Peter Uvin and David MillerThe World Hunger Program
The World Hunger Program is part of the Watson Institute of International Studies at Brown University. The 1970s and 1980s have witness ed a dramatic growth in the number, the size and the level of activity of non-profit/non-governmental organizations (NGOs) in the Third World. They include peasant associations, neighborhood associations, people's movements, community initiatives, urban a ction committees and intermediary NGOs, filling the ranks of what is often referred to as the "associative sector." This phenomenon parallels a slow but profound change in the international aid system. Since the middle of the 1970s, donor agencies have be come increasingly preoccupied with the issue of participation and have looked to Northern and Southern NGOs as the vehicles for carrying forward an agenda of participation. As a result, both the numbers of international and local NGOs and the number of do nor funded projects that have a participation component within them have grown remarkably over the past ten years. This growth in size, number and activities of organized participatory initiatives has become recognized within the development community as" scaling-up." It constitutes an objective, or desire, of many funders, field practitioners and scholars alike.
However, some important questions about scaling-up need to be answered. What role do the participants that expand the ranks of the scaled up o rganizations play within these organizations or within their funded projects? Is there a relationship between donor support of scaling-up and constituent participation? In what way will this new interest in scaling-up affect the rank and file of the parti cipation movement? Has the redirection of development assistance toward scaling-up initiatives been effective? As Edwards and Hulme stated it: "how can [NGOs] increase their development impact without losing their traditional flexibility, value-base and e ffectiveness at the local level?"
Together, these questions constitute a research agenda that, to date, has not received adequate attention. Indeed, most of the literature on grassroots development movements is no rmative in nature; the rest is largely anecdotal. It is as if the usual laws of scientific inquiry, verification and experimentation do not apply when dealing with organizations whose social goals have often eluded the same rigors of measurement administe red among the bottom lines of government and for-profit organization.
This article proposes a first scientific look at scaling up. It does not present any grand theory of scaling up, nor is it the result of detail ed comparative field research. Rather, it represents what can be called a "pre-theory:" the development of some clear definitions and taxonomies, which can constitute the basis for scientific investigation and discussion. Indeed, only when there is an und erstanding of the dimensions of the concept of scaling up can donor and beneficiary, participant and observer, scholar and practitioner, begin to communicate in a way that can address the questions above. This article will also supply the interested reade r with a foray into the existing literature, suggesting paths for further reading.
1. The History of Scaling-up
Schematically, since the 1960s, two basic organizational paradigms for promoting development have dominated th e scene. The first is the top-down, usually state-led model. In it, the central government, with support from the aid system, through its policies, programs and projects, attempts to set in motion development. The population will follow suit, as will `dev elopment.'
For most of the last thirty years, this model has been dominant, with Third World governments, the development aid system and most development scholars. Yet, it also has been the subject of severe criticism. Time has shown the central state often to be very inefficient in `creating' development: too far away from the population, rigid and bureaucratic, often controlled by elite interests and characterized by a condescending attitude towards the poor and the peasants, it does not respond ver y well to the needs of the majority of its population. Subject to internal rivalries and external pressures, plagued by poor budgets and staffed by badly trained and underpaid civil servants, it is a fertile ground for vested interest, rent taking and fac tional enrichment. Hence, the state has proven in most cases to be a poor organizational device to implement effective and just development strategies. Similarly, top-down projects, even if not implemented by the stat e, were proven to be largely unsustainable because of a lack of local institutional involvement.
The second model for organizing development is a reaction against the perceived deficiencies and injustices of the t op-down model. Its origin dates from the 1960s by NGOs and engaged scholars, but it gained only broad acceptance in the second half of the 1980s. The bottom-up, grassroots model seeks to organize development from below. It seeks to increase the capacity o f poor people and communities to take their fate in their own hands, to change the conditions of their lives; in short, to `empower' themselves through `participation' in their own development. This paradigm of participation centers on the creation and th e strengthening of NGOs and community based, grassroots organizations. They are to tap the latent energies of the poor in fostering endogenous development and social transformation.
For various reasons, this strat egy became increasingly accepted by the development community in the 1980s, most of all on the level of discourse, but also, and increasingly so, in practice. The above described limits of the previous state-centered, top-down development strategy were be coming increasingly evident. Project and sectoral evaluations all through the 1980s revealed that projects that were not appropriated by local institutions tended to be much less successful and less sustainable than those in which local communities played a central role. As one Asian Development Bank researcher states it: "NGO provision of services is an effective and growing response to some of the problems that projects have faced in the past." Moreover, a deep econ omic and financial crisis hit most Third World governments in the 1980s. It necessitated that new, internal, sources of financing and investment be found: tapping the financial and human resources of local communities were the evident alternative. Finally, the dominant global ideology of the 1980s was strongly in favor of anything that was not public: private enterprise for the creation of wealth, and private association for its distribution. Hence, reasons of cost ef ficiency and ideological appeal combined to make the more fiscally conservative managers of development aid march down the same road as the more politically liberal advocates of economic and social transformation and popular empowerment.
This "new" participatory, bottom-up, NGO-led, strategy is not without its detractors. Common critiques among many development specialists are that:
most NGO-led programs are too small, underfunded, poorly staffed, slow and localized in the face of the daunting size of the problems of poverty, hunger and environmental degradation, often NGOs actions do not match their rhetoric, for example in the fields of community participation, innovation, cost-effectiveness, reac hing the poorest, most existing NGOs are unable to absorb the massive quantities of money needed to make a difference, the anti-state attitude of most NGOs limits both their capacity to change some of the root causes of underdevelopment (policy level causes) or to have a large impact.
Existing NGO programs were thus considered to be only actions at the margin, capable of providing local relief and service, but not able to tackle the real issues of the eradication of hunger and poverty for hu ndreds of millions of people.
Following this, at the beginning of the 1990s, a new model has been emerging, drawing on the best of both previous strategies. It holds that the state remains necessary, for only it h as the capacity to enable and coordinate the work of NGOs and private initiative on the scale that will make a difference. It can do this by providing certain public services, enabling legislation and subventions in the forms of entitlements. Interacting with the state should be a new breed of large, well managed and competent NGOs that are capable of mobilizing the participation of large numbers of people and channeling large sums of money to large-scale activities. At the state side of the equation, thi s involves a willingness and a capacity to take a partnership position with civil society, to create an "enabling environment" for private initiative. At the NGO side, it implies the need for "scaling-up," i.e. to increase their size, complexity, i mpact and interaction with the state. Presently, much money is poured by development assistance organizations into promoting such scaling-up of Third World NGO activity.
