AG R I C U LT U R E A N D R U R A L D E V E L O P M E N T D I S C U S S I O N PA P E R 5 3 70609 THINKING SYSTEMATICALLY ABOUT SCALING UP: Developing Guidance for Scaling Up World Bank-supported Agriculture and Rural Development Operations T H E C A S E O F C O M P E T I T I V E G R A N T S C H E M E S F O R AG R I C U LT U R A L RESEARCH AND EXTENSION M a rke t a Jo n a s ova a n d S a n j i va C o o ke AUGUST 2012 AG R I C U LT U R E A N D R U R A L D E V E L O P M E N T D I S C U S S I O N PA P E R 5 3 THINKING SYSTEMATICALLY ABOUT SCALING UP: Developing Guidance for Scaling Up World Bank-supported Agriculture and Rural Development Operations T H E C A S E O F C O M P E T I T I V E G R A N T S C H E M E S F O R AG R I C U LT U R A L RESEARCH AND EXTENSION M a rke t a Jo n a s ova a n d S a n j i va C o o ke © 2012 International Bank for Reconstruction and Development / International Development Association The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The �ndings, interpretations, and conclu- sions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to the Of�ce of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Cover Photos: 1) Farmer in Gansu province, China: Liang Qiang/World Bank. 2) Pineapple seedlings grow in a nursery near Accra, Ghana: Jonathan Ernst/World Bank. 3) Preparing fertilizer for the cultivation of aloe plants: Dana Smillie/World Bank. 4) Woman tending plants in Sri Lanka: Lakshman Nadaraja/World Bank. C O NT E NTS III Table of CONTENTS Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii Chapter 1: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Agriculture Critical for Meeting Millennium Development Goals (MDGs) . . . . . . . . . . . . . . . . . . 1 1.2 Scaling Up Agriculture Is Essential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.3 The Experience of Scaling Up Agriculture Operations to Date . . . . . . . . . . . . . . . . . . . . . . . 4 1.4 Task Team Leaders Could Bene�t from Guidance for Thinking Systematically about Scaling Up . . . . . . 5 1.5 Methodology and Outline of the Report. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Chapter 2: Emerging Findings from Analyzing Case Study Experiences of Scaling Up . . . . . . . 11 2.1 Competitive Grant Schemes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 2.2 Emerging Findings from the Case Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 Chapter 3: Proposed Guidance for Scaling Up Competitive Grant Schemes for Agricultural Research and Extension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 3.1 Proposed Guidance for Scaling Up CGS for Agricultural Research and Extension. . . . . . . . . . . . . 20 Chapter 4: Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Annexes Annex 1: Additional Financing for Investment Lending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Annex 2: Overview of Key Literature and Concepts on Scaling Up . . . . . . . . . . . . . . . . . . . . . . . . . 33 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 IV C ONTENTS Lists of FIGURES, TABLES, AND BOXES FIGURES Figure 1.1: Share of New Scaled Up Agricultural Projects (Percentage), FY06–11 . . . . . . . . . . . . . . . . . . . 4 Figure A.1: Organizational Roles in Scaling Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Figure A.2: Which Form of Scaling? A Decision Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Figure A.3: Scalability Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 TABLES Table 1.1: Links between Agricultural and the Millennium Development Goals . . . . . . . . . . . . . . . . . . . . 1 Table 1.2: State of Practice Spectrum Classi�cation System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Table 2.1: Emerging Findings from Analysis of Case Study Experiences of Scaling Up Competitive Grant Schemes (CGSs) for Agricultural Research and Extension At-a-Glance . . . . . . . . . . . . . . . . . . .13 Table 3.1: From a Pilot to Scaling Up CGSs for Agricultural Research and Extension: Considerations for Scaling Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Table A.1: Types of Scaling Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Table A.2: Checklist for Thinking Systematically about Scaling Up Impact . . . . . . . . . . . . . . . . . . . . . . 36 BOXES Box 1.1: Major Scaling Up Initiatives at the World Bank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Box 1.2: Overview of World Bank Instruments Used for Scaling Up Agriculture Projects . . . . . . . . . . . . . . 5 Box 1.3: Scaling Up Innovation: Collaboration with the World Bank and Brandeis University . . . . . . . . . . . . 7 Box 3.1: Peru Agricultural Research and Extension Project (INCARGO) APL (2000–2010) Impact . . . . . . . . . 20 TH INKING SY STEMATICA LLY A B OUT S C A LING UP AC K N OW L E DGME NT S V ACKNOWLEDGMENTS The Guidance for Building Scale into the Design of Selected World Bank-Supported Operations report was written by Marketa Jonasova and Sanjiva Cooke (Agriculture and Rural Development Department). Additional inputs were provided by Jock Anderson, Derek Byerlee, Eija Pehu, Riikka Rajalahti, John Mackedon, Andrea Pape- Christiansen, Sergyi Zorya, Gunnar Larson, Maria Ana de Rijk, Willem Janssen, Erwin de Nys, Diego Arias Carballo, Melissa Williams, Josef Ernstberger, Ru�z Vakhid Chirag-Zade, Veralakshmi Vemuru (World Bank), Johannes Linn (Brookings), Cheikh Sourang (IFAD), and Susan Holcombe (Brandeis University). Overall direction was provided by Fionna Douglas, Mark Cackler, and Juergen Voegele (Agriculture and Rural Development Department). AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 VI AC R ONY M S ACRONYMS AAP Agriculture Action Plan ADCP1 Azerbaijan First Agriculture Development and Credit Project AF Additional Financing AKIS Agricultural Knowledge and Information Systems APL Adaptable Program Loan ARD Agriculture and Rural Development SB Sector Board ARTPII Uganda Agricultural Research and Training Project ASARECA Association for Strengthening Agricultural Research in Eastern and Central Africa ATAAS Uganda Agricultural Technology and Agribusiness Advisory Services Project BOS-KITA School Operational Assistance-Knowledge Improvement for Transparency and Accountability CAADP Comprehensive Africa Agriculture Development Programme CAS Country Assistance Strategy CDD Community Driven Development CGS Competitive Grant Scheme CLTS Community-Led Total Sanitation CORAF/WECARD West and Central African Council for Agricultural Research and Development CRG Competitive Research Grants CRW Crisis Response Window CSA Climate Smart Agriculture CTF Clean Technology Fund DM Development Marketplace EMBRAPA Empresa Brasileira de Pesquisa Agropecuária ESMF Environmental and Social Management Framework GEF Global Environment Facility GFRP Global Food Crisis Response Program GIZ Deutsche Gesellschaft für Internationale Zusammenarbeit ICAR Indian Council of Agricultural Research ICR Implementation Completion Report ICT Information and Communications Technologies IDA International Development Association IFAD International Fund for Agricultural Development INCAGRO Peru Agricultural Research and Extension Project IPR Intellectual Property Rights IRR Internal Rate of Return M&E Monitoring and Evaluation MAAIF Uganda Ministry of Agriculture, Animal Industries and Fisheries MARENASS Peru Management of Natural Resources in the Southern Highlands Project MDG Millennium Development Goals MICs Middle Income Countries TH INKING SY STEMATICA LLY A B OUT S C A LING UP AC R O N Y MS V II MTEF Medium-term Expenditure Framework NAADS Uganda National Agricultural Advisory Services Project NAIS National Agricultural Innovation System NARO National Agricultural Research Organization NARS National Agricultural Research System NGO Non-Governmental Organization NRM Natural Resource Management PAD Project Appraisal Document PER Public Expenditure Review PIU Project Implementation Unit PMU Project Management Unit PPP Public Private Partnership PSIA Poverty and Social Impact Analysis R&D Research and Development R&E Research and Extension SASKI Sustainable Agricultural Systems, Knowledge and Institutions SLM Sustainable Land Management SME Small and Medium-sized Enterprises SOCAD China State Of�ce for Comprehensive Agricultural Development SWAP Sector Wide Approaches TA Technical Assistance TOC Theory of Change TTL Task Team Leader WB World Bank WDR08 World Development Report 2008 on Agriculture for Development WSP Water and Sanitation Program AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 VIII S UM M A RY SUMMARY BACKGROUND As the Millennium Development Goals (MDGs) deadline of 2015 approaches, the need for more effective development efforts that can help reach these goals and improve rural livelihoods at larger scales of magnitude becomes increasingly more impor- tant. In spite of the recent series of shocks that have disrupted the reduction of poverty globally, overall progress continues to be made in this effort (UN MDG Report, 2010). Targeted approaches to poverty reduction, including those that entail scaling up, can continue to have both immediate and lasting impacts. Scaling up means expanding, adapting and sustaining successful policies, programs and projects in different places and over time to reach a greater number of people (quoted in Hartmann and Linn, 2008). Scalability has been an important priority within the Agriculture and Rural Development (ARD) sector at the World Bank for some time. The 2003–2007 ARD strategy Reaching the Rural Poor identi�ed “scaling up innovations and successful invest- ments in rural development� as one of four “main implementation thrusts, � and proposed that the World Bank develop a knowledge product devoted to the subject. The strategy recognized the need to identify and scale up good practice invest- ments in client countries alongside the parallel need to assign greater emphasis on piloting new and innovative approaches that reflect the dynamic economic, social, environmental, and institutional contexts within which rural development takes place. More recently, the need to scale up was reinforced by the Agriculture Action Plan covering the �scal years 2010 through 2012, and by the 2008 World Development Report Agriculture for Development. Both documents depict scaling up support for an agriculture for development agenda as relying on more experimentation, learning, and adjustments, using a variety of mechanisms. The Action Plan calls for increased IBRD/IDA support for agriculture and related sectors from an average of $2.9 billion annu- ally over FY06–08 to between $4.5 and $6.4 billion per year by 2012 across �ve focus areas: (1) raising agricultural productivity; (2) linking farmers to markets and strengthening value chains; (3) reducing risk and vulnerability; (4) facilitating agriculture entry and exit, and rural nonfarm income; and (5) enhancing environmental sustainability and services—with the combination of support in these thematic areas differing across countries and regions. As a result of the emphasis placed on scaling up, between FY06–11, 34 percent of approved agriculture projects (or 110 projects) were scaled up. A phased approach that used a series of investments was the dominant scaling up instrument (56 percent of new scaled up agriculture projects). Other instruments that were used include Additional Financing (AF), Repeater Projects, and Adaptable Program Loans (APLs). OBJECTIVE Although a substantial body of literature, research, and analysis on scaling up is available, it generally lacks speci�c frameworks or practical guidelines that can be used to identify scalable activities or the factors that drive the scaling up process itself. Without such operationally applicable guidelines, there is a serious knowledge gap. The World Bank has therefore undertaken a series of activities designed to examine scaling up at both ends of the state-of-practice spectrum that was articulated by J. Hancock in a working paper published in 2003. The working paper viewed this spectrum as one that extends from innovation, promising practice, lessons learned, good practice, best practice, through to policy principles. The objective of this report is to assess the usefulness of providing guidance for scaling up good practices in core ARD busi- ness lines, and to test the prospects for doing so. TH INKING SY STEMATICA LLY A B OUT S C A LING UP S U M MA RY IX The output of the document is a guide for a systematic discussion on scaling up of Competitive Grant Schemes (CGSs) for ag- ricultural research and extension at key decision points during the life of an Agriculture and Rural Development (ARD) project. This report, which is a companion to the Development Marketplace (DM) report (2011) that focuses on scaling up innovation, addresses the other end of the state-of-practice spectrum—good practices and beyond. METHODOLOGY The preparation of this report entailed �ve main activities. 1. An overview of scaling up concepts and approaches. 2. The selection of a particular sub-area within one of ARD’s core business lines—scaling up CGS for agricultural research and extension. 3. Application of the IFAD/Brookings framing questions (Linn et al. 2010) to �ve World Bank projects that were identi- �ed as addressing that business line—using information provided by the projects’ task team leaders (TTLs) or other member of the project team. 4. The development of sub-area speci�c guidance for a systematic discussion on scaling up based on the �ndings from a series of �ve case studies. 5. Validation of the scaling up guidance for CGSs for agricultural research and extension by World Bank practitioners and other internal consultations. Steps 3 through 5 can be emulated to develop guidelines for scaling up in other business lines. The framing questions were divided into four broad areas. 1. Ideas and vision 2. Drivers, spaces, and pathways 3. The role of the World Bank 4. Additional considerations The framing questions were extended to include “social space� in order to address a number of the speci�c concerns ex- pressed by World Bank TTLs working in the sector. LESSONS FROM ANALYZING CASE STUDY EXPERIENCES OF SCALING UP The projects provide an important example for the development of a scaling up pathway: through deliberate efforts via phasing through APLs in Peru and Azerbaijan; by a non-deliberate evolutionary approach via building on previous institu- tional reforms and lessons in India and Uganda; and through piloting, testing, and replicating in China. All projects initially focused on selected geographic areas and were gradually expanded nationwide. PROPOSED GUIDANCE ON SCALING UP World Bank case studies informed the scaling up guidance for CGSs for agricultural research and extension presented in Chapter 3. Feedback received from practitioners in the Sustainable Agricultural Systems, Knowledge and Institutions (SASKI) thematic group suggests that this guidance can be applied throughout much of the ARD sector, in addition to the particular case of competitive grant schemes. A number of interesting observations were made about the study:  Elements of the IFAD/Brookings framing questions are particularly constructive in bringing out tacit knowledge.  Framing questions are a �rst step toward general agreement on terms.  Applying “importance weights� to drivers, spaces, and pathways could be considered.  