Mindfully.org This Domain & Website Are For Sale. Serious Inquiries Only. Contact Here

Home | Air | Energy | Farm | Food | Genetic Engineering | Health | Industry | Nuclear | Pesticides | Plastic
Political | Sustainability | Technology | Water

Capacity Development for the Environment: 
A View for the South, A View for the North 

Annual Review of Energy and the Environment Volume 25, 2000

Ambuj D. Sagar

Science, Technology, and Public Policy Program, Belfer Center for Science and International Affairs, John F. Kennedy School of Government, Harvard University, Cambridge, Massachusetts 02138; e-mail: ambuj_sagar@harvard.edu

KEY WORDS: Developing countries, environmental management, policy research, policy innovation, technological capabilities


The notion of capacity development (CD) has been receiving increasing attention as a way to assist the South in its environmental management. Consequently, there has been an exploration of various facets of the capacity issue in the literature and an incorporation of CD in environmental programs of donor agencies. Yet, many of these discussions have remained rather broad, and efforts to develop environmental capacity have shown only limited success. Based on an examination of the capacity needs for environmental management in agriculture and industry, and for dealing with climate change, this review suggests that strengthening domestic capabilities for policy research and innovation as well as for managing technological change may be particularly critical to allow for adaptation of policies and technologies for local conditions and needs. Examination of innovative local experiments on environmental management in developing countries can also provide useful lessons on how to develop and utilize capacity that works under the constrained conditions often found in developing countries. Furthermore, it is important to stress that improving the environment in developing countries also requires capacity in the North to examine and reorient Northern policies that impact the environment, as well as capacity for the environment, in the poorer parts of the world. Ultimately, though, the development of sustainable and appropriate capacity for the environment will require not merely donor-driven programs but a systematic effort driven by Southern governments and organizations.


In recent decades, there has been an increase in concern about environmental issues in the North and the South. Several factors have contributed to this: an intensification of economic activities without much consideration for the environment, leading to substantial degradation in terrestrial, aquatic, and atmospheric systems; a greater understanding of the potential human and economic consequences of environmental degradation and of the need for the incorporation of natural resources into national accounts; the emergence of global environmental issues that have catalyzed international policy discussions, resulting in the need to undertake specific commitments; the rise and empowerment of nongovernmental organizations (NGOs) and citizen groups in the North and South, as well as of transnational NGOs (such as Greenpeace) whose global presence serves to link activists from the North and South; the rise of ethical concerns about the environment, biodiversity, and the rights of indigenous peoples; and an increased global movement of ideas and understanding about environmental issues (beginning with Earth Day in 1970) made possible through advanced communication technologies.

In some sense, the global upsurge in policy concern can also be considered to follow from the 1972 Stockholm Conference on the Human Environment that put the environmental issue fairly and squarely on the global and national policy agendas and resulted in the establishment of the United Nations Environment Programme (UNEP) in 1974. Since then, most countries have established government agencies with a sole or primary focus on the environment. At the same time, there has been a realization that environment and development cannot be unlinked in the South. Development (or economic growth, to use the still most popular proxy for development), although seen as the solution to the environmental problem, is itself a cause of environmental degradation, and environmental degradation can impede development. This connection was reflected in influential policy documents such as Our Common Future (1) and the follow-up meeting to Stockholm, the United Nations Conference on Environment and Development (UNCED). To a large extent, concern for the environment has begun to be integrated into mainstream thinking on development (see, e.g. 2), at least on paper, even if the extent of its translation to actual practice is still contested.

By now, the linkages between the natural environment and the well-being of humans and other organisms have been well established, as has the dependence of various forms of economic activity on the environment. Significant effort has been devoted to examining environmental issues that are of concern to the South, and to finding ways to tackle these issues. Among the various (often complementary) perspectives, there has been a substantial and increasing focus on the concept of capacity development for the environment (CDE) as a way to help the South to understand and meet the challenges facing it. Although this concept is inherently attractive, the discussions on developing capacity for the environment have either remained rather broad and conceptual, analogous to the more general literature on capacity development (CD), or are scattered within different sectorial domains.

In examining the CDE literature, this review serves multiple purposes: (a) juxtaposing specific capacity needs that emerge from examination of selected environmental issues with the broader CDE discussions; (b) drawing out themes common to different sectors, as well as the lacunae in current capacity discussions/ approaches; and (c) highlighting areas that seem to be particularly important for future attention in both the research and applied domains in the context of CDE. Through this approach, this review aims to shed light on the why, what, where, and how of CDE.

At the outset, it should also be noted that the CDE literature has emanated mainly from within donor or other international agencies, from research activities funded/solicited by them, or from individuals and groups seeking to target/respond to donor thinking in this regard. Similarly, most of the current attempts seeking to develop environmental capacity have emerged from donor agencies. Not surprisingly, therefore, this review focuses substantially on donor agencies and their role in CDE. However, it also goes beyond this by discussing other, broader, aspects of capacity building that are relevant for the South.


Capacity Issues: Back to the Core

The capacity-building concept itself is not new-the idea (in some shape or form) has been a part of the development literature since the 1950s. However, it has taken many twists and turns over the years,1 its evolution reflecting not only the changing understanding of development processes but also the dominant political philosophies of various times.

The concern over capacity has been reinvigorated recently in part because of a (re)recognition of the importance of domestic capacity for tackling national and global policy issues facing developing countries.2 Like the swing of a pendulum that so characterizes development thinking, the focus in the late 1970s and most of the 1980s on correcting the problems of "too much state" gave way, by the late 1980s, to the realization that government does play a critical role in the development process and that "too little state" can also lead to problems (5). To a large extent, this shift resulted from the lack of success of the economic reform programs of many developing countries in the 1980s (termed the lost decade).

At the same time, dissatisfaction with the high costs and lack of effectiveness of technical assistance programs, especially in Africa, led to substantial and justified internal and external criticism. This forced donor agencies to examine and rethink the issues surrounding technical cooperation. Reflection within these agencies, catalyzed by studies such as a major review of technical cooperation efforts by the United Nations Development Programme (UNDP) (6) and a detailed examination of the economic development programs in Africa by the World Bank (7), resulted in acknowledgement of the essential role of national capacities in successful development.3

With the increasing recognition of the relevance of national capacity, the United Nations General Assembly Resolution 44/211 emphasized that its agencies should focus on capacity building in developing countries and directed them to improve their operational activities, including the coordination of various efforts (10). Slowly, the programmatic efforts of major multilateral and bilateral donor agencies have begun to incorporate capacity building into their technical assistance/ technical cooperation efforts.4

The UNDP, the central funding mechanism for the United Nations system of technical cooperation, also moved to situate capacity building as a central and fundamental element of its activities during the Fifth Program Cycle. At the same time, it focused on understanding in a systematic manner the elements of capacity building and the processes through which these can be built and strengthened (5). According to the UNDP, CD is the process by which individuals, organizations, institutions, and societies develop the ability (individually and collectively) to perform functions, solve problems, and set and achieve objectives. The UNDP's definition and approach to capacity building is based on three cornerstones. It is a continuing learning and changing process; it emphasizes better use and empowerment of individuals and organizations; and it requires that systematic approaches be considered in devising CD strategies and programs (17). In addition, its CD framework considers the four dimensions of individuals, entities, interrelationships between entities, and the enabling environment.

Agenda 21, the outcome of UNCED, similarly takes a multi-faceted approach to the capacity issue: "[S]pecifically, capacity-building encompasses the country's human, scientific, technological, organizational, institutional and resource capabilities. A fundamental goal of capacity-building is to enhance the ability to evaluate and address the crucial questions related to policy choices and modes of implementation among development options, based on an understanding of environmental potentials and limits and of needs as perceived by the people of the country concerned" (18).

The World Bank has also shifted the emphasis of its technical assistance (TA) towards institutional development-by the mid 1990s, TA for institutional development/capacity building was about half of all TA lending by the Bank (19). The Bank also recognizes that different types of capacity may be relevant for organizational effectiveness, including the capability to set goals, evaluate options, adapt to changing circumstances, and build linkages with other organizations as needed (20).

The Development Assistance Committee (DAC) of the Organization for Economic Cooperation and Development (OECD) adopted a new set of guidelines for technical cooperation programs in 1991 based on the recognition that the planning, design, and delivery of these programs can significantly influence local capacities (21). These principles include setting the following strategic objectives of technical cooperation: (a) long-term capacity building rather than short-term performance improvements, (b) stressing the importance of long-term institution building (especially in the area of policy analysis and development management), (c) advancing greater use of local expertise and existing structures, and (d ) encouraging broadened participation by the developing countries, including intended beneficiaries, in all phases of technical cooperation projects. The latter is particularly important because it is felt that participatory development strengthens and empowers citizens, groups, and organizations, in addition to improving the effectiveness and efficiency of development programs and eventually linking to good governance (22).

There has been a parallel evolution in the thinking on capacity building in the arena of science and technology (S&T) for development. Given the importance of S&T in contributing to economic growth through technological change, and in improving the performance on other indicators of development, such as reduced mortality and morbidity, most developing countries have traditionally focused on national S&T capabilities as part of their overall development strategies. Their efforts have included setting up universities and research institutions, establishing national S&T policy organizations, and training personnel. Although these have had mixed results, the theme of capacity building has permeated the organization of such endeavors, especially through the efforts of the United Nations. The International Development Strategy of the General Assembly for the Second United Nations Development Decade made the task of expanding overall capability a central part of the S&T development; the related Report of the World Plan of Action for Science and Technology also emphasized the need to strengthen indigenous scientific capability in developing countries (23). Since then, there have been numerous international discussions on S&T for development, of which the 1979 United Nations Conference on Science and Technology for Development is particularly notable (24).

More recently, a study prepared at United Nations Educational, Scientific and Cultural Organization's (UNESCO's) request by the International Council for Science Policy Studies reemphasized the role of S&T in national economic and social development and suggested that harnessing S&T for this purpose includes not only a creation of a national S&T base, but the capacity to guide, manage, and exploit these S&T resources to tackle the problems and issues specific to the country. In keeping with much of the literature on capacity building, this study also recognizes the importance of the institutional context in which S&T activities take place, and that the effectiveness of these activities depends very much on the appropriate designs of these economic and social institutional arrangements (25).

Thus, the CD framework in donor as well as international policy agencies, while still evolving,5 has begun to converge on a nuanced perspective on capacity building-what capacity is, where it resides, and how it can be built up. Most importantly, there is a recognition that it is important to consider capacity at different levels ranging from the individual to the institutional, that a multiplicity of state and nonstate organizations contribute to national capacity, and that it is important to empower and include all relevant actors in decision-making processes, i.e. it is as important to utilize capacity as it is build it. At the same time, the relationship between technical assistance/cooperation and CD has also been reexamined, with the realization that specific efforts are required for TA activities to build capacity (e.g. through strengthening of institutions or human resources); activities targeted purely towards assisting countries in the completion of specific tasks or moving towards short-term goals may not only not build local capacity, but may actually be counter-productive in that regard. In fact, attention not just to the outcome but also to the process has led "capacity development" to slowly supersede "capacity building" as the term of choice, with the understanding that the former is a process- as well as an outcome-based app roach.

Capacity Development for the Environment

The continuing, and often intensifying, need to move towards sustainable production systems-industrial, agricultural, energy, etc-to reduce stresses on the environment, as well as the need to mitigate and/or adapt to the social, economic, and ecological impacts associated with environmental degradation, poses a significant challenge to developing countries. At the same time, these countries are often hampered in their attempts to tackle the environment issue owing to constraints in technical, organizational, and institutional capacities, and at a more fundamental level, paucity in financial and human resources. CDE aims to focus targeted interventions on these various barriers such that the ability of developing countries to protect their environment can be enhanced even at their present state of development.

The current emphasis on the issue of CDE probably owes its genesis, at least in part, to the ferment in the development community about the relationship between capacity building and national development. At the same time, there has been increasing recognition that issues of environmental management cannot be unlinked from the broader development discourse (1, 28, 29), and that all of this lies within the general social, cultural, political, and institutional contexts in which policies are designed and implemented.

Although some discussions of developing capacity for tackling environmental problems preceded UNCED, the spotlight on the environment at Rio provided the impetus for a thorough and sustained effort to examine the capacity issue. Perhaps the most systematic and analytic exploration of the CDE issue has come out of the OECD/DAC through its Working Party of the Development Assistance and the Environment. Whereas the broad mandate of this group was to encourage DAC members to pay more attention to environmental issues in their program planning and implementation and to improve donor coordination in this regard, CDE was considered a key issue from the outset for two reasons: (a) the importance of institutional and organizational capacity issues made apparent by the uneven performance of development cooperation efforts in the late 1980s and the lack of sustainability of positive outcomes and (b) concern over the nature of traditional technical cooperation efforts that were not very successful and were sometimes even counter-productive to the development of endogenous capacity to design and implement programs (4).

Thus, the Working Party established a Task Force on CDE whose mandate was to develop common approaches, planning and analytical tools for donor programs of technical cooperation for capacity development in the environment (31). The program initiated by this task force has led to a comprehensive analysis of the CDE issue, leading to the establishment of a framework for donor efforts to develop environmental capacity (4) and examination of how CDE principles could be translated into practice (32), as well as the compilation of CDE case studies (33) and an assessment of the DAC members' capacities to support CDE activities (34). The results of this OECD/DAC program have substantially shaped the discussion of CDE.

At the same time, a few other organizations (bilateral actors such as GTZ and multilateral agencies such as the World Bank and the UNDP) also focused on the CDE issue in the developing-country context. The Capacity 21 program of the UNDP launched in 1992 at the UNCED, for example, aims to assist developing countries to build their capacity to integrate the principles of Agenda 21 into national planning and development.

The CDE issue is complex not just because the notion and practice of capacity development itself are amenable to many interpretations, but because the environment is an inherently "messy" public-policy problem owing to a unique combination of attributes: It has multi-dimensional and cross-sectoral aspects; it is difficult to categorize and involve actors whose activities span many legal and geographical jurisdictions; it often has high levels of uncertainty and complexity, and cause and effect are difficult to determine; long time frames are often needed for resolution; and impacts are often external and remote from the source of the problem, but pervasive through society (4). Tackling such problems, therefore, requires approaches that span disciplines and sectors, and embody integrative systems thinking.