It is already clear that the basic, as yet unanswered, issue concerning scaling up will be the one of its relation with participation. It can be put like this: How can a development initiative move beyond the local level and make a larger impact while continuing to foster participation? Can a participatory, bottom-up program, or the or ganization managing it, scale up while avoiding the problems of cumbersome and overstaffed organizations, detached from their grassroots bases, becoming mere sub-contractors of the foreign aid system or of the state, unaccountable to the communities who t hey claim to represent? This paper will not supply an answer to that question; it will, however, provide some clear terms allowing to address it.
2. The Concept of Participation
The term "participation" has become one of t he most debated terms used in the development community. It has been used to explain a broad host of activities that have been done to, for and by NGO constituents. What is a participatory organization becomes nothing short of a philosophical question. We wish to define the term participation more precisely here, presenting a hierarchy of elements that can be identified in the way an organization conducts its affairs that enhances the participatory nature of that organization.
At the bottom (zero-level) of the hierarchy we find participation by the target population as program beneficiaries -- the stated objective of every development program. Immediately above that, we find organizations seeking constituent part icipation in the costs and the work of their programs. This appears to be the most common form of participation supported by development institutions such as the World Bank, or by many Third World governments. On a higher level, we find people's participa tion in training that permits the transfer of organizational, managerial and technical capacities to members of the target community; supposedly, this will prepare them for taking over after the program or project departs. This is also a prelude to the ne xt level of participation whereby target groups participate in the management of development programs, making routine decisions over the day-to-day operations of the program (also called participation in implementation). While the community may make manag ement decisions, the strategic, goal-setting decisions continue to remain in the realm of the executing or funding agency. The next level -- participation in policy-making -- indeed puts decision making around program goals into the hands of the community (also called participation in design). The highest level of participation is ownership by target group members over the resources of the organization or program. Ownership can take many forms including worker-owned companies, producer or consumer coopera tives, land trusts, and limited equity housing schemes.
It appears that, for maximum effectiveness, each of these elements sits along a continuum that is inclusive of the elements below. For example, ownership of a program is preferable when the owner s are not absent from making policy. They make better policy if they have experience managing the programs. Program managers make better decisions when they have acquired relevant skills. Relevant skills are best learned when they are directly related to a participant's hands-on working experience. And workers are most motivated when they directly benefit from the programs in which they are working. In other words, the higher forms of participation are only superior if the lower forms are present too; if not, they can be meaningless (f.ex. cooperative ownership without decision-making power or capacity)
This paper deals with those organizations and/or programs in which program beneficiaries are engaged at least in the management of their programs, i.e. making decisions about the day-to-day operations and the ways to attain operational program objectives. We do not consider programs or organizations that allow for the lower levels of participation to be truly participatory ones. This, then, is q uite a high cut-off point, designed to exclude the many cases of participation in rhetoric, "that do not go beyond taking advantage of cheap local labor for construction of public works, or, for example, involve token consultation with village chiefs in o rder to gain the acquiescence of the population." Examples of the latter include the socialist-type of participation, most rural cooperatives, or the World Bank type of participation.[1 2]
This definition does not mean, however, that participatory programs have been initiated solely by the target group, nor that they are entirely funded or managed by it. This would be unrealistic and self-de feating. No meaningful development comes without outside contact; given the size of the needs of the poor and the hungry in the Third World, local (financial and human) resources alone are not sufficient to respond successfully to the many challenges faci ng Third World communities. The defining criterion for bottom-up development is not that there is no external funding or expertise, but rather that the people concerned are engaged in the decision-making concerning these resources.
In this paper, we de fine NGOs, approximately following Salamon & Anheier, as those organizations that are formally constituted, non-governmental, non-profit, not overly partisan (in a party political sense) and participatory. This is a broad definition including internat ional, national, regional and local membership organizations; funding, operational and advocacy NGOs; community associations, cooperatives, networks, service providers and intermediary NGOs. In much of the American l iterature, the term NGO is used only for intermediary organizations, i.e. those non-governmental organizations (mainly from the North) that provide financial and technical assistance to grassroots organizations; in the US, the term PVO (Private Vol untary Organization) is also used for them. As can be seen, we employ the term NGO in its larger, generic meaning.
If organizations, and the programs they run, are characterized by high levels of participation in policy-making and management or even ownership, then they really constitute cases of self-help, i.e. people taking care of their own problems through their own actions, maybe receiving some outside support. Hence, in this paper, following Koenraad Verhagen, we will often use the term self-help, distinguishing between two types of organizations: "self-help organizations" (SHOs -- membership organizations searching to advance their own interests through their own activities) and "self-help promotion organizations" (SHPOs -- intermediary organizations working at the behest of self-help organizations). SHOs tend to be local and membership based (but they can form national coalitions), while SHPOs can exist on any geographic level and are usually not membership based.
3. A Taxonomy of Scaling-up
Different definitions of scaling-up have been used in the literature. In a recent USAID evaluation of tw o innovative Freedom From Hunger credit projects in Africa, the term scaling-up -- the focus of the study which is explicitly mentioned in its title -- is equated with `expansion', or, more precisely, the need to "reach several times the actual number of members" in the countries concerned. This definition of scaling-up as expansion of membership or target group is probably the most commonly used one.
Social scientists tend to propose more complicated definitions. Robert Berg, for example, talks about scaling-up organizationally, management-wise and financially, while Goran Hyden differentiates between scaling-up organizationally and functionally or activity-wise. "Organizationally" is defined by both authors as "serving larger constituencies," i.e . the same organization, keeping the same goals, grows in size. This, then, is the same meaning of the term as `expansion.' "Functionally" means that the same organization increases or diversifies its range of activities, regardless of size. Howes and Sattar talk about "intensification," referring to the addition of new activities to existing programs.
Clark, in his book on NGOs, makes a different distinction among three types of scaling-up: project replicatio n, building grassroots movements and influencing policy reform. The first two are linked to expansion, but the latter is new. Fisher also defines scaling up as the process of influencing policy; she uses the term "scaling out" to describe expansion. Although he does not use the term scaling-up in this context, Korten clearly discusses a similar type of scaling-up where he advocates so-called "third generation" NGOs. The latter a re distinct from first and second generation ones by their concerns for "bridging the gap between micro and macro" (i.e. quitting the local level) and their desire to attack the (political) root causes of underdevelopment, and not its manifestation s. This involves the development of relationships with governments as well as international partnerships. Therefore, for Korten, Clark, Fisher and many others, influencing politics is considered to be an important form of scaling-up.