Developing a flowchart to ease the project designer’s use of the scaling up guidance could be useful. AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 X S UM M A RY  A focus on learning during the scaling up process is important as interventions evolve during implementation. The les- sons learned through Monitoring and Evaluation (M&E), along with external knowledge should act as a feedback loop for the next design. CONCLUSIONS Applying the IFAD/Brookings framing questions on scaling up to the case studies, together with validation feedback received, suggests the need to develop two new tools for rural development practitioners and their partners seeking to support scaling up impacts in rural development. One tool is an analytical guide for speci�c sub-sectors or a business line highlighting key considerations to guide scaling up and encourage development partners to systematically think about the impact of scaling up. The other tool is an updated list of case studies for a speci�c sub-sector or a business line. Developing these tools in partnership is desirable. The IFAD/Brookings framing questions should be useful to any World Bank sector as well as to donors, governments, and other development agencies to analyze existing cases and develop guidance for future scaling up. They do not provide a blueprint but rather a starting point in the scaling up effort. They should be complemented by relevant analytical tools and further re�ned in light of experience. In the future the World Bank could focus on some areas not covered by this report including (1) a more systematic review of the institutional incentives and/or obstacles in the Bank to scaling up; and (2) the role of different operational instruments, including M&E and Knowledge Management underpinning scaling up. Prospective application of this framework could be explored for other sub-sectors, such as scaling up value chains or irrigation. TH INKING SY STEMATICA LLY A B OUT S C A LING UP C H A P T E R 1 — INT RODUCT ION 1 Chapter 1: INTRODUCTION 1.1 AGRICULTURE CRITICAL FOR MEETING pursue one or more MDGs, scaling up successful smaller MILLENNIUM DEVELOPMENT GOALS (MDGs) projects and replicating their results in broader areas is a Approximately three out of every four poor people living in logical necessity. Scaling up these smaller, more targeted developing countries live in rural areas, and most of them interventions is also necessary in translating immediate depend on agriculture for their livelihoods. In the 21st cen- impacts into longer-term, more lasting ones. The 2008 tury, agriculture continues to be a fundamental instrument World Development Report Agriculture for Development for sustainable development and poverty reduction. The sec- tor contributes to development as an economic activity, as a livelihood, and as a provider of environmental services (World TABLE 1.1: Links between Agriculture and the Bank, 2008). Millennium Development Goals DIRECT AND INDIRECT LINKS Promoting agriculture is imperative for meeting the �rst MDGS WITH AGRICULTURE Millennium Development Goal (MDG1) of halving poverty Eradicate extreme Agriculture contributes directly to this MDG through and hunger by 2015 and continuing to reduce poverty and poverty and hunger agriculture-led economic growth and through improved hunger for several decades thereafter. Agriculture alone will nutrition. In low-income countries, economic growth, which enables increased employment and rising wages, not be suf�cient, but agricultural development has proven to is the only means by which the poor will be able to be a uniquely effective source of poverty reduction. satisfy their needs sustainably. Achieve universal A more dynamic agricultural sector will change the While agriculture is a key component to achieve MDG1, it primary education assessment of economic returns to educating children, intersects all seven of the goals, some directly and oth- compared to the returns from keeping children out of school to work in household (agricultural) enterprises. ers less so (see table 1.1). Agricultural development, for Promote gender Agriculture contributes to this MDG directly through the instance, directly contributes to the empowerment of equality and em- empowerment of women farmers and indirectly through women farmers and indirectly helps reduce child mortal- power women reduction of the time burden on women for domestic tasks. ity by increasing the diversity of food production and by Reduce child Agriculture contributes to reduced child mortality making resources available to manage childhood illnesses. mortality indirectly by increasing diversity of food production and making more resources available to manage childhood Agricultural practices can be both direct causes of and illnesses. important solutions to environmental degradation—for in- Improve maternal Agriculture directly helps improve maternal health stance, more productive agricultural technologies allow the health through more diversi�ed food production and higher- quality diets, and indirectly through increased incomes withdrawal of agriculture from marginal, sensitive environ- and, thus, reduced time burdens on women. ments (World Bank, 2005). Combat human Agriculture directly helps to combat HIV/AIDS, malaria, immunode�ciency and other diseases through higher-quality diets and virus/acquired indirectly by providing additional income that can be immunode�ciency used for health services. 1.2 SCALING UP AGRICULTURE IS ESSENTIAL syndrome (HIV/AIDS), malaria, and other Despite the global economic downturn in recent years, prog- diseases ress continues to be made in reducing poverty (UN MDG Ensure environmental Agriculture practices can be both direct causes of and Report, 2010). sustainability important solutions to environmental degradation. For instance, more productive agricultural technologies The approaching target date for achievement of the allow the withdrawal of agriculture from marginal, sensitive environments. Millennium Development Goals in 2015 lends a certain Develop a global Developing a global partnership for development will urgency to the already widely recognized need for more partnership for help maintain the steady increase in agricultural trade effective development interventions—ones that purpose- development and signi�cant increases in development assistance fully reduce poverty and are instrumental in improving offered to the agricultural sector, which help sustain the bene�ts from agriculture in the longer run. rural livelihoods. For these interventions to be effective Source: Adapted from “Agriculture and Achieving the Millennium at a magnitude that is suf�cient in scope to signi�cantly Development Goals, � World Bank, 2005. Authors. AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 2 C H A PTER 1 — INTR OD UC TION underscored the requirements that needed to be satis�ed related sectors from an average of $2.9 billion annually over to support agriculture for development agendas. The re- FY06–08 to between $4.5-6.4 billion per year by 2012 across quirements included more experimentation, learning, and �ve focus areas: (1) raising agricultural productivity; (2) link- adjustments, using a variety of mechanisms. ing farmers to markets and strengthening value chains; (3) reducing risk and vulnerability; (4) facilitating agriculture In 1997 the World Bank revamped and intensi�ed its rural entry and exit, and rural nonfarm income; and (5) enhanc- development strategy. The 1997 sector strategy Rural ing environmental sustainability and services—with the Development: From Vision to Action resolved to create an combination of support in these thematic areas differing investment climate conducive to rural economic growth and across countries and regions. The Action Plan focused on to empower the poor to share in the bene�ts of that growth. expanding core business lines such as irrigation and land In implementing that strategy, the World Bank continued to tenure (particularly in sub-Saharan Africa), and research and address the greatest challenges for ARD, offering guidance, extension, which together account for over half the IDA/ good practices, lessons learned, sourcebooks, and other IBRD agriculture and related sector portfolio. Agricultural assistance to Bank TTLs and other practitioners working to markets and trade-related programs in particular are areas improve rural livelihoods. As this work continues, one con- in which client country demand for World Bank support has sistent theme that emerges is the need to scale up good increased. Many programs in these areas now have the po- practice. Scaling up has been a topic of inquiry for decades tential to be scaled up. Experimentation and learning will and has seen a recent intensi�cation of attention. Although also continue with innovations in the areas of insurance and small-scale interventions are necessary in the �ght against technology, and new product development in support of global poverty, these projects alone are not suf�cient for environmental services. achieving larger development goals, further highlighting the need to scale up (Cooley and Kohl, 2006). The World Bank, During James Wolfensohn’s second term as the World Bank with its global development mandate; its services spanning President (2000–2005), support for scaling up increased sig- �nancing, knowledge, and coordination; and its close en- ni�cantly.1 Leadership on the topic culminated in organizing gagement with both aid recipients and donors, has a key role the 2004 Global Conference on Scaling Up Poverty Reduction in supporting the scaling up effort, based on a strong country in Shanghai in collaboration with the government of China development model. (with published proceedings in two volumes, World Bank 2004a). As a result, sectors across the World Bank took up In the following ARD sector strategy Reaching the Rural the issue of scaling up. Some of those initiatives are de- Poor, covering years 2003 through 2007 , “scaling up innova- scribed in box 1.1. tions and successful investments� was one of four central “thrusts� and led the World Bank to produce a knowledge base devoted to the subject. The strategy recognized the need to identify and scale up good practice investments in client countries while increasing the emphasis on piloting new and innovative approaches that reflect the dynamic eco- 1 After Wolfensohn’s term ended, he continued supporting scaling nomic, social, environmental, and institutional context of ru- up at the Wolfensohn Center for Development at the Brookings ral development. The case for doing so was expressed most Institution, by positioning the topic as a core focus of its research on development and aid effectiveness. The Center supported comprehensively in a document titled Scaling-up the Impact research and the publication of Hartmann and Linn (2008). Col- of Good Practices in Rural Development (Hancock, 2003). laborative research was also conducted with the International Fund for Agriculture Development (IFAD). Collaboration between More recently, the need to scale up has been reinforced by Brookings and IFAD, institutionally anchored in the Brookings In- stitution (and funded by an IFAD grant), led to the publication of the Agriculture Action Plan covering the �scal years 2010 the Brookings Working Paper at IFAD (Linn et al. 2010). Its �nd- through 2012. The Action Plan sought to make operational ings were discussed on several occasions, including workshops the principles spelled out in the 2008 World Development organized in October 2009 and January 2012 for World Bank staff and Management as well as presentations gven at World Bank- Report Agriculture for Development. The Action Plan hosted Agriculture and Rural Development Days 2011 and Sus- called for increased IBRD/IDA support for agriculture and tainable Development Network Week 2012. TH INKING SY STEMATICA LLY A B OUT S C A LING UP C H A P T E R 1 — INT RODUCT ION 3 BOX 1.1: Major Scaling Up Initiatives at the World Bank Many sectors across the World Bank allocate resources to efforts and results of these initiatives over the past sev- scale up projects and promote an improved understanding eral years are the “Scaling Up School Feeding� pro- of effective and ef�cient ways to do so through knowledge gram and Indonesia’s “BOS-KITA (School Operational products and other activities. The initiatives listed below Assistance-Knowledge Improvement for Transparency are examples of such efforts carried out by other sectors. and Accountability)� Program. The scaling up school feed- ing program was implemented by the Global Food Crisis Community Driven Development—In the last few years, Response Program (GFRP) and Crisis Response Window Community Driven Development (CDD) has undertaken (CRW) and was a response to the global food and fuel price several initiatives, including (1) Scaling Up Community crisis in 2008. This program supported existing school feed- Driven Development: Theoretical Underpinnings and ing programs, linking access to food and education for poor Program Design Implications (2003); (2) Community and vulnerable children living in food security risk areas. Driven Development for Dummies (2004); (3) Scaling Up a By scaling up support for various school feeding programs Community Driven HIV/AIDS Program in Malawi (2004); and around the globe, this project provided meals for millions (4) Scaling Up Local and Community Driven Development: in dozens of countries. A Real World Guide to Its Theory and Practice (2009). The latter serves as a practical guide for CDD practitioners and The BOS-KITA initiative, much like the scale up school presents background, best practices and underpinnings, feeding program, was attached to a larger program to in- analysis and lessons learned, and tool kits to facilitate the crease the overall scope of this project. This project repre- process of empowering communities and local govern- sented a direct link with the government of Indonesia and ments to drive social development. provided up to US$600 million to support the Indonesian government’s commitment to school-based management Scaling up principles related to CDD is important for the ARD through the strengthening of school committees. sector in light of the increasing share of ARD projects being delivered via CDD (44% of new ARD commitments in FY11). Environment/Climate Change—With the climate change agenda receiving more attention, the environmental sector Health—In the last several years the sector has worked to has made strides toward implementing scaling up initiatives. promote scaling up through numerous initiatives. Among One example is the background paper “Scaling Up Climate these is the recently published guide “Scaling Up Nutrition: Change Mitigation Efforts. � This document addresses the A Framework for Action. � Recognizing under-nutrition as questions of how and how fast to achieve mitigation ef- one of the most serious but least addressed health con- forts—in terms of both actual reduction and investment in cerns, this publication outlines key direct nutritional inter- technology. This document discusses the need for scaling ventions which, if scaled up, could signi�cantly decrease up through aggregated programs and provides guidance under-nutrition globally. This report also provides a series of on ways to address these needs and includes examples of principles and actions to take on the global and national level. mitigation efforts. The sector is also implementing other ini- In addition to this report, another key publication is the tiatives such as investments by the Trust Fund Committee of 2009 “Scaling Up Nutrition: What Will It Cost?� This docu- the Clean Technology Fund (CTF) in 13 countries, including a ment explores the overall cost implications of scaling up 13 recent push to double the geothermal capacity of Indonesia direct nutrition interventions. It concludes that an annual through a US$400 million investment plan. investment of US$11.8 billion in 36 countries is needed. Water and Sanitation—Recognizing that sanitation im- This report outlines proposed costs, funding sources, tar- proves the lives of the world’s poor, the Water and Sanitation get implementation areas, and expected outcomes from Department has scaled up initiatives to respond to this need. the initiatives. One such initiative, conducted in collaboration with the Education—There are several successful initiatives in Water and Sanitation Program (WSP), is the Global Scaling the education sector. Two programs that showcase the up Rural Sanitation Project. For more than four years the (Continued) AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 4 C H A PTER 1 — INTR OD UC TION BOX 1.1: Major Scaling Up Initiatives at the World Bank (Continued) (WSP) focused on learning how to combine the approaches When the Solution Is the Problem: Arraying the Disarray of Community-Led Total Sanitation (CLTS), behavior change in Development� (World Development 32, no. 2 [2004]: communications, and social marketing of sanitation to 191–212); (ii) “Toward a Plurality of Methods in Project generate sanitation demand and strengthen the supply of Evaluation: A Contextualized Approach to Understanding sanitation products and services at scale. This project is de- Impact Trajectories and Ef�cacy� (Journal of Development signed to improve the health in rural areas through safe and Effectiveness 1, no. 1 [March 2009]: 1–14.); (iii) “Looking hygienic sanitation. Using the four components of demand, Like a State: Techniques of Persistent Failure in State supply, learning, and performance M&E in India, Indonesia, Capability for Implementation� (Journal of Development and Tanzania, this project has improved access to sanitation Studies, forthcoming); (iv) Blog on understanding impact to an estimated 8.3 million people. trajectories to avoid Type I and Type II errors available at http://blogs.worldbank.org/impactevaluations/guest-post- Development Economics—Recent work by the World michael-woolcock-on-the-importance-of-time-and-trajectories- Bank Development Economics Department (Michael in-understanding-project-effe. Woolcock et al.) advises a more cautionary approach to scaling up. The department’s work includes (i) “Solutions Source: Authors. Transfer refers to the scaling up of principles and processes 1.3 THE EXPERIENCE OF SCALING UP to new and unassociated areas on the basis of one or more AGRICULTURAL OPERATIONS TO DATE useful, preferably successful, initiatives. Expansion refers to As a result of the emphasis placed on scaling up, between scaling up within the geographic area or country and also is FY06 and FY11, 34 percent of approved ARD projects (110 described as quantitative or physical replication. projects in all) were scaled up.2 Using the Hancock (2003) typology, these projects were identi�ed as using either trans- As indicated in �gure 1.1, of the agricultural projects that fer or expansion for scaling up the impacts of the project. were scaled up, a phased approach which used a series of FIGURE 1.1: Share of New Scaled Up Agricultural Projects (Percentage), FY06–11 NEW SCALED NEW ARD SB UP ARD SB FY06-11 PROJECTS PROJECTS # of projects 322 110 Source: Authors. percentage of new 34 ARD SB projects 2 “Agricultural� projects are de�ned as those falling under the responsibility of the World Bank’s Agriculture and Rural Devel- opment Sector Board. TH INKING SY STEMATICA LLY A B OUT S C A LING UP C H A P T E R 1 — INT RODUCT ION 5 investments was the dominant instrument employed to achieve process needs to be developed to aid TTLs in identifying the scale (56 percent of new scaled up agricultural projects). Other most effective scaling up practices through guidance that is instruments that were used include Additional Financing (AF), speci�c to particular business lines and sub-business lines. Repeater Projects, and Adaptable Program Loans (APLs) (see This is especially true in the case of AF mechanisms. box 1.2 for a short description and Annex 1 for a detailed de- scription of AF). All the parent projects that were scaled up Scaling up has multiple de�nitions. Scaling up can be de- had satisfactory progress toward development objectives and �ned in the sense of “more money� or “more impact. � implementation. The success of the earlier projects led to high There is confusion in the use of the term, since in some demand from clients to scale up these examples. aid literature “scaling up� refers to “getting more aid funds out the door� (in total or for speci�c causes). Thinking about scaling up as “more impact� is common but not universal. Themes that run through the de�nitions are scale of impact, 1.4 TASK TEAM LEADERS COULD BENEFIT FROM quality of impact, impact for whom, and sustained time GUIDANCE FOR THINKING SYSTEMATICALLY frames (Binswanger and Aiyar, 2003). A de�nition from the ABOUT SCALING UP 2004 Shanghai Conference on scaling up is simple and used Although substantial progress has been made in increas- in this report: ing and improving the World Bank’s ability to scale up good practices, the need to build upon this progress is ongoing. Scaling up means expanding, adapting and sustaining For instance, there is no requirement at this time to elabo- successful policies, programs and projects in differ- rate on scaling up in project preparation documents such as ent places and over time to reach a greater number of the Project Appraisal Document (PAD). A more systematic people. (Quoted in Hartmann and Linn, 2008) BOX 1.2: Overview of World Bank Instruments Used for Scaling Up Agriculture Projects Additional Financing (AF) is a World Bank instrument Adaptable Program Loans provide phased support for used to implement additional activities or expand activities long-term development programs. They involve a series to increase a project’s impact and development effective- of loans that build on the lessons learned from the previ- ness. The TTL is responsible for modifying additional or ous loan(s) in the series. Ongoing APLs do not usually expanded activities, summarizing the results of the ap- qualify for additional �nancing since their design nor- praisal, con�rming consistency with the project develop- mally assumes that new, revised, and scaled up activi- ment objectives and the Country Assistance or Partnership ties will be �nanced under follow-up phases. However, Strategy. Additional �nancing has strengthened the World successful APLs may qualify for additional �nancing to Bank’s ability to scale up the results and development im- (1) address a cost overrun or �nancing gap encountered pact of successful projects faster and with less processing during implementation of a phase, (2) scale up the APL ’s and transaction costs. Between FY06 and FY11, 28 per- �nal phase, (3) scale up project activities that are per- cent or 31 scaled up ARD projects used AF as an instru- forming better than expected and cannot be postponed ment. Further analysis needs to address (i) whether the to the next APL phase, and/or (4) complement project AF activities are part of a scaling up pathway or funds for activities that are performing better than expected and additional activities that are not part of a longer-term stra- cannot be postponed to the next APL phase. Additional tegic approach, and (ii) whether the decision to provide ad- Financing may also be feasible in the context of a hori- ditional �nancing is justi�ed by M&E results. zontal APL under which only one loan is anticipated for a given country, sub-national jurisdiction, or sector. Repeater Projects are those in which the basic design and Between FY06 and FY11, 6 percent or seven scaled up effectiveness are proven, and which a borrower proposes ARD projects were APLs. to elaborate and scale up. An initial assessment of a cur- rent project’s suitability for a repeat operation should be In addition to the formal instruments described above, scal- conducted by the TTL following a midterm review or at the ing up has occurred in a series of individual investment op- point when disbursements have reached at least 50 per- erations (phased approach)—56 percent of the new scaled cent of commitment. Between FY06 and FY11, 10 percent up ARD projects between FY06 and FY11. or 11 scaled up ARD projects were repeater projects. Source: Authors. AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 6 C H A PTER 1 — INTR OD UC TION This de�nition has the merits of brevity and simplicity. The part required for successful scaling up. These tools, however, of impact is implied by the requirement of reaching greater do not provide speci�c, easy to use guidelines that can be numbers of people. The “successful� descriptor implies that readily utilized by World Bank TTLs. there is some valued end of scaling up, without specifying what that end consists of. It implies more than one pathway The World Bank undertook a program of activities as well to to scaling. It leaves open many questions, including what is close the existing gap in practical knowledge about scaling up. scaled, who does the scaling, how we decide which people These activities look at scaling up in terms of a broad perspec- are reached or how implementing scaling is managed. The tive described by Hancock and others that is referred to as implication is that the de�nition can (and must) be tailored to the “state-of-practice spectrum� (Hancock, 2003). The state- speci�c contexts. In the context of this report, the focus is on of-practice spectrum can be viewed as extending from innova- scaling up as replication or expansion. Expansion and replica- tion, promising practice, lessons learned, good practice, best tion are intentional and planned types of scaling up. However, practice, through to policy principles (see table 1.2 for more readers should not overlook the importance of spontaneous details). At this time the Bank is focusing on two ends of the replication, which may occur when an innovation is broadly state-of-practice spectrum: (1) scaling up of innovations, and perceived by all stakeholders as highly bene�cial and the bar- (2) scaling up of good practices and beyond. The �ndings of riers to adoption are low. Scalability, then, is the potential of a both sets of work have some overlap and feed into each other. practice or change to be scaled up, or expanded, adapted or The work on scaling up the innovation end of the state- replicated (Holcombe, 2011). of-practice spectrum was conducted in collaboration with In 2010, the Brookings Institution and the International Brandeis University. It applied the Brookings/IFAD framing Fund for Agricultural Development (IFAD), with support questions to explore the opportunities for scaling-up inno- from the Wolfensohn Center for Development, undertook vative ideas identi�ed by the winners of the Development an analysis to answer practical questions about how to go Marketplace (DM) on the theme of Sustainable Agriculture about scaling up and about what characteristics distinguish for Development. The DM was developed as a launching scalable projects from other projects. The results of the pad for innovative small grant projects around the world. analysis were described in a report titled Scaling Up the While the DM grant support has been successful in help- Fight against Rural Poverty: An Institutional Review of ing projects demonstrate proof-of-concept, challenges IFAD’s Approach (Linn et al., 2010). The report identi�ed around replication and scale, the key end goals of the com- a conceptual framework and tools that can be used to as- petition, have persisted. The work with Brandeis University sess the scalability of development interventions, identify focuses on some of these challenges and resulted in the appropriate forms, drivers, and spaces. It offers practical publication Lessons from Practice: Assessing Scalability guidance on leading, planning, and managing the changes (2011) (see box 1.3 for main �ndings). TABLE 1.2: State of Practice Spectrum Classi�cation System STATE OF PRACTICE LEVEL OF EVIDENCE GENERAL APPLICABILITY Policy, principle Proven in multiple settings, replication studies, Consistently replicable, widely applicable “truism� evidence quantitative, scienti�c essential for success Best practice Evidence of impact from multiple settings, meta- Demonstrated replicability, limited risk analyses, expert review Good practice, better practice Clear evidence from some settings, several evaluations Promise of replicability, medium risk Lessons learned Positive evidence in a few cases Limited number of settings and experiences Program evaluations, conference workshops Promising practices, state of the art Unproven in multiple settings, anecdotal evidence, High risk testimonials articles, reports Innovation Minimal objective evidence, inferences from parallel New idea, no previous experience; highest risk experiences and contexts Source: Hancock, 2003. TH INKING SY STEMATICA LLY A B OUT S C A LING UP C H A P T E R 1 — INT RODUCT ION 7 BOX 1.3: Scaling Up Innovation: Collaboration with the extension, in order to develop sub-business line specific World Bank and Brandeis University upstream guidance for future scaling up efforts. The re- port found that the application of the Brookings/IFAD Brandeis University focused on the innovation end of framing questions, with some modifications, produced the State-of-Practice spectrum. It reviewed the 22 DM useful guidelines for systematic discussion on scalability projects implemented in the agriculture sector in 2008, during project preparation. using the questions developed in the IFAD/Brookings study, with the objective of developing a framework for assessing the scalability of DM and other projects that 1.5 METHODOLOGY AND OUTLINE seek to demonstrate innovations that promise to have OF THE REPORT an impact on poverty reduction. The framework was Preparation of this report included the following: (1) an over- used to identify three DM projects that showed promise view of scaling up concepts and approaches; (2) selection of of scalability and impact. These case studies helped to one sub-area within a core ARD business line—scaling up provide a systematic information base and a framework CGS for agricultural research and extension; (3) application for the subsequent stages needed, including the form of of the IFAD/Brookings framing questions (Linn et al., 2010) scaling up, identifying likely organizations to lead, provid- on scaling up to �ve World Bank projects addressing this ing an information base for initial planning, and identify- business line (populated by the TTLs or their team mem- ing the institutional, intermediary, and �nancial support bers); (4) development of a sub-area speci�c guidance for a required to ensure success. systematic discussion on scaling up based on the �ndings The report, Lessons from Practice: Assessing from the case studies; and (5) validation of the scaling up Scalability, concludes that innovations should be sim- guidance for CGSs for agricultural research and extension ple, strategic, and readily monitored. Scaling up needs by World Bank practitioners and other internal consulta- (1) local legitimacy and ownership, leadership, and an tions. The above mentioned steps (3) through (5) can be implementing organization with capacity to learn and emulated for developing speci�c scaling up guidance for grow; (2) time to prove the effectiveness of the imple- other business lines. mentation and build the conditions; (3) a champion who For a review of the literature the reader is directed to the can put the innovation on the agenda of key stakehold- companion report to this one prepared for the World Bank ers and play a role in bridging actors and eliminating by the Heller School for Social Policy and Management at roadblocks; (4) �nancial viability; and (5) incentives. The Brandeis University, Mapping the Roads from Development report proposes a Simplicity and Complexity Index of Marketplace Agriculture and Rural Development Projects Scalability (adapted from Cooley and Kohl, 2006) that to Sustainable Practice (2011). Relevant sections are repro- can be used by project managers and key decision duced in Annex 2. makers to assess scalability on an iterative basis again during implementation. The Brookings/IFAD framing questions below (from Scaling Source: Authors. Up the Fight against Rural Poverty: An Institutional Review of IFAD’s Approach, 2010) were developed on the basis of an extensive literature review and analysis of IFAD work on This report addresses the other end of the state-of-practice innovation and scaling. The framework emphasizes the im- spectrum—scaling up good practices and beyond. It builds portance of learning in an “iterative and interactive cycle� of upon previous and ongoing scaling up activities and aims scaling up. In particular, it highlights organizational and insti- to aid TTLs scale up good practices with specific guid- tutional aspects of scaling up; focuses on the drivers of scal- ance tailored to one core agriculture business line. This ing up, and on the �nancial, political, organizational, and other activity tested in detail the usefulness of applying the spaces that permit scaling up; and stresses the importance Brookings/IFAD framing questions to past and ongoing of monitoring and evaluation for learning and adaptation. The cases of scaling up in World Bank support for competi- concept of “social space� was added by the World Bank to tive grant schemes (CGSs) for agricultural research and the framing questions below. AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 8 C H A PTER 1 — INTR OD UC TION In general, there are many possible pathways for scaling  Vision and leadership: A vision is needed to recognize up successful interventions. By “pathway� we understand that scaling up of a (new) idea is necessary, desirable, the sequence that needs to be taken in the innovation- and feasible. Visionary leaders or champions often learning-scaling up cycle to assure that a successful pilot is drive the scaling up process forward. taken from experimental stage through subsequent stages  External catalysts: Political and economic crisis or to the scale ultimately judged to be appropriate for the in- pressure from outside actors (donors, EU, etc.) may tervention pursued. drive the scaling up process forward. Selecting the dimensions. Scaling up pathways can  Incentives and accountability: Incentives are key to follow different “dimensions. � They may simply expand drive the behavior of actors and institutions toward services to more clients in a given geographical space. scaling up. They include rewards, competitions, and They can also involve “horizontal� replication, from one pressure through the political process, peer reviews, geographic area to another; “functional� expansion, by and other evaluations. Monitoring and evaluation adding additional areas of engagement; and “vertical� up against goals, benchmarks, and performance metrics scaling, that is, moving from a local or provincial engage- are essential ingredients to establish incentives and ment to a nationwide engagement, often involving policy accountability. dialogue to help achieve the policy and institutional condi- tions needed for successful national-level scaling up. If scaling up is to succeed, space has to be created for the initiative to grow. The most important spaces are the Defining the desired scale. It is important to de�ne, up- following: front, the ultimate scale to which an intervention should  Fiscal/�nancial space: Fiscal and �nancial resources or could be taken, given the needs of the target popula- need to be mobilized to support the scaled up inter- tion and the nature of the intervention; and to realistically vention, and/or the costs of the intervention need consider the time horizon over which the scaling process to be adapted to �t into the available �scal/�nancial needs to extend in order to achieve the desired ultimate space. scale.  Natural resource/environmental space: The impact of Focusing on key drivers and spaces for scaling up. the intervention on natural resources and the environ- There are two factors that need to be considered in de- ment must be considered—harmful effects mitigated signing the appropriate pathway for any given case: or bene�cial impacts promoted.  Drivers: forces pushing the scaling up process  Policy space: The policy (and legal) framework has to forward; and allow or needs to be adapted to support scaling up.  