant: how is COEP to preserve its organisational autonomy, and its nonpartisan role in public life, while remaining engaged with the critical issues of social justice in Brazil?
• Programme development: engaging with community development initiatives. COEP is likely to place more emphasis on supporting local development initiatives by communities and their organisations. Its state networks and their new municipal counterparts will play the major part in this change. If COEP becomes more prominent in local development, it will face an important issue of governance: what accountability does it have to local communities, and how is this to be exercised?
• Continuous learning. Participants in the network continue to place priority on increasing their knowledge of community development and on access to technical support when they need it. COEP has an opportunity to continue to strengthen individual skills, and to systematise its own knowledge and experience. The question now is how the network will meet this challenge as it continues to grow and to increase its engagement with community development initiatives.

1 Introduction

This paper examines the history and current circumstances of a Brazilian social solidarity network, COEP - o Comitê de Entidades no Combate à Fome e pela Vida (the Committee of Entities in the Struggle against Hunger and for a Full Life) through the lens of organisational and social capacity and change. COEP is committed to building a just and inclusive society for all Brazilians, one without hunger and poverty. Its members include government agencies, parastatals, private sector firms and civil society organisations. COEP is in fact a network of networks, active nationally, in all of Brazil's 27 states, and now also at the municipal level. Its strategies include encouraging its members to support and participate in development projects; organising and participating in campaigns to mobilise public and institutional support; and promoting cooperation among its affiliates in their work and campaigns.

The analysis of COEP's experience is one of several case studies within a larger study, Capacity, Change and Performance, organised by the European Centre for Development Policy Management (ECDPM). The study seeks to understand better what 'capacity' is; what strategies are effective in developing it; and what 'performance' looks like from the perspective of capacity. ECDPM intends that the participants will reap practical benefits for their future development, in that they can use the case studies to reflect on their achievements and challenges.

COEP attracted the interest of the organisers of the wider study for several reasons. First, it is a Brazilian creation, born in the early years of that country's renewed democracy, and very much a part of the broad social movement against hunger. Created in 1993, COEP recently celebrated a decade of work as an autonomous organisation, sustained by Brazilian initiative, energy and resources. In the last two years COEP has begun to reach out to international organisations, hoping to exchange ideas on issues of common interest. The ECDPM project offers one way of doing so. Second, the network has grown to include more than 800 member organisations - it has become a large and complex organisation in a large and complex country. Third, it is an intriguing hybrid - a voluntary association that embodies aspects of civil society, but which operates in the border area between the state, the parastatal sector, private business and civil society. Finally, COEP is addressing the paramount development issues, mobilising citizens and organisations to work for social justice. Its legitimacy as a development actor has enabled it to address these issues, and to keep them on the public agenda.

The analytical framework of the ECDPM study and the approach used to analyse COEP's experience are explained in Annex I. The three core variables - capacity, change and performance - are in turn shaped by four intervening variables: the external (here, Brazilian) context; internal features and resources; relations with stakeholders; and external intervention by actors such as donors or multilateral agencies. The structure of this report reflects the primacy of the core variables. Section 2 examines COEP's origins, growth and development, focusing on the organisational capacity within the network. Section 3 examines COEP's performance - its effectiveness in realising its aims of helping to end poverty, and of engaging its member entities in that task. Section 4 draws on the preceding two to explain the capacities COEP has used in its organisational trajectory and in its programme of work. Finally, section 5 highlights several probable challenges to COEP's capacity in the future.

In keeping with its multinational origins, this case study is intended for at least three audiences. First, this report is addressed to COEP, to be used in the continuing discussion of its future shape and programme. It also has an international audience - the ECDPM team itself, the counterparts of COEP who took part in other case studies, researchers, and the staff of donor agencies. A third, broader audience, not directly connected to the study, includes people in Brazil and elsewhere interested in capacity, organisational development and social change.

2 Ten years in the life of a development network

COEP's trajectory over its first decade was remarkable by any standards. From improbable beginnings in 1993, it has grown to include more than 800 organisations, and is active throughout Brazil. At critical junctures along the way, it has shown considerable resourcefulness and creativity, drawing on its substantial internal resources to respond to the dynamics of a changing environment and its own membership. Four elements stand out:

• COEP's strong self-definition: the core values, principles and purposes of the network were clear from the beginning, and have remained constant.
• The network has had creative leadership with strong legitimacy and a capability for strategic thinking.
• The network has created effective structures for governance and management, well suited to its changing circumstances and profile.
• COEP has managed a dramatic growth in numbers and geographic scope to become a nationwide organisation.

Success apparently breeds success. COEP's performance has attracted more organisations as members, and has retained their energy and resources through its effective governance and management. These qualities have enabled COEP to adapt its structures and processes to suit changing circumstances, while preserving its original values.

2.1 Audacious beginnings: seizing the moment
From a distance, COEP's very existence may seem puzzling, even anomalous. It started as a social solidarity committee dedicated to ending poverty and misery, rooted in a group of powerful public enterprises in one of the world's richest and most unequal societies. Its creation, purpose and character warrant explanation not only because the circumstances of its birth are intrinsically interesting, but also because its guiding principles remain intact after a decade of growth and change, as is the core of its original membership.

COEP was created by a small group of activist intellectuals possessed of audacity and vision and led by sociologist Herbert de Souza, 'Betinho'. They met a sympathetic response from leaders of Brazil's public entities. The synergy of their encounter, at a particular historical moment, enabled them to create COEP, and to nurture it through its early years. According to one COEP activist, Betinho was 'a heroic figure' who helped shape a remarkable period in Brazil's history.

The 1980s and early 1990s saw the growth of popular movements for democracy and social justice, such as the Ação da Cidadania no Combate à Fome e pela Vida and the Movimento pela Ética na Política.¹ The end of the military dictatorship and the restoration of electoral politics in 1984/85 was prompted by, and in turn unleashed, extraordinary popular energy. A new generation of democratic leaders, Betinho among them, drew upon this energy to help Brazilians define 'active citizenship' (cidadania) as an essential part of the new era. The power of these social forces became evident in 1992, with the impeachment of the then President Fernando Collor on grounds of corruption. It was a special conjuncture, with the boundaries and workings of new and recreated institutions still being defined, and with a mobilised and articulate civil society helping to shape the political agenda.

Betinho and his friends saw in Brazil's public entities an opportunity and a challenge: to harness their huge organisational and material resources for the campaign to end hunger, and at the same time to give new content to the idea of a state-owned enterprise, one that would be genuinely public, serving the interests of all Brazilians. This sector had expanded significantly during the dictatorship. In the early 1990s it was still substantial, if threatened by privatisation, a key part of the policy orthodoxy of the day. Together with Luis Pinguelli Rosa of the Federal University of Rio de Janeiro, and André Spitz of Furnas, the electricity utility, Betinho invited the presidents of the major public entities to meet to discuss their integration into the Struggle against Hunger and Misery. On 28 May 1993, 33 of them established o Comitê das Empresas Públicas no Combate à Fome e pela Vida - the Committee of Public Enterprises in the Struggle against Hunger and for a Full Life. Two months later 30 enterprises, representing sectors such as banking, energy, telecommunications, health, agriculture and education, declared their membership (see Annex III). Acknowledging the 'absolute priority' of the campaign against hunger for the federal government, and the power of a nationwide citizens' movement, they promised their 'active and total' participation in the campaign. They have remained active in COEP's national and state networks ever since.

2.2 Responding to the environment of the early 1990s
We can see here a remarkable confluence of broad social forces and personal initiative. The result, COEP, was no accident, but it may well have been one of a kind. Ten years on, one of COEP's founders would say, 'Perhaps we could not have created COEP in another country, or even at another time in Brazil'. How then did it happen?

President Itamar Franco had created political space in response to the popular mobilisation against hunger by affirming the priority of the issue.² The key actors in COEP used that space creatively, inviting - in effect, challenging - the public entities to engage with the issues of social justice and development. Their challenge reflected strategic considerations. The entities commanded not just substantial financial and physical resources, but critically, national reach as well, since their subsidiary structures extended into all regions of the country. This reach, or capiliridade ('capillarity'), was later to prove invaluable, an essential condition for the creation of COEP's state-level networks. The initiative rested on more than intelligent use of political space, however. COEP's founders had earned widespread public legitimacy, especially Betinho, whose integrity and humanity evoked admiration and affection from people in all walks of life.

Not that Betinho and his friends were speaking to a hostile audience. Senior executives responded positively to the invitation to commit their organisational resources and moral support to the campaign. Indeed, the proposal to form a social action committee came from Marcello Siqueira, then president of Furnas. Furnas also absorbed COEP's secretariat function, committing the time of a senior manager, André Spitz, to that role.³ In addition, companies such as Banco do Brasil had been active in the Ação Cidadania for some time, with over 2000 branch committees participating in the movement. Of course, many public enterprises also stood to benefit - those in need of political support could gain credibility by affirming their social responsibility.