Following the OECD Pressure-State-Response framework, the capacity to manage the environment within any country can be thought of as requiring the capability to evaluate the condition of the environment and natural resources (air, water, land, forest cover, biodiversity, etc), the capability to assess the human activities that are driving any changes in the state of the environment (energy production and use, transport, agriculture, industrial production, etc), and the capability to design and implement appropriate societal responses (including those that may be needed to mitigate the environmental pressures or the impacts ensuing from them). CDE can be broadly defined "as the ability of individuals, groups, organizations, and institutions in a society to devise and implement solutions to environmental issues as part of a wider effort to achieve sustainable development" (Ref. 4, p. 12), which is an evolutionary version of the definition of capacity building in Agenda 21 and the UNDP's Capacity 21 program.

This in turn requires a whole range of capacities: to collect and analyze suitable environmental information, to identify particular problem areas, to understand the relationship between a particular environmental change and the human activities that drive it, and to formulate appropriate environmental strategies and programs and implement them, and then to monitor the effectiveness of these programs. In the author's opinion, of particular relevance to the development of environmental policies and specific management capabilities is the role of policy research in determining the various options available, analyzing the opportunities and costs of each in the national context, and focusing on the most suitable framework for implementation. At the same time, the broader macroeconomic context as well as the institutional framework (legal, regulatory, property rights, etc) is also recognized as influencing both the human activities that drive environmental problems and the effectiveness of any societal responses. The ability to carry out scientific and socioeconomic research and use it to provide analytical grounding for developing environmental strategies and implementing them through specific programs cannot be underrated.

Consequently, the general framework proposed by the OECD/DAC is based on five components (where the interrelationships among the components is as important as the components themselves) (4):

1. Functions: strategic planning, effective decision-making, delegating responsibility and empowering individuals and groups, networking and collaborating, mediating and consensus building required to accomplish the task involved in environmental management

2. Actors: individuals, formal organizations, informal institutions, networks of organizations that carry out the tasks necessary for environmental management

3. Normative context: vision, values, organizational strategies, policies that shape CDE

4. Societal context: political, social, cultural, demographic, and geographic forces and conditions-at the global, regional, national, and local levels-to which any CDE program must respond

5. Resources: human, informational, financial, technological, and ecological inputs required to develop capacity

In a parallel to the generic CD literature, the CDE framework suggests not only strengthening the performance of individual organizations, but given the multiplicity of groups involved, improving the capability to promote and manage organizational pluralism. The development of networks, with a particular emphasis on coordination, could improve linkages and alliances among organizations that have complementary skills. Enhancing the participation of informal (e.g. community-based) organizations is also highlighted to supplement the traditional focus on formal organizations. The participation of the former is particularly important for diffusing environmental protection activities down to the local level, for communicating local needs and concerns to policy-makers, and for utilizing traditional practices and knowledge embodied in such organizations to complement "modern" knowledge (4). Furthermore, particular attention must also be paid both to the societal and cultural context (which can vary substantially among and within nations) in which CDE activities are being carried out and to the inherently political nature of many environmental issues (4).

With all this in mind, the OECD/DAC has also developed a set of principles (32), based on a set of case studies (33), to guide the practice of CDE so as to (a) promote sound environmental considerations and criteria in the development process; (b) strengthen institutional pluralism in civil society; (c) integrate environment and development concerns; (d ) be a multi-faceted process that can take into account local norms and values, different organizations and stakeholders involved, as well as local resources; (e) be process-oriented rather than product-oriented, with a particular focus on interaction and learning, consultation with and participation of local groups, and being adaptive; ( f ) take a systemic approach; (g) belong to, and be driven by, the community in which it is based; (h) take gender issues into account in all their aspects; (i) develop appropriate approaches to include all disadvantaged groups in society; and ( j ) choose from a variety of management techniques, analytical tools, incentives, and organizational structures so as to best achieve a given policy objective.

Some potential fields of activity for CDE include (4, 35, 36) (a) assisting in environmental policy formulation and implementation; (b) supporting environmental information and monitoring systems; (c) developing an enabling environment for CD (such as improving environmental legislation or designing policy instruments to provide incentives for environmental management); (d ) improving the ability of individuals, organizations and groups to carry out the kinds of functions listed above; (e) developing and/or creating space for local organizations; ( f ) augmenting various resources; (g) facilitating dialogue between governments and groups; (h) enhancing public awareness and education through improved communication and outreach; and (i) promoting the application of environmentally sound technologies.

This discussion of various aspects of the CDE issue highlights the complexity of the issue-the multiplicity of actors involved, the range of tasks they need to engage in to tackle environmental problems, and the resource and contextual factors that influence their ability to do so. The explicit incorporation of important principles into the CDE framework, namely the integration of environment and development, local participation, national and local ownership of efforts, subsidiary, (i.e. working at the lowest practical level), and a focus on the process, provides not only donors, but also decision-makers within developing countries, with the basic elements of a powerful approach. A number of donors have initiatives that are carrying forward different aspects of the CDE discussion, although very few agencies have developed their own specific CDE guidelines.

Although such a CDE approach is useful in laying out a progressive agenda for assisting developing countries in meeting environmental challenges, neither the framework nor the guidelines provide any help in making strategic choices (34). Furthermore, the notion of CDE (similar to the notion of CD) is broad and all-embracing to the point that it is almost synonymous with the definition of development itself (34), and thus is of little help to practitioners in the field. Still, in practice, CDE may need to take a multi-pronged approach to devise specific programs and projects that are germane to local conditions and needs and build local capacity, but at the same time are in line with CDE's systems-view mandate as well as explore themes and questions that cut across issue areas or sectoral perspectives. The latter would be a particularly useful area to devote some attention and efforts to because an improved understanding of cross-cutting issues could bolster a range of programs. Some understanding on this front could be gained from an examination of current challenges facing developing countries and of the approaches being discussed or taken in this regard.

With this in mind, the next section focuses on three cases-two sector-centered (agriculture and industry), and one cross-sectorial (climate change)-and briefly reviews the relevant sector- or issue-specific literature, with the aim of highlighting some specific implications that emerge for CDE. Through this, not only does this article present specific examples of how the CDE framework may be relevant in individual areas, but it also points out any gaps that might emerge, with the ultimate goal of contributing to broader CDE discussions.


Agriculture and Environment

The so-called Green Revolution is generally considered a success because it allowed a rapid increase in food production in developing countries. For example, average yield of wheat in India increased from 900 kg/hectare in 1964 to 2300 kg/hectare in 1999 (37). However, it has also propagated a model of agricultural intensification, characterized by high-intensity use of chemical inputs (pesticides and fertilizers), monocultural cropping, irrigation, and increased mechanization. This has resulted in a well-recognized litany of environmental problems, including the degradation and erosion of soils; depletion of groundwater; eutrophication of rivers, lakes, and estuarine ecosystems; loss of crop genetic diversity; impacts on ecosystems, humans, and other animal species from pesticide use; and global pollution (38, 39, 40, 41, 42).

Many of the issues on the agriculture-environment nexus are cause for serious concern. For example, agriculture consumes almost 70% of the freshwater supply globally (43), and in highly productive countries such as India and China, this number is almost 90% (44); about 38% of the 1.5 billion hectares of cropland worldwide has been degraded through agricultural practices (45), and soil erosion rates in many parts of the world exceed the replenishment rates by 1 to 2 orders of magnitude (46); an estimated 3 million people worldwide (mainly in developing countries) suffer acute pesticide poisoning every year (47); and the use of nitrogen-based fertilizers, and the industrial fixation of nitrogen to produce them, is a major contributor to the significant human perturbation of the nitrogen cycle, with attendant implications for terrestrial and aquatic ecosystems, as well as greenhouse gas warming and ozone depletion (48).

Although the environmental and economic costs arising from agricultural practices are impossible to assess accurately, the few attempts in the literature suggest that the numbers may be rather large. For example, the annual costs of US pesticide use are estimated to be $8 billion (49), and of soil loss in the United States, mostly from agricultural land, costs about $44 billion (50). Furthermore, not only does modern agriculture have a substantial impact on the environment, but a substantial degradation of the resource base (such as degradation and depletion of soil and water resources) can also place a constraint on crop production (46).

Given these concerns, there has been an immense amount of attention devoted-on paper as well as on the ground-to various aspects of the agricultural production system to make it less environmentally unsound and still maintain productivity. This has included a focus on pesticide use, fertilizer and water consumption, improved agricultural practices, and the deployment of innovative technologies. Consequently, a number of strategies, practices, and options have been discussed and explored in practice.

One major set of literature, informed by substantial agricultural endeavors, focuses on a mode of agricultural production termed sustainable (51, 52), alternative (53), ecological (54), regenerative (40), or agroecological (55). The discussions and practical explorations that fall under this rubric are primarily driven by recognition that the current paradigm of agricultural production carries with it severe environmental and other impacts and is probably unsustainable in the long run. A common feature of this literature is that the discussed approaches comprise a system of technologies and practices to ensure and sustain adequate levels of agricultural production in an environmentally sound manner.7 This last condition necessitates broadening the analytical perspective and utilizing a systems approach to understand and maintain the web of inter actions between the natural resources and various plants, animals, and other organisms that comprise the agricultural ecosystem. Sustainable agriculture thus includes a focus on soil and pest management, agronomic practices, yield improvement through agroforestry, and development of new crop varieties. Thus, it relies on a combination of local knowledge, practices, and resources, as well as modern knowledge and technologies to achieve its goals. Notably, such approaches need not lead to a loss of productivity (62).

Many new technologies and practices are being explored that could form part of an approach to sustainable agriculture. Precision agriculture (PA), for example, could help improve the efficiency of input use through more efficacious application. PA is an information-technology-enhanced management strategy based on the collection of data at appropriate scales, its interpretation and analysis, and implementation of a management response at an appropriate time and scale (63). A regional-level focus may be needed for weather data, though field-level information is needed for soil fertility and pest conditions. Given both the requirement to collect information in substantial detail and the ability to process and analyze it, PA is in reality a high-technology approach that combines sophisticated geo-referenced data gathering with information processing, crop production modeling, and decision support systems.

The application of modern biotechnology to agriculture also has the potential, in theory, to contribute significantly to ameliorating the environmental impacts of agriculture. Manipulative techniques such as tissue culture and genetic engineering allow a faster approach to modifying plant traits according to need, and recent advances in genome mapping techniques allow for determination of genes with specific functions that could then be expressed in plant strains. Crops that are engineered for higher yields could reduce the area under agriculture, and those incorporating pest and disease resistance may assist in the management of pests. For example, corn expressing the Bt (Bacillus thuringiensis) toxin has built-in insect resistance, thereby reducing the need for insecticide applications. So far, though, this potential has not translated into positive gains for the environment-in fact, it is not even clear whether the current economic structure of agriculture or environmental policies will promote the appropriate biotechnology research (51). Some commentators also suggest that biotechnology cannot be thought of as a sustainable technology unless it is embedded into a system of sustainable agriculture (64).

A common feature of the above strategies and approaches to reducing the environmental impacts of agriculture is that they are highly knowledge and information intensive and are designed to take into account spatial and temporal variations in local soil, water, and other ecosystem conditions. Thus, sustainable agriculture in industrialized countries is generally not seen as a return to low-technology practices, but is quite dependent on the use of ecological data and information, sophisticated technologies, and analytical and decision-making tools in farm management strategies-in a sense, the farmer is a high-tech manager (51). Furthermore, whereas such approaches offer the potential for environmentally sound agriculture, their successful implementation will require a coherent and integrated set of policies which has not been the case so far (65). There are a range of measures in industrialized countries to limit the environmental impacts of agriculture, but there is no grand plan to do so (65).

The move towards sustainable agricultural production in developing countries is complicated by the simultaneous existence of two agricultural production paradigms (often in the same country): the input-intensive green revolution variety and low-resource farming. The latter is common in many parts of South and South-East Asia and Latin America and most of sub-Saharan Africa. These small-scale farmers have limited financial and technical resources and often do not benefit from agricultural research. It is estimated that 1.4 billion people are dependent on low-resource farming, often in rain-fed areas, and that such farmers have been slow or unable to adopt many of the recommendations flowing from agricultural research (66). Each of these paradigms needs to move towards economically and environmentally sustainable agriculture, but from different starting points, and with access to different kinds of resources (and in both cases, to far than fewer resources than are available in industrialized countries).

Any strategy for promoting sustainable agriculture in developing countries has to take this into account, and hence will require approaches tailored to the particulars of local production patterns and resource constraints. Still, common elements of the technological agenda for these approaches include (a) solving problems using diversified production systems that take advantage of the beneficial interactions that exist in nature rather than using external inputs, (b) recognizing the location-specific nature of agricultural practices, and (c) exploitin technology and practices based on intensive use of knowledge and genetic resources rather than machinery or agrochemicals, with the knowledge coming from the scientific establishment as well as farmers (67).

A number of indigenous agricultural systems, as well as new experiments mixing tradition with modern approaches, successfully illustrate the potential for sustainable agriculture in developing countries (40, 52, 53, 55, 68, 69). For example, one of the most cited examples of integrated pest management (IPM) comes from Indonesia where a program to manage the brown plant-hopper and other rice pests was developed by the government with Food and Agricultural Organization support, and included countrywide IPM training and implementation, elimination of pesticide subsidies, and restrictions on the use of 57 pesticides. Between 1987 and 1990, pesticide use declined by 50% while yields went up 15%, and ending pesticide subsidies saved the government about $120 million annually (52). However, such examples are still few and far between.

If such approaches to agricultural production offer an environmentally and economically sound alternative to current input-intensive agriculture, then why are they not being adopted in developing countries, especially as farmers themselves stand to gain the most in terms of improved and sustained productivity? There are some insights from analyses of sustainable agriculture efforts: For example, there may be a lack of information on alternative approaches as well as on the hidden environmental and other costs of current production methods; there are transition costs and uncertainties of shifting to different methods of production [in addition, the transitional phase may be accompanied by a temporary lowering of productivity (54)]; government agricultural policies often promote the status quo (especially through subsidies on pesticides and fertilizers) and environmental policies are either not integrated with agricultural policies or are inadequately or weakly enforced; chemical companies continue to promote pesticides (global agricultural pesticide sales in 1995 exceeded $32 billion), and the IPM programs they promote may focus on pesticide management rather than pest management (52).

The shift to sustainable agriculture will require better characterization of the natural environment for adapting agricultural production to local conditions, for better natural resource management, and for monitoring changes. Environmental information is considered key to sustainable agriculture, but the ability to collect and integrate such information is weak in most developing countries and needs substantial strengthening (70). An information strategy, and the capacity to manage information, needs to be an essential part of agricultural research organizations (71).