Spaces: opportunities that can be created, or potential  Institutional/organizational/staff capacity space: The obstacles that need to be removed to open up the institutional and organizational capacity has to be cre- space for interventions to grow. ated to carry the scaling up process forward. Not all the drivers and spaces will have to be consid-  Political space: Important stakeholders, both those in ered or developed with equal depth for all cases. Indeed, support and those against the intervention, need to identifying and focusing on the core factors applicable to be attended to through outreach and suitable safe- a particular case is one of the requirements of effective guards to ensure the political support for a scaled up scaling up. intervention. A few key factors drive the scaling up process forward:  Cultural space: Possible cultural obstacles or support  Ideas and models: There has to be an idea or model mechanisms need to be identi�ed and the interven- that functions on a small scale. It emerges from tion suitably adapted to permit scaling up in a cultur- research or practice. ally diverse environment. TH INKING SY STEMATICA LLY A B OUT S C A LING UP C H A P T E R 1 — INT RODUCT ION 9  Partnership space: Partners need to be mobilized to  Working with limited �nancing mechanisms, not join in the effort of scaling up. identifying policy constraints and working with small  Learning space: Knowledge about what works and implementing partners (such as NGOs) may limit the doesn’t work in scaling up needs to harnessed potential for scaling up later; and through M&E, knowledge sharing, and training.  Lack of effective, timely M&E may lead to poor deci- sions in scaling up. A number of problems can result from not paying the necessary attention to scaling up pathways, including The lessons learned from the application of the Brookings/IFAD the following: framing questions to the analysis of �ve well-documented  Opportunities for scaling up may be missed (“Type 1 cases of scaling up CGSs for agricultural research and exten- error�) or scaling up may be conducted when it was sion in Azerbaijan, India, Uganda, Peru, and China are discussed not justi�ed (“Type 2 error�) [It should be noted that in Chapter 2. the borrowing of these statistical terms in fact de- pends on how the null hypothesis is de�ned; since a Chapter 3 presents upstream scaling up guidance based on Type 1 error is that of rejecting a true null hypothesis, lessons learned from the analysis of the �ve case studies, for this usage to make sense, the null implicit is that additional literature, and expert opinion. “a project is worth scaling up�]; Chapter 4 summarizes key �ndings and outlines possible  Creating “boutique� approaches that only work in the next steps. small due to their high costs;  Setting up special purpose entities (e.g., PIUs) rather than working through ministries may limit institutional options later; AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 C H A P T E R 2 — E ME RGING F INDINGS F ROM A N A LY ZING C AS E STUDY EX PERIENC ES OF SC A LING UP 11 Chapter 2: EMERGING FINDINGS FROM ANALYZIN G CASE STUDY EXPERIENCES OF SCALING UP As described in the methodology section in Chapter 1, this In many countries, innovation funding has shifted away from report applied the Brookings/IFAD framing questions to �ve block grant funding toward the use of innovation funds. These well-documented cases of World Bank supported competi- funds aim to support innovators and their links to public in- tive grant schemes (CGS) for agricultural research3 in order stitutions, private entrepreneurs, and other actors (such as to analyze experiences of scaling up and to judge the use- groups of rural producers). Accordingly, they can be thought fulness of doing so. A competitive grant scheme is a well- of as tools that create platforms for innovation by providing tested instrument drawing on 30 years of experience in Latin incentives for collaboration. Innovation funds allocate grants America, and is therefore considered good practice rather to targeted applicants based on a system for evaluating the than innovation. eligibility, relevance, and excellence of their proposals. Grant schemes can be competitive or noncompetitive.4 One of the The scaling up framework consists of four broad areas: crucial considerations in choosing between a competitive and (1) ideas and vision; (2) drivers, spaces, and pathways; a noncompetitive grant system is the capacity available in the (3) the role of the Bank; and (4) additional considerations. client country. If the capacity of the applicants or those who Overall, this proved to be a useful guide to analyzing scale are expected to support them is limited, a noncompetitive up and serves as a foundation on which to build a more system is likely to be a better choice than a competitive one. honed, business-like, speci�c set of guides. Its usefulness was underscored in a focus group meeting of World Bank Worldwide there is increasing use of competitive grants to agricultural research and extension practitioners belonging �nance agricultural technology development and transfer as to the Sustainable Agricultural Systems, Knowledge and well as community development programs and environmen- Institutions (SASKI) Thematic Group on March 7, 2012. This tal projects. A CGS constitutes a tendering process where chapter reports on the emerging �ndings from analyzing providers of speci�ed services bid for contracts within pre- the case studies, while keeping in mind that scaling up de�ned programs and/or priorities or in response to calls for at the World Bank is driven by client demand and that proposals. The CGS may be “completely open� for global com- aid organizations do not scale up development impact by petition, have regional or national limits, or require particular themselves, as their role is to support partner countries combinations of partners in a bidding “consortium. � Broadly, and stakeholders in this process. the CGS enables governments to address priority research, extension, and other rural development needs and simultane- ously serves to “jump-start� an agricultural services reform 2.1 COMPETITIVE GRANT SCHEMES process. Under the CGS, funds are allocated competitively to Competitive grant schemes are a good choice in many situa- public and/or private research and development institutions to tions. First-generation CGSs were used to improve ef�ciency implement contractually agreed activities. by fostering competition between research and development There are, however, many different types of CGS. They can proposals while stimulating collaboration within the national be focused either on competition or on facilitating partner- agricultural research system, whereas second-generation ships. They can be “bottom-up� or “top-down. � They can be CGSs emphasize the importance of the demand side, requir- based on “full funding� or require “counterpart funding. � ing clearer roles for and better participation from other stake- The design of a CGS depends on a thorough analysis of holders, including farmers and marginalized groups, within the agricultural innovation system. 4 For detailed information on CGSs including project examples, please refer to World Bank publications (1) Designing and Imple- 3 It should be noted that in addition to the CGS, other interventions menting Agricultural Innovation Funds: Lessons from Competi- were scaled up in some cases; however, the focus of this report tive Research and Matching Grant Projects (2012), and (2) Agri- is on scaled up CGS activities only. cultural Innovation Systems: An Investment Sourcebook (2012). AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 12 CHAP T E R 2 — EMER G I NG FIND INGS FROM A NA LY ZING C AS E STUDY EX PERIENC ES OF S C A LING UP what is expected by the stakeholders and what their capaci- 2.2 EMERGING FINDINGS FROM THE CASE STUDIES ties are, and therefore on determining the key objective(s) Scaling Up Pathways of the CGS. The CGS can be particularly useful and effective The projects provide an important example for the develop- when (1) there is a large number of potential service provid- ment of a scaling up pathway: through deliberate efforts ers (2) for new areas of work that can be completed in a short via phasing through APLs in Peru and Azerbaijan; through a time; (3) bene�ts of competition are high in relation to the non-deliberate evolutionary approach via building on previous transaction costs involved in a CGS; and (4) when the desired institutional reforms and lessons in India and Uganda; and results are known and can be used as clear and unambiguous through piloting, testing, and replicating in China. It should be selection criteria for projects to be awarded grants. noted, however, that while there is a perception that scaling Competitive grant schemes are regarded as a powerful up is happening via a sequence of projects, in the case of mechanism for shaping the way agricultural research is con- China, scaling up was conceived in one project, which can be ducted. Among the factors appearing as critical for success considered unusual but not inconsistent with the concept of are balancing open participation with quality, a good gover- scaling up used here. All projects initially focused on selected nance system that can manage multiple relationships and geographic areas and were gradually expanded nationwide. interests, �nancial stability, and well-managed projects that Scaling up efforts rely on and bene�t from periodic progress generate knowledge products desired by stakeholders. In ad- reports that document interim impacts and results quantita- dition, a shared vision of problems and envisaged solutions, tively. This enables project staff to identify unforeseen prob- a demonstrated demand for proposed activities, research lems and to address them in a timely manner before they management and service provision capacity, frequent com- undermine achievement of the project’s larger objectives. munication, decision-making transparency, and institutional Impact assessments are used to quantify the reach of the ownership are key determinants of success. impact, its quality in terms of who bene�ted, and the amount Since 2005, the World Bank has provided a sequence of of time the impact took to achieve. The impacts achieved by several loans where CGS were initially introduced as inno- the Peru project, which are described in the next chapter, are vations and subsequently scaled up, among other interven- particularly noteworthy. Data on impact assessment were tions. These loans have principally addressed the needs of not yet available for the China, Azerbaijan, India, and Uganda small-scale farmers in agricultural technology development projects analyzed, but these will be highlighted in this report and adoption. In all case studies scaled up, CGSs were imple- when they become available. mented in conjunction with advisory services and value chain Based on feedback from TTLs and their team members on linkage interventions. The following �ve cases of CGS imple- the case studies, the following key drivers and spaces were mented in the ARD sector with World Bank support were instrumental in the scaling up process: used in the analysis of scaling up in this report:  Azerbaijan Second Agricultural Development and Credit Project APL (2006) with input from Phase 1 Scaling Up Drivers  China Agricultural Technology Transfer Project (2005)  Ideas and models: the project cohort aimed at scaling up CGS from Phase 1 (except in China) where the  India National Innovation Project (2006) with input experience with the CGSs has been transferred from from the National Agricultural Technology Project other countries with successful outcomes. (1998)  Vision and leadership: Government commitment and  Peru Agricultural Research and Extension Project leadership from key institutions were conducive for (INCAGRO) APL (2000-2010) with input from Phase 1 scaling up.  Uganda Agricultural Technology and Agribusiness  External catalysts: Agricultural technology generation Advisory Services Project (ATAAS, 2010) with input and transfer formulated in a government strategy. from the Agricultural Research and Training Project (ARTPII, 1999) and the National Agricultural Advisory  Incentives and accountability: Incentives include the Services Project (NAADS, 2001) need for institutional change (CGSs to ‘jump-start’ the reform), APL triggers, and innovation as a pre- The emerging lessons from applying the IFAD/Brookings requisite for donor funding. Competitions (including framing questions to the above mentioned case studies are CGSs) grants and subsidies can also be considered discussed below and summarized in table 2.1. incentives. TH INKING SY STEMATICA LLY A B OUT S C A LING UP TABLE 2.1: Emerging Findings from Analysis of Case Study Experiences of Scaling Up Competitive Grant Scheme (CGS) for Agricultural Research and Extension At-a-Glance FRAMEWORK ELEMENT AZERBAIJAN CHINA INDIA PERU UGANDA Scaled up Functional: CGSs were used Geographical (nationwide) expan- Geographically (nationwide) Geographical (nationwide) Geographical (nationwide) scal- primarily for agricultural research sion of the Competitive Research scaling up lessons of 25 years of widespread use of the CGSs for ing up of CRGs (from Phase 1). in Phase 1. In Phase 2 they were Grants (CRGs) to the whole coun- experience (from earlier projects) agricultural research and exten- Functional scaling up by introduc- used for a different purpose – try. Transfer of CRG experience to integrate technology develop- sion (from Phase 1). ing competitive matching grants testing and transferring new from other countries. ment and transfer mechanism, to facilitate commercialization of and improved technologies and �nance research through farmers, and technology uptake for improved market acces- competitive grant schemes. grants. sibility, storage, and small-scale Also, horizontal, functional, and processing. vertical5 Geographical (nationwide) expansion DRIVERS Ideas, models Ideas and models from Phase 1 Learning (technology support) Increasing pluralism in the Indian The idea was to test on a small Closer links and collaboration will including CGSs; effectiveness was project pilot to be replicated agricultural research actors, scale, catch the flaws, remedy be forged by involving multiple assessed by surveys, implementa- outside project area on a country- incentivized by competitive them, and then replicate on a stakeholders in priority setting tion reviews, the Implementation wide scale once approaches are grants. Innovative Consortium larger scale. In a third phase the and market-relevant research, Completion Report (ICR), and deci- successfully pilot tested. approach has been tested and model would then be made and by developing strong National sion was made for nationwide adopted successfully in other institutionally sustainable. Agricultural Research System— scale up. countries (transfer). Geographical (Uganda) National Agricultural nationwide scale up. Advisory Services Project (NARS- NAADS) research–advisory service interfaces at the national, AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 zonal, and local levels. Funding mechanisms are created to allow farmers and extensionists to attract researchers to collaborate with them on applied research on farmers’ �elds. Formal and informal means are established to ensure that farmers have a voice in decisions affecting research priorities, funding, execution, and evaluation. Vision and leadership APL instrument, support by Relevant government institution. Champion in the Indian Council of The scaling up was based on a Government’s commitment to the Cabinet of the Ministers and Agricultural Research (ICAR), later previous agreement between project from the start, interest Ministry of Agriculture. appointed as national project World Bank and Borrower. There from the Ministry of Finance, C H A P T E R 2 — E ME RGING F INDINGS F ROM A N A LY ZING C AS E STUDY EX PERIENC ES OF SC A LING UP coordinator; now head of ICAR is was considerable support after support for one project combin- also on board. Phase 1 for scaling up from the ing research and extension, sector and from Peru Agricultural institution building, country-wide Research and Extension Project reach, directors of research and (INCAGRO) Board. extension provided exemplary leadership, satisfactory record of previous projects. (Continued) 5 See Section 1.5 for de�nitions. 