The creation of COEP can be seen as an example of organisational capability, adept strategic thinking and action, although the founders seem not to have had a formal and detailed plan. Rather, they worked with a coherent set of assumptions and principles that were and still are explicit. The choice of a network with its non-hierarchical structure reflected the intent to encourage organisational flexibility, and participation and creativity from its members. Betinho's colleagues acknowledge his influence, particularly his belief that an open organisational form would enable people to use their imagination to change themselves and society. It was also assumed that the committee could and should grow, with the entities' capiliridade a key asset in this process, even though it was unclear what COEP's future profile would be.

Beyond these organisational issues, clear principles of social purpose shaped COEP. According to André Spitz, Executive Secretary between 1993 and 2003, COEP sought to engage the wealthier and more powerful sectors of society in the campaign against hunger. It was founded as a committee to mobilise people and organisations, and to challenge the culture of indifference. Public enterprises could not be islands of excellence building walls to keep out the surrounding social reality. Poverty was a problem for everyone, and ending it would require a concerted national effort. With their invitation to the entities, COEP's founders were challenging them to change their culture and methods as well - to break with narrow sectoral and competitive logic, to cooperate with each other and with other organisations, and to become truly public bodies.

COEP invited the entities to use their organisational and technical competence to support a social movement. The network simply asked them to do what they knew, but to do it differently, for different people, and to work with and for poor and marginalised communities. They were challenged to think of different uses for existing resources - to allow fish farms in the reservoirs of hydro-electric dams, for example, or community gardens on their landholdings. Banks were challenged to open branches in poor communities, and to establish microcredit and financial management schemes for street vendors. Rural and agricultural development organisations were encouraged to work with small farmers and cooperatives. To their employees, COEP's message was similar: your professional competence is needed, and you can contribute to a better society by adapting your expertise to new circumstances.

Governance and responsible citizenship have always been an integral part of COEP's agenda, yet the organisation has consciously avoided taking partisan political positions. This stance has been rigorously maintained at all levels to preserve the network's autonomy. Respondents in this research firmly believe that it has been critical to COEP's legitimacy and acceptance by different political actors. This principle makes for a diverse and inclusive 'church', which has made the task of building and maintaining consensus all the more demanding, but it has also protected COEP from shifting political winds, and from capture by narrow agendas.⁴

The decision to engage the public entities in COEP was inspired, audacious and, until the invitation was accepted, perhaps improbable as well. Without question, it has had a lasting impact in that it has changed the institutional terrain on which Brazilians respond to the problem of hunger and poverty. Politically, the public enterprises are now 'within the tent', an established part of the public discourse on the issue. Their early engagement also set a precedent, opening the door for other organisations to join COEP. Although the public entities are no longer the sole members of COEP, they remain important. Their annual financial subventions enable the network to function, and their capiliridade makes their subsidiaries the core of COEP's state networks.

2.3 A decade of growth, decentralisation and diversification

First principles - identity, purpose and structure
Several key components of COEP show a clear pattern of continuity in substance and change in form. The early leaders established its governance and management structures, which have been maintained, even though the architecture has become more elaborate. The founding members created an interim structure in May-August 1993, and a year later the network adopted its statutes, signed by 34 entities, which established the basic elements of COEP's organisational architecture:⁵

• COEP described itself as a collegial body, a (voluntary) association and a non-profit entity. Its purpose was to link together and to encourage actions by its affiliated entities, and those of other public and private bodies, in support of the Campaign against Hunger and Misery. It would advance the fundamental objectives of the Republic 'to build a free, just and solidary society', and 'to eradicate poverty and reduce social and regional inequalities'.⁶
• Its proposed actions were cast in terms of projects and programmes to meet these objectives, but also mentioned CONSEA (Council on Food Security), the Ação da Cidadania, and the Campaign against Hunger and Misery.
• The words describing COEP's role were carefully chosen. It would act as a catalyst to 'promote and encourage' the actions of its affiliates. It would 'disseminate information' about its own actions, and 'encourage and publicise' those of its affiliates through 'technical and promotional documents'. Members retained their autonomy, choosing how to participate in COEP. The network could encourage but not command.
• The network had three parts. A Deliberative Council, made up of the President of COEP and chief executives of member entities, decided on strategy. An Executive Committee, with representatives of each member, was responsible for management. The first Executive Secretary was André Spitz of Furnas.⁷ The statutes also provided for technical committees to encourage affiliates' actions.
• Member entities' commitment was voluntary, but nonetheless formalised. All signed a protocol acknowledging the struggle against hunger as a point where the priorities of government and the power of society converge, and reaffirmed their 'active and complete participation' in the campaign.⁸ They also agreed to allow their staff to perform COEP duties on company time - a major in-kind subsidy to the network.

Decentralisation: the growth of state-level COEPs (estaduais)
With the growing membership and spread of COEP, the leadership decided to decentralise the network by encouraging the formation of state-level networks (estaduais). In 1995 the Deliberative Council amended COEP's statutes to provide for these networks, which would be governed by the same principles as the national body, with the same governance and management structures. Their committees were not subsidiary to the national network, but joined COEP as autonomous bodies. Indeed, as more appeared they transformed the membership of COEP's national Deliberative Council and its Executive Committee. In 2001, COEP amended these bodies to include representatives of state committees. This was a deft adjustment: the network accommodated a radical expansion of its membership without adding another layer of hierarchy to its governance and operating structures.

Strategic considerations lay behind the decision to decentralise. The leadership saw state-level networks as a way of bringing COEP closer to the regional realities, and particularly to the different faces and dynamics of poverty,⁹ allowing it to support community development initiatives across the country. Engagement of this kind requires local knowledge, presence and credibility - it cannot be done from afar. Decentralisation was also seen as a way of broadening and diversifying the base of COEP. Expanding the number of people and organisations with a stake in the network would strengthen its autonomy and integrity.

These judgments proved to be accurate.Decentralisation was slow at first, but within eight years there were networks in all 27 states (see Figure 1). This dramatic growth can be seen as a response among people and organisations at the state level to the opening offered by the leadership at the national level. Decentralisation has also brought about a sea change in the organisation, in that its capacity to support community development initiatives now lies principally with the state-level COEPs. Decentralisation is continuing as well - in 2003, COEP accepted its first municipal networks, and more will follow in the years ahead. COEP's statutes affirm that these networks will have the same form and function as their counterparts at national and state levels.

Figure 1. Creation of the COEP state-level networks, 1995-2003.


Administration and management
COEP also reworked its administration and management in order to sustain a larger and more complex network. By 1997, as the growth of the network threatened to overwhelm the management capacity of the organisation, the Executive Secretary recommended establishing an operational arm. As the
secretariat for COEP, this would have an operating budget and staff to support the growing volume of meetings and electronic interactions, and to monitor an expanding portfolio of local development projects implemented by COEP members, especially in the state networks.

The proposal sparked a vigorous debate within COEP. For the first time the network would have to manage money, specifically to meet the costs of its operation. One faction believed that introducing money into a solidarity committee would destroy it; another, including the Executive Secretary, felt that without a dedicated secretariat the organisation would collapse. COEP opted for the latter, and in 1998 set up Oficina Social (Social Workshop) within the Federal University of Rio de Janeiro. Oficina is financed through annual subventions provided by 18 national entities - a clear statement of their commitment to the work of COEP as a whole (see Annex II).

Setting up a secretariat may not seem like a spectacular achievement, but one respondent argues that the creation of Oficina was the critical event in COEP's development. Operationally, its budget has allowed COEP members to meet regularly, and to build the interpersonal trust essential to COEP's workings. Politically, through their financial and other material contributions, the 18 national members have made a vital statement of their commitment. They have also taken a huge administrative burden off the shoulders of COEP's leadership, allowing them to devote more time to maintaining communications, resolving differences among members, and supporting the state networks.

In some ways the most visible part of COEP, Oficina provides logistical support, organises and finances meetings, and arranges the teleconferences that have been a regular part of the network's internal communications since 1999. It is actively involved in development projects supported by affiliates, assists in monitoring, and administers the pool of project seed money for state-level COEPs. Oficina is also responsible for sharing knowledge and experiences. It maintains a database of projects - almost 850 of these are accessible on COEP's website.¹⁰ It publishes a series of cadernos ('notebooks') containing information on COEP's projects and programmes, and commentaries on development issues.¹¹ It has also produced a series of more than 30 videos, 'Imagens da Oficina Social', providing a lively and accessible audio-visual record of COEP's work. These database, print and audio-visual resources have also contributed to raising COEP's public profile.¹²

Creative paradoxes: people and institutions, change and continuity
These changes in COEP's organisational profile show an intriguing interplay between the institutional and the personal dimensions of the network. It was created by individuals using their acumen and good public standing to seize a political moment. Its growth at the state level reflected both the good judgment of the leadership and the commitment of participants. But COEP could only survive and prosper when these personal qualities were joined with material contributions of its member entities.

