Farrington, 1991). As they seek solutions, policymakers must ... market intelligence, farm management, and in negotiating ... thousands more farmer technicians or leader farmers .... percent with an intermediate diploma or certificate. .... with comp
Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce ...... concept of an innovation system has guided this more holistic approach to planning knowledge ...... business and entrepreneurial s
Nov 2, 2006 - (www.rimisp.org) to the preparation of the World Development Report 2008 âAgriculture for Developmentâ. This work ..... Ecological â the core concerns are to reduce negative environmental and health externalities, to enhance and u
example, the percentage of the national budget allocated to extension (the National Agricultural Advisory ...... Over the years, the size of the public and NGO extension systems has fluctuated from around 600 to .... female farmers as well as differ-
A.3 Typology of Attitudes and Practices Affecting Key Innovation Processes ...... Organizing principle. Using science to create inventions. Accessing agricultural knowledge. Finding new uses of knowledge for social and economic change ..... sectors s
Jul 2, 2007 - public extension; the frequent encumbrance with public duties in addition to those related to knowledge transfer .... advisory services can play an important role in promoting the spread of farmer-based innovations. ..... decentralizati
After independence the government steadily increased its control over the sector by promoting state-linked co-operatives ... ONCAD, performing poorly and laden with debt, was disbanded and its functions divided among other organizations, ...... tors
Feb 23, 2006 - 37. Production of basic food crops is the key driver for increased agricultural production. Basic food crop production grew at an average rate of about 3-4 percent annually. Maize and millet production showed the highest increases, fol
Reforming Agricultural Trade for Developing Countries, Volume 1: Key Issues for a Pro-Development ... Sustainable Land Management: Challenges, Opportunities, and Trade-Offs ...... handled by Patricia Katayama, Aziz Gokdemir, and Nora.Missing:
for the Consultation of. âModernizing the World Bank's Operational Policy on Guaranteesâ ... World Bank Guarantee Products and ..... EPC Contractor: Salini/ ...
responsibilities among management, the board of directors, controlling ... BANK GOvERNANcE LESSONS FROM ThE FINANCIAL CRISIS ..... They also run the risk of being perceived as short-term politi- cal gestures if there is no policy or institutional fra
the sample, it does suggest that these excesses did not play ...... international financial openness is defined as the sum of its gross stocks of foreign assets and.
diversity of views and the intensity of their expression make a review of the available evidence both necessary and timely. Because of the considerable overlap between community-based and community-driven development projects, evaluation evidence is
In this section we analyze the various contributions of Cooperative Game Theory to ... The main management instruments in regulated open access are ... 1 The EEZ was instituted in the Law of The Sea Convention signed at the end of the U.N. Third ....
They kill people, destroy infrastructure, weaken institutions, .... internal opposition groups without intervention from other states; internationalized internal armed ...
conditional political and income effects predicted by the theory are also significant and large. For example, both ... being central to the earthquake resistance of adobe homes. .... none of these quakes reached a 7.0 on the Richter scale. Japan ...
The World Bank. Number 8. Cultural Heritage in. Environmental Assessment. Cultural heritage, also termed cultural property, cul- tural patrimony or cultural resources, can be defined as the present ... Cultural heritage is legally protected in almost
of investment alternatives from an environmental and ... sence of such consideration, those steps in the project ..... described in the EA report prior to appraisal.
Jul 12, 2000 - sexual behavior and is a multidimensional and dynamic concept. Explicit and implicit rules imposed by society, as defined by one's gender, age, economic status, ethnicity and other factors, influence an individual's sexuality (Zeidenst
ficiency and corruption, delivering services through rural kiosks ... government-run departmental kiosks or ... model is based on self-service through the. Internet ...
Aug 22, 2004 - Haverkamp, D.R. Harris, and others and The Breastfeeding and HIV. International ..... Horton, Susan, Tina Sanghvi, Margaret Phillips, John Fiedler, Rafael Perez-. Escamilla ..... Association 102(3): S81âS85. Strauss, John ...
coastal waters and estuaries. It is intended to assist those involved in managing coastal wastewater discharges and planning appropriate levels of pollution ...
(in the broader sense) in Dhaka is similar to Kolkata, less than in Mumbai, and greater than in three other major Indian cities, Delhi, Chennai, and Bangalore (Figure 3.1). Figure 3.1: Estimated proportion of population living in slums in Indian citi
AGRICULTURAL KNOWLEDGE & INFORMATION SYSTEMS (AKIS) GOOD PRACTICE NOTE
Decentralizing Agricultural Extension Lessons and Good Practice Prepared by the AKIS Thematic Team with inputs from Gary Alex, Chris Gerrard, Derek Byerlee, C. AnnorFrempong, Daniel Moreau, David Nielson, Dely Gapasin, Eliseo Ponce (Philippines), Graham Kerr, Jaakko Kangasniemi, Jaakko Kangasniemi, Joko Budianto (Indonesia), Mahamood Kamara, Marie-Hélène Collion, Matthew McMahon, Melissa Williams, Solomon Bekure, and Willem Zijp
Work in progress for public discussion
The World Bank Rural Development Family Agricultural Knowledge & Information Systems (AKIS)
Agricultural Knowledge & Information Systems is a thematic team focusing on agricultural extension, education, and research within the Rural Development Department of the Environmentally & Socially Sustainable Development Network of the World Bank.