Furthermore, attention to agricultural research capacity is critical because this offers a primary defense against the uncertainty about resource and environmental constraints (42). In fact, such constraints may be one of the most serious challenges facing agricultural research in the twenty-first century (72), and this will require bridging the "island empires" of agriculture and environment (73). Agricultural research systems have been the recent focus of much attention because of the stagnation in the funding of international agricultural research in general8 and the slowdown (or decline in some cases) in its growth in developing countries, especially given that the agricultural research intensity of developing countries is about one fifth that of developed countries (75). Not only is the size of developing country national agriculture research systems important but so is their focus: At present, most developing countries have a limited capacity to incorporate natural resource and environmental issues into their research agendas. Strengthening the capacity to do this is critical (76). For example, high-technology approaches such as precision farming will exceed the implementation capacity of most developing countries, but it may be possible to increase the efficiency of inputs through simpler approaches (77). Thus, exploring such alternatives, but keeping in mind local resource constraints, may be an important area of research. Generation and diffusion of agricultural innovations require a range of institutional innovations, for example, in market institutions and in organizations that generate and disseminate knowledge to farmers. The latter, national research and extension programs, are particularly relevant because they are the locus of agricultural research and its dissemination and hence play a key and direct role in technological change in the agriculture sector. The key criteria determining the effectiveness of agricultural research systems can be thought of as integration of research with technology transfer, relevance of research efforts to farmers' needs, systems' responsiveness to poor farmers, and mechanisms for adoption of new technologies and practices, which determine their practical usefulness and implementability (78). This requires better links between research, extension, and farmers' organizations (79), with a special focus on small farmers. Although there has been much attention paid to the importance of farmer feedback to research and extension, there is often still limited participation of farmers in setting the agenda of these agencies (80).

In many cases, traditional research and extension services are poorly equipped for developing and disseminating technologies and practices necessary for sustainable agriculture. NGOs play a particularly useful role in overcoming this constraint (67). Strengthening linkages to the international agricultural research centers for obtaining information about natural resource management and other technical advances, as well as suitably influencing the direction of the CGIAR research, is also important (44). In fact, one of the main lessons that emerges from an examination of successful cases of sustainable agriculture is that diverse linkages and partnerships-among research institutions, extension agencies, NGOs, farmers' groups, and international o rganizations-are key to such efforts (52).

Whereas agricultural programs generally focus on building up and strengthening research and extension organizations, such "supply-side" approaches to the dissemination of improved crop varieties or agricultural practices may be less effective than "demand-side" factors such as contracting of research by the actual users of these technologies, which suggests a need for focusing on innovative ways to promote dissemination through user demand (80). Focusing on small farmers as users of technologies may be particularly important. In fact, small farmers should be considered not just as users but as partners in agricultural research and extension. Putting the knowledge, problems, analysis, and priorities of farmers first as a complementary approach to traditional research could yield rich dividends for sustainable agriculture (66), especially as farmers' knowledge, although often valuable and uniquely suited to local conditions, is often marginalized by traditional research (66, 81, 82). Thus, technological innovation in agriculture in developing countries has to be seen as a partnership among various actors.

Agricultural policies can have many unintended environmental consequences (72) that need to be systematically studied. For example, agricultural inputs-fertilizers, chemicals, power-in developing countries are often heavily subsidized (83), which sends the wrong signals to farmers. Domestic groups as well as international organizations can play a major role in influencing government policies; the World Bank, for instance, has been promoting the phaseout of such subsidies in developing countries. Improving the interface between researchers focusing on natural resources management in agriculture and policy-makers has been shown to lead to positive changes in the relevant policies, a strategy with considerable potential that has received little practical or analytical attention (84).

The role of biotechnology in sustainable agriculture, although currently a contentious issue, is likely to assume increasing importance for developing countries. Agricultural biotechnology will certainly be of use to developing countries-for example, through improving the nutritional characteristics of food (85, 86), as shown by the recent success in engineering a vitamin-A-enriched rice (87). However, most of the research and development (R&D) efforts in biotechnology are located in the North and hence are shaped by industrialized countries' needs and markets and are driven almost exclusively by profits.9 The result has been limited focus on research relevant to agriculture in developing countries (89).

If biotechnology is to serve developing countries in improving the environmental performance of agricultural systems, most of the gains will be achievable through research relevant to specific local needs and conditions. This will require a better understanding of local problems and of the implementability of potential solutions, a process in which developing countries at the very least have to be active partners, but can do so only if they have the appropriate analytical and technical capacity. Given that agricultural biotechnology is in its nascent stage, developing countries could still play a role in guiding the agenda, but the window of opportunity is limited by the speed with which the field is evolving. Thus, it is critical to take a strategic approach to understanding both the kinds of capabilities required to place local needs on the biotechnology agenda and the process used to build up such capabilities. Whereas transfer of technologies may be a short-term answer for developing countries, in the long term, the ability to absorb and assimilate knowledge, and develop indigenous technologies and products is crucial (90). There has been some development of local research capacity in biotechnology in the South, assisted by training and capacity-building efforts such as those sponsored by Rockefeller's rice and cassava programs, the Biotechnology Action Council of UNESCO, and the International Cooperation program of the European Commission (90).

Two last points should be mentioned in this discussion of agriculture and the environment. First, much of the discussion of the evolution of agriculture in developing countries centers on its production aspects. To the extent that increased production improves the economic well-being of farmers, this emphasis is perfectly logical. In the end, however, the focus has to be on the efficiency of the agricultural system, including storage and distribution. Reducing post-harvest losses and improving distribution will result in upstream gains for the environment by ultimately increasing the amount of agricultural produce on the market per unit of harvest. Unfortunately, these issues are often sidelined by the obsessive attention to agricultural production driven by a supply-side-oriented focus on food security among researchers, food policy analysts, and industry (see e.g. 91, 92, 93, 94), despite the facts that there is not expected to be a shortage of food in the short term (95, 96), and that agricultural production is generally not the constraint to eliminating hunger (97, 98). This creates an impending doom mentality and puts the policy focus overwhelmingly on food production and the perceived need for a rapid increase in it, issues that capture the policy debate at the cost of others. [For example, although there is general agreement that post-harvest losses are locally significant, even the precise amount and nature of loss are not well characterized (99).] Such a narrow framing of the agriculture issue is inappropriate for developing countries, and analysts and researchers in developing countries need to broaden the focus to a full systems approach.

Second, moving toward sustainable agriculture in developing countries requires farmers in developing countries to make appropriate choices in agricultural production and practices. These choices are also influenced by the larger context in which they operate-farmers' production systems are increasingly linked to the global market. Under increasingly liberal regimes for international trade, crop choices are often directed by external demand and supply pressures (100). The growing commercialization of agriculture is leading to an increase in the availability and use of inputs (seeds, agrochemicals, and machinery), the choice of which may be directed by increased competition (101). Others suggest that moving away from traditional crops to those more suitable for the market can cause changes in existing community institutions, leading to poorer environmental management (102). Such trade issues may be of particular importance in the developing country context and merit further study by analysts and researchers in such countries; as Dragun & Tisdell (103) mention, research on the globalization impacts on agriculture and the environment has been all too sparse.

To summarize briefly, reducing the environmental impact of agricultural production in developing countries will not be a trivial task. It will require harnessing knowledge and technologies to suit a range of production paradigms ranging from rain-fed subsistence agriculture to irrigated, input-intensive, large-scale production. Given the resource constraints in developing countries, blindly following the industrialized country trajectory of improved environmental management through application of high technologies may not always be possible or sensible; thus, the South will have to be innovative in its approach to sustainable agriculture. Effective development and diffusion of suitable technological and other innovations will require better information about the interaction of agriculture with the environment, and active partnerships between various stakeholders such as farmers, NGOs, extension agencies, agricultural researchers, and policy-makers. It will require tailoring policies to actively promote sustainable agriculture and modifying those that present a barrier. This requires a focus not only on domestic policies, but also on the international context that influences choices in agricultural production systems. Finally, it will require a reorientation of the analytical framework to examine the agricultural system in toto rather than merely its production aspects.

Industry and Environment

Industrial activity is responsible for a wide range of environmental impacts, of which possibly the most disruptive to human health and the ecosystems result from the release of numerous organic and inorganic chemicals, heavy metals, and gaseous pollutants (see 104 for detailed examples). Even though the magnitude of industrial activity in most developing countries is only a fraction of that in industrialized countries, both the nature of the industry and the fleeting attention paid to environmental aspects of the industrial system in developing countries have led to substantial natural resource degradation and environmental pollution (2). For instance, it is estimated that factories are responsible for over 70% of China's total pollution, including 70% of organic water pollution, 72% of SO2 emissions, and 75% of flue dust (105). Most developing countries still need to expand, and are in the process of expanding their industrial sector substantially to drive economic growth (as opposed to many industrialized countries in which the contribution of industry to the national economy is shrinking while that of the service sector is growing). For example, industrial chemical production has been shifting to the developing world-between 1990 and 1996, this industry grew at an annual rate of 5.6% in developing countries in comparison to 1% in industrialized countries (106). Thus, ensuring that industrial expansion does not come at the cost of the environment should be, and increasingly is, a high-priority issue in most developing countries (see e.g. 2, 107, 108).

The environmental performance of industrial systems has received substantial attention in the North, with the ultimate aim of "greening" them (109). Importantly, it has been increasingly recognized that not only do industrial processes lead to environmental and human health impacts (both during extraction of resources that serve as inputs to industry and during the production process through air, water, and land releases), but also the use of many products and their ultimate disposal (take, for example, the automobile) lead to further adverse consequences. This led to the development of the life-cycle approach that is increasingly being applied in most Northern countries, and the environmental protection paradigm in the North has slowly shifted from pollution control to pollution prevention. The latter is different from the traditional approach to environmental management in that it is based on an approach that integrates environmental thinking into product design and management of the production processes. A variety of tools and techniques, such as life-cycle analysis and design for the environment (110), have been developed to assist in this goal. An integration of environmental considerations into the production system, however, does need active promotion through government or other policies because the "natural" technological evolution of industry has not, by and large, resulted in the development or deployment of environmentally-protective technologies (111).

The initial narrow and almost exclusive focus on the firm has also been broadened to an integrated systems-perspective examination of the nexus of industry and environment (commonly referred to as industrial ecology) (112, 113) whereby the firm is considered as being embedded in an industrial system in which materials, goods, technology, and information flow among various organizations (114). Thus, it is understood that the technological and environmental trajectory of a firm depends not only on its internal characteristics but also on the structure of the system of which it is a part and the linkages between the firm and other organizations. In a natural evolution, the focus is now beginning to broaden beyond the industrial production system to include consumption aspects, with a focus on the determinants of patterns of consumption (115).

Within developing countries, environmental management efforts have focused mainly on the industrial production system, i.e. on reducing emissions from manufacturing facilities as well as improving their environmental management to reduce energy and materials use during production. There has been an increasing awareness in these countries also of the need to prevent, rather than control, pollution for long-term sustainability. Although there are a range of donor and domestic efforts to improve the environmental performance of industry in developing countries, the most promising seem to be the efforts to incorporate cleaner production approaches-the linchpin of pollution prevention-into industrial production.

Cleaner production (CP) means the continuous application of an integrated preventive environmental strategy directed to both processes and products in order to reduce risks to humans and the environment, while promoting increases in productivity and competitiveness (116). For production processes, cleaner production includes conserving raw materials, water, and energy, eliminating toxic raw materials and inputs, and reducing the quantity and toxicity of all emissions, effluents, and wastes before they leave a process. For products, the strategy focuses on reducing impacts along the entire life cycle of the product, from raw material extraction through manufacturing and use, to the ultimate disposal of the product. Cleaner production is achieved (a) by applying knowledge to improve existing production technology, develop new technologies, change existing products and create new ones and (b) by changing policies, procedures, and institutions to ensure that individuals perceive positive incentives and receive rewards for taking preventive actions (116).

The UNEP CP Programme, established in 1989, has played perhaps the most prominent role in the promotion of CP in developing countries, especially for small and medium-sized enterprises. It has catalyzed the establishment of a CP network of participants from various organizations and has developed capacity for environmental management through increasing awareness of the preventive strategies, conducting CP assessments, and developing and disseminating information, training materials, and case studies (117), with a special focus on industries such as metal finishing or tanning rather than on individual firms [similar to efforts directed towards specific sectors dominated by small and medium-sized enterprises in the North (118)].

Despite a major focus on enhancing environmental awareness, education, and training-considered key components of capacity building for CP (119)-it is felt that current activities to promote CP in developing countries have not been able to fulfill their potential. Possible shortcomings include an undue emphasis on increasing the supply of CP strategies (rather than creating a demand for them), a primary focus on existing rather than new installations, and a focus on technical experts and environmental policy-makers rather than on industrial policy-makers, the financial sector, or equipment manufacturers (117). At the same time, there still remain broader institutional, informational, attitudinal, and policy barriers to the implementation of pollution prevention approaches in developing countries (120, 121). These include environmental policy that focuses on end-of-the-pipe controls (and may rarely be enforced in any case), risk-averse behavior in firms, and inadequate knowledge and expertise about pollution prevention (122).

This suggests a need to further strengthen efforts promoting CP/pollution prevention but also to strategically reorient them in light of knowledge gained from past experiences. Three areas stand out: (a) the need to target policy-makers to modify incentive structures and regulations to promote pollution prevention in preference to pollution control, as well as to integrate environmental protection into industrial policy; (b) the need for innovative mechanisms to create support for environmental management within firms; and (c) the need to improve the capacity for managing technological change and innovation within firms. I will focus on the latter two briefly here.

Creating bottom-up demand for pollution reduction/prevention within firms may be a particularly critical issue in developing countries. A recent strand of thinking has highlighted the potential role of information in assisting is such a process (105, 123, 124, 125). One of the first experiments in this domain, initiated in 1987, was the Toxics Release Inventory in the United States, which has been rather successful by many measures. Between 1988 and 1997, industrial on- and off-site releases decreased 42.8%, from 3.40 to 1.94 billion pounds, for chemicals reportable in all years (126). At the same time, public and advocacy groups have used the Toxics Release Inventory data to obtain information about their local environment and to participate in, and influence, relevant local and national debates.

A more recent program in Indonesia, PROPER, used public dissemination of the environmental performance of firms to successfully encourage improved practices. Not only did public disclosure provide a public reputation incentive for firms to reduce their emissions (thus reducing the need for reliance on government enforcement of regulations), but the systematic collection and presentation of data also educated owners and employees about the performance of their firms (123). Part of the program's success lies in its simple design utilizing a rating system that was understood easily by communities, and focusing on data (water pollutant loads) that was already being collected (123). Such public information about the environmental performances of firms may also provide signals to financial markets, even in developing countries (105, 123). The success of this program has inspired a similar effort in the Philippines, and elicited strong interest from Colombia, Mexico, Thailand, and Bangladesh (127).