COEP's current governance structures and statutes¹³ show the recurring balance between adaptation and preservation of basic principles:

• In March 2000 COEP changed its name from the Committee of Public Entities to the Committee of Entities, reflecting the growing number of members from the private sector and civil society. This diversification resulted from the decision to decentralise, and is expected to continue as COEP builds municipal networks.¹⁴
• Several key operating principles, previously well understood but not codified, are now spelled out in the statutes. Article 6 confirms that participation in COEP is voluntary. Article 8 forbids discrimination on the basis of religious creed, colour, race, gender or political and philosophical belief.
• It also prohibits entities or their representatives from seeking personal gain from COEP activities, or promoting or discriminating against political parties.
• A new governance and management body, the Administrative Council, is charged with monitoring and ensuring compliance with COEP's statutes, and may recommend the exclusion of a member that fails to observe them.
• Complementing the statutes is a 'Regimento Interno', a code of conduct that sets out principles and procedures for the conduct of meetings and the overall functioning of the network.

In 2003 COEP's leadership added another component to its operational infrastructure. 'Rede Mobiliza' ('Mobilisation Network') is an independent registered NGO whose purpose is to support the work of COEP, and the campaign against poverty in general. Whereas Oficina is financed by COEP's national entities, Rede will enable people and organisations to participate in or support COEP and its work. Its members will be individuals, it may engage volunteers who may not be linked to one of COEP's affiliates, it may receive donations, and it may negotiate and administer external grants and contracts. Its link to COEP is effected by an overlap of elected officials (see Figure 2).

With hindsight, it is evident that consistency of form and function has been a key principle for COEP since the first state networks were created. As the organisation grew, its principles have been codified and an oversight body established. As the membership has increased, an apparently paradoxical link between participation and a tight structure has become evident: to realise its design potential for agility and participation, the network has had to define more sharply the constitutional principles encouraging that. It has had to set up an oversight body to ensure these principles are observed, as well as secretariat/support bodies with the necessary budgets.

One other force has shaped the network. Asked to identify key events in COEP's history, several respondents mentioned the spread of the Internet in the mid-1990s. A fortuitous symbiosis developed between two critical enabling factors, one constitutional and one technological. The decision to decentralise COEP's structure by creating state networks, occurred as the members' capiliridade acquired electronic form. The use of the Internet required, of course, investments in hardware, software and training, but once these were in place, COEP had a rapid, inexpensive and pervasive communications infrastructure. On a smaller scale, but important nevertheless, members such as SEBRAE (the agency that supports small enterprises) have provided television conferencing facilities to the network. As the state-level networks mushroomed in the late 1990s, television conferences provided an effective complement to face-to-face meetings and email.



Growth of this nature in a voluntary organisation represents solid institutional performance. No one is obliged or paid to take part in COEP's forums and its work, yet people continue to 'vote with their feet' and participate in large numbers. A consideration of 'performance' in a development organisation should also take account of what it does, however. Let us therefore now look at how effective COEP's work has been.

3 Performance: What has COEP accomplished?

The survival of COEP and the growth in its membership and national reach in its first decade are no small achievements. Successive governments have recognised its mobilising capacity and have drawn the network into their councils - CONSEA under President Itamar, Comunidade Solidária under President Cardoso, and CONSEA once more under President Lula. Acknowledging this, we now extend the discussion of performance to the ensemble of its activities and ask: how effective has COEP been in realising its mission?

There is compelling evidence - both informed opinion and quantitative data - that COEP has created and pursued its strategies effectively, and has played its chosen role adeptly. COEP can point to real achievements:

• It has used its influence well, persuading hundreds of entities to commit themselves to the campaign. In doing so it has maintained its legitimacy as a development actor and has succeeded in creating a public space where its members and participants can speak to the major issues of development and social justice.
• In a decade of public and institutional mobilisation it has both encouraged and drawn upon the active participation of people from all levels of its member entities.
• It has supported hundreds of development projects, which have helped to create and consolidate the links between COEP's members, communities and their organisations.
• Finally, COEP can reasonably claim to have applied its influence effectively to its own sphere, helping to change the culture of its members to become more socially active and responsible organisations.

There is less clear evidence, however, about the social effectiveness of the development initiatives COEP has supported. Nor does it have a base of information and analysis from which to assess the longer-term impact of its work, whether at the level of communities or of national discourse. This is not to suggest that any such assessment would be negative - on the contrary, the signals that do exist are positive. Rather, the analysis remains to be done, and this is both challenge and opportunity for COEP. This section recaps assumptions relevant to COEP's effectiveness, and reviews the evidence available, as well as the gaps. We conclude that COEP could use its considerable capabilities to gain a better understanding of its real achievements. It has an opportunity to analyse the impact of its work more systematically, so as to guide and support the next phase of its development.

3.1 COEP's role as 'influencer' and catalyst
In examining COEP's programme to combat poverty, it is important to be clear about its role in this undertaking, and about the type of power it commands. These factors shape our understanding of its effectiveness. COEP offered a definition of its role in its plan of action to support the government's Fome Zero (Zero Hunger) programme, presented to President Lula in April 2003. 'To accomplish its mission, COEP mobilises organisations and people, promotes partnerships, encourages the practice of innovative projects, builds capacity for social action, and disseminates knowledge and information on initiatives to combat hunger and misery'.¹⁵ By its own description, COEP is a catalyst for innovation and mobilisation, encouraging, cajoling, documenting and celebrating work well done. It provides guidelines for its affiliates on what they could do to advance Fome Zero - but the affiliates retain the prerogative of choice. COEP aims to enhance their initiatives, encouraging joint action whenever possible and using the knowledge and experience of the network to support the people doing the work. By presenting its affiliates' proposals, COEP is asserting that this is a collective effort. By joining COEP, the entities have affirmed that they are in this together, and for the long haul.

The power COEP uses to discharge this role is strongly informal, best understood as influence. A sympathetic observer neatly described the network as 'an influencer'. Neither its leadership nor its membership can prescribe what individual affiliates should do. Because the affiliates retain the prerogative of choice, they are also accountable for their own actions, and they have the operational capacity for implementing the projects and campaigns. COEP has clear principles, procedures and structures for its membership and programme (including a formal affiliation process) but it has no jurisdiction over its affiliates and their resources. Whether and how they act depends on the use of informal power within COEP, and on their readiness to accept that power as legitimate. In practice, this power looks like influence - individual persuasion, collective pressure, serendipitous negotiation of diverse positions, clever use of opportunities, personal trust and chemistry among colleagues and friends, communication of ideas and strategies, and so on. This is not unusual - power is often exercised in this way in civil society, for example. It is present in COEP by design - this is how the network is intended to operate. Power of this kind means, however, that COEP often functions best as an indirect or behind-the-scenes presence in its affiliates' development work. It may play an important role, but it may not always be visable or understood, even by members of the communities involved.

Why then does it matter that we get a reading of the effectiveness of COEP's programme? The leadership enjoys the confidence of its growing membership, and its mobilising capacity continues to attract the attention of government. Surely these are good proxy measures of an effective organisation? The answer is that they are good indicators, but an account of the performance of a development organisation must surely address its success in realising its mission. COEP, to its credit, has resolutely engaged with the big development issues of the day - poverty, inequality and social transformation. It follows that those with an interest in COEP - its members, supporters and the communities affected by its actions - have the right, even the obligation, to find out what difference it makes to those issues.¹⁶

Two respondents, founders of the network, underlined the importance of this issue. COEP's leadership also needs to be able to speak convincingly about the network's relevance and the effectiveness of its work, particularly to communities and their organisations. COEP is on the threshold of another wave of decentralisation, and is a prominent supporter of the Fome Zero initiative. Questions about the effectiveness of its programme thus bear on its strategic directions: how can an assessment of the effects of COEP's work help to inform its choices about its short- and medium-term future?

The leadership expects the network to continue to decentralise and to place more emphasis on community-level work in the future. This more local focus would rely primarily on COEP's existing state networks, its nascent municipal networks, and the infrastructural support of Rede Mobiliza. Such a change would make the network still more locally grounded. Closer links to communities and their organisations would enable state and municipal COEPs to participate in local development initiatives, and to mount support campaigns both to sensitise the public and mobilise financial resources.

A shift of this kind points to a higher profile for COEP itself as an actor in local development, with a more direct and public role. This role is still to be defined but could include COEP participants becoming more active development agents, especially as interlocutors between communities and COEP members. This function is particularly important, since the quality of these relationships is a key factor in effective projects. One respondent noted that communities that are organised and clear about their agendas are usually better able to negotiate. Sympathetic and knowledgeable people within the entities, such as COEP participants, can play an important complementary role, not in organising communities or speaking on their behalf, but in hearing and understanding the issues they face and what they hope to achieve, and in encouraging creative responses by the entities.

It is thus important that COEP and those it works with know where it has succeeded in its development agenda and where it has not. All involved stand to gain from an assessment of who has benefited and who has not, and why; what role(s) COEP has played, why, and how well. Accordingly, we asked our respondents what, in their view, COEP has achieved.

3.2 What has COEP achieved?
This section summarises the respondents' judgments and other relevant information. The data provide clear and positive messages; they also set the stage for more systematic examination of the evidence and the issues.