List of Acronyms
The Case for Decentralization Principles in Decentralization Reform Coproduction of Goods and Services Regional Governments vs. Local Communities National Frameworks for Decentralization
2 3 4 5 5
Issues with Decentralization of Public Sector Extension Recognizing Multiple Extension Functions Administrative Decentralization Decentralized GovernanceIntroducing Accountability Fiscal Decentralization of Extension Services Transition to Decentralized Systems
6 7 8 9 10 11
Good Practice for Decentralization in AKIS Projects Centralize or Decentralize Programs As Appropriate To the Service Adapt Strategies to Local Institutional Environments Strengthen Central Support Services for Extension Provide Mechanisms for Policy Formulation in Mixed Systems Expect to Continue Public Sector Financing Fiscal Transfers for Research and Extension Plan for Transition and Local Capacity Development Ensure Monitoring and Evaluation of Decentralized Systems
13 13 14 14 14 14 15 15 15
Additional Readings and References
Foreword AKIS is the Agricultural Knowledge and Information Systems Thematic Team, composed of World Bank staff working in or interested in research, extension, and education programs. The overall team objective is to enhance the effectiveness of Bank support to agricultural knowledge and information system development, and thus contribute to the Bank’s objectives of alleviating poverty, ensuring food security, and improving sustainable management of natural resources. The AKIS team emphasizes policy, institutional, and management issues associated with agricultural research, extension, and education, recognizing that other thematic teams will focus on technical issues. The Team mission is to “promote the development of sustainable and productive agricultural research, extension, and education systems in Bank client countries.” In 1999, the AKIS Thematic Team identified decentralization as an important issue in agricultural research and extension systems, and made it one of the topics for the 1999 AKIS Retreat. As input to the retreat, Graham Kerr and Mahamood Kamara summarized data on the status of decentralization of extension systems, drawing from a “Study on Decentralization, Fiscal Systems, and Rural Development”. In the retreat, C. Annor-Frempong and Solomon Bekure (Ghana), Eliseo Ponce (Philippines), Joko Budianto (Indonesia), and Matthew McMahon (Latin America) shared experience with country decentralization initiatives. A discussion paper was prepared by Gary Alex drawing on the 1999 AKIS Retreat, review of other experi-
ence with decentralization, and discussions in two AKIS team “shared learning seminars”. This Good Practice Note summarizes the findings and implications for decentralization of agricultural extension services, and is intended as a contribution to the exchange of ideas and experience within the AKIS Thematic Team. “AKIS Good Practice Notes” are designed to disseminate views, experiences, and ideas that may assist World Bank Team Leaders, national counterparts from Borrower counties, and other partners to prepare and implement projects to strengthen agricultural research, extension, and education programs. The Good Practice Notes contain lessons learned from innovative experiences in World Bank projects and elsewhere, and make this information readily available for comment and use by project teams. This “AKIS Good Practice Note “ was prepared by the AKIS Thematic Team with inputs from Gary Alex, C. Annor-Frempong, Solomon Bekure, Joko Budianto (Indonesia), Derek Byerlee, Marie-Hélène Collion, Dely Gapasin, Chris Gerrard, Mahamood Kamara, Jaakko Kangasniemi, Graham Kerr, Matthew McMahon, Daniel Moreau, David Nielson, Eliseo Ponce (Philippines), Melissa Williams, and Willem Zijp.
Marie-Hélène Collion Chair, AKIS Thematic Team
Abbreviations AKIS IGFT NGO LGU MOFA NAES
Agricultural Knowledge and Information Systems Inter-Governmental Fiscal Transfer Non-Governmental Organization Local Government Unit Ministry of Food and Agriculture National Agricultural Extension System
ver the past two decades many countries have undertaken to decentralize government functions and transfer authority and responsibilities from central to intermediate and local governments, and often to communities and the private sector. Decentralization is potentially important to agricultural knowledge and information systems, but decentralization is not an end in itself, and successful decentralization strategies must address three challenges—establishing a national framework for decentralization, developing subsector approaches, and enhancing capacities of various participants for coproduction of decentralized goods and services. Agricultural extension services are under increasing pressure to become more effective, more responsive to clients, and less costly to government. Decentralization is an increasingly common aspect of extension reforms. Field extension advisory services are well suited to decentralized approaches, but a comprehensive extension system requires a range of extension support services and programs, some of which (strategy formulation, training, monitoring and evaluation, specialized technical support) are often best carried out at the central level. Decentralization strategies appropriate to AKIS projects frequently include institutional arrangements that: • Decentralize extension services where possible, with emphasis on giving users control over program planning, implementation, and evaluation. • Provide for adequate centralized support systems for decentralized extension ser-
vices, especially support for training, subject matter specialists, and production of extension materials. Adapt strategies to local institutional environments to accommodate country legal frameworks, political traditions, administrative structures, and social and agroecological conditions. Extension strategies can emphasize decentralization when there is already a strong political decentralization in the country, but should proceed cautiously when decentralization is not yet well established. Determine on a case-by-case basis whether decentralized services should be managed by local governments, community/producer organizations, or local governments in conjunction with producer/community organizations. Provide clear division of responsibilities between the different levels of government and other program participants. Develop procedures for policy formulation and priority setting in mixed systems to reconcile central government financing and policy objectives (poverty alleviation, food security, and environmental conservation) with local peoples’ priorities that emerge from the decentralized program governance. Provide for needed fiscal transfers from central government to decentralized implementing agencies to finance decentralized extension services, recognizing that over the short term decentralization rarely reduces requirements for central government financing.
Decentralizing Agricultural Extension
Structure fiscal transfers to give users maximum influence over programs and to promote institutional pluralism in service provision. This empowers users and develops capacities in a range of public and private providers, such that the most competent institutions are able to provide the services. Provide for extensive planning, promotion of the rationale and principles behind reforms, and training in new operational procedures before launching decentralization reforms.
Provide for needed investments in development of local capacity (local governments, executing agencies, community or producer groups), as such implementation capacity is critical to success of decentralization reforms. Establish effective systems to monitor and evaluate decentralized programs, and ensure that the data are available at all appropriate levels. Central monitoring should be sensitive to equity issues and the possibility of local elites capture of programs, thus excluding services to the poor, women, or minority groups.