Finally, coming fr om a different tradition, Bernard Lecomte, a French grassroots specialist with decades of field experience in Africa (and co-founder of the "Six-S" movement) writes about different phases in the maturing of self-help organizations -- phases mainly charact erized by increased capacity to innovate, generate local resources and improved organizational capacity. His scaling-up is a matter of autonomy, self-reliance, independence.
All in all the literature of scaling-u p is reminiscent of the Loch Ness monster. It has been sighted enough to make even the most skeptical give it a measure of respectability; and its description is as varied as the people who have written about it. We believe that this variety of def initions is important. It allows us to look at the phenomenon in a number of different ways, giving us some insight into the complexity of the associative sector itself. It also suggests that there are different types of scaling-up, which often go togethe r but are not identical. In the typology below, we look at scaling-up in terms of either structures, programs, strategies or resource base.
The first type of scaling-up is where a program or an organization expand s its size, through increasing their membership base (in the case of grassroots organizations) or their constituencies (for SHPOs) and, linked to that, their geographic working area or their budgets. This is the most evident kind of scaling-up, equaling ` growth' or `expansion' in their basic meanings. We call it quantitative scaling-up. It happens when participatory organizations draw increasing numbers of people into their realm.
A second type of scaling-up is where a community-based program or a grassroots organization expands the number and the type of its activities. Starting in agricultural production, for example, it moves into health, nutrition, credit, training, literacy, etc. This we label functi onal scaling-up. It takes place when participatory organizations add new activities to their operational range.
The third type of scaling-up refers to the extent to which participator organizations move bey ond service delivery and towards empowerment and change in the structural causes of underdevelopment -- its contextual factors and its socio-political-economic environment. This will usually involve active political involvement and the development of rela tions with the state. This process, similar to a graduation to higher generations in Korten's parlance, we call political scaling-up.
Finally, community-based programs or grassroots organizations can increase their organizational strength so as to improve the effectiveness and efficiency of their activities. It can be done financially, by diversifying their sources of subvention, increasing the degree of self-financing, creating activities that genera te income (cooperative enterprises, consultancy), or by assuring the enactment of public legislation earmarking entitlements within the annual budgets for the program. It can also be done institutionally, by creating external links with other development actors, both public and private (including the enterprise sector), and by improving the internal management capacity of the staff (such as through training or personnel development) allowing the organization and its programs to grow, to be flexible, to be sustainable. This we label organizational scaling-up.
These different types of scaling-up form part of the dynamics of successful community based development, in all sectors. Take the case of the fight against hunger. In that context, quantitat ive scaling-up essentially refers to the strategy for community mobilization employed by community-based programs: how did these organizations bring about community self-help in the field of food and hunger? How did they draw increasing numbers of people into their programs? How did they manage their expansion? The case of functional scaling-up asks the question of the type of activities organizations have undertaken to combat hunger. This is the more "technical" question on what type and sequence of acti vities organizations or programs undertook against hunger: increase food production or income, organize credit systems, cereal banks, etc. The analysis of political scaling-up asks to what extent have organizations attempted to influence local, regional, national and international factors that cause hunger? Do programs attack the political and economic forces that often underlie hunger such as unequal landholding, unequal access to credit and inputs, corruption and clientism, or the exploitation of ethnic groups, women and children? Finally, the question of organizational scaling-up leads us to the issues of civil society and sustainability. To what extent have organizations created legitimate and diverse social structures, allowing local communities to m anage their own affairs in a sustainable manner? Can existing programs continue, even after the (gradual) withdrawal of foreign aid? Are leadership and management capacity being built that can serve as the starting point for action in other fields? This i s a most important question, often neglected in the usual project approach to development or hunger eradication.
Following the taxonomies laid out above, we will now look more closely at how scaling-up occurs.
4. Quantitative Scaling-up
We distinguish five basic paths for quantitative scaling-up, i.e. for increasing the size of NGOs and their programs. Each of these different paths of quantitative scaling-up, as well as the way the development aid system see ks to support them, presents its own problems. Schematically, they are:
spread: increasing numbers of people spontaneously adhere to the organization and its programs, perceiving them to serve their interests/ preferences. Spread is the almost o rganic process of growth of organizations or programs by drawing more people into them. It can happen at any level of participation: ownership, decision-making, policy-making. It can also take place, and often does, without any external interference in th e form of development projects, etc. It is most typical for grassroots organizations and for community-based programs, growing from a handful of villages working together around one charismatic leader to peasant organizations with tens of thousands of mem bers in hundreds of villages -- as has been well documented for some Senegalese grassroots organizations.
replication: a successful program (methodology and mode of organization) is initiated else-where. Of ten, in the case of so-called pilot programs, this process is deliberate: a model approach is tested on a small scale, and, if successful, promoted on a larger scale. Replication can be done either by the original or ganization or by another one -- in which case it is akin to integration (see below).
Replication is very popular with Northern and Southern NGOs, and, to a lesser extent, official bilateral and multilateral development cooperation agencies. For some, r eplication of pilot programs offers the potential of a more scientific approach to development, whereby programs (or various approaches to one program) are first "tested" in the field before more money and people are committed to it. For others, replication represents a possible solution to a deep seated conflict within the NGO community. On the one hand, NGOs have a strong philosophical commitment to participatory, community based development; on the other hand, the y need to respond to external pressures for a controllable, fast, results-guaranteed, mode of functioning. NGOs do not have the time to sit and wait until well-developed grassroots organizations, capable of dealing with external resources and mobilizing l arge internal energies, present themselves. Thus, so as to save time and to encourage development faster, they take recourse to the duplication of models that they have used successfully elsewhere.