13 14 TABLE 2.1: Emerging Findings from Analysis of Case Study Experiences of Scaling Up Competitive Grant Scheme (CGS) for Agricultural Research and Extension At-a-Glance (Continued) FRAMEWORK ELEMENT AZERBAIJAN CHINA INDIA PERU UGANDA External catalysts Structural transformation. Project responds to the govern- The Ministry of Economics Agricultural Research and ment of India objective of and Finance has supported the Extension deemed as desirable, agricultural technology generation program strongly. Comprehensive Africa Agriculture and transfer. Development Programme (CAADP) framework, Public Expenditure Review (PER) evidence. Incentives and accountability Project is a tool for improve- CGS enables different actors APL triggers. Provision of donor funding, ment of the larger State Of�ce to be considered by ICAR and Innovation as a precondition to for Comprehensive Agricultural prompted institutional change in receiving funding. Development (SOCAD) program. ICAR; project help desk facilitated partnerships and support. SPACES Fiscal/�nancial Resources indicated in Country Project linked to the major Availability of World Bank The �scal space was discussed Project cost to be reflected Assistance Strategy (CAS) and funding institution with funding resources leveraged ICAR budget, with the Ministry of Economics in Medium-term Expenditure allocated by the World Bank and portfolio covering the entire state-dominated Research and and Finance. Framework (MTEF), development the government. country. Development (R&D) setback, of alternative resource mobi- �scal crisis prompted economiz- lization strategies, conducting ing and dependence on external economic modeling on impact of grants, collaboration for R&D, scaling up (e.g., impact on wages, growing importance of private prices). sector and partnerships. Natural resource management/ Environmental Management Plan In addition to observing the gen- Addition of Global Environment The Project Environmental and Value-added of GEF-�nance environment was prepared. eral safeguards policy provision, Facility (GEF) component. Social Management Framework component on sustainable land the project launched a number (ESMF) were expanded to cover management component, focus of environmentally oriented the whole country. on Sustainable Land Management sub-projects with the intention to (SLM) technologies by research, scale up positive technologies or promoted by advisory services. procedures. The Ministry of Agriculture, Animal Industries and Fisheries (MAAIF)-led platform for Sustainable Land Management (SLM) is a key ingredient for sustaining the project’s impact on SLM and climate risk mitigation and for a future Sector Wide Approaches (SWAP) for coordinat- ing climate change interventions in Uganda. (Continued) TH INKING SY STEMATICA LLY A B OUT S C A LING UP CHAP T E R 2 — EMER G I NG FIND INGS FROM A NA LY ZING C AS E STUDY EX PERIENC ES OF S C A LING UP TABLE 2.1: Emerging Findings from Analysis of Case Study Experiences of Scaling Up Competitive Grant Scheme (CGS) for Agricultural Research and Extension At-a-Glance (Continued) FRAMEWORK ELEMENT AZERBAIJAN CHINA INDIA PERU UGANDA Policy No restrictions for scaling up. The overall policy framework was Preparing the policy context was Legal framework in place in already in place when the project one of the components in Phase 2, Phase 1. Program is the succes- started. A critical additional policy and this was the one that was sor to long-term International came in place when China passed rated lowest. The policy space Development Association (IDA) a law on farmer cooperatives in did not open up enough to allow engagement in both technology 2007, which was conducive to for a third phase. Between Phase generation and dissemination in working with farmer associations 1 and 2, progress was excellent. Uganda, delivered through sepa- as partners in the project. rate operations. New features were introduced in Phase 2: strengthening NARS-NAADS research–advisory service interface and joint priority setting, planning and monitoring at the national, zonal, and local levels. Driven by government desire to achieve national coverage in a short time. Institutional, organizations, staff Successful implementation by the The institutional and organi- Project built capacity for govern- The implementation unit was Project used existing orga- capacity Project Management Unit (PMU). zational setting was in place ing and managing CGSs. expanded with 8 regional of- nization structure conducive already. The China State Of�ce �ces in Phase 2. The plan was to for scaling up, made sure of for Comprehensive Agricultural institutionalize the CGS model in National Agricultural Research Development (SOCAD) is part of Phase 3. Organization (NARO) links to the Ministry of Finance and has NAADS (one of each in each other AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 its structure all the way down board); zonal level collaboration is (provinces, municipalities, and a must where applied research is counties) with of�ces generally developed and extension brought attached to the �nance depart- to farmers, and zonal-level plat- ments at each level. form imperative for collaboration and scaling up. Political Phase 2 APL included in CAS. The design of the project was ICAR champions. Key role of the INCAGRO Board, Matching grants are important well tuned into the policy environ- which was effective in the transi- politically; funding goes to the ment. This was fully in line with tion from the �rst to the second constituency and results in Chinese central government phase but not in the transition of inclusive development. Parliament policies and thus the project was second to third phase. approved the project. considered a fully Chinese project exactly addressing the national policies. C H A P T E R 2 — E ME RGING F INDINGS F ROM A N A LY ZING C AS E STUDY EX PERIENC ES OF SC A LING UP Cultural No obstacles. An ethnic minority action plan Cultural aspects were incorpo- NAADS is demand driven, cannot was prepared addressing rated in the project design. be imposing on farmers. concerns in several areas where ethnic minorities were part of the project. Social (added by the World Bank) Women included in participation. The project included a spe- Social assessment identi�ed key The project put in place special Gender issues are included in ci�c component for Targeted issues for the project’s imple- mechanisms to deal with women NAADS activities. Technology Transfer interventions mentation. Consortia focused on and marginalized groups. As a to vulnerable groups and women poverty and gender issues, among result, women and marginalized in particular. other things. groups were more targeted in the �rst than in the second phase. (Continued) 15 16 TABLE 2.1: Emerging Findings from Analysis of Case Study Experiences of Scaling Up Competitive Grant Scheme (CGS) for Agricultural Research and Extension At-a-Glance (Continued) FRAMEWORK ELEMENT AZERBAIJAN CHINA INDIA PERU UGANDA Partnership Project was prepared in coordina- The project was in itself a part- Several partners including IDA tion with partners in Azerbaijan. nership project mainly building on support institution building for ag- Public Private Partnership (PPP). ricultural research and extension. However, the project also created partners by supporting the cre- ation and improving the strength of farmer associations. Learning Strong Monitoring and Evaluation The project developed and ap- CGSs were expanded after ben- An excellent M&E system was Phase 1 implemented strong M&E (M&E) for lessons and decisions plied a monitoring and evaluation e�ts were realized (and initially put in place. INCAGRO has and learned from best practices on what to scale up; lessons framework to extract lessons and documented); scenario planning produced substantial information around the world. The project will were well documented in the disseminate the lessons from this conducted with stakeholders about its experiences. Its learning develop an integrated information Implementation Completion project, particularly in two areas: con�rmed the need for pluralism; ability was great. and communications platform. Report (ICR) and the �rst (1) the understanding of and ICAR institutionalized learning Agriculture Development and rationale for public funding, and organization mode, including Credit Project (ADCP1). (2) the improvement of proce- learning by doing in Consortia, dures for the use of public funds. help desk portal for preparing pro- posals and disseminating lessons, effective M&E , and communica- tions system for outreach. PATHWAYS Deliberate effort via phasing Piloting, testing, and replicating Non-deliberate evolutionary Deliberate effort via phasing Non-deliberate evolutionary through APL. in one project. approach: building on previous through APL. approach: building on previous institutional reforms and lessons. institutional reforms and lessons. WORLD BANK ROLE APL instrument was valuable The project bene�ted from the Contributing factors to scaling The APL tool! This was very use- The World Bank team brought for scaling up from Phase 1 to World Bank’s pool of experi- up included policy dialogue, ful in planning the process. knowledge of global good Phase 2. ences and expertise in these Technical Assistance (TA), practice in advisory services areas. It was closely supported leveraging funds, committed Task and facilitated understanding of and monitored by ARD providing Team Leaders (TTLs), scenario the model on the part of a wide expertise at the design and planning, World Bank leader- array of stakeholders in Uganda. preparation stage and throughout ship, and CAS. Impediment were The addition of the Global implementation. procurement procedures and Environment Facility (GEF) to the �nancial management. Noted project added value. need for capacity strengthening was needed in this respect Source: Authors. TH INKING SY STEMATICA LLY A B OUT S C A LING UP CHAP T E R 2 — EMER G I NG FIND INGS FROM A NA LY ZING C AS E STUDY EX PERIENC ES OF S C A LING UP C H A P T E R 2 — E ME RGING F INDINGS F ROM A N A LY ZING C AS E STUDY EX PERIENC ES OF SC A LING UP 17 Scaling Up Spaces  Social space: All projects incorporated focus on gender  Fiscal space: No major �scal constraints were en- and marginalized groups in their design. countered in scaling up; projects were embedded in  Partnership space: Collaboration with other in-country Country Assistance Strategies (CAS) and where there partners took place. In particular, the value-added of would be a signi�cant �scal impact (as in the case of partnership with IDA and GEF was highlighted. Uganda), project costs were reflected in the Medium-  Learning space: From the start, the projects developed term Expenditure Framework (MTEF). However, an effective, integrated M&E system, information ex-post analysis should examine whether in all projects technology/communications platforms for document- �nancial sustainability and scaling up were supported ing lessons based on which scaling up interventions by �scal resources after project completion. were made. The use of help desk for proposal prepara-  Natural resource management space: The value-added tion is a noteworthy intervention. of the Global Environment Facility (GEF) component was highlighted in two projects. World Bank Role  Policy space: In all projects the necessary policy space  World Bank’s expertise and funding, policy dialogue, was created in Phase 1 (legal foundations). Linking na- Technical Assistance (TA), committed TTLs, and value tional research and extension institutions (in the case added of the APL. of Uganda) and creating a parallel system for research Conclusion from the Case Study Exercise and for innovation (in the case of Peru) are the two is- sues that might affect the scaling up process and need The application of the IFAD/Brookings framing questions to be addressed. on scaling up to the case studies favorably supports their  Institutional/organizational/staff capacity space: Strong likely practical value. In addition, the questions are easy-to- institutions (including Project Management Unit (PMU) remember as the concept of scaling up comprises only three integrated in a government agency and established re- elements: pathways, drivers, and spaces. gional of�ces), and building on existing organizational To ensure its operationalization, the scaling up guidance structures were conducive to scaling up CGSs. could be developed for each ARD sub-sector using the IFAD/  Political space: Except for Peru APL Phase 3, no signi�- Brookings framing questions or, where applicable, for its sub- cant political obstacles got in the way of systematically business lines. Based on the TTL feedback, the de�nitions expanding CGSs and other interventions nationwide. of drivers, spaces, and pathways could be elaborated upon, In particular, competitive matching grants were politi- with examples added. cally important in Uganda.  Cultural space: One of the key factors of success of the Armed with these case study �ndings, we can now move projects was adaptation of the CGS design to be in line to synthesize the insights into tentative guidance for TTLs with prevailing cultural norms of the targeted areas. contemplating scaling up worthy pilot efforts. AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 C H A P T E R 3 — P R OP OSE D GUIDANCE F OR SC A LI N G UP C OMPETITIVE GR A NT SC H EMES 19 Chapter 3: PROPOSED GUIDANCE FOR SCALING UP COMPETITIVE GRANT SCHEMES FOR AGRICULTURAL RESEARCH AND EXTENSION The previous chapter applied the IFAD/Brookings framing 2. Strengthening of the legislative structure has taken questions on scaling up to �ve cases of scaled up CGSs sup- place in support of farmer/producer organizations and ported by the World Bank in Azerbaijan, China, India, Peru, Intellectual Property Rights (IPR). and Uganda. This chapter uses the lessons learned from that 3. Stakeholders including farmer groups and associa- analysis to propose a set of general guidelines that can be tions have influenced agricultural services priorities useful to practitioners in scaling up CGSs. and activities. 4. In addition to strengthening agricultural research, As highlighted in Chapter 2, CGSs are being implemented signi�cant progress in revitalizing demand-driven across the world, including in several sub-Saharan African agricultural advisory services is being made. countries either as freestanding projects or as components of World Bank-supported projects. A review of the �ve case 5. The introduction of CGS has led key decision makers in studies in this report concludes that CGSs have had a positive countries to re-examine the effectiveness and adequacy impact and helped implementing countries to address priori- of their national agricultural services and to develop ties in the agricultural sector. In particular, available evidence a long-term national vision and strategies for their from project supervision reports concerning ongoing projects Agricultural Knowledge and Information Systems (AKIS). and implementation completion reports demonstrates that Areas of concern are these: the vast majority of these grant-funded sub-projects have re- sulted in new technologies and/or widespread dissemination 1. More effective communication is essential for and adoption of agricultural technologies. continued public support and to improve chances for sustainability. Based on analysis of selected case studies, many CGSs were 2. Engagement of the poorest of the poor has been initially conceived in order to provide support to the develop- challenging due to geographical remoteness aggra- ment and reform of an agricultural research system. Although vated by weak infrastructure and local language and research institutes and stations often constitute the primary literacy constraints. grant recipients, there is an increasing emphasis on widening 3. Parallel or coordinated investments into enabling fac- the eligible participants to include universities, public and private tors are important, such as infrastructure, market de- extension providers, non-governmental organizations, or private velopment, �nancial services, and regulatory issues. organizations, either as partners or as primary applicants. Grant proposals are now also expected to have a signi�cant element Scaling up is very much about “institutionalizing, � govern- of on-farm technology adaptation, testing, and dissemination, ment program change, or “mainstreaming� and “policy and expected impact is an important selection criterion for grant change. � These could, in theory, imply that a project could be approval. The primary target groups are farmers, but the project more successful by influencing policy and program—putting design also considers marginalized groups. As highlighted in in place institutional capacity, and then phasing out (exiting), Chapter 2, measuring relevant scaling up outcomes and im- rather than adding a new phase project. pacts (see Peru example in box 3.1) is key to scaling up. Frequently, there is also an unstated CGS objective with Some important sector-wide impacts based on the lessons respect to the institutional dimension, which could affect a learned from the �ve case studies may be summarized project’s scaling up. Where the public sector research and below: extension establishment is inef�cient, but �rmly entrenched and resistant to change, a CGS can be set up to circumvent 1. A number of agricultural services projects that include the existing system by creating a parallel one that is more CGSs have resulted in strengthening the institutional effective and responsive to clients in producing results. The base of agricultural Research and Extension (R&E). AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 20 CH A PTER 3 — PR OPOS ED GUIDA NCE FOR SC A LING UP C OMPETITIVE GR A NT S C H EM ES BOX 3.1: Peru Agricultural Research and Extension Project (INCAGRO) APL (2000–2010) Impact By improving the ef�ciency and effectiveness of the agri- has many new competitive funding schemes supporting cultural technology system, the Project has contributed to agriculture, with at least two other schemes managed increased agricultural productivity and farmers’ income. In by the Ministry of Agriculture and others. In interviews, addition, a major social bene�t has been the strengthening these funders credit INCAGRO for much of the success of producers’ organizations, especially as they have been of competitive funding schemes, including their role in explicitly designated under Phase 2 to take the leadership developing competitive funding strategies; in building of the strategic alliances in which NGOs played a domi- the