The Case for Decentralization In a decentralized system, the folks at the bottom are in charge. This is very different from the past, and very difficult for a lot of us to accept. — Graham Kerr
Decentralization (see Box 1) as transfer of authority and responsibility for government functions from central government to intermediate and local governments, and often to communities and the private sector has become widespread over the 1980s and 1990s. Countries with diverse systems and traditions of government have pursued decentralization initiatives for many reasons, including especially the failure of government to meet expectations under centralized approaches to economic management and service provision. Decentralization is a complex phenomenon involving a variety of
approaches to organizing public administration. Though not yet widely applied to agricultural research and extension, decentralization strategies are potentially important to these agricultural knowledge and information systems. Decentralization is frequently viewed from one of two different perspectives: 1. The democratic view emphasizes the aspect of empowering local people to control and direct their own public programs; and 2. The administrative view emphasizes the efficiency gains resulting from improved administration and effectiveness of public programs due to local control. Decentralization is generally expected to: encourage local financing and ownership of programs, result in more efficient and equitable allocation of government resources, provide incentives for production and service delivery, ensure lower-cost service delivery, build local capacity, and respond more effectively to local needs.
The Case for Decentralization
Box 1 Definitions Definitions important to understanding decentralization include three reform strategies that are not true decentralization: • Deconcentration involves central government dispersing staff responsibilities to regional offices without changing the basis for authority and control. This is not true decentralization and can actually increase central control and influence. • Delegation is the transfer of responsibility for public functions to lower levels of government or to other organizations, which implement programs on behalf of the central government. • Privatization is government transfer to the private sector of managerial, fiscal, and decision-
For rural programs, decentralization offers hope for correcting the urban bias that results from the geographic dispersion of rural people, the difficulties for them to organize to promote their interests, and the discrimination against agriculture inherent in many country policy frameworks. Decentralization of agricultural extension and research seeks to increase user participation in technology programs and make programs more accountable to users. Enthusiasm for decentralization needs to be tempered with some caution. In small countries, decentralization may be unnecessary and in very large countries decentralization to the state or provincial level may still leave programs distant from user influence. Definitive evidence of the impact of decentralization is limited and not everyone benefits from any reform. Furthermore, decentralization does little to improve intraregional disparities, may bring oppressive elites into power, and can lead to greater inequalities in allocation of government resources. Thus, decentralization has the potential to increase access to and cost of services, but specific targeting mechanisms and strong central oversight are needed to avoid inequities in service access and quality.
making control, while retaining normal regulatory authority. True decentralization is considered to involve a mix of three reform strategies: • Administrative decentralization is the transfer of authority over regional staff from the central government to regional or local governments. • Political decentralization (or democratic decentralization) is the selection of local government officials by local election rather than by central government appointment. • Fiscal decentralization is the transfer of responsibility for raising and spending program funds to lower level government units.
Principles in Decentralization Reform Decentralization takes many forms with varied mixes of fiscal, administrative, and political decentralization. Privatization, deconcentration, and delegation initiatives can complement and reinforce an overall decentralization policy, but these do not constitute, and can in some cases work against, effective decentralization. Four requirements for successful decentralization are: • Providing local people with substantial real influence over the local political system and local developmental activities; • Ensuring availability of financial resources adequate for decentralized institutions to accomplish their tasks; • Ensuring adequate administrative capacity in local units to carry out their tasks; and • Establishing reliable mechanisms for accountability of politicians and bureaucrats to local people. Deconcentration is nearly always the first— and necessary—step in any process of decentralization. This puts staff from central administrations in closer contact with local people, problems, and conditions and provides a chan-
Decentralizing Agricultural Extension
nel for local interaction with government. Unfortunately, decentralization reforms frequently stop at this point with central authorities retaining control over deconcentrated administrative structures. Administrative decentralization represents a more fundamental reform that replaces existing centralized structures with a new administrative structure of local government. Transfer of power to decentralized offices increases local participation in decisionmaking and allows programs to be tailored to local needs. Political decentralization makes decentralized bureaucracies accountable to locally elected officials and officials accountable to the people. Elections, referenda, and local participatory decisionmaking arrangements give people direct control over government programs, but short of these formal political processes, a variety of mechanisms (reflecting “participation” more than “decentralization”) can give people influence over government programs. These include: incorporating local representatives into governance and advisory boards, client surveys, polls, and program “report cards,” and rapid rural appraisal techniques. Fiscal decentralization is often seen as a way to reduce central government budgets by off-loading tasks a central government can no longer finance. In practice, however, decentralization is likely to result in higher costs for central budgets. Fiscal decentralization may transfer authority for expending funds, raising taxes, or borrowing, but intergovernmental fiscal transfers (IGFTs or “grants”) are usually the key means of financing decentralized programs. Concern over local administrative capacity frequently leads central governments to impose controls that are costly to administer and that restrict local flexibility in managing funds. Experience would indicate that local governments are generally capable of assuming substantial responsibility, and decentralized programs can provide different financing packages to communities with different levels of capacity.