The two most famous cases of replication are probably both from Bangladesh: the Bangladesh Rural Advancement Committee (BRAC) for in-house, controlled replication of pilot programs, and the Grameen Bank, whose methodology and mode of organization have been replicated by tens of different organizations (including governments) throughout the world. The Grameen Bank even publishes a newsletter essentially devoted to the promotion and discussion of replication experiences in the rest of the world. In a recent analysis of th ree such cases of replication (in Sri Lanka, Malaysia and Malawi), David Hulme concludes that successful replication is possible, if it is done in a "learning process" manner, i.e. not the taking over "as is" of a foreign blueprint of poverty lendi ng, but rather the "use of the Grameen Bank as the initial model for a small scale action research project that can be gradually modified to suit different socio-economic and physical environments." Indeed, replicati on as a path for scaling up is only likely to work if it is done flexibly, without undue copying of structures and procedures from elsewhere -- replication must not mean duplication. The same structures and procedures that encouraged participation in one place can stifle it elsewhere, if copied mindlessly.
nurture: a well staffed and well funded outside agency (can be a public one), using a specific incentive-based methodology, nurtures local initiatives on an increasingly large scale. Nurture is a strategy for promoting scaling up that is often employed by bilateral and multilateral organizations and some of the larger NGOs. The popularity of nurture-type programs with these organizations can be explained by a mix of ideological commitment and need for efficiency. On the one hand these organizations are increasingly attached to community involvement, but on the other hand they are characterized by a top-down mode of organization behavior, large budgets to be spent within fixed time periods, the availability of highly trained personnel, the obligation to work in large areas, aiming to touch many people. For such organizations, nurture-based schemes are the solution, for, on paper, such programs allow them to achieve large-scale community participation while spending their available money and employing their experts.
Nurture-based paths of scaling-up have been employed by a variety of agencies. The Aga Khan Rural Support Programme in Northern Paki stan, for example, typically follows such an approach, and, indeed, in some years managed to involve approximately 300,000 farmers. Communities are asked to organize and decide on the construction of a public infrastructure (school, road, hospital, bridge ) they wish to undertake with AKRSP co-financing. Afterwards, the programme will continue working with the newly created community organizations in other fields of rural development. Multilateral institutions that wish to work in a participatory manner of ten take recourse to nurture-based schemes too. In all these cases, community organization is undertaken as a precondition for external support, whether in the form of credit, grants, technical assistance, etc.; a degree of community participation in the daily management of the activities undertaken is usually allowed too. Finally, some Third World governments, especially in Asia, have sought to use nurture-based strategies for development.
aggregation: a number of distinct organizations combine their resources. This can involve both horizontal processes, through partial or full merger of activities and resources, and vertical ones, through the creation of federative structures, that represent them on high er geographic levels and/or deliver services to themselves.
There are few cases known of real fusion between previously independent non-governmental organizations (unlike in the for-profit sector), and more of instances of collaboration, coordination, joint representation, joint programs, etc. The main reason for that has to be sought in the fierce ideological and financial competition that reigns between NGOs.
integration: a program is integrated "into existing structures and systems and in particular government structures after it has demonstrated its potential." Integration can come about as a result of demand by the NGO, which persuades a government agency to take over a successful program it launche d, or as a result of independent state action. Such integration, although going against the ideological and organizational dynamics of most NGOs, is important in the light of both increasing impact and assuring sustainability, i.e. the capacity of the program to continue after the departure of the NGO. "Integration" is mainly an option for SHPOs, such as BRAC in Bangladesh or the Savings Development Movement in Zimbabwe. It is difficult to achieve, but desired by increasing numbers of NGOs, for it offers the fastest possibility of significant scaling up. To be successful, it requires that the bureaucracy adopts a client-centered, learning-process approach, similar to the one NGOs use. Integration can take place even in situations where the political regime is authoritarian, if relatively high-up civil service individuals are willing to take the risk.
Grassroots organizations can pass through more than one of these processes during their lifetime of quantitative scaling-up. In a case study of farmer irrigation associations in Indonesia, for example, scaling-up began with spread. The original a ssociation, being successful in furthering its members' interests, attracted more members of the village; neighboring villages visited it and joined it. This is the usual process of almost organic growth, in which program success and direct experience are the main motors for scaling-up. After some years, however, a new path of scaling-up came into being as the organization that had started the program (a university center in this case) replicated the same approach used in the original association in other regions of the same country. Finally, at some point the original organization became too burdened with the managerial demands of the rapidly growing farmers' associations and their new needs for training and lobbying. So it transformed itself into a full -time national farmers' union -- a case of organizational scaling-up for the university center and aggregation for the farmers' associations.
5. Functional Scaling-up
Organizations and their programs take on new activ ities because of
push factors: their constituents, and the programs they have been running, demand it. We refer to a "push" situation, when an organization expands to take on new projects as a result of member or client demand: the organization 's members or constituency requires new programs or services. Sometimes such demand emanates from groups who are not (yet) members of the NGO. Hence, for example, development NGOs move into the human rights field, while at the same time human rights NGOs take on women's activities.
pull factors: new sources of funding are available. Organizations expand as a result of a "pull" by forces external to the clients or membership of an organization -- mainly the availability of funds. From the point o f view of the development aid donors and governments, the rationale for pulling a grassroots organization in new activities lies in the fact that the organization already has an administrative structure in place with a concomitant track record. In many ca ses, a structure already in place is half the battle in initiating a new activity. This is particularly true in regions with a dearth of trained and experienced program administrators. There exists a strong temptation on the part of the organization to ta ke everything on that comes along; particularly if the funding will help it sustain itself. Often little or no planning lies behind such an approach other than that of responding to targets of opportunity. This can have devastating effects. Sometimes even the creation itself of NGOs or of SHOs is the result of such pull factors.
In so doing, like private enterprises, grassroots organizations can grow through horizontal integration: add new activities to existing programs, such as adding a housing component to an income-generating program, or, in a more "integrated" manner, linking agriculture to forestry or to livestock or to health and education. The BRAC experien ce demonstrates the importance of horizontal integration, adding institution building and training to income generation activities.
vertical integration: add other activities related to the same chain of ac tivities as the original one, for example when an organization, after years of managing soup kitchens, develops an alternative food purchasing and distribution network ("downstream integration"), or when a program working to improve the techniques of rura l artisans helps them to build retail outlets ("upstream integration"). The most common case of vertical integration is the addition of savings and credit mechanisms to the original mobilizing efforts of grassroots o rganizations, responding to the lack of access to credit by the rural poor in the Third World.