capacity of producer organizations to follow rigorous nant role under Phase 1. The impact study realized in 2010 funding protocols and implement projects; preparing a showed an Internal Rate of Return (IRR) of 23 percent. cadre of professionals that now manage other funds; developing a pool of competent service providers; and Although the project implementation arrangements did not creating competitive funding mechanisms and tools explicitly target the rural poor, its expansion nationwide has that were directly adopted by the new funding agen- included the poorest regions of the country. Speci�c mea- cies. These are actually some of the strongest signs of sures have also been designed during Phase 2 to facilitate INCAGRO’s sustainability, if not as a program then as the participation of vulnerable and traditionally marginalized a concept. Furthermore, the INCAGRO funding process groups. This worked well for the participation of indigenous was replicated by Dutch bi-laterals, the IFAD-sponsored communities, which bene�ted from 116 sub-projects in ag- Peru Management of Natural Resources in the Southern ricultural extension (out of a total of 334 in Phase 2). The Highlands Project (MARENASS) project, and in other participation of women’s organizations was lower, with 40 sector projects and interventions in Peru. sub-projects implemented. Participation in INCAGRO ex- tension sub-projects has positively impacted the economic Note: Implementation and Completion Report (ICR) and and personal lives of the indigenous people and women’s the World Bank Independent Evaluation Group (IEG) evalu- organizations involved in the 156 sub-projects. ations were available only for the Peru project; however, impact will be also quanti�ed for other case studies when Peru is now explicitly pursuing a countrywide, plural- ICRs and IEG evaluations become available. istic, demand-driven agricultural innovation system, in which competitive funding has a key role. The country Source: Information provided by the Project TTL. expectation is that when the two systems are compared, the up guidance can be reaf�rmed as valuable in contributing to existing system will be encouraged to reform. good practices in the agricultural sector, beyond the particular case of use of competitive grants in agricultural services, and For example, the Peru INCARGO APL Phase 2 underestimated should be made available through the ARD website. A num- the risk of the development of a parallel system of research and ber of interesting observations were made about the study: innovation and did not suf�ciently address the question of the  Elements of the IFAD/Brookings framing questions institutional and �nancial sustainability of the INCAGRO model are particularly constructive in bringing out tacit and the exploration of different options for this with the govern- knowledge. ment of Peru early on during the APL (World Bank, 2011a).  Framing questions are a �rst step toward general agreement on terms. 3.1 PROPOSED GUIDANCE FOR SCALING UP CGS  Applying “importance weights� to drivers, spaces, and FOR AGRICULTURAL RESEARCH AND EXTENSION pathways could be considered.  Developing a flowchart to ease the project designer’s The World Bank case studies informed the scaling up guid- use of the scaling up guidance could be useful. ance for CGSs for agricultural research and extension as presented in table 3.1.  A focus on learning during the scaling-up process is important as interventions evolve during implementa- Based on the feedback of the Sustainable Agricultural tion. The lessons learned through M&E, along with Systems, Knowledge and Institutions (SASKI) Thematic Group external knowledge should act as a feedback loop for validation meeting conducted on March 7 , 2012, the scaling the next design. TH INKING SY STEMATICA LLY A B OUT S C A LING UP C H A P T E R 3 — P R OP OSE D GUIDANCE F OR SC A LI N G UP C OMPETITIVE GR A NT SC H EMES 21 TABLE 3.1: From a Pilot to Scaling Up CGSs for Agricultural Research and Extension: Considerations for Scaling Up (Living Document) EXPERIENCE AT THE EXPERIENCE AT THE PILOT PHASE FULL SCALE APPLICABLE ACTIONS DRIVERS—FORCES PUSHING THE SCALING UP PROCESS FORWARD Ideas, models to be scaled up 1 Create awareness of the possibilities Farmers and other value chain actors  Articulate demand for new knowledge and technology. Producers, of more pluralistic approaches to more aggressively and purposively traders, and others must be able to express their demand and must Research and Development (R&D) but pursue a range of more pro�table new be able to adapt and adopt new knowledge and technology. proceed with some caution, given the technologies.  Organize effective supply of new knowledge and technology from uncertainty perceived about just how the public research system, but also from other sources, such as an untried model may work. indigenous knowledge, private sector research, and transfers from abroad. 2 The pilot project design articulates Research-extension alignment: tech-  Create an integrated platform at all levels of National Research how the research and extension, in nology generation and uptake might Organization and an institution responsible for advisory services particular the state line departments often best be combined in one project. operations, including their decentralized of�ces. This infrastruc- of extension, would be linked, or ture might support joint systems such as M&E, knowledge base, how research would be linked with workflow, and a joint internal and external portal. the other mechanisms for technology transfer, preferably through inclusion of research and extension in the same project. 3 Demonstrate results in a �rst phase Government informed by pilot experi-  Adopt a scaling up mind-set in pursuing a design aligned with the (e.g., promising technology adoption ence may desire to achieve national vision. and increased incomes of farmers and coverage in a staged manner.  Pursue an appropriate legal framework and institutional reform. experience gained). The pilot provides  Engage the private sector in R&D. opportunity to �ne-tune the approach.  Emphasize strong producer organizations.  Analyze the capacity to absorb additional R&D funding in small countries. Vision and leadership—A vision is needed to recognize where scaling up is necessary, desirable, and feasible. Visionary leaders or champions often drive the scaling up process forward. 4 Pilot CGS to foster the engagement Mainstreamed engagement of the  Identify and empower diverse high-level champions for CGS and of the private and other sectors in private sector and producer organiza- reforms to secure commitment. technology transfer elements. tions, NGOs, and universities in  Develop a broadly owned Agriculture Sector Strategy and articu- agricultural research and extension late an Investment Plan (Scenario Planning may help). via CGSs.  Develop long-term perspectives and long-term research goals in the context of the Strategy and a rapidly changing global environment.  Strengthen capacity for developing vision for agriculture and agricultural science and for sharing it with stakeholders.  Develop the ability to plan and implement research programs to realize the vision.  Predispose agricultural research system reform, including strength- ening of the National Research Organization’s role, to be a catalyst of change in the national agricultural innovation system.  Foster effective linkages with international, regional, and national research centers of excellence.  Recognize intellectual property rights and identify where reform of linked industries (e.g., seeds) is required.  Identify early on whether there is room for scaling up, where there is demand for a larger scale, what size limitations may prevail, and consider how best to integrate the desired activities. External catalysts—Political and economic crisis or pressure from outside actors may drive the scaling up process forward. 5 Government responds to structural Agricultural technology generation  Work with champions to secure buy-in from the government and problems in the agricultural sector, and transfer formulated in a govern- other key institutions and players, including the private sector. such as insuf�cient productivity ment strategy with wide buy-in from  Work with the �nancing agency to exert pressure for accountability. growth. Financing agency wishes for key stakeholders with all committed to  Provide cogent technical assistance to �ll critical capacity gaps. more accountable and productive use the role of a CGS.  Build a strong communication strategy of the national agricultural of funds. innovation system and exploit the flow of knowledge through it. (Continued) AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 22 CH A PTER 3 — PR OPOS ED GUIDA NCE FOR SC A LING UP C OMPETITIVE GR A NT S C H EM ES TABLE 3.1: From a Pilot to Scaling Up CGSs for Agricultural Research and Extension: Considerations for Scaling Up (Living Document) (Continued) EXPERIENCE AT THE EXPERIENCE AT THE PILOT PHASE FULL SCALE APPLICABLE ACTIONS DRIVERS—FORCES PUSHING THE SCALING UP PROCESS FORWARD External catalysts—Political and economic crisis or pressure from outside actors may drive the scaling up process forward. 6 Look for good practice around In the developing world, innovative  Build a wide stakeholder base in the Bank. the world. practices in LAC informed interven-  Look to Organization for Economic Co-Operation and Development tions in AFR. But TTLs should look (OECD) benchmarks, e.g., comparing agricultural research intensity widely, including to experience in ratios has helped to prioritize sector investment in some countries. more-developed countries. Incentives and accountability—Incentives are key to drive the behavior of actors and institutions toward scaling up. They might include rewards, competitions, and pressure through the political process, peer reviews, and other evaluations. Monitoring and evaluation against goals, benchmarks, and performance metrics are essential ingredients to establish incentives and accountability. 7 Pilot demonstrates that CGSs incentiv- Created closer links and collaboration  The sustainability of project activities will depend, to a large ize and engage diverse innovation by involving multiple stakeholders in extent, on how bene�ts accrue to farming communities and other actors, particularly farmers, private priority setting and market-relevant stakeholders, so as to give them an incentive to maintain and sector, universities, and other con- research, and by developing strong further develop the investments made under the project. It will also tributors to agricultural innovation. research–advisory service interfaces depend on effective governance structures to ensure that farmers at the national, zonal, and local levels. play a critical role in all aspects of the program. Sustainability Created a funding mechanism to will also be influenced by the effectiveness and ef�ciency of the allow farmers and extensionists to interface between agricultural research and advisory services, in attract researchers to work with them particular at the local level, and the degree that together they can on collaborative applied research on engage directly with the farm community that stands to bene�t farmers’ �elds. most in the �rst instance from such technical progress. Innovators Established formal and informal active in the R&D system can be incentivized in many ways, most means to ensure that farmers have a obviously by directly sharing in the bene�ts of technical progress, voice in decisions affecting research such as through sharing in property rights associated with pro�t- priorities, funding, execution, and able innovations. This is easy for the private �rms involved, who evaluation. can protect and market materials protected by patents or secrecy (such as in hybrid crop cultivars). But ordinary personal excitement about useful discoveries can provide adequate incentive for some players to be productively engaged. Accountability is best handled by high levels of transparency in the CGS system adopted, aided by effective and timely M&E arrangements.  Consider prizes for innovative R&D and adoption (output-based scaling up), advance market commitment such as Harvest+ has used to interest companies in vitamin-rich cultivars.  Ensure a strong M&E system that identi�es problem grants and puts in place rule-based transparent procedures to close or �x them.  Consider provision of seed money and credit line availability to complement CGSs, perhaps through credit guarantees. SPACES—OPPORTUNITIES THAT CAN BE CREATED OR OBSTACLES THAT NEED TO BE REMOVED TO OPEN UP THE SPACE FOR INTERVENTIONS TO GROW Fiscal/�nancial—Fiscal and �nancial resources need to be mobilized to support the scaled up intervention; and/or the costs of the intervention need to be adapted to �t into the available �scal/�nancial space. 8 Fiscal/�nancial implications are small A full scheme will have much larger  Build overall commitment to public and private funding of R&D. and secondary to the testing purpose. �scal/�nancial implications, and  Conduct �scal capacity review and if applicable integrate in the thus requires serious buy-in by policy Medium-term Expenditure Framework (MTEF). makers and the Ministry of Finance,  Review options for reallocating ineffective R&D funds to CGS. as well as careful consideration of the  Review options for co-�nancing from private sector and producer commitments to be made by producer organizations. organizations and other private-sector entities.  In decentralized systems may need deliberate harmonization to achieve national coverage. 9 Donors are happy to �nance/own pilots. The country is expected to own the  Country co-funding in increasing amounts over time becomes full scale and to eventually take over central to the country taking ownership and sustaining the scheme. the �nancing, although concerned Pay more attention to co-�nancing and the need for progressive donors may well stay the distance increase in co-�nancing. for quite long-term support if success  While it is important to focus on establishing funding cycles and is clear. tools, eventually the emphasis must shift to developing a more sus- tainable system based on private cost recovery, funding partners, and ongoing government support.  Ensure that CGS embeds aspects of continuity. Schemes differ and many activities may be deemed ineligible for follow-on funding.  Consider decentralization, including funding from other sources at lower levels of government and from engaged civil society organizations. (Continued) C H A P T E R 3 — P R OP OSE D GUIDANCE F OR SC A LI N G UP C OMPETITIVE GR A NT SC H EMES 23 TABLE 3.1: From a Pilot to Scaling Up CGSs for Agricultural Research and Extension: Considerations for Scaling Up (Living Document) (Continued) EXPERIENCE AT THE EXPERIENCE AT THE PILOT PHASE FULL SCALE APPLICABLE ACTIONS SPACES—OPPORTUNITIES THAT CAN BE CREATED OR OBSTACLES THAT NEED TO BE REMOVED TO OPEN UP THE SPACE FOR INTERVENTIONS TO GROW NRM/environment—The impact of the intervention on natural resources and the environment must be considered—harmful effects mitigated or bene�cial impacts promoted. 10 Sustainable Land Management/ Global Environment Facility (GEF)  Consider Sustainable Land Management (SLM) platform for sus- Natural Resource Management/ �nancing may be a possibility, but also taining the project’s impact on SLM and for a future Sector Wide Climate Smart Agriculture (SLM/ governments increasingly committed Approach (SWAP) to coordinate SLM/NRM/CSA interventions in NRM/CSA) interventions are not to a green agenda may be able to the country. necessarily attractive for impatient support.  Put in place Environmental Impact Assessment with associated public funders; investments imple- manual and ensure that CGS management has access to environ- mented sporadically by NGOs may be mental expertise. insuf�cient. 11 Pilot operations rarely pose environ- With scale comes the possibility of  When scaling up, because potential negative environmental impact mental threats, given the modest signi�cant environmental impacts, is larger, ensure a stronger environmental management framework scale of intervention, but awareness such as related to exploitation of soil that effectively monitors at the larger scale. of such risks should be made part of and water resources, so awareness  In addition to managing it, proactively mainstream environmental routine scrutiny. must be heightened. sustainability in the grant call—incentivize innovation in this direction.  Create windows for projects with positive environmental effect. De�ne criteria upfront; cut off non-compliant options. Policy—The policy (and legal) framework has to allow or needs to be adapted to support scaling up. 12 Continued reform in the National Build political will for this new funding  Put in place enabling policies and institutions and include a strong Agricultural Innovation System (NAIS) approach, surely helpfully informed public-sector framework for research. sets the scene for innovative pilots. by transparently documented results,  Create a parallel system for research and innovation when the through a strong M&E system that public-sector research and extension establishment is inef�cient demonstrates institutional productivity but �rmly entrenched and resistant to change. The expectation is and social impacts. that when the two systems are compared, the inef�cient system will be encouraged to reform.  