(1) COEP has persuaded 800-plus entities at the national, state and municipal levels to commit themselves to the campaign against poverty.
The network has used its influence well. It has mobilised substantial resources - expertise, money, services, matériel, and people's energy and time - to support public campaigns and practical development initiatives in all parts of the country. The plan of action in support of Fome Zero is a good indicator of the scale of action that COEP is now capable of mounting. Its aggregate weight is formidable. Two examples show the scale of organisational commitments. One COEP member, an electrical utility, estimates its 2003 expenditures on COEP projects at R$10.5 million (€2.75 million). The projects are carried out in concert with municipalities, so that the utility's resources have a multiplier effect in several sectors: health, education, urban horticulture and integrated development schemes. Another member, a financial institution, provides training in financial management for small enterprises. It aimed to reach 50,000 people in 2003, a number that will rise to 600,000 by the end of 2006.

COEP is not a funding agency, but has leveraged the resources of its members by encouraging them to work together and with other development organisations. Although it has not achieved its success overnight, it can justifiably claim real progress. Its members have indeed committed substantial resources to the mandate of the network.

(2) COEP has encouraged, supported and participated in hundreds of development initiatives undertaken by its affiliates. By June 2004, COEP had supported no less than 841 projects,¹⁷ including emergency relief operations and 'structural projects' (projetos estruturais) to address the underlying causes of development problems. While acknowledging that COEP has not usually played an operational role in these projects, respondents argued that the network has nonetheless significantly improved them. In some cases COEP took the initiative in bringing operational bodies together and challenging them to take on an innovative project (see box). In others, it has encouraged people to bring forward good ideas, and has publicised innovative practices. Respondents cited COEP's introduction of reference projects as a key intervention, in which examples of innovation were made available to both members and other development organisations. Respondents also appreciate Oficina's efforts to disseminate information through the online project database and the cadernos.



COEP has also promoted professional development for its participants. When it became apparent in the late 1990s that members' project work needed improvement, Oficina engaged university affiliates to offer courses in project design. These actions point to a readiness to invest in organisational learning, and to enable COEP's participants to strengthen their development knowledge.

(3) COEP has organised many campaigns to mobilise opinion within its affiliates, and among the public. COEP describes itself as a network that mobilises people and organisations. It has both relied upon and encouraged the spirit and practice of active citizenship of people from all levels of its member entities.

The creation of the state-level networks has dramatically broadened the spread of COEP's messages. To show the effects of COEP's institutional mobilisation, respondents noted that in 2003 it took less than a month for national institutions and state networks to prepare COEP's plan of action to support Fome Zero. A decade of effort had obviously succeeded, and both were ready to respond. This change is far enough advanced, some respondents argued, that COEP's challenge vis-à-vis the entities is now to consolidate the effects of this mobilisation by making the commitment to social action and cooperation one of their core values.

Respondents also pointed to COEP's success in mobilising the public, although the indicators are less specific. COEP regularly makes its collective public presence known, both through its contribution to campaigns such as Natal pela Vida (Christmas for Life), and its own efforts, such as the annual Week of Mobilisation to commemorate Betinho and his work. Respondents are convinced that efforts such as this have helped to keep poverty and hunger on the national agenda. Such campaigns are not only directed at the public. Working with civil society organisations, COEP has campaigned for a national law designating 9 August (the date of Betinho's death) as a National Day of Mobilisation for Life. This would require all branches of government annually to review and publicise their efforts to combat hunger and promote active citizenship. An electronic petition on the COEP website continues to accumulate signatures, and the network is seeking the endorsement of a majority of deputies in the National Congress.

One final aspect of COEP's work in mobilising public opinion holds considerable potential for the future. As suggested earlier, by engaging hundreds of organisations in the campaign against hunger and poverty, COEP has changed the texture of public discourse on these issues. By establishing and maintaining its credentials as a legitimate actor in public life, the network has created a public space for debate about relevant policy and practice. Its commitment and non-partisanship are not in doubt. It can legitimately convene different actors for public discussion of the issues, both from within its own membership and beyond. This record may enable COEP to participate in future policy discourse, without becoming dedicated to policy advocacy.

(4) Through these initiatives, COEP has changed the culture of its member entities. COEP's members are now substantially more responsive and proactive on issues of social justice, and their corporate social responsibility, than they were a decade ago. COEP has used its influencing capability not only in public life, but within its own internal sphere as well. Gradually, respondents argued, the entities have become integral to public debate and action on poverty, and that experience has changed them. Betinho's original judgment has been vindicated, and his vision at least partially realised. Respondents cited a number of examples. Several entities now have policies on social responsibility and staff units responsible for corporate action. One agency now has 1200 volunteers available for work on social responsibility; less than ten years ago there was a committee of ten or twelve people. By financing Oficina's budget, 18 entities are investing in COEP's capacity to attract and support people, and have themselves benefited, insofar as their employees have acquired new skills, knowledge and contacts. Collectively, the members have allowed or encouraged specific COEP initiatives, for example by agreeing to launch innovative projects that individual agencies had been reluctant to support. In these instances the entities are accepting the exercise of COEP's informal power. Finally, both members and external observers point to the widespread cooperation and partnerships among the entities as COEP's major achievements. Gradually, the practice of working together has become a habit, and has eroded both internal bureaucratic constraints and the boundaries between organisations.¹⁸

3.3 Assessing impact
Beyond these considerable achievements, it is difficult to assess the broader, longer-term effects of COEP's work on the big issues. Indeed, two respondents identified this as a serious gap in the network's knowledge, and a continuing challenge. In effect, the challenge is to assess systematically the validity of one of the core strategic assumptions of COEP's founders - that the substantial resources of Brazil's entities could be harnessed in the campaign to end poverty, and that they could make a substantial difference. It is clear that significant resources have been mobilised. The complementary question remains: to what effect?

There are problems to be overcome if COEP wishes to answer this question. First, there exists no systematic analysis of the overall results of the projects and campaigns with which COEP has been associated. There is substantial information on individual projects. Many of the 800 or so project summaries in the database, for example, refer to proposals and reports, and some have been the subject of detailed examination.¹⁹ Yet any synthesis, such as a scan of project reports, is lacking. In these circumstances, even establishing the parameters of an impact assessment would be a major task. The breadth and diversity of the activities would make it conceptually difficult as well. Second, COEP's role in these activities is often indirect and intermediate. It has typically been a contributor, rather than the prime mover or operational agency, and sometimes has not even been visible to people on the ground.²⁰ Thus any assessment of COEP's role is intrinsically difficult, because its responsibility is typically diffuse or indirect, its resources and 'value-added' often intangible, and its power and authority usually informal.

Nevertheless, there are at least two ways in which COEP could use its capabilities to begin a systematic assessment of its impact. First, it could conduct a desk study of issues related to impact, using a sample of the projects and programmes in the database. This would allow a scan of opportunities, problems and available data, and could be useful in setting the parameters for a more detailed analysis. It would give COEP an overall sense, for example, of how well-conceived and effective are the projects undertaken or supported by its affiliates. Such an analysis would complement Oficina's work in documenting, monitoring and assembling project data. The capability to do this work surely exists within COEP's member educational and research institutions.²¹

A second option would be to organise a more detailed, community-based participatory research programme to examine the impacts of several initiatives across the country, including an analysis of the roles of COEP and other actors. This analysis could be done to inform future development work by the communities themselves, and to guide COEP in its efforts to build closer working links with local communities. Such an undertaking would have to be negotiated with communities and their organisations, perhaps including pilot studies to test the approach, spread over several months. Indeed, if it works, it might become a continuing part of COEP's practice of reflection and systematisation of knowledge. Although the capability to do this probably exists within COEP's networks, it may also be helpful to engage Brazilians from outside COEP who have experience in participatory research.

Both of these initiatives would use applied research to strengthen COEP's programme. They would also augment COEP's ongoing efforts to disseminate examples of effective development practice among its participants, in order to improve their knowledge and skills.

4 Capabilities and capacity: What makes COEP work?

This section assesses COEP's capabilities, the key factors in its capacity and performance. The hypotheses guiding this research identified several success factors:

1. a creative and legitimate leadership that has used available political space well;
2. a capability for strategic thinking and action, nurtured by its leadership;
3. solid commitment from institutional members and individual participants;
4. a clear role, mandate and values for the network;
5. a substantial pool of skilled, educated people with professional competencies, and technical, material and organisational resources, both within COEP's member entities and in society at large.

Our research showed these hypotheses to be essentially correct. They identify resources that are largely internal to COEP, and which are also largely intangible. According to most respondents, it is the first three that matter, yet closer examination suggests a more complex picture of the influence of each of the five factors. The analysis of capacity also throws other issues to the surface. These success factors can indeed help to explain capacity in COEP. There is, however, a quality to the network that merits further scrutiny. In addition, the interplay between the capacity of the network and the circumstances in which it emerged, generates other questions. To what extent is capacity in COEP a response to social demand, or a product of organisational supply? Is COEP's remarkable experience relevant to social justice activists elsewhere, or is it a 'special case'?

4.1 Success factors
We begin by examining the factors that may explain COEP's capacity, and their interaction. First and foremost are the intangibles.