The issue of functional scaling up is relevant for the effectiveness and efficiency of organizations (or their clients or members) to achieve the goals that they have set out for themselves: undertaking the "right" kind of activities or programs is of course as crucial for any grassroots organization as it is for any government or international organization. But the functional scaling up issue is also crucia l for the sustainability of any organization, i.e. its capacity for long-term survival. Indeed, as Gray states it for the case of Paraguayan farmers' organizations: "if small farmer organizations are to be sustainable they must ultimately work to d eliver significant economic improvements for their members and their region, and be able to generate at least a respectable portion of their own operating costs." This means that, whatever the objective and the ideol ogy of the organization, if it is to survive it must undertake activities that improve its members' or clients' economic fate, and that increase its own capacity for self-financing. Those participatory organizations programs that fail to do so are doomed to cease to exist.
6. Political Scaling-up
Political scaling-up consists of the deliberate building a political power base for furthering the goals of an organization or organizations through the politic al process. It is usually based on the assertion that "to stem the mounting tide [of poverty] necessitates attacking the root causes of poverty at the macro-level." The reasons for political scaling up can be more mi nimal and reformist too. Gudrun Lachenman, for example, in her excellent article on peasant movements in Senegal, argues that "the Senegalese peasant movement consciously aspires to influence other spheres [than the grassroots -- PU] mainly in order to be able to work at the grass-roots level without being `disturbed' by the state." Finally, it should be observed that political scaling up is not limited to attempts to influence national governments, but also regional or local government structures; in many countries, given the "soft state" nature of government, the lower levels are more important determinants of local outcomes than the central level.
To a certain extent, NGO s, even if they are not politically scaling up, are always in contact with the state, through registration requirements, import exonerations or taxes, legal harassment or protection, benign or malign neglect, confrontation of friendship with its officials , etc. Moreover, "psychologically," the state is the point of reference for almost all NGOs: whether it is being combated, or its deficiencies are being compensated, or its objectives implemented, "in the end the state acts as the fundamental referent." What political scaling up refers to, then, is to a conscious strategy to interact with the state so as to address state-level variables of local problems. This involves developing strategies to go from the micro to th e macro-level with the objective to bring about governmental policy changes (or the implementation of existing policies).
More and more people believe that political scaling up is the main challenge awaiting NGOs, be they grassroots, membership organiz ations in the South, or intermediary organizations throughout the world. It can take place through a variety of paths, which we will discuss below. The path chosen will depend strongly on the original objectives and strategy of the NGOs, and on the nature of the state they are dealing with. Clearly, if NGOs came into being to provide relief to children, for example, their willingness and capacity to work with the state will be different from if they were born to figh t human rights violations. In Latin America, there exist many NGOs whose mandate is to help the poor to link up to the services of government. Similarly, if the government, and the coalition of interests that back it, is repressive and strongly anti-developmental, the style of collaboration will be different from the case of an open, democratic, development oriented state. Finally, if the N GO works "in territory unoccupied by the government," or if the goals and the programs of the NGO are consistent with national policies and structures, easier and more fruitful interaction is to be expected than if this is not the case.
information and mobilization: mobilization of the members or communities so as to bring them to participate in the body politic. It is designed to create general public awareness and sympathy for the propositions of an organ ization, and hence to put pressure on politicians. Such mobilization can be done through letter writing, through the organization of meetings, conferences and workshops, through the distribution of documents and brochures and through the media. More on th e grassroots level, methods employed by NGOs, such as Participatory Action Research, seek to conscientize poor people of the local, regional and national causes of their poverty, and bring them to organize and act. I n Africa, such forms of community mobilization are rare. In Asia and Latin America, on the other hand, there exist a large number of participatory organizations that are created explicitly for this function: to mobilize people and lobby for change (land r eform, dam construction, indigenous rights, etc.).
Few Northern NGOs like to put resources into the strengthening of the capacities of Southern NGOs and SHOs to scale up politically. Instead, they prefer to fund projects. There are various reasons for that. One is their fear that their position vis-a-vis governments will be endangered if they engage too openly in politics. Southern governments could expel them, while Northern governments could withdraw their advantageous tax-free status. Another is tha t they usually prefer visible concrete actions in the field. It attracts more funds from the general public. Yet, since some years, Northern NGOs increasingly serve as lobbyists and public relations agents for their southern partners, working together wit h them to lobby for policy change. Northern NGOs only have some capacities to pressurize their own governments, and, to a lesser extent, international organizations: they can do this either by mobilizing their members, if they have any, or by public educa tion campaigns. In the future, development education and direct government lobbying will likely become the most important functions of Northern NGOs. Ideally, in so doing, they collaborate with Southern NGOs. This, then, brings us to the next form of political scaling up -- networking.
networking: non-permanent collaboration between various organizations on political issues of joint interest, involving the creation of links, usually between Third World NGOs, but also between them (or coalitions of them) and Northern ones, for the purpose of influencing the body politic. The links are sustained through meetings, newsletters, e-mail, or exchange of personnel. One of the oldest forms of such networking was in the field of human rights, with Amnesty International, or FIAN (located in Germany, and dealing with the right to food) as typical examples. One interesting case outside of the human rights sector is the Philippine Development Forum, described by David Korten describes as an NGO of the "fourth generation." Its originality is that it is composed of organizations of both the U.S. and the Philippines who employ their various resources (financial, managerial, access to policy-makers at all sides) for the achievement of commonly defined objectives. It has a very small administrative body; rather it draws upon the resources of its membership. The usual form of networking, however, is less structured and more of an ad-hoc nature than this. Haitian NGOs pass information about disappearances and torture since the coup d'Etat to their Northern partners who publish them so as to influence public opinion; Oxfam lobbies the World Bank to suspend a planned dam in Brazil; the UUSC lobbies US senators to sponsor legislation for cleanup of former military bases in the Philippines. Some organizations have come into being that concentrate on the facilitation of political networking. Examples incl ude IRED and FIDA (both located in Switzerland). In the US, among others, InterAction and World Resources Institute (through its publication The NGO Networker) are active in this field.