Thus, involving communities, civil society, and media, as well as markets may offer a paradigm for assisting in guiding industrial systems that may be well suited to developing countries where regulatory agencies are often unable (or unwilling) to carry out all the functions required for such management (128). Of course, the availability of environmental information is often a major constraint in developing countries-the incentive to gather and provide information is often dependent on the organization of a society and its institutions (129). If information is to be used as part of a strategy to control industrial pollution, the collection and provision of detailed, accurate, and trustworthy information become critical capacities (111, 123).

Such information is useful not only in developing indicators of environmental performance but also more generally in assessing the state of the environment and the processes that affect it (129). This in turn requires institutions that understand the information needs for specific goals (for example, which pollutants are particularly implicated in adverse health impacts from air pollution), as well as the willingness and the ability to collect this information systematically and in an unbiased manner. The manner in which this information is disseminated is also important, both in terms of the transparency of its presentation and organization as well as its ease of public availability.

It should also be recognized that providing incentives to industry to improve environmental performance is a necessary, but by no means sufficient, condition for it do so. Sustained progress in reducing environmental impacts of industrial systems requires the integration of this concern into the core activities of the firm, and is based on a model of transformation of industrial processes and products driven by technological change and innovation. Note that the CP paradigm is not based on some absolute standard of clean-it is almost impossible for any process to be completely clean-but rather on a relative approach suggesting that the goal is to improve production and make it cleaner than in the previous iteration. This also puts a focus on continuous improvements, as highlighted earlier, indicating that continuous improvement and innovation are required. In addition, an innovation-driven approach can eliminate the tension between environmental management and competitiveness by enhancing resource productivity and overall performance of the firm (130, 131). Such an approach, by putting technological innovation at the center, forces a reconceptualization of the firm's role in the process, as well as that of the government and other organizations. In this view, the firm is an active agent of technological change for meeting desired environmental goals, and the role of government policy, among other things, is to provide incentives to firms to develop the appropriate capabilities for innovation, as well as to create an enabling environment to assist them (see Section 4.1.2 for a more detailed discussion of the capacity to manage technological change).

The increased entry of multinational corporations into developing countries may also offer incentives and opportunity for improved environmental management within national firms. Multinational corporations generally have more well-established environmental management systems and practices (132), which may spur their local counterparts to improve either through internal or public/government pressures. Spillover effects such as diffusion of better environmental information systems, technologies, and management practices may be relevant not only for local suppliers but for other domestic industries as well. From the perspective of strengthening domestic capabilities for technological innovation, however, it should be noted that multinational corporations do not perform R&D on any significant scale in developing countries (133, 134).

If CP (or pollution prevention) is viewed-as it must be-as being based on continuous improvements and innovation, then many of the current donor approaches to CP (such as the UNEP effort) or pollution prevention (see, e.g. 135) may not be very effective. These efforts are mostly based on sectorial case studies and databases/compendiums of technologies, with the expectation that the exposure to, and availability of, such information, and assistance with implementing new technologies and processes will help firms in improving their production. And it surely can, but without necessarily building up the technological capacity to allow improvement on a continuous basis within the firm.

Furthermore, pollution prevention ("green") approaches, such as those involving process and product modifications, often may be specific to, and embedded in, the production system and therefore cannot be sold or transferred as separate components or technologies. However, pollution control or other clean-up ("brown") technologies are easy to sell because they are generally add-on or stand-alone units (such as sulfur dioxide scrubbers, wastewater treatment plants, etc). This can lead to a potential tension between the promotion of technologies, knowledge, and activities needed for a shift towards green production systems and the efforts of the traditional environmental goods and services industry [a large industry that is expected to grow substantially in non-OECD countries (136)]. The majority of the environmental technologies supported by Northern government export-promotion activities are pollution-control (end-of-the-pipe) technologies rather than pollution prevention (clean) technologies (137).10 This is not to say that clean-up or pollution-control technologies are irrelevant for developing countries-clearly they will be needed-but it should be recognized that in the long term, a pollution prevention approach is preferable to a pollution control approach because the former promotes dynamic efficiency, whereas the latter offers a static solution.

To briefly recapitulate, the substantial and growing environmental impacts of industrial production will require technological and policy innovations that are suitable (as in the case of agriculture) for a range of production units going all the way from informal and semi-formal, small and medium enterprises to highly organized large firms. Perhaps the greatest need is for creating a demand for environmental management within firms-especially as consequences of their activities often don't affect them directly. This may require highlighting the linkage between environmental management and enhanced resource productivity, and improved overall performance. Other innovative programs, such as public disclosure schemes, can complement government regulation and other policies to provide incentives for environmental management. It is important to maintain a focus on pollution prevention rather than pollution control as the approach of choice, which requires efforts within the firm and through public policy to strengthen the domestic capabilities to manage technological change.

Climate Change

Most of the discussion surrounding CDE has generally focused on domestic issues. Less attention has been devoted to capacity building for global environmental issues, i.e. what kind of capacity is needed, towards what goals it needs to be deployed, and how it may be built up. The discussions on this front are still relatively general in nature, and there is not much literature on the topic, even though in many cases the capacity to identify, understand, and tackle local or national environmental issues may be different from the capacity to deal with environmental problems that are global in origin, and need global solutions. This section uses climate change as a case study to examine this issue. Climate change has emerged, in recent years, as the most significant global environmental issue, and for good reason. The climate system inexorably links the modern industrial economy in its present form with the environment, and through its global scope, links all countries. The scale, scope, and complexity of this issue, combined with uncertainty in our understanding of the problem and its potential impacts, thus poses a unique challenge to policy-makers worldwide. The distribution of (a) responsibility for greenhouse gas (GHG) accumulation in the atmosphere (the driver of climate change), (b) the potential impacts, and (c) the expected ecological, economic, and social implications varies substantially among regions and nations. This variation results from the large differences between countries' GHG emissions and the geographic distribution of the diverse impacts (because their manifestation is often dependent on local geography, ecology, and economy). There are also significant differences in the socioeconomic and other institutional conditions between countries that govern the implementability of any mitigation and adaptation strategy. Some combination of national responsibilities, capabilities, and potential impacts has influenced (and will continue to do so in evolving ways) the current and future burden-sharing arrangements. The result is a problem of immense complexity, not made any simpler by the fact that it needs a multi-national resolution; issues of economic security and national competitiveness thus also get intertwined into the discussions.

Substantial resources have been devoted to analyzing various aspects of the climate problem. These efforts include improving understanding of the scientific aspects of the climate problem, assessing the potential impacts of perturbations and shifts in the climate regime, and assessing the overall economic and social dimensions of climate change, as well as analyzing various mitigation and adaptation strategies to cope with the potential socioeconomic impacts. From a national capacity-needs perspective, one could divide the problem into three (not completely unrelated) categories-climate science and assessment, negotiations, and implementation-that offer a useful framework for analysis (139).

The first category comprises the capacity to understand the scientific nature of the problem and its local impacts, as well as the national socioeconomic implications of any mitigation and adaptation policies. The second category refers to the capacity to engage in international negotiations on the climate issue and to move towards the objectives of the climate convention while presenting and protecting the national interest. The last category pertains to the capabilities required for successful and effective implementation, within countries, of any GHG mitigation or climate adaptation.

Although the climate problem may be global in origin, it has been increasingly realized that a significant knowledge and understanding of not only local biogeophysical, but also social, economic, and institutional conditions, is important for a better comprehension and definition of the nature and magnitude of the implications of climate change.

To give a simple example, shifts in crop yields will result from the ways in which variability in climatic parameters, such as temperature and moisture, will interact with specifics of the local farming conditions, such as the nature of the soil, ground- or surface-water availability, agricultural practices, crop choices, and plant varieties. However, the eventual socioeconomic impacts will also depend on factors such as market conditions, local or regional economic circumstances, and the hunger vulnerability of population groups (140); thus, an integrative approach is needed to assess the potential human dimensions of climate-agriculture linkages. In a different arena, Pfaff et al (141) highlighted the importance of local institutional conditions and needs for the appropriate and effective dissemination of climate forecasts. Using the case of the Peruvian fisheries during the 1997-1998 El Ni˝o episode, they illustrated that the mode of dissemination may unwittingly disadvantage some groups more than others. Thus, a better sensitivity to how forecasts may be useful-or used and misused-in developing countries could suitably orient the focus of researchers as well as their expertise and products.

Such issues take on particular importance for developing countries where the infrastructure, the organization of economic activities and practices, the economic, legal and other institutional frameworks, the socioeconomic conditions and structures, and the financial, human and other resources are vastly different from those in industrialized countries. Because the overwhelming majority of climate science and assessment is carried out by researchers from industrialized countries, the specific conditions and needs of developing countries may not be incorporated into the research. Furthermore, even a sensitivity to the need to do so is not enough-substantial effort is still required to develop an understanding of relevant aspects of any developing country, especially given that the societies and economies of each developing country are different. Such a situation holds not just for scientific research, but also for policy analysis. For example, global models of carbon abatement that are used to study national responses to global policy measures often make incorrect assumptions regarding the structure and dynamics of less-industrialized economies and pay little attention to specific considerations of informal economies, market disequilibrium resulting from persistent energy shortages, the use of biomass as a fuel, and developmental choices that imply very different investment patterns and long-term technology paths (142).

Thus, many analyses dealing with the climate issue require not only natural science research or economic modeling, but also social science research and the integration of various perspectives in a multidisciplinary approach. This is a challenge for research communities in both industrialized and developing countries as well as for international institutions. There has been some progress on this front through the increased sophistication of climate research. For example, the increasing resolution of global climate models, combined with an increasing entrainment of disciplinary specialists (such as agricultural scientists, water resource specialists, human health experts) and social scientists into the climate area, has allowed some progress on this front.

Even the focus of the research community may be skewed by the proclivities of researchers. For example, although explorations of the equity issue seemed to have engendered some interest in the early years of the debate, only recently has attention again been devoted to this issue, possibly driven by the renewed focus on developing countries following the Kyoto Protocol and these countries' insistence on keeping the issue in the discussions (143, 144). Given the disjointed and sparse analysis of equity, perspectives that are of importance to many in the South have received short shrift (143). Issues such as adaptation and vulnerability, which are critical for the South, are only now becoming the focus of concerted research efforts, whereas issues such as the role of cultural preferences in determining consumption patterns in the North have been mostly sidelined.

Given the linkages of the climate issue to almost all aspects of society and economy, significant resources have been devoted to research programs in most industrialized countries. It is estimated that the annual climate research budgets of the United States, Japan, and the European Union add up to $3 billion (145). National and international programs (such as the United States Global Change Research Program and the International Human Dimensions Program respectively) have also begun to direct attention to the human dimensions of climate change, although substantial effort is still required on this front (146). The South has lagged far behind, although major countries such as Brazil, China, and India have made some effort to build up their capacity in this area (although at a completely different scale-for example, the five-year budget of the Indian Climate Research program is approximately $2.5 million). There is some effort to build up scientific research capacity in the South through programs such as START (System for Analysis, Research and Training) and APN (Asia-Pacific Network), although their success has been limited.

The United Nations Framework Convention on Climate Change (UNFCCC) also requires the Parties to submit national communications containing inventories of domestic GHG emissions as well as the steps taken or envisaged to implement the Convention (in line with Article 12 of the UNFCCC). Accurately assessing GHG inventories is important for developing countries, not only for implementing the Convention in good faith, but because current or historical emission levels may eventually play a role in determining future allowances or obligations (especially given the precedent set by the Kyoto Protocol). Also, the capability to estimate GHG emissions will be useful for other, future activities (such as baseline determination and monitoring for GHG mitigation projects). The development/strengthening of this capacity in developing countries is a major focus of capacity-building efforts on the climate issue and includes current efforts such as the US and UNEP Country Studies Program as well as the Global Environmental Facility.

A better understanding and definition of the nature of the problem, of course, is only the first step towards its resolution. For most Southern countries, a central element of the climate debate revolves around the need for a fair resolution of contentious questions such as who should reduce GHG emissions and by how much, how these reductions will be achieved, and how the burden of impacts, adaptation, and mitigation will be shared. Negotiating on such issues requires the ability to understand and articulate national needs and concerns, to place them on the international agenda, and to support and propagate national positions through targeted analysis (145), especially given that global climate discussions have been characterized by sophisticated strategizing to best serve national interests, rather than a cooperation in mutual confidence (147). This requires a focus on the capacity for detailed strategic policy analysis and use of the knowledge gained for developing national positions-this is the second category in the classification of climate-related capacity needs mentioned earlier.

Even the most rudimentary national policy on climate change requires an assessment of the national and subnational implications of the impacts of climate change and proposed abatement strategies in the context of historical GHG-emitting activities. Countries therefore need to develop a clear understanding of why the climate issue is important for them, and what implication alternative scenarios would have for their national economies. Such studies require a whole range of skills-data collection and analysis, emissions scenarios, climate modeling, impact analysis, and technical and economic analysis of abatement strategies. Unfortunately, there has been almost no attention paid to developing the capacity to perform these functions, to coordinating and utilizing them to formulate national positions and policy proposals, or to doing this in a reiterative manner as the negotiations evolve (145). As a result, the South has had little constructive influence on the direction of the negotiations in recent years.

For example, although developing countries highlighted the primacy of the principles of equity and justice from the beginning of the climate negotiations (148) and were successful in incorporating them in the Climate Convention, there has been little effort on the part of Southern governments to be proactive in developing, in any substantive detail, frameworks that are based on the principles of the Convention that bridge the divide between the positions of industrialized countries and developing countries and that show the way towards long-term participation by all countries. The Brazilian proposal before Kyoto (149) that eventually became the basis for the Clear Development Mechanism is perhaps the only concrete and innovative plan incorporating equity and other principles of the Convention. As a result, equity has become an often recited but almost hollow mantra used by developing countries, mostly it may seem, as a way to avoid taking on commitments.

It is the capability to be proactive in the negotiations, utilizing analyses and strategies that take a long-term perspective, that may be the most immediate need for developing countries, and yet one that is often overlooked in most discussions of capacity building in the climate context. This lack of emphasis may be explained in part by the evolution of the climate negotiations whereby the Berlin Mandate oriented the focus of the debate to the binding commitments to be taken on by Annex-I nations, resulting in the Kyoto protocol adopted at the third Conference of the Parties (COP-3). Thus, attention was mostly devoted to issues relating to the short-term targets and timetables for the Annex-I countries before COP-3, and developing countries were content, effectively, to sit by the sidelines having avoided the prospects of commitments for the time being. Only the post-Kyoto heat, with the United States effectively tying its ratification of the Kyoto Protocol to "meaningful participation by key developing countries" (150), has stirred developing countries to think about negotiations with the long-term in mind.