1. A creative and legitimate leadership
Our account of the founding and growth of COEP has highlighted the public legitimacy of the small group that created it, and the influence of Betinho's personal charisma and imagination. Betinho may have been the motive force - there is no substitute for genius, and he was especially gifted - but he did not work alone. The creativity and audacity of the founders bore fruit because complementary institutional processes were also at work. Inspiration works best in the company of perspiration. Betinho had operational capacity in the person of André Spitz, a senior manager with Furnas. Furnas' management supported his work with COEP, and continues to do so. Furnas' commitment to the initiative provided an important example to other entities, and the proposal from Betinho's group found a receptive audience.

Since 1993 COEP has benefited from its non-hierarchical structure and its open, participatory style, retaining imaginative leadership at the national level, and nurturing creative leaders among its state networks. After Betinho's death in August 1997 COEP struggled to compensate for the loss of his personal and symbolic presence, but appears to have preserved the legitimacy of its leadership, and of the network as a whole. In this respect, all respondents argued that COEP's non-partisan political stance has been of fundamental importance. It gives the network a distance from immediate political agendas, national and local, and has sustained COEP's autonomy and integrity. At the same time, the network is in no sense neutral on Brazil's big development issues. Poverty and social justice are profoundly political, because they touch on deep-seated patterns of power, interests and the use of organisational and social resources. Hence, the network operates in a politically charged environment, and it does so with considerable political savvy. One respondent neatly summarised these inherent tensions: 'COEP is not political but it acts politically'.

Other characteristics of COEP's leadership should be noted as well. Respondents noted with pride that COEP has had no major problems of misappropriation of funds or abuse of power. The network has always held ethical behaviour and transparency as core values, which have been codified in its statues and reinforced by an oversight body. The network still relies on informal power to get things done and to move its strategic discussions forward. This takes the form of personal friendships and contacts, influence, knowledge and status. At the same time, the leadership has to account for its decisions - informal power in this context is not unaccountable power. The system seems to work: all respondents spoke well of COEP's leadership, both the individuals involved (especially but not only at the national level) and the style of leadership.

2. A capability for strategic thinking and action
COEP seems well able to analyse and adapt to changing circumstances. Hence, we treat strategic thinking as a distinct capability, rather than as, say, an aspect of creative leadership. The leadership has in turn encouraged this quality. This capability is important for any organisation, but several respondents considered it to be the critical one for COEP, because the network must constantly navigate tricky political waters. COEP is not and has never been static - it has constantly changed its organisational form in order to maintain its essential qualities.²² It is notable that the profound changes in COEP over the years have come about as things have been added to the network, to respond to new challenges and circumstances. Its basic principles and original raison d'être have remained intact, and established structures have not been dismantled. One respondent summed up the major changes as 'diversidade, decentralizar, projetos e municipalizar'.²³

The measure of this capability is not so much the magnitude of the changes, as the fact that the network continues to reflect on its condition and its position. It evidently has the will to act when it believes changes are necessary, and a capability to identify and respond to challenges before they become crises. Examples include the establishment of Oficina Social in 1998, even though it was controversial at the time, to ensure that COEP's growth did not unbalance the network, and of Rede Mobiliza in 2003 to enable individuals to assist COEP. Finally, the anticipated shift in COEP's programming, to forge closer practical links with communities, is deliberate. According to one respondent, the decision 'recognises COEP's dualistic nature. We are strengthening our non-governmental side within our public setting'. One participant sees COEP becoming more systematic in its approach to change. In the early years, there was a strongly reactive quality to its organisational evolution; more recently, the network has taken a more considered approach, although there is no blueprint in evidence.

Creative leadership and strategic thinking are enhanced by, and augment an evident readiness to learn. The national entities have financed the documentation of COEP's development projects and training programmes, such as the university course on project methodology. Respondents repeatedly spoke of their need for a better grounding in the theory and practice of social development, as well as for technical support. As the network grows, this issue will surely remain a priority for participants. COEP has invested in a range of educational activities, but - perhaps because of its scope and diversity - it has not yet established a coherent strategy.

3. Solid individual and organisational commitment
This interest in learning leads us to motivation and commitment, both institutional and personal. COEP's membership is institutional, but what an entity does, and what it contributes to COEP, depends very much on the energy, interest, commitment and imagination of the person(s) representing it. This holds true in both national and state networks. So much so, that respondents who know COEP well said without hesitation that it is the commitment of the people within COEP that holds the network together and makes it work. Nevertheless, personal commitment, the most intangible element of capacity, comes into play only within an institutional context, in concert with the financial/material resources and organisational processes of COEP's affiliates, not to mention the constitutional principles and structures of the network itself. These institutional features comprise the enabling operational conditions in which COEP's people apply the energies and passion that they bring to their work.

In terms of commitment and motivation, COEP looks like a voluntary organisation. Its institutional members decide what they will contribute, and the way that works out in practice depends on how the individual participants from the entity carry it out. From this perspective, COEP is a part of civil society, despite the fact that most of its members are public and parastatal organisations and a minority are private firms and NGOs. This voluntary association, and a structure that encourages participation, gives it special qualities. COEP's participants want to work in the network - no-one compels them to do so. At the same time, the leadership has been welcoming and encouraging, a quality that respondents praised highly.

The growth of state-level COEPs provides a good example of the ethos of the network. The national entities encouraged the creation of these networks in the mid-1990s, aware that employees of their state subsidiaries would play key roles. Their endorsement created an opportunity for activists in smaller cities to act on social issues in their own communities. The rapid growth of the state networks shows how energetically people and organisations seized that opportunity. The leadership has also provided experienced advisers as backup where needed. Through the Oficina Social budget, the national entities have also ensured that the state networks can participate in COEP's planning meetings. Oficina's cadernos and videos describing innovations in development work, as well as the training programmes, are examples of COEP's commitment to participants throughout the network.

Participants in COEP have usually been ready to work towards consensus so that the network functions effectively. Respondents say that COEP's social and geographic diversity, combined with a non-hierarchical structure, make consensus difficult to achieve. Equally, once established, that consensus is durable because it is based on will, not compulsion. There are strong personalities in the network, but there has been remarkably little internal conflict. Respondents offered several explanations: the apparent widespread agreement on COEP's purposes and core values; its structure (there are 'no bosses and no employees'); the rewards for taking part are non-financial; and the primary principle is solidarity, not egotism. Others praised the style of individual leaders, especially at the national level. The structures and culture of the network require a particular type of leadership: non-authoritarian, accessible, ready to listen, ready to encourage people and accept diversity, willing to work towards consensus while maintaining the integrity of the network's mission, ensuring that people have the space to voice their ideas, and sufficient autonomy to get things done, within COEP's broad programme guidelines. One respondent observed that 'We don't have an articulated strategy, but we do have results'.

Several respondents spoke of COEP's activist culture - people take part because they want to get things done. Regular face-to-face and electronic meetings are important, but there seems little interest in prolonged organisational wrangles. The material base of this voluntary association, derived from its institutional membership, plays a part, although it is not immediately obvious. Individual participants are paid employees of their entities, which are also the vehicles by which they contribute to COEP's programme. Programme discussions thus do not involve allocating funds to one issue or another, because COEP has no programme funds in this sense. Hence, programme decisions do not impinge on the jobs and salaries of its participants. In a similar vein, COEP's participants are not constantly raising funds to 'feed the machine' and arguing about their use. The leadership has secured annual subventions for Oficina, to be sure, which are critical to the network's operations, but this is a periodic rather than continuous undertaking, and one that the leadership attends to. Important as it is, it appears to be a sidebar to COEP's discussion of programme directions and organisational development.

Finally, COEP's design rests on a link between voluntary commitment and organisational and professional competence. When COEP's founders invited the entities to join a social action committee to combat hunger, they asked them to do what they knew, but to do it differently, for different people, and to work with and for poor and marginalised communities. The decision to make training and other forms of professional development available to network members was wise, and participants have welcomed it and have benefited from new skills. COEP's open culture has given many people a chance to use their knowledge outside the bureaucracies where they work.

Our respondents' comments on COEP's capabilities focused on these first three qualities - creative and legitimate leadership, an evident capability for strategic thinking and action, and individuals' motivation and commitment. The two remaining factors were usually regarded as residual influences. We now examine COEP's values and mandate, and its technical, financial and professional resources.

4. Values, mission and role
We hypothesised that COEP's values, mission and role were a unifying ethical reference point for the organisation. Our respondents gave this factor little attention as a distinct element in COEP's success - it seemed tacitly accepted as part of its identity. Nevertheless, references to COEP's principles, and to the values guiding the participants, were diffused throughout the respondents' comments, especially in the discussion of commitment, and of the creation of the network, its principles and the legitimacy of its founders. Our respondents embraced the ethical framework of the organisation. Indeed, this is what attracted many of them to COEP. Several of them attributed the relative absence of conflicts within COEP to explicit agreement on its aims and purposes, and to a less overt but still influential understanding of the rules of appropriate behaviour and interaction among participants. Several respondents also noted that as COEP has grown, the leadership has properly decided to codify its principles in the 'Regimento Interno', and to establish an oversight body, the Administrative Council. They saw this formalisation of principles as entirely reasonable. Because respondents share a broad consensus on COEP's values, their influence in the examples above appears unexceptional.