Coproduction of Goods and Services Significantly, most government programs involve a mix of goods and services that vary in: the economies of scale or scope inherent in their production and financing; their economic characteristics (whether public, private, toll, or common pool); users’ ability to withdraw from their use (exit) or influence their production or supply (voice); and spatial and temporal considerations in the value of the good or service. These factors determine the institution likely to have a comparative advantage for production of the good or service. Many programs are best implemented through “coproduction” or partnerships between various actors—central government, local government, private sector, civil society, and the individual—each providing the good or service for which it has a comparative advantage. Coproduction requires clarity in division of labor and clear “contracts” between different partners. Privatization, delegation, and devolution strategies complement decentralization and, like decentralization, broaden the institutional base for administration and execution of technology programs; reduce the burden on central governments for provision of services (responsibilities in which they have been less than fully successful); and increase stakeholder participation and influence over programs. Advantages of these complementary strategies are that: • Full privatization relieves government of responsibility for production of private goods and services with few externalities. Extension services for commercial crops grown by wealthier farmers, information on postharvest handling and processing technologies, and marketing of machinery or production inputs often fall in this category. • Private provision of publicly financed services takes advantage of private providers’ greater efficiency and flexibility in executing programs. Government contracting of
The Case for Decentralization
NGOs or private extension providers is a common example. Delegation and devolution maintains some government authority and financing, but gives implementing institutions operational flexibility and ability to specialize. Governments may delegate extension responsibilities to research institutes or devolve responsibility for commodity extension to a commodity group.
Regional Governments vs. Local Communities Decentralization strategies frequently must choose between decentralizing authority to local governments or to local community groups. The principle of subsidiarity would indicate that authority should go to the lowest level at which economies of scale and scope are not compromised, and all costs and benefits are internalized. Giving local communities responsibility for programs is an attractive option in that it integrates rural people fully into program design and implementation. Furthermore, much local knowledge and commitment of resources can only be accessed by increasing local participation and decentralizing program authority to the community level. Even though shifting program responsibility to the community level facilitates client participation and control over programs, strategies that accomplish this (rural investment funds) have the disadvantage that they bypass and restrict development of local government capacities. Community participation in program management in collaboration with both local and national governments may be a preferred option (for example, central government financing for extension services managed by local
government units and implemented by producer organizations). Such strategies make it possible to coordinate programs across a broader area, help ensure that interests of the poor and remote areas are represented, facilitate scaling-up of successful initiatives, and help overcome local authoritarian enclaves. National Frameworks for Decentralization Decentralization requires an institutional setting with an established legal framework, an active civil society, adequate capacity in decentralized institutions, and a system of accountability. Thus, the three challenges in implementing decentralization strategies are development of: • National commitment and legal frameworks for sharing responsibilities between administrative levels and with the private sector and civil society. Such a framework requires political action to establish and empower locally elected governments as a base for decentralized service provision. • Subsector strategies and institutional arrangements based on analysis of the nature of the specific services and the comparative advantages, capacities, and potential of each participating institution. • Institutional capacity to enable local government units, national institutions, private sector, and civil society to play their respective role in service provision. Institutionalizing mechanisms for coordination of coproduction is the key challenge in decentralizing sector programs. Development of these enabling policies, strategies, and capacities generally must proceed concurrently as opportunities arise.
Decentralizing Agricultural Extension
Issues With Decentralization of Public Sector Extension
Pubic extension services are being forced to change. In the 1990s agricultural extension services were attacked for being inefficient, irrelevant, ineffective, and poorly targeted. The need for reform was obvious and national systems responded with three major strategies— privatization, decentralization, and program revitalization. Although cost reduction has been the force behind many changes, the principal objective of reforms should be an attempt to improve quality of services to clients (see Box 2). Decentralizing extension services, when implemented effectively, can transform exten-
sion and address a range of generic problems. Decentralized extension brings decisionmaking processes closer to clients and makes programs more responsive to user needs. Service providers become more accountable to clients and better oversight increases efficiency of operations. Decentralization itself can introduce a new dynamism in programs and can promote diversity in service providers and program approaches, thus serving as a first step toward privatization. In addition, reforms to revitalize and privatize programs can accompany decentralization reforms, which generally involve:
Latin America—“Modernization” of Agricultural Extension
In Latin America, many countries are undertaking agricultural extension reforms that generally go beyond decentralization to include a range of innovations, characterized as “modernization” of extension. Reforms have led to the demise of public sector extension agencies, which were often bureaucratic and unresponsive to the needs of the farmers. New programs are targeted, decentralized, privately implemented, and have greater farmer cofinancing and participation. Concerns for fiscal restraint and financial sustainability have led to programs designed for restricted coverage and targeted to priority client groups. Delinking financing from implementation has improved program management and helped develop pluralistic systems. Opening up to private sector service providers has created a market for extension services, while involving farmers in planning and priority setting has empowered farmers and begun to institutionalize cofinancing of extension. Farmers are substantially more committed to extension programs when they pay a portion—even a small portion—of program costs. Reforms are headed in the right direction and there is virtually no move to turn back. Still, with
the exception of Chile, the reformed systems have yet to prove themselves sustainable and there is need to further consolidate progress. Second generation problems include: • Lack of national strategies can lead to an aggregation of disparate, locally planned extension activities. • Quality control, supervision, and oversight systems are weak in many decentralized extension systems. • Systems for knowledge flow to decentralized extension are poorly developed, especially with regard to research-extension linkages and extension agent training. • Extension agents contracted by municipalities or farmer organizations have little opportunity for career development. • Program ownership by farmers is poor for programs receiving heavy subsidies from governments. Source: Based on a presentation by Matthew McMahon, World Bank Task Team Leader in the Latin America and Caribbean Region at AKIS Retreat 1999.