Environmental professionals in developing countries, often already struggling to respond to domestic priorities, may be overwhelmed by the administrative and other demands placed on them by the Convention and other global environmental issues (145). In fact, as noted in an independent evaluation of the pilot phase of the Global Environmental Facility, some developing countries voiced concerns that "emphasis on global benefits is beginning to divert scarce local, institutional and financial resources away from national environmental development issues" (Ref. 151, p. 34). Furthermore, systematic linkages between policy-makers and analysts are often absent. Attention to the capability to carry out policy research and strengthen linkages among science, policy research, and decision making in developing countries would go a long way to ameliorate the situation, as would stronger linkages between developing countries for cooperating on analyses for informing negotiations.

The third set of capacity issues is relevant mostly for effective implementation of programs and projects for mitigation and adaptation in developing countries. First, it may be useful to determine the domestic policy framework under which the private sector and other implementing organizations can engage in activities consistent with the national positions and obligations under the FCCC. Capacity relating to implementation of the Kyoto Protocol flexible mechanisms may include a wide range of activities that countries are proposing to support (152). In the case of the CDM, for example, these could include capabilities for strategy development and needs assessment, for development and use of methodological and analytical tools for project work, and for information dissemination and exchange (153), all of which will require institutional and technical expertise.

The Special Report on Technology Transfer prepared by the Intergovernmental Panel on Climate Change, upon examining the experience of transfer of technologies relevant to the climate issue, concludes that current efforts and established procedures will not be sufficient to meet future challenges posed by the climate issue, and that enhanced capacity will be required in developing countries (154). Capacity issues in the South include government ability to promote an enabling environment, information availability and dissemination, and the capacity to manage technological change (155), although industrialized countries can also take steps to promote the development and transfer of relevant technologies (see, e.g. 156). Possibly the most important of these is the capacity to manage technological change, something that is relevant more broadly as well, and will be discussed in greater detail in Section 4.1.2. In sum, one could say that Southern capacity needs include (35, 139, 157) (a) capacity for compiling and communicating information regularly, (b) scientific and analytical capability to understand the bio-geophysical impacts of climate change and their implications for national economies and societies, (c) capability to generate possible policy alternatives, as well as to analyze their applicability in various national contexts and their implications for national economies and societies, (d ) capacity to utilize the above knowledge to assist national decision-makers in articulating and protecting their national interest in the negotiations under the climate convention, (e) management capacity for implementation of possible climate protection strategies and programs, and ( f ) capacity to manage technological change (i.e. to choose, adapt, maintain and develop innovative technologies).

Many of these needs have received only sparse attention until recently. In fact, it is fair to say that much of the past effort to build capacity in developing countries has focused on the need to fulfill specific obligations under the FCCC. A recent decision (10/CP.5) at COP-5 to focus on CD (158), however, offers some hope that a more comprehensive approach will emerge.

CAPACITY DEVELOPMENT: Filling in the Blanks

Developing Capacity in the South: Some Key Issues

A multitude of issues and perspectives relevant to CDE in developing countries emerges from the brief review of the three cases in the previous section. First, it is clear that much of the CDE literature is on target: Responses to environmental challenges require an integrative and systemic approach; in the case of both agriculture and industry, it was seen that environmental considerations have to be integrated into the production system. Often, analytical and practical efforts have already recognized and begun to act on this. Linkages between various actors in any issue domain, as well as organizational diversity to design and implement innovative approaches, are key to successful incorporation of environmentally sound practices, and experiments that recognize the local context and constraints are much more likely to be successful.

Some cross-cutting aspects of the capacity issue highlighted in the CDE literature are especially important for effecting a shift toward environmental sustainability. For example, gathering, analyzing, and disseminating environmental information in addition to integrating it into decision-making emerges, not surprisingly, as being critical (129, 159, 160),11 as does increasing environmental awareness within the public and policy-making community.

Some gaps remain, however. The remainder of this section covers some issues that seem to be of relevance across the areas examined in the previous section, but that have not received much detailed examination in the CDE literature.

Role of NGOs

The CDE literature recognizes the relevance of NGOs (the term is used broadly here to include community-based organizations) as key actors in fulfilling sustainability objectives. In a sense, this is quite appropriate because developing countries have a long tradition of traditional organizations and institutions predating the arrival of modern organizations (or modern labels) (4, 164) that have, among other things, played a significant role in managing natural resources (165).

The role of NGOs in catalyzing or assisting in environmental management in developing countries has become increasingly prominent. A number of factors have contributed to this, including:

1. the rise of environmental consciousness in industrialized and developing countries;

2. the increasing realization of the links between social, economic, and human development and the environment [both pointed out by the World Commission on Environment and Development and seen empirically as a result of economic liberalization and structural adjustment (see, e.g. 166)], leading to a growing interest in environmental issues among traditional development organizations and social movements;

3. the UNCED process that not only gave official imprimatur and heightened visibility to NGOs, but also the networking and coalition building at UNCED that allowed for continued international linkages and coordination (164) (especially with the advent of the Internet);

4. numerous examples of NGO-driven or assisted environmental management at the local level in developing countries (see, e.g. 167) through efforts tailored to local conditions, constraints, and needs;

5. generally increasing popularity of NGOs with governments and official aid agencies in response to the "New Policy Agenda" governing development policy (168), characterized by neoliberal economics (whereby the role of government is minimal, mainly as an enabler) and liberal democratic theory (whereby NGOs act as a counterweight to government and are seen as vehicles for democratization) (163).

NGOs fill a niche by delivering services that both the market and government fail to provide; thus, they are "supplements, complements, alternatives, critics and watchdogs to government" (169:18). NGOs carry out a variety of tasks at both the micro and macro level: They engage in experimentation, serving as pioneers or catalysts for government action; they disseminate information, and are trusted by, and mobilize the public; they promote policy reforms; and they act as facilitators or monitors (170, 171). The capacity to engage in these multiple activities makes NGOs' roles powerful, especially regarding the environment. NGOs also are generally good at learning from their experiences and from their clients and adapting to evolving conditions (172), which helps them maintain their local relevance. Given the (correct) view in CDE approaches that local participation in, and ownership of, environmental programs is a key issue, the potential importance of NGOs and other local organizations in contributing to environmental problem-solving needs to be further stressed. In fact, a review of donor activities on CDE suggests that NGOs do indeed perform well by promoting CDE among environmental resource users, establishing linkages with other institutions, promoting awareness of environmental issues, and playing a role in advocacy and environmental policy dialogue (34).

Given the strengths of NGOs and their suitability to many development issues, though, they may often be viewed as a "magic bullet" (173, 174). This calls for a cautionary note. First, there is a vast range of capacities, ideologies, and approaches among NGOs, and they vary tremendously in their effectiveness. Not all NGOs necessarily promote public participation, and not all of them engage in promoting environmentally sound practices.

For example, studies of the efforts of NGOs in agricultural change suggest that, contrary to what might be expected, many groups not only introduce conventional production-oriented technologies (175) but also replicate them unquestioningly, although there are also numerous cases of NGOs implementing these technologies with greater sensitivity to local cultural and social contexts. [In fairness, though, it should be noted that many NGOs also take a broader view of the agricultural system than research and extension agencies, and promote technical change in agriculture through agroecological approaches (176)].

If NGOs are to be a mainstay of CDE activities, careful attention needs to be paid to their performance. The recent literature on the subject highlights the lack of systematic monitoring and evaluation of NGO performance, as well as criteria for their accountability. This is a key issue for effective participation of NGOs in environmental problem-solving. There is increasing evidence that NGOs do not perform as effectively as expected in terms of cost-effectiveness, sustainability, popular participation, and flexibility and innovation (174), and their managerial and financial capabilities are weak (34).

However, evaluating NGO performance is a difficult proposition because of the kinds of activities they are often engaged in, their lack of control over the contextual factors such as macroeconomic performance, state policies, and the actions of other agencies, and because of the requirements and constraints often placed upon them by their funders (174). Efforts to improve the effectiveness of NGOs are further confounded by the wide variation among organizations (171). In some cases, NGOs also lack the technical skills to carry out in-depth environmental analyses (34)-a particular area of concern because of the hurdles to building research capacity in NGOs, such as their inability to attract and retain researchers and to identify appropriate research advisors or collaborators (177).

NGO accountability is another not very well-studied area, especially because most of them are answerable to multiple actors-downwards to their partners, beneficiaries, and staff and upwards to their donors, trustees, and governments. CDE programs need to pay attention to various nuances of the accountability issue to ensure that it does not impede effective NGO participation.

A related issue is the question of NGO agendas. Given the increasing amount of funding being channeled through NGOs [in 1993-1994, at least 5% of the total aid from OECD countries was channeled through NGOs, up from 0.7% in 1975 (163), though most of it goes to Northern NGOs], their agenda may be co-opted by funding agencies or at least be guided or driven by the programming of funders (163, 178). The increasing emphasis on service delivery by NGOs also conflicts with their ability to engage in institutional development. Furthermore, many NGOs in developing countries still show little interest in CDE activities (34).

It should also be noted that NGOs have only limited capacity-in terms of numbers, resources, and coordination-which implies that by themselves they cannot transform whole sectors. This will require that their ideas of technology, methods, and clients be taken up by governments. This upwards diffusion of the learning and knowledge from NGOs should, in fact, be an important component of CDE in developing countries.

An important area of NGO participation in CD is in global environmental issues. Such issues, as mentioned in Section 3.3, can have substantial implications for developing countries through environmental impacts or because of the actions required to tackle the problem. Developing countries are at a particular disadvantage here because of their limited capacity to assess the generally highly complex and variegated scientific, social, economic, and ecological aspects of these issues, and NGOs can play a very important role in assisting decision-makers on various fronts (and, in fact, in many developing countries, they already do). In part, they can do this through internal analytical capability, but more often than not, Southern NGOs are able to obtain information, analyses, assessments, critiques, and policy proposals from their counterparts (or other actors) in industrialized countries through various channels [what Keck & Sikkink (179) refer to as transnational advocacy networks], as well as work with them to shape the international policy agenda (164, 180). Enabling such activities requires specific support for NGOs with an outward international orientation. By contrast, the current CDE approach generally focuses on NGOs that have a local orientation.

Policy Research and Innovation

The capability to develop environmental policy, and implement it effectively, is a core element of environmental capacity. Although degradation of the environment can cause significant human health, ecosystem, and economic impacts, such issues are still marginalized by decision-makers in developing countries who are often more concerned with macroeconomic policies, infrastructure development, provision of basic services, or national security issues. Environmental issues are characterized by substantial complexity and uncertainty and are often intimately linked to a range of human and economic activities, thus impeding policy-making. Ministries of the environment are often considered second-tier ministries, have limited resources, and do not have much political clout in national policy-making, thus further limiting their ability to develop policies or to implement them vigorously. Very few countries have made any effort to integrate the environment into their economic policies (34).

At the same time, donor agencies often promote imported organizational and institutional models (which may be inappropriate to the local context) and off-the-shelf environmental laws and frameworks in developing countries (181). Their approaches are often driven by the policy agendas dominant in the North.12 Currently there is a heavy emphasis on market-friendly policies and a reduced role for the government (184), regardless of whether the markets and institutions in these countries function as desired or can be made to do so in the relevant time period. Such an approach may also prescribe uniform solutions to local problems, which as pointed out by Hirschmann, can often be counter-productive (185). It also may not leave much room for the flexibility that is often important for effective implementation (186). At the same time, the strengths of local institutions and norms remain under-utilized. Also, given that there is a huge diversity of actors within spheres of economic activity, a spectrum of policies may be needed; for example, policies to make subsistence, rain-fed farming more environmentally sustainable may be very different from those required for irrigated, commercial agriculture, and policies to reduce the environmental impact of small and medium-sized enterprises will be very different from those required for large firms.

In this context, the capacity to research policies suitable to national (or local) problems and conditions becomes extremely valuable (187, 188). Policy research can be thought of as provision of intelligence to policy-making communities (187). Beyond this, however, the development of policy research capacity, and the availability and dissemination of the resulting analytical output, are likely to stimulate the demand for research within the policy-making community and eventually improve its absorptive capacity. Well-researched analyses of policies and the policy-making process will also raise the level of public debate on environmental issues, which should further strengthen the demand for improved environmental management.

Policy analysis can assist in choosing among competing policies prior to selection and implementation as well as help guide policy development as it unfolds (189). Thus, the various ways in which policy research can offer guidance in the policy process includes not only problem recognition and agenda setting, definition of policy alternatives, and selection of policy and its implementation and evaluation, but also support for, and analysis of, preexisting policies (190). This process might include examination of policies in other countries and their relevance to the national or subnational context, adapting suitable options and incorporating them into the national policy framework, or developing a different or hybrid approach if existing models seem inapplicable.

For example, an important aspect of the policy research agenda could be to develop innovative methods of environmental protection under local constraints and using local institutions. It is no doubt true that environmental sustainability requires appropriate policy frameworks, appropriate institutions, and appropriate technologies, as Lutz et al (191) point out using the case of agriculture. But a more immediate need may be to explore the paths to local optimization under current constraints. There are a number of innovative experiments on this front. The example of PROPER from Indonesia, discussed above, highlights how collection and public dissemination of information can lead to improved environmental performance of industry even under weak government institutions. In India, environmental activists have been utilizing public interest litigation to mobilize the judiciary to direct government agencies to improve their enforcement or pass judgments that fill lacunae in present regulations (192). The regeneration of small-scale water harvesting in many parts of India provides a heartening example of community-led programs that build on traditional local institutions even in the absence of the "right" macro policy framework or functioning legal institutions (167). The last example is particularly important because it illustrates the power of a bottom-up approach, whereas much policy research seems to focus on top-down efforts (as do CD efforts often). An important challenge for policy research in developing countries therefore is to understand the factors that contribute to the success of such local experiments and then to explore ways to systematically apply the lessons in other locales. This is not to say that a focus on the broader policy frameworks and institutions is not needed, but that a move towards environmental sustainability need not be solely dependent on appropriate shifts in these frameworks and institutions.

At the same time, an analysis of the policy-making process itself, although often overlooked, may be useful because it is only through such examination that one is able to ascertain patterns in the ways issues are brought on (or not brought on) to the policy agenda, and alternatives are presented to, and selected by, policy-makers. This could serve as a precursor to identifying shortcomings and gaps in the national policy-making process. Such a systemic understanding is critical if one is to effect fundamental changes in the way policy is shaped and crafted. Also, because the dynamics of the policy-making process vary from country to country depending on a variety of economic, social, cultural, and institutional variables (including historical factors), such analysis must necessarily be carried out within the national context by those who understand this context.