5. Technical, material and organisational resources
Last on the list are the more tangible factors at work within COEP. Respondents' views on the resources available to COEP were more contradictory. When we asked them to rank the importance of the five success factors in shaping capacity, this one invariably came last. However, respondents made two arguments that assign it much more significance. When we asked about key events in the life of COEP, several argued that the rapid adoption of the Internet in the mid-1990s was the critical technical condition that allowed the creation and growth of the state-level networks. Electronic communication was an essential link in the organisational chain. Oficina Social has relied on the Internet to make COEP's project database widely accessible. One of most acclaimed projects in COEP's ambit, Canal Saudé, the health education channel, offers electronic distance education.

Respondents made a similar point about the appearance of Oficina Social at a critical stage in COEP's evolution. The growing network needed a supportive infrastructure that could finance meetings and communication, and assist in the development of individual competencies. The scale of COEP required a substantial response, and the national entities have provided it, in the form of their subventions to Oficina, as well as in-kind support, namely the paid time of their employees. As our discussion of commitment suggests, this defining quality of a voluntary organisation owes its influence in COEP precisely to its combination with the less glamorous institutional resources of the member entities. From yet another perspective, COEP's very dependence on these material subsidies is an indicator of its success in institutional mobilisation.

Our respondents' lack of emphasis on this factor does not fit well with one of the strategic assumptions guiding COEP's creation: that the substantial resources of the parastatals should and could be harnessed in the campaign against poverty. Moreover, they argued that COEP could claim major achievements in this field. From the perspective of its initial logic, then, as well as its track record, COEP could well be described as a capacity utilisation organisation. The clarity with which respondents analysed COEP's original premises, and the conviction with which they argued the case for its effectiveness, thus seem at odds with the marginal weight they assigned to this success factor.

One additional point should be added to this discussion, drawn from the author's observations over more than two years of work with COEP. A visitor can only be impressed by the presence of highly qualified people working within COEP's networks in far corners of Brazil. Their presence and the quality of their work are testimony to their commitment, but also a signal of the depth of Brazil's reservoir of qualified people, and of the organisational strength of the entities that employ them. This also verifies the validity of another assumption about the value of the entities' capiliridade.

Placing this factor in comparative perspective shows its importance. Let us hold constant the special political or cultural qualities of Brazil that may have allowed COEP and sustained it, as well as the influence of inspired individuals. It is hard to imagine COEP working as it does without its affiliates' substantial material contributions, or without the deep pool of professional competence within the network. The test surely is to look elsewhere in countries of the South, and ask where similar enabling capabilities may be found. The list would be a short one.

4.2 Capacity as a product of supply or demand?
To make a final observation about the interrelationship between COEP's capabilities and its broader environment: capacity in COEP cannot be neatly categorised as a response to social demand, or as a product of organisational supply. It appears that there has been a more complex interplay among social need, a strategic opportunity seen and seized, and a widespread response to an organisational vehicle once it was created. There is no doubt that COEP is acting upon a social priority in Brazil, the fact of poverty and injustice. Clearly, in the early 1990s citizens' movements exerted pressure for public action on these issues. Yet, supply factors also played a role. Creating COEP required the initiative of imaginative people who made a compelling case for a particular type of public action. The subsequent development of the network has required sustained strategic and operational competence within its leadership. COEP was established as a means of engaging organisational resources and enlisting individuals' commitment in the campaign against poverty. Once that vehicle existed, organisational supply created its own demand: people and organisations across Brazil have responded to the opportunity that COEP's presence and record offered them. They have enlarged the network dramatically, thus deepening and broadening its capacity to act. Finally, if we view their commitment as an aspect of social demand, their obvious personal engagement is all the more effective precisely because they can call upon a substantial institutional supply of material and technical resources.

4.3 Capacity and paradox
This discussion of success factors provides a plausible explanation of capacity within COEP, but this may not be the whole story. We set out here some additional observations on capacity that may be useful to the network. Our research identified numerous paradoxes within COEP's culture, structures and dynamics, which may be nodes of creative tension within the network. For example:

• COEP is not a formal organisation in the conventional sense, yet its members have formally affiliated themselves to the network, and its principles, structures and procedures have clear constitutional expression.
• Informal power is the 'current' that makes these formal structures work. It is evident in the style of leadership, and in the energy the participants bring to the network.
• The membership of the network is institutional, but it is the commitment and competence of the people involved that make it go.
• COEP's structure is non-hierarchical, intended to recognise diversity, give space for participation and promote organisational agility. It is nonetheless tightly structured, with deliberately consistent form and function among its national, state and nascent municipal networks.
• COEP's original ethical principles and purposes have been preserved, as have its basic structural premises, yet its organisational profile has been radically reworked. There remains a constant tension between preservation and renewal.
• Intangibles like leadership, creativity, confidence and legitimacy give COEP its energy, and attract new participants. Yet, it can only do what it does because institutional members make sizeable financial and other in-kind contributions, in particular the paid time of their employees.
• COEP is a voluntary organisation whose members retain their institutional autonomy in programming. Similarly, the people who participate in COEP choose to do so. Yet COEP's capabilities exist because its participants hold salaried positions secure enough to let them 'get on with the job'. They do not have to scramble for funds to feed an administrative machine, or to dance with donors about programme directions.
• COEP does not accumulate or dispose of its own material or financial resources. Yet, it guides the application of substantial organisational resources towards national campaigns and community development initiatives and touches hundreds of thousands of people. For all this, it may not be known or seen by many of the communities involved.
• Politically, COEP is consciously nonpartisan. Yet it operates with evidently good political judgment in a charged institutional milieu, and is engaged with the big development issues - poverty and social justice - which are inescapably political.
• COEP's ability to maintain a nonpartisan stance contributes substantially to its legitimacy and independence. These qualities continue to attract the attention of political actors, so that the preservation of organisational autonomy is a constant challenge.
• Guided by its mission and values, COEP works for the betterment of communities throughout Brazil. Collectively, it does not work with them. It is COEP's member entities, including the individuals active in COEP, who do so. Hence, there is a structural disconnection between COEP as an organisation, and the people it is trying to assist. Its affiliates may be COEP's means of working to combat poverty, but their prominence obscures the presence and role of the network. COEP per se is thus not directly accountable to the people in whose interest it works.

COEP's objectives - its 'project' - are the eradication of poverty and the construction of social justice. These are its nominal and real purposes, and they motivate its people. Something else is also at stake, however: governance and democracy. The success or failure of COEP's efforts to harness the entities to its cause will tell us whether these important institutions can become accessible and responsive to all Brazil's citizens and communities, and whether they can advance the public interest. In this sense, COEP's long-term success or failure will be a partial barometer of the quality of Brazilian democracy, and of the limits and possibilities of active citizenship, cidadania. It is, of course, to the everlasting credit of Brazil and COEP that this issue is even on the public agenda. 'Mature democracies' such as Canada would do well to watch, learn and emulate.

This list of paradoxes could probably be extended. COEP's leaders and participants generally seem comfortable with these tensions and manage them well. These tensions, and the way they are acknowledged and managed, suggest another set of social processes at work, possibly forces within social and organisational cultures that are not evident to an outside observer. If these and other creative tensions are indeed present in the network, their workings and implications require another level of analysis beyond the preliminaries here. It may be, for example, that these paradoxes are in fact the source of COEP's evident collective energy. Their presence thus offers COEP participants an opportunity for further reflection on why and how their network works.

4.4 Relevant elsewhere, or a special case?
Readers may ask whether COEP's experience is relevant to activists working for social justice in other countries. To open this discussion is not to suggest that COEP's form and content are replicable elsewhere. On the contrary - as one of COEP's founders commented, creating COEP might not have been possible in any country other than Brazil, or at any other time in Brazil's history. Moreover, the material base of the network - the wide capiliridade of the member entities, augmented by the electronic web of communications and the substantial pool of technical and professional resources - signals a one-of-a-kind quality, especially for the South. Nevertheless, at least three aspects of COEP's experience seem relevant to other settings:

• COEP's founders 'seized the moment', and made a creative political and social intervention to use available political space. They proposed an organisation with a purpose and form evidently well suited to the cultural and institutional conditions of Brazil at the time. A different society would require a different response to harness and build on popular energies and the desire for change; and of course the political space and terrain may be quite different from those that existed in Brazil in the early 1990s.
• COEP's formation reflected a specific strategy to address poverty. It was clearly understood as a problem for the entire society, not only for people living in poverty. It was integral to the public interest, and thus warranted mobilising resources from all parts of Brazilian society. More wealthy and powerful economic actors could and should make an important contribution to a broad social project. The vehicle created - COEP - also offered an opportunity for individuals to act on their own commitment to cidadania, enabling them to advance the public interest by applying their professional skills to issues of social justice.
• Democratic governance and active citizenship have been key themes in COEP since its inception. One of its founders' purposes was to change the culture of Brazil's public entities - to make them more genuinely public institutions, serving the interests of all citizens, transparent and accessible in their operations. The same intent now extends to the private corporations within COEP's membership

These issues are not uniquely Brazilian. Brazilian citizens and organisations have addressed them in their own way. COEP's experience may be relevant less for its specific contours than as an implicit friendly challenge: how would the citizens of another country respond?