Issues with Decentralization of Public Sector Extension
Administrative decentralization—moving responsibilities for extension to local levels of government; • Political decentralization—expanding user influence on program priority setting, planning, and management; and • Fiscal decentralization—giving financial management responsibility to local governments or requiring cofinancing from local governments and producer groups. Extension services differ from research in two important ways that affect their potential for decentralization. First, extension advisory services (field extension services) come in direct contact with clients and provide services that have a high private-goods content. These characteristics make field extension services a much better candidate for decentralization than research, which typically has a longer-term payoff. Local producers are more willing to commit resources to pay for effective extension services from which they realize immediate direct benefits. Still, there remains a need for other extension services to address “externalities”— environmental problems, food quality or safety concerns, or social equity issues (that is, special needs of small farmers)—that are in the public interest, but are not a priority for individual producers or decentralized institutions. This requires continued central support for extension. A second difference between research and extension is the scope and scale of programs. Research institutions are generally smaller and more concentrated. Extension programs typically operate across the country, provide information on a wide range of technologies from various sources, and draw on traditional knowledge and farmer innovation to improve producer organization, management, production, and marketing functions. The broad demands on extension require strategies that incorporate a variety of approaches to providing services. Despite the apparent suitability of extension service provision to be decentralized, they are often highly centralized. A World Bank study •
of 19 countries found that in the early 1990s 13 countries or regions showed almost no evidence of decentralization of extension services. Colombia, Jiangxi (China), the Philippines, and NusaTenggarra-Timor (Indonesia) were relatively highly decentralized, and Poland and Tunisia showed some decentralization. The study found that: • When extension is decentralized there is a fairly good balance in fiscal, administrative, and political decentralization; • Political decentralization (the role of elected officials) lags other elements of decentralization; and • NGO involvement is moderate and farmer participation is significant in extension. Underlying these conclusions was the fact that institutional development and civil society provide important support to decentralizing extension services. Recognizing Multiple Extension Functions National agricultural extension systems (NAESs) must incorporate a range of extension activities that vary in suitability to decentralization. Field advisory services, as the traditional extension methodology, are compatible with decentralized program strategies and in some cases are suited to private service provision or complete privatization. Other services to support field extension agents and complement field advisory services are often better suited to centralized production. Functions best centralized are those that: support national strategies and financing mechanisms; involve economies of scale and scope; serve a number of administrative regions; or require greater technical input and networking than can be managed at the local level. Services needed in a comprehensive extension system include: • Extension policy, strategy formulation, and planning (centralized); • Training programs for extension agents (centralized or decentralized);
Decentralizing Agricultural Extension
Ghana: Decentralization to Improve Program Coordination
In 1992, the Ghana Ministry of Food and Agriculture (MOFA) had nine technical departments operating essentially parallel programs in each of the country’s 110 districts. The Local Government Act (1993) transferred responsibilities from central government to districts and municipalities, and gave district assemblies power to establish technical departments (including agriculture). Lack of political will delayed implementation for nearly two years, but decentralization finally became effective in 1998, and ministry staff and budgets were transferred to district assemblies and regional administrations. Reforms left the following problems that interfere with efficient provision of extension services: • Most central technical departments continue business-as-usual and have not adopted new roles in policy formulation, planning, coordination, monitoring, and provision of technical support. • Some district assemblies delayed integration of ministry staff into district assembly structures. • Misunderstandings persist regarding implications of decentralization for personnel administration (for example, promotions, use of resources). • Data collection for statistical purposes and program monitoring has been hampered by lack of clear guidelines from the central ministry. • Subject matter centers are largely nonfunctional due to inadequate staff and logistical support
• • •
Technical specialist support to extension agents (centralized); Production of extension publications, audiovisual materials, guidebooks, and other materials (generally centralized); Monitoring and evaluation to support program quality enhancement (needed at all management levels); Training programs for farmers (generally decentralized); Market information services (centralized); Encouragement for (and possibly some controls on) private sector extension (privatization with mixed centralized/decentralized controls);
and responsibilities are not clearly defined for division of labor in the operation of breeding farms and stations. Staff transferred from other departments into the extension service have not received orientation training. Monitoring the use of government resources in the regions and districts is not adequate. Lessons from Ghana’s experience include: Political commitment was essential to start reforms, but loss of commitment delayed implementation. The Ministry was able to initiate the program only because of the existence of a conducive legal and policy framework. Clarity in objectives for decentralization, adequate preparation, and educational campaigns are necessary for effective implementation of decentralization reforms. Educational programs are important to promote attitudinal change, but a one-shot orientation is not enough and sustained training (defining communication channels and job descriptions) is needed for all staff. Implementation of reforms requires dedicated and committed leadership for program oversight.
Source: Based on presentation of a paper by C. AnnorFrempong and Solomon Bekure of The World Bank Ghana Country Office at AKIS Retreat 1999.
Mass media campaigns, including radio, television, agricultural magazines, newspapers, and letters (generally centralized, but may be decentralized or privatized); and Internet and/or telephone dissemination of information and fielding questions from farmers, agribusiness, or extension agents (centralized).
Administrative Decentralization Deconcentration is intrinsic to extension services that are provided in dispersed fields and communities throughout a country. Cropping systems, markets, agroecological zones, and eth-
Issues with Decentralization of Public Sector Extension
nic and cultural characteristics of farmers can vary widely within a country, and moving administration closer to field services can substantially improve program management through better understanding of local conditions. Administrative decentralization goes further by making extension programs directly responsible to local authorities. The challenge in any successful decentralization reform is that of maintaining overall program quality and coherence. (see Box 4). Decentralized extension programs are limited if the decentralized administration lacks awareness of new technologies, sources of assistance, and extension methodologies. Although decentralized administrations can effectively integrate local institutions, organizations, and technologies into an extension system, major benefits from formal extension often come from integrating external knowledge into the local system. Lack of coordination between local administrations can be a problem. If many localities
Box 4 Colombia—Successful Decentralization in a Conducive Political Context Relatively radical reforms in decentralizing extension in Colombia led to a tripling of extension staff and municipal coverage of extension, an increase in number of beneficiaries by two and a half times, and a doubling in funding for the extension program. Success with these reforms was due to existence of representative local governments, willingness of the center to decentralize, and availability of adequate resources. Reforms were basically an administrative decentralization. Political decentralization was to be effected through system governance by municipal committees, but this has not yet begun as planned. Financial sustainability of the system remains in question, as the reforms add considerably to the cost of extension. Quality control is also a major operational problem, reflected in the need for closer links with research, better training for extension agents, and better oversight of extension activities.