As with other environmental activities, the environmental policy research process is interdisciplinary and may need to utilize the skills of various individuals or organizations (generally with shared interests and often with complementary expertise). One effective, albeit difficult, way to develop environmental policy research/analysis capacity is to establish institutes with this as their central goal. A formal organization can attract and retain researchers, train new scholars, and provide a focus for activities. But although independent policy institutes/think tanks are a familiar part of the landscape in the United States (193) and to some extent in other industrialized countries, they are less common in the South. Despite this limited presence at present, Myers regards them as an important innovation in capacity building in developing countries (194). According to him, policy research institutes can assist decision-makers by serving in a "fire-fighting" capacity (i.e. responding to the short-term needs of policy-makers) and carrying out long-term research, both of which can also create demand for research from policy-makers. Additionally, these institutes can also focus on problems that are anticipated to become important in the future.

Such policy research efforts can fulfill multiple roles: (a) They can promote the use of data and analysis to replace ad hoc policy-making (and the need for data and analysis in turn promotes systems for collecting and managing information); (b) they can serve to build capacity in other organizations and build networks among researchers, decision-makers, and civil society, thus improving the policy debate about development (and environment); and (c) they can be pioneers in diffusion of policy innovations (194). On the down-side, policy elites in think tanks can unduly influence or even preempt important public debates on complex issues (195), which can contribute to policy capture by domestic interest groups or external funding agencies.

Policy research capacity does not have to reside only in specialized institutions-it can also be promoted within existing organizations such as the government, academia, NGOs, etc. The latter two are particularly important because they allow for policy analysis exercises to be connected to basic research (academia) or field-based knowledge (NGOs). Hybrid models seem to be more common in developing countries, especially in the environment area, whereby institutions take on multiple roles-research, implementation, policy advocacy-as illustrated by the Center for Science and Environment, the Tata Energy Research Institute in India, and ENDA Tiers Monde in Senegal. There are also examples of transnational networks that are multi-functional in that they link individuals and organizations that carry out research, but they also serve the research-enabling function of disseminating data and information and are active in advocacy (e.g. the Third World Network and the Climate Action Network).

Despite the fact that building or strengthening capacity to carry out environmental policy research and analysis would seem to be a core part of CDE activities [the OECD/DAC has stated this as one of the principles underlying its new guidelines for technical cooperation (21)], it has been mostly ignored by donors. They show little propensity to fund national policy research institutes, despite evidence of this group contributing positively to environmental problem-solving (34), or policy research activities in NGOs (196). Even on the analytical front, there have been almost no efforts to examine research institutions in the South (let alone policy research institutions) and little is written on the development and management of such organizations in the developing country context (197).

This is not to say that policy research capacity has been completely ignored in the development arena. There has been some attention paid to developing capacity for policy research, with much of it being focused (appropriately) on Africa, leading to some major initiatives13 that have been quite successful.14 In these efforts by the development community to establish and nurture policy research, it is notable that the predominant focus is gen erally on (macro)economic policy research. This focus stems from the perceived need to develop policy research capacity to assist in implementing economic reform programs. As this capacity is strengthened, it may be able to provide a base for other kinds of policy research activities, although this may be a slow process.15 Even though they may not directly lead to capacity for environmental policy research, experiences of such programs, and the lessons learned from them, may be extremely valuable for the design of environmental policy research efforts.

All in all, supporting environmental policy research has not been a main focus of CD efforts by donors. Even in the case of global environmental issues such as climate change, most of the capacity building efforts have been targeted to assist developing countries in fulfilling their immediate commitments under the FCCC.

The policy research focus of most developing countries' governments (to the extent that there are systematic or programmatic efforts) has been on macroeconomic issues, both because of the evolving policy frameworks in most countries and the changing rules of the game internationally. Some researchers and organizations in developing countries (such as the ones above) have, through their entrepreneurial efforts, been somewhat successful at building environmental research capacity. To what extent these individual efforts will flourish and allow developing countries to set their own environmental policy agendas (particularly on global environmental issues where this is of critical importance, as mentioned in Section 3.3) remains to be seen.

Last, but not the least, the capacity for policy research need not contribute only to effective policy-making, but should also be invaluable for analyzing and improving the CD process itself. Among other things, therefore, environmental policy research could assist in assessing capacity gaps as well as examining methods and strategies to fill these gaps. In fact, such assessment of national capacities is critical in developing any national capacity building program (8). Local individuals and organizations are the most suited for such assessments because of their familiarity with the context and with past failures and successes that may yield rich lessons for future efforts. Monitoring and evaluation of capacity building activities is also essential to evaluate the performance of ongoing programs and to improve them (4, 8, 13, 34). Domestic research capacity could also assist in the development of suitable indicators and local monitoring of CDE efforts. Through ways such as this, policy research capacity could help strengthen the domestic capacity to develop capacity.

Managing Technological Change

It is clear from the above case studies that global and domestic environmental issues are determined by, and in turn influence, national technological trajectories. For issues such as climate change, the need to constrain GHG emissions to prevent dangerous interference with the climate system will increasingly direct the choice of energy technology, fuel mix, and industrial practices and hence the direction of the energy sector in developing countries, especially as they begin to undertake binding commitments under the Climate Convention. For other environmental issues, significant shifts in patterns of industrial production may be required to abate local or regional pollution. To do this in a sustainable fashion will require a transition from the business-as-usual approach. Transfer of new technologies, often the solution of choice both in developing countries and international circles, can provide a gain in industrial and environmental performance, but the process is not as simple as going to the store and picking up an energy-efficient air conditioner. Selecting a technology that is suitable for local conditions and needs, adapting it accordingly to allow it to operate at design parameters (or at least at the best possible level), and maintaining it require substantial skills and information.

Furthermore, in order to not be caught in an import-to-upgrade cycle, accumulating the knowledge and expertise to improve the performance of a technology can be crucial. Incremental innovations to continuously fine-tune and improve the performance of equipment and production processes can help improve overall firm efficiency and productivity substantially, as the Japanese showed so successfully (200). Finally, firms should also be able to engage in radical innovations. This all requires that the firm be more than a passive operator of technologies, but rather an active manager of technological change that in turn requires technological competence and the capacity to innovate (201). Developing the indigenous capacity to manage technological change and to innovate must therefore be thought of as a central feature of CDE, and not as something remote from the environmental management arena. Furthermore, the capacity to manage technological change also builds the capacity to respond effectively to shifts in regulations or consumer practices (202).

The vast literature examining the experience of developing countries over the past few decades suggests that building up certain kinds of technological capabilities (TCs)-not a trivial task but far from impossible-is critical to fostering technological dynamism and the ability to innovate (201, 203, 204, 205, 206, 207, 208). The range of capabilities center around operating, mastering, and improving technological systems and components thereof. Thus, TCs include the skills and information-technical, managerial, and institutional-necessary for operating machines, understanding the intricacies of blueprints, designs, and patents, improving production processes and equipment, and eventually designing new machinery, products, or production methods (205). Thus, TCs are needed at the firm level for mastering the skills for setting up and running an industry efficiently on a day-to-day basis, for implementing minor innovations that in a cumulative fashion can contribute significantly to productivity increases, and for developing major innovations. Furthermore, such capabilities are also necessary for suitable selection, acquisition, adoption, diffusion, and eventual innovation of newer technologies (201, 203, 205). Hence, these capabilities can also improve the quality and effectiveness of foreign direct investment.

The development of TCs takes place at many levels. At the simplest level, it is mastering the operation of a simple piece of equipment. Deepening of these capabilities involves moving from a knowledge of the operation of a machine to its adaptation and improvement, substitution of inputs, modification of products, learning product design and the principles behind the process and equipment, new product and process development, and finally innovation in a basic sense, i.e. "knowing why" (209). Such capabilities are not created overnight, but must be nurtured over the long term. This requires, among other things, the development of human resources for production, investment, and innovation capabilities through primary, secondary, and tertiary education, apprenticeship programs, specialized education programs that match the needs of the industry, and pre-employment as well as on-the-job training programs (203). Linkages between firms and technology support organizations such as universities and technical colleges, consultants, contract R&D providers, basic R&D laboratories, information clearinghouses, and industrial extension services are critical to technology development (210), as are linkages with subcontracting and other sourcing firms, equipment suppliers, clients, and industry groups. [In fact, growing realization that a range of organizations-technical, educational, financial-with a distribution of different kinds of knowledge and abilities, and complex interconnections between them, are all necessary for innovation, and thus the study of the innovation process has increasingly focused on these "national systems of innovation" (211, 212, 213).]

Although the role of technological capabilities in industrial dynamism seems quite clear, the process by which they can be nurtured is fuzzier. Clearly, public policy can play a significant role in such developments. Appropriate institutional frameworks, for example, a stable and robust macroeconomic environment, and a dependable and predictable legal and financial system can promote technological capability building, but these are national policies that are driven not only by technology policy. Other more specific policy actions could develop institutions and infrastructure that directly target accumulation of TCs. Some that may be relevant to the environment could include, for example, the promotion of agencies that provide intelligence about technological options to reduce transaction costs; and R&D institutes, engineering consulting services, and extension services to not only assist in the adoption and diffusion of suitable technologies but also to provide advice on issues such as environmental metrics and standards and quality control. Furthermore, specific efforts to boost the linkages between firms and other organizations (by providing support for industry associations or environmental fairs) could be very helpful. Governments can also encourage technological capability in specific sectors by setting up requirements for improving the quality of technology acquired by firms, assisting them in developing contracts that encourage more than a simple technology transfer, and providing financial stimuli such as conditional investment grants or R&D tax incentives. Among all of these targets, the ones that seem to be the most useful are human resources development, research institutes, and intellectual property rights regimes, though substantial effort needs to be devoted even to these to allow them to achieve the desired function of promoting TCs (214).

Finally, perhaps the most contested area of government policy vis-Ó-vis TC development is the role of industrial policy in promoting such capabilities. The neo-liberal school of thought argues that the market is the best means of allocating resources and that government intervention distorts the functioning of such markets, whereas many researchers feel that active and strategic government intervention is critical for encouraging the development of technological capabilities through a range of policies, including providing protection to local industries in the early development stages (205, 207, 208, 215, 216). Any protection may also require integration with wider industrial policy, linking it with export performance (216) or limiting its duration (207).

Drawing upon other lessons from the East Asian economies, it is seen that government policies can also play a significant and important role to channel investment into areas that are deemed important for the economy (216). A well-designed national technology policy can play a vital role in all stages of importing and incorporating new technologies, going all the way from the planning stage to the eventual diffusion of the new technology (204, 217). In fact, Enos & Park go so far as to assert that, "there is no stage in the process of incorporating an imported technology that a conscientious government should neglect; there is no stage that should not have its appropriate public policy" (218). Although such policies can work, a review of the literature suggests that it is extremely difficult to bring off the infant-industry protection approach successfully, and this requires, almost more than anything else, the government's ability to learn from experience, act in a judicious and effici ent manner, not be captured by particular special interests, and respond to the evolving conditions (214). This requires flexible and intelligent public policy that unfortunately may be too much to ask of many Southern countries. Still, this does not let Southern governments off the hook. Pleading an inability to develop TCs is no solution.

Although there seem to be conflicting lines of theory and evidence on how to best develop indigenous technological capabilities, there is no question that such capabilities are central to technological dynamism and the capacity of innovation. Thus, at the very least, this point should be kept center-stage during discussions of improving the long-term environmental performance of technological and industrial systems in developing countries.16 Further research may also be desirable to better understand how developing countries can be effective in promoting TC building, especially in an era of evolving trade regimes and emerging new technologies (214, 219).

With the above discussion as a backdrop, it should be understood that developing countries' environmental problems cannot be solved by buying "environmentally sound technologies" off-the-shelf. In consonance with this, Agenda 21 views environmentally sound technologies as "not just individual technologies, but total systems which include know-how, procedures, goods and services, and equipment as well as organizational and managerial procedures. This implies that when discussing transfer of technologies, the human resource development and local capacity-building aspects of technology choices, including gender-relevant aspects should also be addressed" (Ref. 220, ch. 39). Unfortunately, all of this is easier said than done, and the term environmentally sound technology is often reduced to a technical fix that will solve environmental problems (202). In practice, many of the nontechnical issues fall by the wayside in acquisition of technologies, especially since "success" requires a proactive approach by firms, governments, and a whole range of institutions in developing countries.

As suggested above, an examination and analysis of successful new industrializers suggests a vital, activist and often interventionist role for the government. This view goes against the current framework within which donors operate that relegates the government to an enabler, leading to disagreement between many of the lessons from the research literature and some policy prescriptions of donor institutions (see e.g. 221A). Thus, it may be that the policy framework within which donors operate may not always support their stated goals of CD. Discussions of this are absent from the CDE literature, which suggests a need for reflexive thinking in donor institutions and their researchers.

Turning the Lens Around: Developing Capacity in the North

Most of the analysis, literature, and efforts underlying CDE operate on an implicit assumption: that managing the environment in developing countries requires building capacity only in these countries and that industrialized countries offer the models to be imitated (222). Thus the emphasis is almost invariably on activities and actions in the South-on improving environmental policies, on improving public awareness, and on integrating economic and environment policies there. In a sense, this is not surprising-seen from the South, the current process of development is often characterized by a lack of reflexivity on the part of the "developers"; and hence Northern policies, and their impact on development and the environment in the South, do not receive much scrutiny in mainstream analyses. This has led to a near absence of analysis of the influence of the policies of Northern governments and the framework and operation of international institutions on the South's environment and its capacity for managing the environment, and by extension, on what capacity needs to be developed in the North to protect the environment in the South.17 But given the extensive linkages between the North and the South, through the continuing globalization of economies and policies and through the global environment, developing capacity in Northern countries to analyze these connections, and then act accordingly, could be of great value to the South in its struggle towards environmental sustainability.18 The rest of this section will briefly dwell on some interactions between government policies in the North and the environment, and capacity for the environment, in developing countries. In the area of agriculture, for example, global markets couple the North and the South. Over the past few decades, this coupling has become tighter even as the overall structure of these agricultural markets has evolved. Economies that have tended to depend less on agricultural trade have made the largest gains in global agricultural market share while economies that are more firmly based on agriculture have lost markets. Thus, developing countries' share of the worldwide agricultural exports has declined over the past few decades; between 1961 and 1993, their share has declined from over 40% to 27% (225). Aggressive export subsidies by industrialized countries have played a role in this shift (225).