5 Looking ahead

To be useful, this analysis should also point to items on the strategic menu that may challenge the network's capacity to remain relevant and true to its purposes. Three issues emerge from respondents' comments about COEP's particular qualities and accomplishments. They speak to the way COEP approaches broad development questions.

How is the network to preserve its organisational autonomy and integrity as a nonpartisan entity? This first issue is political, and apparently a permanent fact of COEP's life. The network understands this challenge well: How can it best participate in and support government programmes to counter poverty, without becoming a part of government? Whether the network is a part of government, or a creature of civil society, is of course partly a matter of perception. Different observers have located COEP at different points on society's institutional terrain. The challenge today is cast in a slightly different light. Senior members of the government have encouraged COEP's work and have formally recognised its support for the government's programmes, notably Fome Zero. Such official endorsement is surely welcome. COEP's challenge now is to sustain its support, while maintaining enough structural distance from government to preserve its existence as an independent organisation.

From another perspective, this challenge is also an opportunity. Precisely because COEP has maintained its stance as a nonpartisan actor in public life, it can offer Brazilians a respected public space dedicated to the anti-poverty campaign. COEP can exercise a convening role, providing a forum in which different views and actors may debate and be debated. By adroit use of this space, COEP may, if it chooses, engage with public policy without adopting overtly partisan positions or becoming a specialised 'policy shop'.

COEP's leadership anticipates building closer working relationships with communities and their organisations. This second issue arises from the community-oriented programme strategies COEP will likely pursue in the near future as it continues to support the government's Fome Zero programme. These
relationships will be based on COEP's support for local development projects and programmes. Besides being valuable in themselves, such projects may offer COEP a focus for campaigns to mobilise local financial resources and to promote public understanding of development issues. The municipal networks now appearing will be especially important in this shift, because they hold the promise of being one step 'closer to the ground' than the state-level networks.

There are potential difficulties in this reconfiguration, however, some obvious, others less so. The intention is that COEP as COEP will adopt a higher-profile role in the interaction with communities, with its participants acting as interlocutors between communities and the entities. Questions arise immediately: how does one describe or understand COEP as a development actor? Banco do Brasil, say, is well enough known as a bank, or Furnas as an electricity utility, or Embrapa as an agricultural research and extension agency. What identity, role or métier does COEP bring to the process? Individual employees of a bank or a utility, moreover, have internal guidelines for their work with the public and their responsibility. Who speaks for COEP? How is COEP to present itself to communities, and how is it to be accountable to them? How can COEP assure consistency in style and content across its different networks? These questions do not arise at the moment, because COEP as an organisation plays a different role behind the scenes. The issue is important because it touches on the quality of entities' interactions with communities, but also it opens up an important governance question for COEP. How can it overcome the current disconnect between the network and the people it works for, by strengthening its work with them? Just as COEP's larger 'project' holds implications for the quality of governance in Brazil as a whole, so the impending refocusing of the network's attention on its community-level activities will illuminate its own public accountability.

COEP's historical legitimacy may help it to work effectively at local levels, easing the construction of new accountabilities. At the same time, it cannot assume that legitimacy earned in national or state contexts will readily transfer to community settings. This will have to be earned and preserved by the people on the spot. In a different milieu, an historical asset may be a vulnerability. These are uncharted waters for COEP, demanding careful navigation. A programme of participatory research with communities to examine the impact of COEP's work (see section 3.3) may be a means of exploring the issues deliberately and in depth, and of building trust and knowledge among the people involved.

Together, these first two strategic issues generate another question about COEP's positioning: what is the appropriate balance among COEP's community orientation-in-the-making, its potential role in influencing policy, and its historical focus on mobilising member institutions and strengthening their capacity as actors in social development?

Participants in COEP are seeking to improve their competencies in areas relevant to the new programme directions. This third strategic issue arises directly from respondents' descriptions of the challenges they face as development workers - their desire for more knowledge about community development practice, as well as technical backup for specific activities. COEP's investments in learning have sharpened people's sensitivity to this issue. As participants have immersed themselves in practical development work, and as they have benefited from training programmes and the examples of reference projects, their demand for professional development has increased. This is a good sign, not a problem, although it does create more demands on COEP's leadership at all levels, and on the network's support structures. The prospective change of emphasis towards municipal networks will bring demands for new skills as well. If COEP embarks on mobilising financial resources to be applied to local community development projects, for example, this will require a whole new array of skills in management and communication. These are not difficult in themselves, but they would require a major departure from past practice and probably a cultural shift as well. Until now, the network has not been a fundraising and grant-making body.

To its credit, COEP's leadership has been thinking about ways to handle such challenges, and is reaching out to similar organisations in other countries for a dialogue on issues of common interest. This case study is one result of that initiative. This report is intended as a modest contribution to a remarkable organisation.

Annex I: Notes on method

Work began on this case study in June 2003. ECDPM and COEP agreed to undertake a case study as part of ECDPM's multi-country study, with John Saxby as the principal researcher and author. The author submitted a proposal to ECDPM, and developed the hypotheses and research plan between September and November 2003 in discussion with both organisations. Most of the field research took place in December 2003. Between January and June 2004, the author prepared a first draft of the paper, revised it after receiving comments from ECDPM and COEP, presented a second draft to an international conference organised by COEP in March 2004. After the conference, the report was revised again for publication.

The analytical framework
The ECDPM project uses a seven-part analytical framework (see inside front cover). Its three core variables are capacity, endogenous change and adaptation, and performance. Interacting with these are social and organisational processes and relationships grouped as four intervening variables: internal features and resources, stakeholders, the external context and external intervention. A recent ECDPM Occasional Paper complements this overall framework by elaborating the notion of 'capacity'.²⁴

We adapted this framework to COEP's circumstances as follows. First, we took the three core variables as the principal reference points. These are, in the words of the ECDPM paper, 'the heart of the framework'.²⁵ The dynamic between change and capacity is most critical, with COEP's (presumed) capacity enabling the network to change and adapt over the years. Hence, in our hypotheses (summarised below), we identified five success factors, or aspects of COEP's capacity, to explain the network's apparent effectiveness. This analysis of capacity requires a prior account of COEP's historical evolution - its improbable beginnings and its remarkable growth and diversification - presented in the first part of this report.

The discussion of performance complements that of change and capacity. In essence, 'performance' is evidence of purposive change and adaptation, and of capacity at work. The case of COEP blurs the boundary between 'change' and 'performance' somewhat. As explained in the text, growth and adaptation in a voluntary organisation, especially on the scale experienced by COEP, are themselves indicators of very effective performance.

We have also used an interpretation of 'performance' that is slightly different from that in the ECDPM framework. There, performance is understood in terms of 'execution, implementation, accomplishment, functioning and delivery of value'.²⁶ This definition seems more suited to an operational agency than to COEP, which we understand as a mobilising and influencing organisation (with operational capacity and responsibility in the hands of its individual member entities). Hence, we focused our inquiry on accomplishment.

Within this triad of core variables, the key notion of 'capacity' warrants further explanation. We found the ECDPM Occasional Paper, One More Time, especially useful in the way it broke down 'capacity' into sub-concepts, which we used in framing the discussion of COEP's 'capabilities and capacity' in section 4. The paper identifies five 'Cs': commitment, competencies, components, capabilities and capacity:²⁷

• Commitment - the loyalty, ownership, motivation and confidence that energise people. From this perspective, efforts at developing capacity are successful when they release people's creative energies and skills that are often dormant, underutilised or suppressed. No amount of technical expertise or organisational engineering can replace commitment.
• Competencies - the personal and professional skills of individuals. These have often been the focus of people and organisations working to build capacity. Training programmes are the classic route.
• Components - the features and assets of an organisation, including structure, leadership, mandate, strategic positioning, technical systems, internal and external communications, legitimacy, behaviour and culture, resources of all kinds, geographic profile, and so on. Components are not themselves capacity, but they can be important enabling or limiting conditions for it.
• Capabilities - the specific abilities of an organisation to harness commitment, competencies and components. Capabilities can be organisational, technical, human, or a mixture of all three. They can be focused inwards or outwards, and should support the performance requirements of the organisation.
• Capacity - the overall ability of an organisation to perform. It brings together commitment, competencies, components and capabilities in a coherent broader system.

We then linked these core variables to the four intervening factors identified in the ECDPM framework: the external context, relations with stakeholders, internal resources and features, and external intervention. Of these, the fourth, 'external intervention' (by, for example, bilateral donors or multilateral agencies) is of limited relevance in this case, because COEP is a wholly Brazilian creation. Only in the last two years has it acquired an international dimension, by exploring links with counterparts in the Americas and elsewhere. It is possible that this aspect will become more important in the future - in March 2004, for example, the network hosted its first international seminar - but it had little impact on the organisation in its first decade.

The three other intervening factors appear not as separate categories of information and analysis but as themes running through the account of change, capabilities, capacity and performance. We adopted this approach partly because of the primacy assigned to the historical account of change in the network, partly because these factors are better treated together than in isolation. These variables seem easier to understand as aspects of an historical account rather than as distinct processes. Thus, the influence on COEP of the external (Brazilian) context in 1993, for example, was very different from that in 2003. And, crucially, so is COEP itself. For this study of capacity, then, the issue is the way these two factors influence each other - specifically, the way COEP used its internal resources to respond to a changing external environment (as well as, of course, to its internal environment). Hence these factors are woven into the account of COEP's evolution, and they reappear in the analysis of capabilities and capacity, which offers an explanation of that evolution.