Finally, this practice of networking for advocacy has beco me standard practice at the level of international conferences: at the UNCED in Rio, for example, there were 1,418 officially registered NGOs present, and hundreds more had participated in various regional preparator y meetings. This was partly due to the work of international networking NGOs such as, among others, the Asian NGO Coalition for Agrarian Reform and Rural Development (ANGOC), based in Manila, with 20 members, most of which are national NGO coalitions them selves, or the Environment Liaison Center International, based in Nairobi, with 340 NGO members. This whole process of networking has become facilitated by the spread of information and communications technology, esp ecially e-mail and its connected services.
aggregation: creation of federative structures designed to influence policy making, based on the principle that "l'union fait la force." Grassroots organizations create regional structures; regional organizations create national structures, and even, in some regions, national organizations create international structures. Examples include Euronaid, a coalition of European NGO s, ASOCORE (Association of Peasants in Central America) which brings together seven national farmers' organizations for the purpose of influencing governments' strategic planning as it impacts upon small farmers or CONCERTATION which is a regional coaliti on representing more than one hundred intermediary organizations from seven countries; INTERACTION in the US or EURONAID in Europe, ANGOC in Asia, and so on. The most advanced case must be the Philippines, where the Caucus of Development NGO Networks (COD E-NG), created in 1990, is composed of 10 national networks, each composed of regional unions and hundreds of small NGOs. In total, 1300 NGOs belong to CODE. Among its main objectives are advocacy and influence of po litics.
direct entry into politics: whereby (a) grassroots organization(s) either creates a political party or joins an existing one. This is the last and least traveled road to political scaling-up. It can be done on an individual basis, with the leader of an organization, based on his personal prestige, entering the political arena with a program along the lines of his work with the organization. Hence Ari Aryaratne of Sarvodaya Shramadana movement in Sri Lanka, or Abdoulaye Dioup of the "Ami cale du Wallo" in Senegal are both cases of grassroots leaders who entered the national political arena. It can also be done on an organizational level, with a grassroots organization, usually in coalition with other organizations, creating a political pa rty. Such processes have been documented in various Latin American countries, where NGOs and grassroots organization coalitions have created political parties that have successfully competed in local and national elections.[ 56]
Similar to the latter is another variant, whereby organizations put up candidates from their own ranks, possibly within existing political parties, usually at the local level, but, in the case of important and large organizations with a region al or national impact, even on the higher levels. Such is the case of AJAC-COLUFIFA in the Casamance region of Senegal, which systematically puts up candidates in local elections at the level of the `conseil rural' -- not with any intent of gaining politi cal control in the region, but rather to assure sympathetic (or at least non-obstructive) attitude from the powers-that-be and gaining a degree of participation in the decision making process.
A slightly less ambitious and visible version of the forme r is the one where grassroots or NGO leaders enter into high positions in the public administration. In Chile, for example, after the restoration of democracy under Allwyn, many NGO and grassroots leaders entered in high civil service positions, some even becoming ministers. In many countries, governments have extended invitations to successful leaders of community-based organizations to enter into the civil service. This is linked to "integration" as a path to quant itative scaling up.
One final remark: to a certain extent, it can be said that grassroots organizations, as well as intermediary NGOs, are, by their very existence outside the state and its top-down development interventions, political statements. The French political scientist Bayart talks in this respect of "popular modes of political action" at the local level: the act of organizing, of conceiving another development, is by definition political, and as such often considered a threat by the powers-th at-be. If successful, development NGOs are agents of redistribution and empowerment, which cannot fail to have political repercussions, on the local, regional and national levels.
7. Organizational Scaling-up
Organizational scaling-up addresses the issues of sustainability of the organization, assuring that when an organization grows, it is able to sustain its programs without complete dependence upon non-renewable resources (fi nancial, technical and physical). This requires the assurance of strong management procedures (financial, organizational, personnel and programmatic) that allow organizations to handle unanticipated consequences that are waiting to confound work (i ncluding, for example, such events as the death of the founding father) and the creation of structures that allow for flexibility, participation and accountability on a large scale. In short, the objective of organizational scaling-up is financial and managerial autonomy. It is an objective not measured in terms of full achievement but, rather, by degrees of approximation. It is the corollary of, or requirement for, all other types of scaling up. As Hodson sta tes it: "there is no escaping the organizational reality that any attempt to scale up impact, whatever the strategy, has serious and difficult organizational implications." Failing to realize this can lead to serious organizational problems if not destruction. Generally, we can distinguish several objectives/paths of organizational scaling-up.
One of the first objectives many organizations set is to diversify and stabilize their funding sources, to go beyond the major donor who supported them at the beginning. This is often as much a matter of quantitative scaling-up as it is one of organizational scaling-up. As important as the number of funding sources is their type: What are the strings attached? How secure is the funding? The best objective one can hope for in funding grassroots organizations is building a portfolio of long-term, stable, non-discretionary program funds (endowments, for example, or public entitlemen ts). Standard donor practice, however, is rather the opposite, favoring short-term, discrete, project-related infusions of cash. More recently, a potentially large source of funding has become the government: often u nder pressure from donor ideologies in favor of NGO action, and sometimes resulting from a realization of the comparative advantages of NGOs, Third World governments are increasingly implementing and funding programs, usually of a service delivery type, t hrough local NGOs. This represents an important opportunity for NGOs: clearly, the resources that can become available from governments are much larger and more stable than those most NGOs can mobilize by themselves.
Another objective or target many organizations eventually adopt is to increase their degree of self-financing. This can among other ways be done by the creation of lucrative activities such as joint ventures, cooperative or other enterprises, f ee-for-service activities (including sale and rental of materials), consulting to other actors such as the government and foreign aid agencies, or by subcontracting to the state or the foreign aid system. As quite some NGO leaders are recognized as expert s in their fields (community health care, rural extension, micro-enterprise, literacy campaigns, etc.), they are increasingly asked to act as consultants or subcontractors in publicly and aid funded projects.
On a different level, grassroots organizat ions that scale up have to promote skills development. This involves both managerial skills (on which most foreign NGO's assistance focuses) and the much more neglected technical skills. NGOs need to strengthen their capacity to manage large organi zations. They have to adopt policies of hiring, training, motivating, paying and promoting personnel, of managing and controlling budgets, of decentralization and innovating management, and of avoiding or overcoming dependency on one leader. This seems ev ident: what started as the initiative of one dynamic leader in one village with one activity cannot be managed in the same way once it has become an organization touching, say, 50,000 persons in 120 villages, with tens of different activities. Yet, very o ften, few changes in management procedures are made. NGOs also need to improve their technical skills: as they deal with more people and their varied needs, as they take on more activities, as their public visibility increases and they interact with gover nment, as they take on consulting for foreign agencies, they need to have at their disposal a variety of technically competent people. This is sometimes difficult, for it necessitates going beyond ideological attachment to professionalism -- all the while keeping participation alive.