Build a strong M&E system from the beginning. 13 Farmers’ Associations are in their Farmers are organized into associa-  Legislate to enable legal farmers’ organizations. infancy and might not absorb great tions federated at the national level  Foster institutional development of farmer organizations. amounts of funding nor responsibility by commodity and region with a cross-  Facilitate social assessment of cooperative associations of in the project’s initial start-up phase. cutting apex body. farmers. Furthermore, their legal status might  Produce detailed guidelines to put producer organizations on a be often unclear and does not allow stronger legal basis and make them more independent. The project them to act as a true legal entity in might design a transitional period during which the capacity of dealings with private companies. such associations to manage and control their own assets, set the terms of their own contracts, and generally direct their own affairs would be gradually raised.  Funding of R&D on commercial crops by producer organizations is desirable. Institutional, organizations, staff capacity—The institutional and organizational capacity has to be created to carry the scaling up process forward. 14 Pilot demonstrates that considerable A consolidated system for policy,  Situate the grant scheme inside the institution that has the highest capacity exists in universities, NGOs, regulation, operation, and �nancing probability of sustaining the scheme. Government entity may favor and the private sector that can be of a pluralistic agricultural technol- institutional sustainability, reduce costs, ensure political buy-in, harnessed through CGS. ogy system is in place. It is based on and make it possible to leverage available capacity for �nancial institutional pluralism both in imple- management, procurement, monitoring, and evaluation. Proximity mentation and �nancing. Farmers and to government may also be desirable to the extent that public other stakeholders have an increased institutions are key stakeholders within the grant and innovation participation in the governance of the system. It would also allow ownership beyond project areas and system, particularly strengthened by beyond the project implementation phase. A permanent focal point increasing the share of the funding or champion within the responsible Government entity should be they provide. responsible for policy, regulation, public �nancing, and interna- tional cooperation for agricultural technology programs.  Ensure that governance of CGS has multi-stakeholder representa- tion but independence from the political process. (Continued) AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 24 CH A PTER 3 — PR OPOS ED GUIDA NCE FOR SC A LING UP C OMPETITIVE GR A NT S C H EM ES TABLE 3.1: From a Pilot to Scaling Up CGSs for Agricultural Research and Extension: Considerations for Scaling Up (Living Document) (Continued) EXPERIENCE AT THE EXPERIENCE AT THE PILOT PHASE FULL SCALE APPLICABLE ACTIONS SPACES—OPPORTUNITIES THAT CAN BE CREATED OR OBSTACLES THAT NEED TO BE REMOVED TO OPEN UP THE SPACE FOR INTERVENTIONS TO GROW Institutional, organizations, staff capacity—The institutional and organizational capacity has to be created to carry the scaling up process forward. 15 Covers one region/a few subjects. Covers the country/many subjects.  Prepare the scheme for wider geographical coverage, for example, by developing regional offices in addition to a national office. A set of criteria can be considered to select the decentralized macro-regions, which could be accompanied with administrative offices established in each macro-region to facilitate the administration of project interventions. These cri- teria could include: (1) the key variables used for the definition of the target population of the Project, as related to the produc- tive efficiency and degree of market definition and relationships with the market (presence of agro-industrial and product chains in relation to the demand, public service operators, and their capacity for technological innovation), economic and infrastruc- ture context; (2) the potential for farming and forest resources to respond to the globalization of the market (potential of impact of services for innovation); and (3) aspects that harness the success of the intervention strategy of the project  Put in place a larger subject matter expertise base, for example, through a bigger pool of peer reviewers and more subject matter specialists in the management of the competitive fund. 16 Small sub-projects �nanced via CGSs Capturing integration and economies  Mainstream a consortium approach and competitive selection but at high overhead cost per project. of scale bene�ts with fewer sub- process throughout country research institutes. projects with increased funding and  Some of the issues that should be addressed when developing a lower percentage project overhead consortium are objectives of the consortium; governance structure; cost. research program guidelines (program components and activities); intellectual property rights management; knowledge management and information sharing; capacity building and human resources development; strategy for commercialization and utilization of research outputs; bene�t and cost sharing arrangements; �nancial management and procurement arrangements; and sustainability and withdrawal measures following a project’s completion. Information requirements in these areas should be clearly spelled out in the call for proposals; the operational manuals and guide- lines for establishing, �nancing, and operating the consortia should be developed. 17 Ad hoc competitive grants managed Institutional capacity for running  Modular project design is conducive to scaling up: in the context of by a Project Management Unit (PMU). CGSs is in place: an institutional CGS, modules can be at various levels; a CGS for a speci�c sector environment that is conducive to the or sub-sector; a set of themes for which competitive applications flow of knowledge, to collaboration, are called, or modular agreements among collaborators bidding experimentation and implementation for CGS support. Ensure there are uniform procedures for the sub- of innovations. project selection, contracts, and monitoring.  Explore institutional options for sustainably managing and �nanc- ing CGS over the long term within the Ministry of Agriculture, Ministry of Science and Technology, or perhaps as a separate autonomous body. Since country situations vary so greatly, this means that the circumstances that prevail should be carefully assessed to set up the scheme where it has the best chances of sustainable success.  Developing local capacity is crucial to facilitate the importation and �tting of appropriate technologies to local conditions. In this con- text, a strategy of developing local and decentralized institutional capacity, seeking stronger and more quali�ed human resources, and producing adapted and available technologies to adequately address speci�c and localized problems should be followed.  Ensure strong and renewable governance representing broad stakeholder interests. 18 Strong intellectual/methodological While such support is still there it  De�ne operational procedures and disciplines clearly. Improvising support, e.g., by an institutional would usually not grow proportionally. or deciding on the spot has obvious limitations. development specialist. (Continued) TH INKING SY STEMATICA LLY A B OUT S C A LING UP C H A P T E R 3 — P R OP OSE D GUIDANCE F OR SC A LI N G UP C OMPETITIVE GR A NT SC H EMES 25 TABLE 3.1: From a Pilot to Scaling Up CGSs for Agricultural Research and Extension: Considerations for Scaling Up (Living Document) (Continued) EXPERIENCE AT THE EXPERIENCE AT THE PILOT PHASE FULL SCALE APPLICABLE ACTIONS SPACES—OPPORTUNITIES THAT CAN BE CREATED OR OBSTACLES THAT NEED TO BE REMOVED TO OPEN UP THE SPACE FOR INTERVENTIONS TO GROW Political—Important stakeholders, both supporting and opposing the intervention, need to be attended to through outreach and suitable safeguards to ensure the political support for a scaled up intervention. 19 Pilot generates enthusiasm for CGS CGSs are generally strongly politically  Ensure that government is committed to investing in R&D as a among a few key decision makers supported as they become known national priority. aware of the bene�ts of introducing for delivering useful results, which  Secure political backing for scaling up including via decision mak- competition and contestability into in turn can contribute signi�cantly to ers, leaders, champions, etc. agricultural R&D activities and by inclusive economic development.  Government policy: a major challenge is to mainstream the CGS developing quick wins with credible concept and methodology into national research and extension impact assessment. programs and to integrate them into government policies, strate- gies, and funding.  Arrange credible impact assessment and strong communication skills with key stakeholders.  Overcome vested interests in the traditional system.  Ensure broad commitment from governments and stakeholders: to be successful, a CGS needs the sustained support of all affected govern- ment ministries, departments, research and extension institutions, and other relevant donor, non-governmental, and private entities. It is particularly important that the government demonstrates its com- mitment by con�rming that it is ready to establish an independent administrative unit (PMU) and a transparent �nancial system that will allow the CGS to be managed in a way that prevents conflict of interest and political interference. In addition, program management staff must have access to senior of�cials in the agricultural sector to help secure counterpart funding as necessary and to have the flexibility needed to change program requirements and structures to address new problems and opportunities as they emerge.  Develop a strong communications strategy for broadcasting suc- cesses; EMBRAPA has done this with remarkable success in Brazil, for instance. Cultural—Possible cultural obstacles or support mechanisms need to be identi�ed and the intervention suitably adapted to permit scaling up in a culturally diverse environment. 20 First-generation CGSs foster collabo- Second-generation CGSs emphasize  Build a culture of scienti�c quality through rigorous peer review ration primarily within the National the importance of the demand side mechanisms. Agricultural Innovation System (NAIS), with a broadened and more plural  Ensure CGS management unit includes skills in gender and social by (usually rather cautious) exploratory community of stakeholders. analysis, and where needed, skills in indigenous language and engagements with the private and cultural interaction. other sectors.  Ensure mind-set is properly incorporated; get buy-in from all stake- holders for collaboration, rigor, and competition.  For more upstream research, the dialogue may well be focused more in the Ministry of Science and Technology than the Ministry of Agriculture. Social (added by WB)—A critical base on which interventions can get scaled up is the social capital created and nurtured. 21 Social and gender assessment capac- Second-generation CGSs emphasize the  Conduct Poverty and Social Impact Analysis (PSIA) and social ity is incorporated in pilot project. importance of the demand side, as well assessment. The social assessment might identify the following as as welcoming growing trust between key social development issues/principles which should underpin the previously non-communicating public, project’s strategy and implementation: (1) participation, (2) inclusion, private, and third sector entities, an ef- (3) poverty, (4) gender, (5) capacity building, and (6) information, fective building of social capital among education, and communication campaign. previously non-cooperating entities.  Farmer empowerment is a prerequisite for demand-driven advisory Social and gender analysis is services. To ensure that as many farmers as possible participate in mainstreamed and as necessary and bene�t from processes intended to improve the availability and creates special windows for poorer use of technologies and information, farmers should be categorized groups (indigenous peoples, especially in their progression from subsistence to market orientation, using women, as in Peru; youth may warrant criteria that take into consideration the asymmetries in power, explicit attention). resources, and capacity. Farmer categorization would promote the learning effects between group members and also help tailor the services to better suit the speci�c needs of different farmers, and group marketing and outgrower schemes will help the disadvan- taged persons take better advantage of market opportunities. Include component to strengthen producer organizations. (Continued) AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 26 CH A PTER 3 — PR OPOS ED GUIDA NCE FOR SC A LING UP C OMPETITIVE GR A NT S C H EM ES TABLE 3.1: From a Pilot to Scaling Up CGSs for Agricultural Research and Extension: Considerations for Scaling Up (Living Document) (Continued) EXPERIENCE AT THE EXPERIENCE AT THE PILOT PHASE FULL SCALE APPLICABLE ACTIONS SPACES—OPPORTUNITIES THAT CAN BE CREATED OR OBSTACLES THAT NEED TO BE REMOVED TO OPEN UP THE SPACE FOR INTERVENTIONS TO GROW Social (added by WB)—A critical base on which interventions can get scaled up is the social capital created and nurtured.  Make value-chain governance mechanisms a prerequisite for nonproduction research and related advisory services on a demand basis.  Build capacity to carry out social assessment, both ex-ante and ex- post, especially among communities where there is low capacity to participate in CGSs.  Improved quality of scaling up (development with equity) may be fostered by engaging more diverse thinkers in innovation processes. Partnership—Partners need to be mobilized to join in the effort of scaling up. 22 Some public-private partnerships Researcher-investor-farmer partner-  Recognize complementary assets of different partners to achieve are established, but the number is ships and targeted technology transfer synergies—not partnerships for partnerships’ sake! small and there is continuing mistrust accumulate in a positive way to  Provide incentives for public and private partners to work between public and private sector. The bolster con�dence, and reduce the together. The simplest conditions for successful collaboration are missing element is a lack of adequate perception of such arrangements as where all parties perceive they will individually bene�t, and do investment in partnership building and being too risky. so more successfully than when working alone. Mutual trust is inadequate attention to the dissemi- clearly critical! nation of potential bene�ts.  Invest in facilitation and match-making activities with the help of service providers (brokers), platforms, help desk or formal coordina- tion organizations.  Provide training programs to overcome skill gaps and cultural bar- riers between participants from public and private sectors; this is particularly the case for service standards.  Hire facilitators to help build trust, as necessary. The situation pre- vailing varies greatly around the world, so local needs assessment is usually required to determine readiness in the private sector, especially among small and medium-sized enterprises (SME) for adequate participation. 23 Issues of intellectual property rights The emerging partnerships with the  Seek agreement on IPR policies. and ownership are analyzed and op- private sector and NGOs raise issues  Traditionally, agricultural research innovations are generated in tions reviewed. relating to IPR, contractual arrange- public sector research institutions with limited interaction with the ments, and bene�t sharing. private sector. One of the key functions of the public sector is to provide for a regulatory framework and its enforcement capacity that enables the private sector to have a reasonable chance of obtaining a return on its investment. This requires protection of intellectual property rights and a credible contracting system. Governments should make efforts to build capacity in IPR by estab- lishing an Intellectual Property Of�ce and an Intellectual Property Service Center or strengthen existing of�ces.  