promote a single commodity, the result might be overproduction and low prices. Similarly, separate localities might finance the same feasibility studies, training programs, or extension materials. Implementing an integrated watershed or regional development plan might prove impossible if programs in each administrative region are completely independent. Other potential problems include the lack of career opportunities for extension staff in decentralized programs, and difficulties with monitoring and evaluation when local administrative units lack ability to compare targets, results, and achievements with other areas. Extension program quality depends fundamentally on good linkages with other programs—specialized training for extension agents and farmers, technical backstopping by subject matter specialists and information services, other extension services (mass media, fairs), and other development programs (credit programs, market development programs, input supply). Some of these linkages can be maintained at the local level, but many require higher level coordination to ensure efficiency and quality support. Decentralized Governance—Introducing Accountability Decentralizing extension by involving farmers and local government in governance of programs can improve program accountability, increase user ownership of programs, ensure relevance to local needs, improve planning and information flows, and strengthen user capabilities. Transferring program responsibilities to locally elected officials directly decentralizes program governance and accountability to local people. Perhaps equally important are alternate mechanisms that increase user participation and influence over program content and operations. Reforms that enhance farmer influence over program governance include: incorporating farmers into governance and oversight committees; adopting participatory extension
Decentralizing Agricultural Extension
approaches; involving farmers in identifying priorities, planning, and monitoring; working through farmer groups; and using participatory evaluation and feedback mechanisms for program evaluation. Decentralizing governance holds particular promise for making extension programs (and agents) accountable to users. Farmers know whether they are receiving valuable services and should have the power to demand good performance by their service providers. When farmers have authority to influence decisions on program funding, hiring and dismissing staff, and staff incentives, they are truly empowered to improve services. At a minimum, all extension programs should seek farmer feedback on the relevance, quality, and usefulness of extension services. A concern in decentralized extension systems is the degree to which governance mechanisms are representative of all farmers in an area. Women, small farmers, and ethnic or cultural minorities are often underrepresented in governance groups, but may be more in need of public services than those actually representing local interests. Disadvantages of farmer governance are the high up-front costs of participatory approaches, difficulties in ensuring true representation of participating groups, risk of aggravating conflicts or unduly raising expectations, and the possibility of program capture by elites. Table 1
As the traditional view of extension as a function of government agencies gives way in the face of multiple service providers, an expanding agenda, and a better understanding of farmer information and innovation systems, decentralized governance of extension services should become both easier and more important. Fiscal Decentralization of Extension Services Government inability to sustain financial support for large extension systems has been a motivation for the many reforms that attempt to reduce public sector funding, introduce private financing, or eliminate government programs that compete with the private sector. Typically, these strategies tend to decentralize extension financing (see Table 1). Although an objective of many decentralization reforms has been to reduce government expenditures, local governments generally have limited resources and limited ability to raise funds. Central governments therefore must usually continue financing for extension services through intergovernmental financial transfers (IGFTs), and must also finance the considerable costs of reform and local capacity development. This increases total financing requirements for extension, at least over the short term. Over the longer term, decentralizing extension services might reduce government financing requirements by: (1) increasing efficiencies through
Strategies Employed to Reduce Public Expenditures for Extension Delivery of Services Public
♦ ♦ ♦ ♦ ♦ ♦
Source: Adapted from Rivera (1996).
Funding for Extension Public Private ♦ Cost-recovery and fee-forDeconcentration service extension Delegation to parastatals ♦ Levies Cost reductions Better targeting beneficiaries Contracting private suppliers ♦ Transfer to the private sector of extension services or NGOs ♦ User contracting of extension Grants for user contracting for extension services services
Issues with Decentralization of Public Sector Extension
better oversight and greater flexibility in funding decisions and (2) increasing cofinancing by being more responsive, and demonstrating greater benefits, to users. Cofinancing grants (IGFTs) to local governments or farmer groups are an important element of fiscal decentralization, but they present two significant problems: • Many local organizations lack capacity to plan, manage, and evaluate extension programs and lack the contacts and financial management capacity to procure needed services; and • Resource-rich farmers are better able to cofinance services and capture program benefits, even if program objectives are to assist weaker elements of rural society. Still, many new initiatives are using subgrants of various types for local subprojects, and future program design can draw on this experience (see Box 5). Decentralization programs must address these two problems. Training and orientation, program promotion, and support services are critical to enable target clients and local organizations to take over extension responsibilities
under new decentralized systems. Later, as programs are implemented, a strong monitoring and evaluation system is needed to provide management with information necessary to understand who is benefiting from the program and what real impact it is having. Transition to Decentralized Systems Planning is important in any transition to decentralized service provision. Implementation problems are common, as reforms change established financing, operating, and reporting systems, and challenge ingrained views of the center as the legitimate source of authority. Reforms may shift program oversight responsibility to new, nontechnical agencies, or rupture links with other technology programs. Central ministries may oppose reforms and fight perceived loss of power from transfer of programs to local government (see Box 6). Short-term problems include the need to: revise terms of reference for all staff, establish new reporting and supervisory relationships, adjust salaries and compensation plans, organize and equip offices, agree on budgets, and
Competitive and Demand-Driven Extension Projects
Many development groups are experimenting with competitive grants, demand-driven rural investment funds, and community action programs to finance local development programs. These may be targeted exclusively to extension or technological innovation, or may permit this as one of a larger menu of possible investments. These programs permit extension programs to be tailored to local needs and conditions, but with central government financing and oversight. Programs use a variety of implementation strategies with almost any mix of the following procedures: • Design of subproject: by central government, local government, service providers, producer or community groups, or (often) a mix of these; • Subproject selection: by competitive process, allocated according to government development
priorities, or by local request conforming to established criteria; • Service contracting: by the central government, local government, or producer/community organization; • Financing: by contract or grant: and • Term: often one year with option for extension, but may be for duration of the subproject. Programs may or may not be decentralized, as central governments seem to continue to play significant roles in management and decisionmaking. This may decline over time, as governments become more confident of local capacities. Most programs are relatively recent and still evolving, and to date there is little evidence as to their sustainability and long-term impact.