Moreover, it is generally recognized that NGOs, in order to expand their impact (and even just so as to simply survive), must develop procedures and structures allowing for organizational learning. This involves procedures of monitoring and evaluation -- serious, critical evaluation, and not the type used to fill fundraising brochures -- and feeding the lessons learnt back into action. Thi s implies the need for quality documentation and archival systems, computerization, standardized procedures of programming, monitoring and evaluation, self-criticism and openness, etc. These are not easy to achieve, and indeed, even Northern NGOs, staffed by well-trained professionals, with large budgets and in easy contact with research institutions, still score badly on this point.
Fifth, organizations that scale up have to create institutional variety,b oth internally and externally. Internally, they have to diversify their structures of self-help, going beyond the original mode of organization and into other realms. This will make the self-help process more resilient to external threats and more capable of responding to new challenges. In the case of some Senegalese farmers' associations, for example, the original village structures have been complemented by federative structures up to the national level, commercialization cooperatives, rural banks and, most recently, a variety of rural enterprises. Externally, they have to build operational links with a large variety of other actors: local, regional and national government, other NGOs of various types, training and research institutions, private enterp rises, and banks. Already in Uphoff and Esman's famous study of 1974, a comparison of 18 Asian NGOs concluded that horizontal and vertical linkages between institutions were the key to the strength of these organizations.[67 ] This is not always easy, given the ideological differences between NGOs (and between them and the private for-profit and public sector) and the fact that all these actors are often in competition for the same scarce resources.
Finally, the greate st organizational challenge of all for NGOs, whether working at the grassroots or on the intermediary level, whether scaled up quantitatively, functionally or politically, is to maintain participation and accountability. Indeed, all types of scalin g up (including the other aims of organizational scaling up), in various ways, have a tendency to promote institutionalization, formalization, professionalism, bureaucratization, distance between leaders and rank-and-file members, increased outside influe nce. These tendencies are quite unavoidable and necessary, if organizations want to move beyond the highly local, spontaneous, level and impact on large levels -- indeed, some of these processes, such as institutiona lization and professionalism, we have explicitly argued for in the previous paragraphs. Yet, all these processes strongly risk decreasing participation and accountability if no explicit attention is devoted to maintaining and reinventing forms, procedures and fora for beneficiary participation and control. There are no simple or universal rules as to how to do that.
In the case of intermediary organizations, the issue of organizational scaling-up takes on a parti cular twist. Indeed, for SHPOs, whether from the South or from the North, the aim for them is ironically not so much to scale up but to "scale down," i.e. to attempt to become superfluous, or redundant, or, in the words of Fowler, to "build the management and institutional capacity of intended beneficiaries" and not of the SHPOs. Foremost, SHPOs shall strengthen the capacity of their clients (SHOs) to diversify funding sources and increase self-financing, for example, and not primarily their own. To meet this challenge, at least two conditions are needed. One is a coherent and shared vision or ideology of what is to be achieved in terms of grassroots development and empowerment. The second is what Fowler calls "management for withdrawal", i.e. the development of structures and practices that are geared not at perpetuating or enhancing the hold of SHPOs over the communities, but rather a program of phased disengagement. Of course, in the short run , also SHPOs need to diversify funding sources, increase self-financing, develop skills and management techniques, etc., i.e. scale up organizationally.
This paper in no way pretends to present the definitive word on scaling-up and participation. Rather, it provides some basic definitions of what is meant by scaling-up, forms it may take and strategies for making it happen. There are many NGOs with serious community development agendas who have experienced considerable scaling-up. How well they have faired in scaling-up while maintaining a strong commitment to participation will be a primary interest of future research. Alongside, the role of government is clearly c rucial: it comes back, in different manners, in every type of scaling up. Here too, more research is needed. During the next two years, the World Hunger Program wants to sponsor a series of case studies that, using the terminology developed in this paper, looks at various organizations related to food and hunger and attempts to learn from them. We invite anybody who wants to share in this research venture to do so.
 Edards & Hulme, 1992: 14.
2 For important exceptions, see Tendler, 1982, Esman & Uphoff, 1984, Uphoff, 1992, Jacob & Delville Lavigny, 1994. On the history of scholarship of the "third sector" in the Western world, see Seibel and Anheier, 1990.
3 Tandon, 1991: 2, Bayart, 1989; Vergara, 1989.
4 Brown, 1991: 809.
5 Anheier, 1990: 363; Scheider, 1985; Korten, 1990;
6 Bowden, 1990: 151.
7 Morgan, 1990: 7.
8 Bebbington & Farrington, 1993: 202.
 Excellently discussed in Tendler, 1982; Smillie, 1993: 15-16; Farrington & Lewis, 1993: 55; UNDP, 1993; Bebbington & Farington, 1993: 208, 212; Martin, 1986: 226 ff.; Johnson & Johnson, 1990: 114; Friedmann, 1992 in his important construction of a theory of alternative development, criticizes NGOs, and authors such as Korten, for their anti-state vision (f. ex. p. 7, 141, 158, 164).
10 We were inspired by Oakley, 1991: chapters 1 and 6; Tendler, 1982: introduction
11 Hall, 1986: 99, 95. See a good study by Martin, 1986, on primary health care and participation.
12 Gould, 1985; Hall, 1986.
13 As done by Anheier, 1990: 367.
14 Salamon & Anheier, 1991, 1992; also Clark, 1994: 4.
 A recent publication by PACT, 1989 complicates things even further by using the term NGOs for grassroots, membership organizations, and VROs (Voluntary Resource Organizations) for SHPOs. The book contains interesting case studies of the relations between these types of NGOs in Bangladesh, Indonesia, Nepal, Sri Lanka and Thailand.
16 For the definition of "local" see the excellent article by Uphof, 1993.
17 Clark, 1991
18 Fisher, 1993
19 Korten, 1990
20 Lecomte, 1991
21 Praderwand, 1993: ch. 1; Descendre, s.d.: part 1; see too Lachenmann, 1993.
22 This approach seems to be popular in Asia. It is followed by BRAC, for example, but was already adopted 70 years ago by James Yen's Rural reconstruction movement: Mayfield, 1985: 15
23 Part of a learning process approach as described by Korten & Klauss, 1984. For the case of BRAC, see Howes & Sattar, 1992: 108.