The real agricultural prices of developing country exports (on a per-unit basis) in the international markets have declined consistently and substantially since the world food crisis years of the early 1970s. In 1979-1981 and 1993, developing countries' net barter terms of trade deteriorated by nearly 40%. Government support and protection for the agricultural sector in industrialized countries, which provided incentives to production well above those provided by the international markets, combined with their export subsidies, were substantial contributors to the decline in terms of trade, despite productivity gains, for developing countries (225). It is estimated that formal subsidies to agriculture in OECD countries amounted to $335 billion in 1995 (226). It is also suggested that oligopolistic control of international commodities markets by large trading corporations may play a role in the decline of commodity prices (227).

Another reason why the prices of developing-country exports have declined is that many of them have had little success in shifting from unprocessed primary commodities towards value-added products. Import protection in industrialized countries in the form of tariffs and tariff escalation is a major constraint to the vertical diversification of agricultural products by developing countries. Such barriers to competing in, or gaining, international markets impose significant economic losses to, and thus constrain, developing countries' agricultural sectors. The resulting financial situation is unlikely to promote the long-term perspective required of sustainable agriculture. If anything, such conditions are likely to propagate the inefficiency of developing country agriculture (228).

Subsidies to the energy and transport sector in OECD countries are also substantial. Although exact estimates are quite difficult-single assumptions can add up to tens of billions of dollars to the total (226)-research suggests that the energy sector receives annual subsidies of about $70 billion per year (83, 226) while road transport receives subsidies of about $550 billion per year (83). These subsidies have a direct and significant influence on the consumption patterns of individual and industrial actors in these sectors, and eventually on the global environment through their GHG emissions. The transport sector, for example, is the fastest growing contributor to greenhouse gas emissions from industrialized countries. Industrialized countries are responsible for over two thirds of the global GHG emissions on an annual basis (and even more on a historical cumulative basis). The trajectory of industrialized country GHG emissions, and the role of subsidies in influencing this, are important issues for developing countries because these countries are likely to suffer substantial environmental impacts from a changed climate.

In fact, consumption patterns in the North are responsible for substantial amounts of natural resource flows internationally. Although some attention has recently been paid to assessing and quantifying these flows (229), the implications for the South of this aspect of the Northern "ecological footprint" have not been well studied.

Northern policies can also influence the development of environmental capacity in the South. For example, Section 3.2 discusses the export of pollution-control technologies (or even cleaner technologies rather than a systematic CP paradigm) to developing countries, often with direct consequences for pollution prevention programs. With the promotion of the former, and capital investment into such technologies, the possibility of a lock-in effect increases, which could have implications for environmental management trajectories that are inimical to the kind of capacity that is more desirable in the long term.

At the same time, the nature of financial flows between the North and the South has changed substantially over the past decade or so, with the bulk now being accounted for by private foreign direct investment. Public development aid dropped from $56.9 billion in 1990 to $47.9 billion in 1998 while over the same period, private flows to developing countries increased from $43.9 billion to $227.1 billion (230). Aid contributions from OECD/DAC member countries are now at their lowest level ever: As a percentage of their combined GNP, it dropped from 0.34 in 1992 to 0.23 in 1997 (231), well below the 0.7% target discussed at Rio. Overseas Development Assistance is important because it can serve a catalytic function for CDE, and the reduction of these funds can impede CD efforts, especially as financial resources are often the biggest constraint to capacity building. Increased levels of foreign direct investment may present a mixed picture for CDE; as mentioned earlier, multinationals with better technologies, environmental management systems, and training programs may not only provide positive spillover effects for suppliers, but also may spur better environmental management in other national firms. At the same time, however, increased foreign direct investment may also lead to increased production and consumption of polluting goods (232). The pressure to attract foreign direct investment, however, does not seem to lead to a "race to the bottom" (except in some cases); in fact, countries that have transparent and effective environmental programs are often quite successful in attracting private investment, and hence this may spur CDE efforts (232). The environmental effects of portfolio investments and debt are not well studied, although it is argued that the higher returns and shorter planning horizons encouraged by speculative capital flows may be biased against accounting for long-ru n effects of environmental unsustainability (232). In a related vein, a recent study suggests that the financial crisis in Indonesia had an adverse impact on the environmental performance of industry (233).

Capacity retention is documented as a serious problem in the South-the strengthening of human resources is reversed through attrition of personnel with a concomitant loss of embodied knowledge and expertise. This is a particular problem for many African countries (14, 234). Although the migration of highly skilled workers to the North (the "brain drain") and its consequences for the South have been acknowledged as a policy issue of significant import, there has been little attention paid to furthering a better understanding of the problem or of ways to mitigate it. The available data clearly suggest that the educational composition of total migration is skewed towards the better educated, and can cause a substantial loss of human resources. For example, immigration rates of those with a tertiary level education exceed 50% for many countries-in the case of Guyana, 70% of those with a tertiary education have moved to the United States (235). Immigration policies in OECD countries tend to favor better-educated people, and therefore are responsible in part for the migration of highly-educated workers (235). The relevance of this phenomenon to capacity retention is direct and clear, yet seldom mentioned and little researched.

This section has focused on the adverse impacts of industrialized country policies on the South's environment and on CDE. This is not to say that there are no beneficial impacts of Northern policies in this context, but it is more that a focus on the former is lacking in the literature although such analysis should be part of the debate on how to improve the environmental conditions in the South.

Developing Capacity in the Capacity Developers

Donor Agencies

Donor agency policies can have a direct and substantial influence on CDE efforts. Perhaps the best example of this is the tying of aid, which refers to the (common) practice by donor agencies through which recipients are restricted in how they can allocate the resources received as grants or loans. There are three kinds of aid tying: (a) to specific commodities/services to be procured, (b) to specific projects/programs, and (c) to the country/region where procurement has to take place (236). These can have a number of direct and indirect costs to the recipient, including an increase in project costs by 20-30%. In 1996, 22% of the total ODA (excluding technical cooperation and administration costs) from the DAC was tied (237). Over 25% of the total donor aid from the OECD in 1996 was in the form of technical assistance (237). A major portion of technical assistance is normally tied (236).

Although the OECD's DAC has called for an end to the tying of aid (238), the practice continues unabated in large part because "foreign aid is big business for donors' consulting firms and universities and their lobbies at home are insistent on their own firms implementing aid activities" (239).

Tied aid can adversely affect CDE in many ways. In the case of environmental projects, such a practice may prevent local providers of environmental goods and services from competing for projects. To the extent that they may have a chance to participate, it will often be after the initial project scoping, design, and coordination phases, though participation in these can impart important skills. Perhaps the aspect of ODA that can be most adverse to CDE is technical assistance, i.e. provision of technical consultants and experts. Whereas sometimes foreign advisors do serve important functions that cannot be carried out by local experts, the "international" rates paid to them use up a substantial fraction of the project cost19-local experts can be hired at much lower "domestic" rates. Hiring locally also supports technical capacity in the recipient country that promotes local ownership of the project. Domestic experts would also have a better grasp of the local context, thereby potentially increasing the effectiveness of the advice. [However, as Cohen (239) points out, this is a two-edged sword: Local experts, once exposed to donor agencies and their higher salaries, may have little motivation to stay in the country or work for low-paying public sector jobs, leading to a retention problem that can severely hamper local professional capacity building.]

Reliance on Northern experts is not exclusively a practice of bilateral agencies. A recent survey of the countries in receipt of funding for biodiversity projects from the Global Environmental Facility revealed that often more than half the project costs were taken up by the World Bank's consulting fees (240). A World Bank evaluation of some of its own agricultural projects found that large numbers of technical assistance personnel included in the projects reflected initiatives from the Bank during project design, sometimes against the stated reluctance of the recipients; furthermore, evidence from this study indicates that in most cases, expensive foreign experts are not worth the price (241). As Edoho (Ref. 14, p. 247) comments rather bluntly on the overgenerous salaries and fringe benefits for these experts, "[E]xpatriates who live a modest life in their home countries always live like a 'pope' under technical assistance schemes in Africa, especially if technicians can easily pass for engineers, and supervisors qualify as managers." Such practices waste resources, cause resentment and frustration in national personnel, do not lead to sustainable capacity, and at worst, can be capacity destructive (4).

Ironically, even research on CD for the South is rarely supported in the South. Almost all the studies of CD and reviews of existing programs, which have been funded by multilateral and bilateral donor agencies, have utilized internal experts or external Northern consultants to analyze the issue. For example, the substantial UNDP effort on CD was anchored by a major (now commonly cited) study by the Harvard Institute for International Development. This geographical skew in the analytical effort is reflected in the bibliographies of the discussion papers and articles reviewing aspects of the capacity issue (see, e.g. 15, 16, 34 as well as this article). In fact, support for research activities on capacity building (and hence the ability to develop research capabilities) often seems to follow the dictum that "those who have, get more of it," leading to a situation in which a few prominent research institutions/individuals dominate the research arena. It is not even a matter of the South lacking expertise on the issue (who better to understand the capacity issue, especially the constraints to developing capacity, than Southerners?), but the entry barrier to the funding circles for such work is often a high one. Donor agencies need to broaden their criteria for selecting experts, and Northern academics and consultants need to cede their dominance in research on, and for, capacity building if they are sincere about CDE, and allow Southern researchers more opportunity to participate in this process as equals (which ironically is one of the supposed goals of CDE programs).

To the extent that there is some analysis that turns the lens around, it focuses on the effectiveness of donor agencies in developing environmental capacity in developing countries through their own evaluation processes. A recent comprehensive review of donor activities for institutional CDE is rather unflattering (34). It highlights a number of shortcomings on the part of donors to carry out CDE or even to promote environmentally sustainable development. Among other things, it suggests that:

* while donor agencies have successfully integrated environmental concerns in their overall policy statements and strategies, and established some capacity (such as the establishment/strengthening of specialized environment units and funding frameworks, enhancing the role of environmental specialists in policy-making, and building up, and stressing, institutional development expertise) to carry out CDE, there is still a lack of translation into practice in many aspects of program design and implementation. For example, the World Bank and Sida are the only agencies that require environmental assessments as part of the project. Thus, "quite simply, development assistance agencies fail to set the examples of good practice that they advocate in wider policy circles" (Ref. 242, p. 90).

Furthermore, the continued existence of separate environment funds within agencies goes against the integration of environment and development envisioned by the World Commission on Environment and Development and the Rio Summit. Donor-supported CDE initiatives have been undertaken in parallel with the more generic CD activities. All of this suggests "a fundamental policy inconsistency among donor agencies" (243). It remains to be seen to what extent the environmental approach of the donors will follow the CDE framework and to what extent it will remain old wine in new bottles.

* Mainstreaming of the environment into sectoral projects has been far from adequate. In the case of water projects, environmental impact assessments have been mostly ignored in the planning and implementation phases, and the emphasis is more on development, rather than management, of water resources. In the case of energy projects and programs, donor (and recipient country) focus has been on commercial energy production rather than on traditional or renewable energy sources or demand-side management, and donor support has been driven more by global, than local, environmental considerations. For example, only 3% of the UNDP energy-related commitments between the mid-1970s and the mid-1990s were directed toward renewable energy (244). Although 2 billion people in the world depend on traditional fuels for their energy needs (245), efforts to improve energy services to them are minuscule compared with the global energy R&D budget (246). A World Bank-led Regional Programme for the Traditional Energy Sector has been initiated to focus on this gap.

* Funding for environmental activities still remains far below the amounts envisaged at UNCED. "Environment" constitutes 8-15% of donor bilateral assistance programs. In the case of the World Bank, environmental projects constituted only 1.3% of cumulative lending in 1996 (247). The problem is further exacerbated by the declining levels of ODA.

* Although a number of donor agencies have attempted to improve their performance through institutional reforms, many of them are still hampered by deficiencies. These include a lack of learning20 even with a surfeit of information (partly because of the lack of systematic organization of this information), lack of intra-institutional coordination, limited capacity to specialize in environmental issues or "mainstream" environment into broader programs, and an unduly narrow focus on environmental organizations and channels in recipient countries (although targeting other channels may be equally, or more important, in improving environmental performance-for example, cooperation with industry agencies may be more effective at preventing pollution than similar efforts directed at environment agencies) (248). There is also insufficient attention paid to learning, and disseminating, lessons learned from past experiences. As Potsma says, there is an insufficient amount of looking backwards and looking within when it comes time to discuss plans by which to enhance a given organization's capacity (249). Monitoring and evaluation of CDE programs has also been wholly inadequate and is in an "indicator vacuum" (34), although there have been attempts to develop measures of program performance through the development of suitable indicators (13).

Overcoming these constraints will require significant change in the way donor agencies think, not only among project officers, but also in the top management. It will require the development of new tools and instruments to (a) carry out, for example, institutional assessment and process-based project evaluations, (b) move the debate and the process from being donor-driven, as it currently is, to being driven by developing countries, and (c) learn to deal with local politics (26).

On a related note, the dominant influence of donor agencies on the policy research agenda on various issues relevant to the South needs to be highlighted. These institutions govern the direction and nature of such policy research, either through in-house capacity or through disbursal of funds for external research programs. As Alice Amsden says, "[W]hat makes the World Bank so powerful is that it has no real rival. The Bank has become a virtual monopoly, if not in its lending, then in its research work" (Ref. 250 p. 633). Thus, the ability to build capacity in developing countries to carry out their own research and design their own policies and programs may depend on the ability of donor institutions to cede their monopoly on the research and analytical activity that underpins any long-term capacity. Alternatively, they could build on the suggestion to use their "huge salary base to send one or two thousand professionals to work at the country level, earning generous local wages while undertaking relevant research and participating on equal terms in national debates" (Ref. 250 p. 152; emphasis added). The Bank and many other donors have moved toward such decentralization to some extent, but the power of these institutions still gives them an unequal footing in the debates. Further introspection and action by donors may tell us how willing they are to reform themselves in order to increase their effectiveness in developing capacity in developing countries.

International Science and Policy Institutions

International institutions can serve a crucial catalytic function in strengthening scientific and policy research capacity in developing countries, not only by helping raise resources for programs there, but much more through the strengthening of Southern involvement in international efforts and the increased participation of Southerners in international networks.21

However, the formation and implementation of the major international research programs (i.e. the International Geosphere-Biosphere Program, the World Climate Research Program, and the International Human Dimensions Program), have been dominated by scientists from industrialized countries who typically constitute 80% of the participants involved (253). This can lead to a reduction in the effectiveness of these programs: Global research programs do not necessarily reflect or address regional needs and may not be appropriately implemented at the national and regional levels. Whereas developing countries remain dependent on scientific findings and policy advice from industrialized countries, they may not always trust such information and/or analysis (253).