Decentralizing Agricultural Extension
The Philippines: Cautions in Decentralizing Extension Services
The Philippine Local Government Code (1991) provided that delivery of basic services, including agricultural extension, be devolved to local government units (LGUs-provinces, cities, municipalities, and communities). Prior to reforms, extension services were too generic to be useful, poorly managed, and not adapted to local conditions. Following reforms, rapid restructuring of extension transferred 10,000 staff to provincial and local governments. Reforms were to promote participatory, bottom-up planning, but the hastily implemented reforms resulted in: • Excessive fragmentation of planning and delivery of extension services; • Increased NGO collaboration with local governments on agricultural development projects, but on an area-specific basis with poor overall coverage. • Exacerbation of the already weak research-extension linkages; • Use of IGFTs to increase salaries for LGU staff with no benefits for nationally devolved extension staff; • Politicization of extension services at the municipal level, stifling professional growth and development of extension staff; and • Inability of the Department of Agriculture to provide technical assistance and financial support to local government units.
establish mechanisms for fiscal transfers from the center. Over the medium term, planning, priority settings, and coordination become important issues. Over the long term, quality control (training, specialist support, technical oversight) becomes the major problem. Adequate planning is essential to translate policies into operational guidelines, to develop capacity for decentralized implementation, to provide essential support services, and to ensure incentives for reform. Piloting reforms in a few areas can help to identify problems and develop plans for a smooth transition. Extensive promotional work fosters a wide appreciation of the reasons for decentralization, and training and operational manuals ease introduction of new operating procedures. Reforms
The Philippine reforms were preceded by substantial planning and some pilot efforts. Decisionmakers recognized that rapid reform would encounter implementation problems, but concluded that a gradual reform process could easily be stalled by political and bureaucratic opposition. This experience demonstrates the importance of adequate preparation by both national and local governments and the need for reforms to recognize that: • Policies are important, but must be translated into operational guidelines before implementation; • Adequate training is required for all officials at both local and national levels; • Phased implementation of decentralization reforms may help avoid serious mistakes; • Research-extension linkages are critical; • Incentives are important for all major players, including (when possible) all extension service staff; and • And, finally, reforms should be designed for sustainability by proper costing and attention to future funding for decentralized services. Source: Based on a presentation by Dr. Eliseo R. Ponce, Director, Bureau of Agricultural Research in the Philippines at AKIS Retreat 1999.
should avoid decentralizing heavy bureaucracies to local governments, where they may be even less productive and a greater fiscal burden than they are in the center. Good practice in extension system design must recognize the diversity required in a comprehensive extension system. Management of field advisory services might easily be decentralized, but other administration functions are difficult to reconcile with a decentralized program. Monitoring and evaluation, specialized technical support, training, and program strategy development generally require support at a higher level, either national or regional within country. Decentralization initiatives need to include provisions for strengthening central services that enable decentralized programs to
Good Practice for Decentralization in AKIS Projects
function and maintain service quality. Although central programs can becoming supply-driven, various arrangements that enable producers or local government units to procure centrally produced services introduce a demand-drive to production and facilitate private sector participation in service delivery. Although decentralization holds considerable promise to improve extension services,
it requires local administrations and central ministries to reinforce and support each other. New models and mechanisms are needed to coordinate activities of central units, local government, and producers. Key to this is reorienting central governments from an attitude of control to one of supervision, facilitation, and support.
Good Practice for Decentralization in AKIS Projects Decentralization can be a useful tool for improving the function of technology systems, but decentralization must be seen as a tool and not an end in itself. The country context must guide any approach to reforming technology services, there is no single model for decentralization. If a country lacks a legal framework for decentralization, extension alone cannot bring about reform, and opportunity for sustainable program decentralization might be quite limited. Good practice in decentralizing agricultural extension services can be summarized as follows: Centralize or Decentralize Programs As Appropriate To the Service The major benefit from decentralizing extension programs may be the empowerment of producers to express their demands for technological innovation. Decentralization strategies increase user ownership and financial support for services, ensure greater program accountability, and develop constituencies for extension. These reforms strengthen the demand side of exten-
sion systems, but simply decentralizing extension to local governments does not guarantee that beneficiaries will have more control over the decisionmaking process. Program execution and authority should rest at the level most appropriate to effective user influence over the program, with the appropriate level likely to be: • The district, municipality, or community for extension and adaptive research; • The national, provincial, or state level (depending on size of country and state or province) for extension support services and strategic and applied research; and • The national (and international) level for basic and strategic research and maintenance of a core agricultural science capability. Reforms should stress decentralized governance of extension and empower producers to make decisions on the extension agenda by participatory planning, user cofinancing, and involving community organizations in implementation. Both extension services and adap-
Decentralizing Agricultural Extension
tive research are well suited to decentralization as they offer fairly immediate benefits to users. Decentralized extension administration facilitates beneficiary participation and encourages partnerships between various players in the technology system, especially facilitating the involvement of NGOs. Private provision of services and privatization of some extension functions can improve cost efficiency. Decentralized extension also offers important opportunities to introduce fee-for-service and cost-sharing arrangements. Involving users in the evaluation of extension staff and programs and linking remuneration of extension agents to performance can be powerful means of enforcing accountability in decentralized systems. Adapt Strategies to Local Institutional Environments Each country has its own legal framework, political tradition, administrative structure, and social and agroecological setting. Decentralization initiatives must conform to local needs and potentials. Local governments may be an appropriate locus for service provision if they have adequate capacity and are accountable to the local population. Community and farmer organizations are potentially important entities, whose role must be considered when decentralizing extension programs. These organizations often have limited institutional capacity, restricted membership, and unstable leadership. They might require significant investments in institutional strengthening before being able to assume major program responsibility. Extension program decentralization must reflect the general status of political decentralization in a country. If decentralized political institutions are well established and strongly supported, decentralization of AKIS programs can proceed with some confidence. Alternatively, if a country’s institutions are still highly centralized, options will be limited and it may be advisable to limit reforms to deconcentration, introduction of participatory extension ap-