24 Howes & Sattar, 1992; Abed, 1986: 500;
25 Hulme, 1993
26 For an interesting example, see Montgomery, 1988: 82, 87, 96 and the experience of the Joint Commission for Rural Reconstruction in Taiwan.
27 Montgomery, 1988: chapter 5; Korten & Syi, 1988.
28 Quote from Mackie, 1992: 70; see too Bebbington & Farrington, 1993: 213; Clark, 1994: 2
29 Lovell, 1992: 187; Howes & Sattar, 1992: 109; Brown, 1991: 817.
30 Allison &Kak, 1992: 165-6; Montgomery, 1988, chapter 5; Morgan, 1990; Tendler, 1982: 9 & 91; PRIA, Building ....: 9 and Friedmann, 1992: 158 for a more theoretical perspective.
31 Many books deal with this issue. Among many others, see Garcia-Zamor, 1985, Thomas, 1985, Khan, 1985 and Chowdhury, 1989: part 1, for the case of Bangladesh, Midgley et. al., 1986, David Korten, 1980 for discussions of learning-process, and his wife Frances Korten, with Joseph Siy, 1988, for a famous example of successful learning approach in the Philippine National Irrigation Authority.
32 Downs & Solimari, 1989: 209. Indeed, history documents many cases -- for example the Grameen Bank in Bangladesh -- of scaling up of NGO programs being the result of one or a few civil servants taking the risk of supporting or taking over small projects.
33 Kiriwandeniya, 1992: 115-117.
34 In the Philippines, so-called GRINGOs (government-run,-inspired or initiated NGOs) are the result of massive inflow of foreign funds for NGOs after the fall of Marcos: Constantino-David, 1992: 138.
35 Howes & Sattar, 1992: 108; Abed, 1986: 501.
36 Abed, 1986: 501
37 Bray, 1991: 132.
38 Clark, cited in Smillie, 1993: 32; Brown, 1991: 810
39 Lachenmann, 1993: 76.
40 Steiner, 1991: 33; Breslin, 1991: 7; more generally, Hyden, 1983.
41 Calderon, Piscitelli & Reyna, 1992: 25.
42 PRIA, Holding....: 29; Fals Borda, 1992.
43 PRIA, NGO-Government..... for an in-depth, concrete discussion of these items; Smillie, 1993: 19. Also Steiner, 1991: 74; Anheier, 1990: 47-51.
44 As mentioned by Tendler, 1982: 101; to be observed in the health and urban renewal sectors.
45 Farrington & Lewis, 1993: 10-11, 33-34.
46 Tendler, 1982: 94; Mayfield, 1985; PRIA, Multiparty cooperation ....
47 Fals Borda, 1988: 45; Fals Borda & Rahman, 1991.
48 Smillie, 1993; and OECD, 1993: passim; Dolan, 1992; Benjamin & Freedman, 1989.
49 Clark, 1992: 195; Berg, 1987: 23.
50 Hall, 1992; at the Brazilian side, a federation of 13 rural trade unions lobbied the state and national governments for the same -- hence, advocacy "from below" and "from above" (here: abroad) converged. For a case study of the same process in Peru, see Dawson, 1992.
51 Kakabadse & Burns, 1994: 3. Note that some of these NGOs are bodies representing the private industry: business councils, chambers of commerce, trading organizations, etc. Their interests and ideology are often conflicting with those of the development NGOs. For a discussion of the same process in the International Conference on Nutrition, see Messer, 1993.
52 Starke, 1990: 77 ff.
53 About the potential and limits of modern technology in development, see Annis, 1992; Hills, 1992; Siefert, Gerber & Fisher, 1989;
54 PRIA, Holding...: 22.
55 Constantino-David, 1992.
56 Morgan, 1990.
57 Loveman, 1993; in the rich countries, a number of leaders of environment NGOs ended up in top government positions: Starke, 1990: 66.
58 Bayart, 1985.
59 Fowler, 1992, speaks about self-development, and sees three areas for organizational scaling up: identity, relationships and performance.
60 Hodson, 1992: 129; see the discussion by the founder of BRAC of the "enabling requirements for successful scaling up": all are organizational in nature: Abed, 1986: 503.
61 For the case of British NGOs, see Billis & MacKeith, 1992 and Hodson, 1992.
62 Anheier, 1990: 364 documents that in three African countries, only approx. 25% of the NGOs enjoyed stable funding.
63 This is standard practice in Europe, where most schools and hospitals, for example, are funded by the state and operated by NGOs: Van der Heijden, 1988.
64 Allison & Kak, 1992: 164.
65 Indeed, maintaining participation while politically and quantitatively scaling up is one of the central challenges facing NGOs: Covey, 1992. For an excellent analysis of these processes in Paraguay, see Bray, 1991.
66 See the excellent study by the Smillie & Helmich, 1993. Organizational learning is all the more important for SHPOs, which are in a position to train grassroots organizations. For an interesting evolution of learning, see Mayfield, 1985: p 30 ff. and 178 ff. on the International Institute of Rural Reconstruction in the Philippines. BRAC is a prime example of a learning organization: Abed, 1986.
67 Esman & Uphoff, 1974: XI-XII; Brown, 1991: passim.
68 Friedmann, 1992: 142 ff.
69 See Leat, 1990, for a discussion of this issue in the West.
70 Fowler, 1988; Bray, 1991.
71 For the case of BRAC, see Lovell, 1992: conclusion.
â€˜Think Large and Act Small: Toward A New Paradigm for Development NGOs,' World Development, 28, 8 (Sept. 2000): 1409-1419. (with Pankaj Jain and L. David Brown)
â€˜Scaling Up, Scaling Down: the Role of NGOs in Overcoming Hunger' in Marchione, T. (ed.) Scaling Up, Scaling Down: Capacities for Overcoming Malnutrition in Developing Countries, New York, Gordon & Breach, 1999.
â€˜Paths to Scaling Up: Alternative Strategies for Local Nongovernmental Organizations,' Human Organization, 55, 3 (Fall 1996): 344-354 (with David Miller)
â€˜Scaling Up the Grassroots and Scaling Down the Summit: the United Nations and Local NGOs.' Third World Quarterly, 16, 3 (Sep. 1995).
â€˜Fighting Hunger at the Grassroots: Paths to Scaling Up,' World Development, (June 1995) Vol. 23, No. 6. (chosen as UNICEF Nutrition Paper of the Month, May 1995)
More Information on Growing Importance of NGOs
More Information on Introduction to NGOs