We should note too, our adaptation of the intervening variable 'relations with stakeholders'. We have chosen to focus on legitimacy, i.e. the rightfulness of COEP's existence and its authority, as the critical aspect of this issue. The concept of legitimacy directs attention to the moral basis of the power exercised by COEP - especially relevant because the network has influencing rather than jurisdictional or even operational power. Three groups of stakeholders are central to the assessment of COEP's power and legitimacy: government in Brazil, especially the federal government; the people who participate in COEP - the senior management of COEP's entities and their employees; and the people and organisations of Brazil's communities. Hence, the research explored the extent and basis of COEP's legitimacy within public life, the commitment its participants bring to the network, and the place of these factors in COEP's links with communities. 'Legitimacy' appears as a thread throughout the report, just as do 'external context' and 'internal features and resources'.

Hypotheses
The hypotheses guiding the research were framed in terms of the three core variables described above. They were based on the author's knowledge of COEP, derived from an advisory and consulting relationship since 2000. First, COEP's positive performance as an organisation is evident in its achievements in two important areas: its remarkable growth, change and adaptation over a decade; and its effectiveness in realising its organisational purposes. We should emphasise that this research did not evaluate COEP's development impact. Assessing impact, and COEP's own analysis of the longer-term results of its work, are part of the discussion, however.

Second, we hypothesised five success factors to explain COEP's overall capacity, its potential to perform. These are mostly intangible forces, primarily within the network:

1. a creative and legitimate leadership that has used available political space well;
2. a capability for strategic thinking and action, nurtured by its leadership;
3. solid commitment from institutional members and individual participants;
4. a clear role, mandate and values for the network;
5. a substantial pool of skilled and educated people with professional competencies, and technical, material and organisational resources, within COEP's member entities and in Brazilian society at large.

Data gathering
The two main sources of information for this study were interviews and documents. The latter included publications and documents used principally within COEP. The author and COEP's research assistant, Patricia Baldarelli, consulted 19 people via group and individual interviews. Respondents included four people working within COEP's national structures, 14 from the state-level networks, and one university faculty member familiar with COEP's work. The interviews ranged in length from one hour to four and a half hours (the latter in two sessions), with the usual length being two hours. The interviews included questions seeking information about COEP, as well as the respondents' assessment of issues within the network. The interview format comprised mainly open-ended questions, and the interviews took place in Portuguese. We agreed before the interviews that we would not attribute respondents' information and opinions unless they asked us to do so. The interview schedule is available on request.

The documents and websites consulted are listed in the references. Staff of Rede Mobiliza kindly provided information on COEP's finances (see Annex II) and on COEP's membership.

Annex II: Financing COEP

Compiling a full picture of how COEP's structure and programme are financed would be a major piece of research in itself, beyond the scope of this case study. We can provide some brief but useful indicators, however.

(1) Budget of Oficina Social: As the secretariat for the network, Oficina's budget shows COEP's cash operating costs, including its requirements for communications, meetings and publications, as well as for project monitoring and support. The following table shows Oficina's budgets for 2002 and 2003 (in Reais, R$100= €26.27, figures rounded):



Related in-kind costs are considerable, but would have to be costed. Individual entities provide substantial subsidies, in the form of (for example) graphic design and printing services, and teleconferencing facilities.

(2) In-kind personnel costs: COEP's member organisations allow their employees to do COEP work on company time, thereby contributing salaried time to the network. This does not account for all the time people commit to COEP, during evenings, weekends and vacations. A respondent gave the following estimate of costed time allotments for the executive committee of a state-level COEP, acknowledging that actual figures would vary with the size of the committee, the scale of its programme, individual workstyles, and so on:

Number of people: 20 (from 20 entities).
Hours committed per week: average 1.5 hours per person = 30 person-hrs/week in total.
Monthly time commitment, total: 120 person-hours, or three 40-hour weeks.
In-kind personnel costs @ estimated average monthly salary R$2500: R$1875, or R$22,500 per year.

(3) Programme resources, cash and in-kind contributions: These figures are also extremely difficult to estimate, either in aggregate or on a project-by-project or entity-by-entity basis. This is partly because much of COEP's work has historically been premised on encouraging the entities to make in-kind contributions to the campaign against poverty. Indicators of the scale of these resources are noted in the text, such as the action plan for Fome Zero. It would be useful to include a more accurate assessment of these resources in any assessment of the impact of COEP's work.

Annex III: List of signatories to COEP, 1993



References



Websites consulted:

ECDPM: www.ecdpm.org; www.capacity.org
COEP: www.coepbrasil.org.br/mobilizacao
Banco do Brasil: www.bb.com.br
Banco do Nordeste: www.bnb.gov.br
Embratel: www.embratel.com.br
Furnas: www.furnas.com.br
Petrobras: www2.petrobras.com.br

-----------------------------------------------------------------------------------------------------------------------

Notes

1 Citizens' Action in the Struggle against Hunger and for Life, and the Movement for Ethics in Politics, respectively. For two Brazilian accounts of COEP's origins, see Fleury et al. (2002) and Miranda (1994).
2 President Franco accepted a proposal from Lula da Silva (then leader of the Partido dos Trabalhadores, the Workers' Party, now President) to create the Council on Food Security (CONSEA) as the strategic body to guide the campaign. COEP has worked with CONSEA from the inception of the network. Bishop Mauro Morelli, president of CONSEA, was also president of COEP's first Deliberative Council.
3 Several respondents emphasised the importance of Furnas' initiative, which encouraged others to join the committee.
4 Betinho was known to be on the left of the political spectrum, but made the case for the priority of the campaign by describing poverty as morally unacceptable and socially unsustainable.
5 The 1993 statutes provided the basis for the creation of state-level networks in 1995. The most recent version, adopted in 2001, made provision for municipal COEPs.
6 Quoted in Fleury (2002: 259), from Article 3 of the 1988 Constitution.
7 This voluntary (unpaid) position, held by an employee of Furnas, is the most prominent example of the contributions in kind made by COEP's members. See Annex II.
8 Although COEP is not a registered society under Brazilian law, legal opinion holds that this protocol is binding on its signatories.
9 The diverse nature of poverty had long been acknowledged. In Betinho's words, 'Hunger has a name and an address'. COEP drew on The Map of Hunger, a 1993 policy document on food security prepared by Anna Peliano of the Institute for Research on Applied Economics (IPEA).
10 www.coepbrasil.org.br/mobilizacao, accessed 28 June 2004. As part of its contribution to the government's Fome Zero programme, COEP opened its project database to other organisations, encouraging them to contribute to and to draw upon it.
11 By 2004 Oficina had published 13 cadernos. One COEP member takes care of the graphic design and printing of the cadernos - another example of an in-kind contribution to the network.
12 Although this profile may be more commonly understood as that of Oficina. One of the challenges for COEP is to create a clearly identifiable public image for the network, distinct from its members and from support bodies like Oficina.
13 See COEP (2001).
In August 2003, out of a membership of 762 organisations, 256 (just over one-third) were non-state entities. Of these 256, 100 were private firms, and the remainder civil society organisations - NGOs, professional and business associations, religious bodies, non-profit educational institutions, cooperative associations, and labour unions.
14 In August 2003, out of a membership of 762 organisations, 256 (just over one-third) were non-state entities. Of these 256, 100 were private firms, and the remainder civil society organisations - NGOs, professional and business associations, religious bodies, non-profit educational institutions, cooperative associations, and labour unions.
15 O COEP no mutirão contra a fome, vols I and II, Rio de Janeiro, April 2003. This statement is clearly consistent with the description of COEP's role in its 1994 statutes.
16 According to one respondent, COEP has an activist culture, and its participants use their time to act on social issues. Evaluating projects would probably be a low priority for most people in the network. As argued in section 4, this activism is precisely one of the reasons why COEP is effective.
17 Banco de projetos mobilização, www.coepbrasil.org.br/mobilizacao/, accessed 28 June 2004.
18 Miranda (1994), p.34, and interviews.
19 Krutman (2004). Oficina staff also monitor projects supported by COEP, such as those of the cotton-growers' cooperatives in the northeast.
20 Some of COEP's leaders thought that COEP had erred in promoting Oficina as it had - not that Oficina's role was unimportant, but rather that in emphasising Oficina's contributions, COEP underplayed its own significance and unwittingly diminished its own profile.
21 One approach would be to build on the research by IPEA, COEP and Oficina Social for 'Pesquisa Ação Social' in 1999. This summarises the organisational makeup of the network at the time, the type of activities undertaken, and an assessment of the achievements and challenges.
22 In the words of one activist, 'Evoluir para continuar o mesmo' ('Evolving in order to remain constant'). Many respondents said that there is not 'one' COEP. The network's reality changes according to who is speaking about it.
23 A loose translation would be: 'diversity, decentralisation, projects and municipalisation'.