A recognition of this situation, in fact, has been the impetus behind the START initiative, whose mission is to develop a system of regional networks of collaborating scientists and institutions to (a) conduct research on regional aspects of global change, (b) assess the impacts of regional findings, and (c) provide regionally important integrated and evaluated information to policy makers (253). Ironically, 7 of the 9 members of the START Scientific Steering Committee are currently from industrialized countries.

The complexity and uncertainty associated with most global environmental problems call for heavy emphasis on scientific assessment and analysis. However, scientists do not illuminate, inform, and make recommendations in a vacuum-the conduct of science is shaped by politics and culture, and the national origin of the analysts matters (254).22 If science and analysis are shaped by societal and cultural variables, as a social studies perspective on the creation and use of knowledge suggests, then a multiplicity of voices from different countries and cultures is essential to illuminating the range of perspectives that shape the science, knowledge, and analysis relevant to the climate debate.

In the climate issue, uneven participation in the Intergovernmental Panel on Climate Change, the authoritative body that seeks to provide a state-of-the-art review of issues in the climate debate to the global community, is well known (even though the extent may still be surprising; Table 1). Developed-country experts far outnumber participants from developing countries, and there are too few objective criteria for selection of participants. In fact, the selection process remains somewhat arcane-networks and personal connections still often seem to influence invitations for participation. To be fair, there has been some effort at improving the situation in the Third Assessment Report, although there is still a long way to go.

Table 1. Number of participating authors in Intergovernmental Panel on Climate Change (IPCC) reports by countrya

Report USA India China UK Total
IPCC 1990 110 5 8 62 210
 Working group I
IPCC 1995 210 5 7 61 512
 Working group I
IPCC 1995 212 20 20 60 582
 Working group II
IPCC 1995 30 7 2 5 97
 Working group III

aThere are many levels of authorship in IPCC reports; among the numbers in this table do not distinguish among them. Since an individual author may have been involved in more than one chapter of a working group, there may be some inadvertent double counting of authors. Thus, actual numbers may be slightly different from those above (145).

To address this participation gap, international programs have to focus on understanding some of the structural and other reasons why such a bias exists and what its implications are likely to be. Some reasons may be obvious, such as a difference in resources; some may be less obvious, such as policy and administrative expectations and requirements assumed by international programs, which reduce the scope of efficient interaction and administrative compatibility with international programs). Two major gaps that may result have been pointed out elsewhere (145).

The Relevance Gap

This is an imbalance in the kinds of research performed and a limited focus on issues of relevance to developing countries. It creates an international research agenda that may exclude the needs and concerns specific to these countries and reduces the motivation for their researchers to participate in international efforts.

The Perception Gap

This is a variance in the views of Northern and Southern analysts about the roles of research, analysis, and assessment in the international discussions surrounding issues such as climate change. This variance in the perception of international efforts in terms of "What are we doing?" "Why are we doing it?" and "How are we doing it?" may have serious implications for acceptability of international efforts.


The need for appropriate capacity to protect and manage the environment in the South has gained recognition in tandem with the more generic discussions on capacity building sweeping the development arena. Topics such as the kind of capacity necessary and the levels at which it is required, as well as the ingredients helpful (and often critical) to the success of CDE approaches have been much discussed. The CDE literature has also explored various facets of the capacity issue-the actors involved and their functions, the resources required, the societal and normative context, and principles to improve the effectiveness of CDE activities. Such explorations have resulted in a welcome shift of the focus from the outcome alone to the capacity development process itself and have strengthened the foundations for the next generation of CDE efforts. Yet there is a long way to go before environmental protection and management in most developing countries reach satisfactory levels.

Almost all of the systematic research and analysis on CDE has been donor driven, and the logical focus of these efforts has been on the activities of such agencies. Whereas this has been useful in reorienting the thinking among donors that are assisting developing countries in improving their ability to manage the environment, the discussions have often dwelt on broad concepts rather than on the specifics relevant to various sectors and environmental issues in the developing country and the global context. The detailed examination in this review of two economic sectors and a global environmental problem highlights some basic cross-cutting issues that merit further consideration in the CDE literature and efforts. Of particular importance are the capabilities for technological and policy innovation in order to allow the adaptation and development of technologies and policies suitable for local needs and conditions.

Even though the management of the environment requires appropriate policy frameworks, institutions, and technologies, at present these are often far from optimal in developing countries. The real challenge therefore is to develop the capacity to protect the environment under current constraints. There are numerous existing endeavors-many of them local in origin, design, and implementation-that are attempting to manage the environment in developing countries through innovative approaches. Although the experiences from some of these have influenced the thinking on environmental protection, further research focusing on a range of such approaches, with the aim of learning how to build and maintain appropriate capacity, and the utilization in CDE efforts of any insights gained, will likely yield rich dividends.

Currently, most CDE efforts have been launched or driven by donor institutions. These look mainly Southwards and ignore the linkages between Northern or international policies and the environment or environmental capacity of the South. However, improving the environment in developing countries requires capacity not only in the South, but also in the North to examine and reorient domestic policies there that have impacts in the South. The relevance of capacity to "turn the lens around" needs to be reflected in the research agenda on CDE. To the extent that there has been some reflexive analysis within the CDE literature, it has focused on donor activities to improve their effectiveness by making them more consistent with the CDE framework (a topic of great relevance, given that current donor activities in the environment area are not often in accord with this framework). Further attention to developing capacity in these agencies as well as other capacity development organizations would be useful. The role of the private sector in relation to CDE also needs to be more completely examined.

Ultimately, for programs on CDE (and for that matter, CD in general) to succeed, it is necessary for the importance of environmental capacity to be recognized and accepted in developing countries, but this is not enough by itself. It is essential for the South to give primacy to this issue and treat it as fundamental (rather than as an add-on or supplementary activity whose shape and size depend on the whims of donors) to its efforts on environmental management, and to take the lead in developing appropriate capacity. Indigenous support for capacity activities should be forthcoming, as should support for analytical research to develop effective programs that can assess, target, and fill existing gaps. Ultimately, the development of useful and sustainable capacity in the South will require a systematic effort driven by government and other organizations there. For this, the onus lies on the South.


I would like to thank Michele Ferenz and Ged Davis for valuable comments on previous versions of this manuscript.


1 The efforts to strengthen the abilities of countries to carry out the tasks targeted institution building in the 1950's and 60's, focussing on the design and operation of institutions, but little attention was paid to political or cultural context or to non-public organizations in the private profit or independent sector. By the late 1960's, and the early 1970's, due to the below-expectation performance of existing institutions, the development community emphasized institution strengthening for existing individual organizations mainly using techniques to improve the internal functioning through better management or training/upgrading of human resources. This was followed by the concept of development management for improved administration and implementation of development programs by concentrating on delivery systems of public programs, especially to reach "special publics." This was a reaction against the perceived "top-down, trickle-down" approaches and emphasized responsibility of states to induce and manage socio-economic change. Importantly, this approach involved more strategic thinking and political content than its predecessors, and displayed evident interest in involvement of non-governmental organizations and community-based organizations. By the 1980's, an institutional focus had appeared again through the approach of institutional development, this time the focus being on organizational effectiveness as an outcome of the interaction between internal management and external (domestic and international) environment. This recognized the long-term process of restructuring and change and highlighted the links be tween sectoral and macro-policy issues as well as recognizing broader sectoral perspectives and the critical roles of groups/networks or organizations (3, 4).

2 This increasing attention to the issue of capacity has led, apart from attempts at integrating it into mainstream donor activities, to a veritable explosion of programs and networks focusing specifically on the topic. Among these are the OECD/DAC Informal Network on Institutional and Capacity Building, the Global Development Network supported by the World Bank, the International Forum for Capacity Building to strengthen Southern NGOs (emerging from the International Working Group on Capacity Building consisting of bilateral and multilateral donors and Northern and Southern NGOs), and the Capacity Building and Governance program at the European Center for Development Policy Management.

3 In fact, the CD issue has theoretically been on the development cooperation agenda since the early 1980s (8). Fukuda-Parr suggested that the policy statement adopted by the UNDP "New Dimensions of Technical Cooperation" redefined the technical cooperation agenda to focus on measures to enhance the utilization of resources rather than augment the level of resources through budgetary aid, capital investments, food aid, etc. However, this did not shape the operational activities of the UN or influence the rest of the development community to pursue capacity building. Even where the objectives of capacity building were adopted by the agencies, the tools stayed the same (9).

4 At the same time, there has been a growth of academic and gray literature on the capacity issue (see, for example 3, 11, 12, 13, 14, 15, 16), driven in part by the changing focus in donor, and other international agencies (and hence the patterns in development research funding), and in turn, giving further impetus to the issue.

5 As some observers have pointed out, even the generic CD concept is still elastic and ill-defined (15, 26). Bossuyt refers to it as a "slippery concept" (26). Whereas this is not surprising given the still-evolving understanding of the issues involved, it is felt that the concept has been diluted by definitional expansion through careless use of the term (27). Given that much of the CDE literature draws on the CD concept, it also tends to sometimes fall into the same trap.

Even at UNCED, though, the focus on capacity came late in the day, only following the discussions on technology transfer (30).

7 The difference between sustainable and conventional agriculture can be illustrated by the approach to reducing crop losses to pests. Whereas conventional agriculture relies on pesticides, sustainable agricultural production generally relies on an approach to managing (rather then controlling) pests, with the recognition that some losses to pests are unavoidable. This approach utilizes a judicious combination of biological control processes-natural enemies, improved plants, etc-and chemical and physical inputs applied at appropriate times and scales, as well as cultural practices, such as intercropping [hence the term integrated pest management (IPM) (56, 57, 58, 59, 60)]. Unfortunately, in practice IPM is often reduced to a focus on single technologies rather the intelligent integration of multiple options (61).

8 Industrialized countries have been cutting back on their contributions to the international agricultural research system, despite having reaped substantial benefits from the productivity gains obtained from crop varieties produced by the Consultative Group International Agricultural Research (CGIAR) centers using germ plasm often obtained from developing countries. It is estimated, for example, that the $71 million of US government support of wheat improvement research at CIMMYT (a CGIAR center) since 1960 has led to a gain for the US economy of between $3.4-$13.7 billion (74).

9 As Guerinot (Ref. 88, p. 243) states in the context of nutritional enhancement of foods, "the road to better nutrition is not paved with gold and hence agribusiness has not centered its efforts on the nutritional value of food."

10 On the face of it, exports of cleaner industrial technology would seem to embody a pollution prevention approach, but this may often be reduced to a market opportunity to sell hardware (albeit less polluting than what it replaces) rather than the promotion of integrated, long-term pollution prevention (see, e.g. 136, 138).

11 Access to information has been acknowledged as an important tool in managing and protecting the environment (161). The need for environmental information management capacity, highlighted by UNCED in Article 40, is receiving some attention among donor agencies. For example, UNEP has established the Environment and Natural Resources Information Network to help collate, store, manage, and disseminate environmental information and data in developing countries and to assess environment and development issues for decision-making, policy-setting, and planning (162). Similarly, one of the main aims of the UNDP Capacity 21 program is to assist countries to gain access to information and use it in the decision-making process. Still, little is known about the utilization of information in the policy process in developing countries or how information resources and capacities are best organized to ensure effective use of information in policy decisions, although preliminary exploration suggests that the potential of information in policy-making is not realized and that information use is hampered by inadequate capacities (159).

12 This has been generally true for all models of "development" where the prescribed policies reveal the preferences of the developers (whether experts from the North or domestic elites in the South). Redclift (182) quotes the famous anthropologist Claude LÚvi-Strauss in this context: "[cultures] appear to us to be in more active development when moving in the same direction as our own, and stationary when following another line" (183).

13 The African Capacity Building Initiative, aimed at building and strengthening local capacity in policy analysis and development management in sub-Saharan Africa, is perhaps the most prominent example of a capacity building initiative that recognizes the importance and role of indigenous capability to carry out policy research in developing countries (198). Another example is the African Economic Research Consortium (AERC), established in July 1988 to strengthen local capacity for economic policy research in sub-Saharan Africa, and funded by a variety of multilateral and bilateral donors, as well as foundations. More recently, the Global Development Network, also launched by the World Bank and supported by it, the UNDP and other institutions link research and policy institutes involved in the field of development. The initiative is still in its early phase.

14 In the case of AERC, for example, the network of researchers grew from 40 initially to over 150 by 1997; AERC-supported research has increasingly appeared in international and regional journals signifying the professional maturation of the network; university departments participating in the AERC graduate degree program had trained over 200 students by 1997 (189). As one evaluation put it, it would be "no exaggeration to claim that the Consortium provides the cement holding African economic research together" (199).

15 For example, the AERC has adopted a cautious approach to expanding its own thematic research areas beyond the topics of medium-term macroeconomic adjustment to longer-term development issues.

16 Notably, whereas most of the TC literature has emerged from a study of industrialization, and hence tends to focus on industrial production, much of the discussion on TCs can also be applied to other sectors, such as agriculture (214). In fact, given the greater location-specificity in the case of agriculture, TCs may even play a more critical role in agriculture than in industry.

17 CDE in Northern countries is also needed for protecting their own domestic environments (222).

18 In a related context, Julius Nyerere, when asked how Oxfam could best help Tanzania, replied, "Take every penny that you have set aside for Tanzania and spend it in the United Kingdom explaining to the people the facts of poverty and its causes" (224).

19 Cohen (239) gives an example of a USAID soil and water conservation project in Kenya where the expatriate advisor salaries and contractor profits consumed 56% of the project budget.

20 Many agencies are characterized by "single-loop" learning rather than "double-loop" learning (248). Single-loop learning consists of learning from mistakes by detecting and correcting errors in ways that allow an organization to maintain its policies and norms, whereas double-loop learning involves modification of the organization's policies and norms so as to avoid future errors (248A).

21 Donors rarely pay attention to building research capacity in developing countries. Sweden is an exception-commitment to research capacity building has been integral to its overseas aid program for many years through the Swedish Agency for Research Cooperation (251).

22 Although the relation between science and policy is often presented in the simple terms of "speaking truth to power," in reality it is rarely so (254). As Jasanoff (253) states, one can question the notion of science as an "impartial adjunct to policy" at several levels: the objectivity of science ("science speaks truth"), the nature of the policy-making process ("science speaks, and politics accepts, the truth"), and the ability of science to define the truth as being independent of the power that turns to it for guidance.

If you have come to this page from an outside location click here to get back to mindfully.org