proaches, and contracting for private service delivery. Strengthen Central Support Services for Extension Some technology system functions are best provided centrally, due to economies of scope or scale in their production or the nature of the service involved. These include system strategy development, extension service quality control (training, subject matter specialist support, and production of extension materials), and monitoring and evaluation. Quality of field extension advisory services is dependent on some of these programs, which must be strengthened in any initiative to decentralize extension services. Decentralized extension requires central support to plan and coordinate programs, facilitate effective institutional linkages, formulate national policies, and ensure that national priorities are addressed. Successful models for research-extension linkages in decentralized systems are still needed. Centralized monitoring and evaluation are also important as a basis for national planning and for comparing program performance and impacts. Provide Mechanisms for Policy Formulation in Mixed Systems Central government financing of extension programs must be directed toward national policy objectives (poverty alleviation, food security, and environmental conservation). Reconciling national objectives with priorities emerging from decentralized programs requires flexibility and planning and budgeting systems that integrate both sets of priorities. Expect to Continue Public Sector Financing In the short term, decentralization rarely reduces requirements for central government financing for extension, and decentralization initiatives should count on increased central
Good Practice for Decentralization in AKIS Projects
government funding, at least during their initial stages. This may be defensible, because technological innovation in agriculture remains, in many cases, a public good and warrants public funding. However, where government budgets face fiscal restraints, the higher costs associated with decentralization and private sector service delivery (by NGOs and private firms) raises issues of sustainability. The question then becomes how to best manage the same amount of funding, possibly implying a decreased service coverage balanced by improvement in effectiveness of services. Although decentralization may initially increase the level of public funding required, over the long term it offers opportunities for producers to cofinance some technology services. Leveraging funding from users and an increase in efficiency of service provision can increase total financing available for extension programs. The feasibility and desirability of user cofinancing will vary depending on the local situation, but some financial participation—no matter how small—increases beneficiaries’ ownership and interest in technology programs. Fiscal Transfers for Research and Extension Fiscal transfers from the central government are almost always required to finance decentralized extension. These should be structured to give users maximum influence over programs and to promote institutional pluralism in service provision. Empowering users to contract the most relevant and cost-effective services, and developing a range of public and private providers, should result in the most competent and efficient institutions being selected to provide services. Plan for Transition and Local Capacity Development Decentralization reforms imply fundamental changes in bureaucratic relationships and program operations. These transitions are difficult
and must be preceded by careful planning, promotion of the rationale and principles behind reforms, training in new operational procedures, and preparation of new operating manuals and transition guidelines. Rapid introduction of reforms may be advisable to facilitate a clean break with past practices, but decisionmakers should expect delays in establishing reforms and new operating procedures. Naming a transition team to manage reforms and address problems may facilitate the transition process. Building local government capacity is a prerequisite for decentralizing extension services, because local governments (especially in Africa) often do not have the technical and managerial capacity to take on responsibility for managing technology programs. Local institutions taking on responsibility for extension might require investments in staff development, facilities, and management systems. Policymakers must allow time—perhaps several years—for local institutions to develop this capacity and experience needed for effective program management. Ensure Monitoring and Evaluation of Decentralized Systems Decentralized systems, especially in their early stages, require strong centralized monitoring and evaluation systems to provide policymakers with necessary information to understand how well programs are functioning, who is benefiting, and what impacts the programs are having. The monitoring and evaluation system must feed information to a planning system that can adjust program guidelines and priorities as necessary. Decentralized implementing institutions require their own M&E systems to support decisionmaking, program management, and planning at the local level. In practice, M&E systems at the various levels should be integrated in a comprehensive system to provide for information needs at each level and to maintain quality and reliability of data.
Decentralizing Agricultural Extension
Additional Readings and References
Feder, G., A. Willett, and W. Zijp. 1999. Agricultural Extension: Generic Challenges and Some Ingredients for Solutions. Policy Research Working Paper No. 2129. World Bank, Washington, D.C. Manor, J. 1999. The Political Economy of Decentralization. World Bank, Washington, D.C. McLean, K., G. Kerr, S. Piriou-Sall, and M. Williams. 1997. Decentralization: Progress in 15 Countries?” Slide Presentation on the Decentralization, Fiscal Systems, and Rural Development Research Program. Rural Development Family, World Bank, Washington, D.C. Parker, A. 1995. Decentralization: The Way Forward for Rural Development? Policy Research Working Paper No. 1475. Agriculture and Natural Re-
sources Department, The World Bank, Washington, D.C. Rivera, W. 1996. “Agricultural extension in transition worldwide: structural, financial, and managerial strategies for improving agricultural extension” in Public Administration and Development. Vol. 16:151-161. Umali, D., and L. Schwartz. 1994. Public and Private Extension: Beyond Traditional Frontiers: World Bank Discussion Paper No. 236. World Bank, Washington, D.C. Wiens, T. and M. Guadagni. 1998. Designing Rules for Demand-Driven Rural Investment Funds: The Latin American Experience: World Bank Technical Paper No. 407. World Bank, Washington, D.C.