Consider establishing a unit (within NAIS or other government entity), which would be adequately empowered to carry out intel- lectual assets validation, valuation, protection, and management. It should be headed by a person of high seniority and standing in the agricultural research community with adequate expertise in IPR. The regional sub-units should be headed by competent technology management professionals well trained in the area of IP review, IP registration, technology valuation, licensing processes, royalty audits, communication and outreach efforts, dispute resolution, and enterprise incubation. In these units major emphasis should be put on developing procedures for validating and up scaling technologies to commercial use. (Continued) TH INKING SY STEMATICA LLY A B OUT S C A LING UP C H A P T E R 3 — P R OP OSE D GUIDANCE F OR SC A LI N G UP C OMPETITIVE GR A NT SC H EMES 27 TABLE 3.1: From a Pilot to Scaling Up CGSs for Agricultural Research and Extension: Considerations for Scaling Up (Living Document) (Continued) EXPERIENCE AT THE EXPERIENCE AT THE PILOT PHASE FULL SCALE APPLICABLE ACTIONS SPACES—OPPORTUNITIES THAT CAN BE CREATED OR OBSTACLES THAT NEED TO BE REMOVED TO OPEN UP THE SPACE FOR INTERVENTIONS TO GROW Partnership—Partners need to be mobilized to join in the effort of scaling up. 24 Public agencies traditionally distrust- If distrust prevails, more imaginative  Establish clear realistic timeframes and service standards; public- ful of private entities may be willing dialogue may be necessary to over- sector processes are often considered too slow. to experiment when the flows of come public-sector fears of signi�cant  Consider mechanisms for guaranteeing credit. Establish business �nancial resources are modest, which �nancial resources being entrusted development units (e.g., for a new and promising technology, assist can help in the learning process. to private �rms potentially active in patenting, look for partner to develop them). Develop a tool kit of the NAIS. instruments to be used in public research entities.  Brokers, platforms formed, regulators, investors in the platform should make partnerships live.  Promote open transparency about contracts, informing what is being done with public money.  Foster open innovation (as in Information and Communication Technologies [ICT]). Innovate on top of the developed technology. Learning—Knowledge about what works and does not work in scaling up needs to be harnessed through monitoring and evaluation, knowledge sharing, and training. 25 Usually focus on testing a methodology/ The focus is on impact, but with  Evaluation mechanisms must become more focused on impact and approach and be less concerned with continuous learning and feedback. less on internal procedures. impacts per se, although documenting  Make sure that the methodology (as described in the operational some is useful in planning for scaling up. manual) is maintained in a rigorous manner and not compromised in order to scale up quickly (risking throwing away the most impor- tant part of the pilot experience).  The full-scale scheme should select sub-projects principally for expected impact, whereas the pilot may have selected also for learning purposes.  Three major steps are crucial to sustain CGS in the future: (1) Develop an evaluation design that provides credible results with reasonable costs; undertake a detailed analysis/evaluation of the impacts, results, and achievements of ongoing or completed CGS and the individual sub-projects; (2) clearly de�ne the net bene�ts from the CGS and the individual sub-projects funded through grants; and (3) widely disseminate the results of the evaluation to encourage governments, donors, private sector entities, and bene�ciaries to buy into CGS and ensure continued �nancial and institutional support.  Include communication skills in CGS management unit and build strong communication capacity.  Facilitate exchange among grantees and with other CGS programs.  Build knowledge management and sharing strategy.  Ensure transparent policies and operational procedures. 26 Semantic issues abound, especially In spite of International Fund for  One view is that scaling up plus learning constitutes “Evolution.� when new processes and procedures Agricultural Development (IFAD) (and New development efforts should ideally emerge from such evolu- are being introduced. World Bank) endeavors, many se- tion. In this way it is hoped that development efforts can be more mantic issues persist as development successful than in the past. Perhaps future sector work should look practitioners can be slow to converge at this in a wide-ranging way to explore more transformational on common use of terminology. ways forward. Meantime, it is hoped that checklists such as this one may contribute to convergence. PATHWAYS Pathways are appropriate actions and steps that should be followed to ensure that a project is taken to an appropriate scale. They can follow different “dimensions�:  Expand services to more clients in a given geographical space  Involve “horizontal� replication, from one geographic area to another  Use “functional� expansion, by adding additional areas of engagement, and  Employ “vertical� upscaling, that is, moving from a local or provincial engagement to a nationwide engagement Each pathway will have its own drivers and spaces (see IFAD/Brookings framing questions in Chapter 1). (Continued) AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 28 CH A PTER 3 — PR OPOS ED GUIDA NCE FOR SC A LING UP C OMPETITIVE GR A NT S C H EM ES TABLE 3.1: From a Pilot to Scaling Up CGSs for Agricultural Research and Extension: Considerations for Scaling Up (Living Document) (Continued) EXPERIENCE AT THE EXPERIENCE AT THE PILOT PHASE FULL SCALE APPLICABLE ACTIONS PATHWAYS 27 The pilot may affect a sub-region or The full-scale scheme is meant to  Deciding on the right size is key. Is there suf�cient absorption sub-sector but will affect the sector as affect the whole sector, and if suc- capacity in the R&D system? Should fresh capacity development be a whole only in a marginal way. cessful, may well have profoundly built into the operation? Just how gradual or otherwise ambitious wide effects. the project should be requires careful assessment by the design and supervision teams. The interaction of the scheme with other government initiatives must be mapped and understood.  Increasingly engage regional research organizations (e.g., CORAF, ASARECA) to facilitate scaling up where they exist. But globally, the main CGS business will be primarily domestic.  Consider outsourcing some grants outside the country where suf- �cient expertise does not exist within the country. WB ROLE 28 As variously noted, the International Several sectors across the World  Several World Bank Instruments are available for scaling up: e.g., Financial Institutions including IFAD Bank are further exploring the scaling Additional Financing, Repeater Projects, Adaptable Program Loans, and the World Bank have been active up agenda. But notwithstanding phased approach (sequence of individual operations). in scaling up for quite some time. progress in other sectors, CGS will continue to be important in the ARD portfolio. Scaling up in the World Bank has mostly been in Community Driven Development (CDD) and on geographical bases. Note: This guidance was compiled based on �ve populated case studies. We welcome additional relevant input from practitioners using CGSs for agricultural research and extension their future operations. Source: Authors. TH INKING SY STEMATICA LLY A B OUT S C A LING UP C H A P T E R 4 — C ONCL USIONS 29 Chapter 4: CONCLUSIONS The application of the IFAD/Brookings framing questions on  Spaces: (1) management capacities of potential imple- scaling up to the case studies, as well as scaling up valida- menting organizations (institutional space); (2) enabling tion feedback, con�rms the need to develop two new tools policy and/or legal frameworks; (3) constituencies for rural development practitioners and their partners seeking (political incentives and policy space); (4) political and to support scaling up impact in rural development. One tool security issues; and (5) prospects for �nancial sustain- is an analytical guide for speci�c sub-sectors or a business ability and stability in flow of resources. line highlighting key considerations to guide scaling up and to encourage development partners to systematically think The Brandeis report moreover makes the valid point that de- about scaling up impact. The other tool is an updated list cisions about whether or not to scale up need to be made of case studies for a speci�c sub-sector or a business line. before the demonstration project has given evidence about Developing these tools in partnership is desirable. whether the innovation is effective and ef�cient. It also rec- ommends development of a scaling up tool kit for managers The IFAD/Brookings framing questions on scaling up can be implementing the Development Marketplace (DM) projects. useful to any World Bank sector as well as to donors, gov- ernments, and other development agencies for analyzing We conclude that scaling up is already a widely used op- existing cases and developing guidance for future scaling up. erational approach at the World Bank. For example, between However, they are not a blueprint, but rather a starting point FY06 and FY11, 110 Agriculture and Rural Development in the scaling up effort. They should be complemented by Sector Board projects (34 percent of the total new ARD relevant analytical tools and further re�ned in light of gained project portfolio) were scaled up. The phased approach (se- scaling up experience. ries of investment operations) was the dominant scaling up instrument. All the parent projects that were scaled up had Not all elements of the IFAD/Brookings scaling up frame- satisfactory progress toward development objectives and work might be applicable to different institutions and ty- implementation. The rationale for scaling up was the success pologies of projects. For example, the World Bank team has of the earlier projects that led to high demand from the cli- found that “social space� (i.e., social capital created and ents. All this information should be well documented in the nurtured) was an important element to add to the scaling annual ARD Portfolio Reviews and disseminated in order to up framework. Based on past local experiences, programs facilitate cross-regional knowledge sharing about scaling up. work best where community cohesion exists. Lack of so- cial cohesiveness is closely linked with failed sub-projects In the future, the World Bank could focus on some areas (Hancock, 2003). not addressed in this report including (1) a more systematic review of the institutional incentives and/or obstacles in Another example is the innovation state of practice spec- the World Bank to scaling up; and (2) the role of different trum, where the companion Brandeis report (2011) found operational instruments, including M&E and Knowledge only the following four drivers and �ve spaces, rather than all Management underpinning scaling up. Prospective appli- IFAD/Brookings framing questions, applicable: cation of this framework could be explored for other sub-  Drivers: (1) clarity about potential driving or imple- sectors, such as scaling up value chains or irrigation. menting organization(s) for replication or expansion; In the meantime, ARD plans to join an IFAD-proposed Learning (2) mediating or influencing organization that is Alliance on scaling up, which may include activities such as ready to support the transition from test to scaling mutual peer reviews, joint products, and learning events. up; (3) champions; and (4) incentives for scaling up. AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 A N N E X 1 — ADDIT IONAL F INANCING F OR I N VESTMENT LEND ING 31 Annex 1: ADDITIONAL FINANCING FOR INVESTMENT LENDING Additional Financing (AF) is a World Bank instrument for imple- borrower, as compared to preparing a repeater or other mentation of additional or expanded activities that scale up new project? Can the magnitude or scope of the scale a project’s impact and development effectiveness where the up be easily accommodated in the context of the ongo- Task Team Leader (TTL) describes all modi�ed, additional, or ing project, relying on the borrower’s existing imple- expanded activities; summarizes the results of their appraisal; mentation capacity and other project arrangements? and con�rms their consistency with the project development  What is the impact of the additional loan on the con- objectives and the current Country Assistance or Partnership tinued economic justi�cation of the project? Strategy. Additional Financing strengthened the Bank’s ability  What is the impact of any changes in �duciary, safe- to scale up the results and development impact of successful guard, or other aspects since appraisal of the ongo- projects more quickly and with much lower processing and ing project that might affect the justi�cation of the transaction costs than was possible previously. additional loan?  Will the proposed scaling up require improvements in The Bank may provide an additional loan to �nance the im- the results framework and monitoring systems, or is plementation of additional or expanded “activities that scale- this an opportunity for making such improvements? up a project’s impact and development effectiveness� (OP 13.20, paragraph 1). When a borrower requests AF to add or  Are the proposed additional or expanded activities expand activities under the original project, the TTL, in addi- expected to be completed within three years of the tion to assessing the suitability of the project on the basis of current closing date of the original loan? its performance, normally examines the following aspects:  Do any of the modi�ed or new activities raise the  Is additional �nancing for the added or expanded activi- environmental category of the project or trigger any ties the best mechanism to maximize development new safeguard policies? If so, does it make sense to impact and results in the relevant area or sector, com- address these impacts in the context of the ongoing pared to other lending tools such as a repeater project, a project? How would they be addressed under the AF? completely new operation, or non-lending instruments?  Are the proposed activities consistent with the project  Is the borrower committed to scaling up the project Development Objectives? Are they strategically activities? Would preparing an additional loan bring aligned with the current Country Assistance Strategy procedural or other cost-effectiveness gains for the or Country Partnership Strategy? AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 A N N E X 2 — OV E RVIE W OF KEY L IT E RAT URE A N D CONC EPTS ON S C A LING UP 33 Annex 2: OVERVIEW OF KEY LITERATURE AND CONCEPTS ON SCALING UP There is an abundance of principles, criteria, guidelines, Cooley, L., and R. Kohl. 2006. Scaling Up—From Vision steps, tasks, and conceptual models developed in the litera- to Large-scale Change: A Management Framework for ture on scaling up. In particular, conceptual models from the Practitioners. Washington, D.C.: Management Systems key publications on scaling up highlighted below do allow an International. analysis of common elements that need to be considered The Cooley and Kohl (2006) three-step and 10-task frame- in developing criteria for assessing scalability. While all of work offers a template for planning and for implementation. these concepts and approaches are important, their inher- This model sees scaling up as a process that begins with ent value changes with each project. In addition to the IFAD/ good planning. It is grounded in the public administration and Brookings report (Linn et al., 2010) featured in this report, development management literature. Ideally planning starts there are additional publications providing analytical frame- with initial testing of the innovation. In this model, most of work for scaling up: the work is done prior to any action to scale up.  Linn, Johannes F ., Artntraud Hartmann, Homi Kharas, Step 1. Developing a Scaling Up Plan Richard Kohl, and Barbara Massler. 2010. Scaling Up the Fight against Rural Poverty: An Institutional Review Task 1: Developing a Vision: What is being scaled up? of IFAD’s Approach. Brookings Global Economy and How is the scaling up to occur? Who is doing the Development Working Paper 43, October. Washington, scaling up? Where is the scaling up to occur? D.C. (For the International Fund for Agricultural Task 2: Assessing Scalability Development). Task 3: Filling Information Gaps  Hartmann A., and J. Linn. 2008. Scaling Up: Task 4: Preparing a Scaling Up Plan A Framework and Lessons for Development Step 2. Establishing the Preconditions for an Effective Scaling Effectiveness from Literature and Practice. Wolfensohn Up Process Center for Development, Working Paper 5. Washington, D.C.: Brookings Institution. Task 5: Legitimizing Change (“getting the issue on  Cooley, L., and R. Kohl. 2006. Scaling Up—From Vision the agenda�) to Large-scale Change: A Management Framework Task 6: Constituency Building (“building bridges�) for Practitioners. Washington, D.C.: Management Task 7: Realigning and Mobilizing Resources Systems International.  Hancock, J., 2003. Scaling-up the Impact of Good Step 3. Implementing the Scaling Up Process Practices in Rural Development: A Working Paper to Task 8: Modifying and Strengthening Organizations Support Implementation of the World Bank’s Rural Development Strategy. Mimeo, Washington, D.C.: Task 9: Coordinating Action World Bank. Task 10: Tracking Performance and Maintaining Momentum Noteworthy excerpts are featured below. AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 34 A NNEX 2 — OVERVIEW OF KEY LITER ATURE A ND CONC EPTS ON S C A LING UP The Cooley and Kohl Model stresses the importance of an Which Form of Scaling? A Decision Guide Intermediary Organization to support the adopting organiza- The guide indicated in �gure A.2 from Cooley and Kohl (2006) tion which is doing the scaling up as indicated in �gure A.1. should be used in conjunction with “Types and Methods of Scaling“ illustrated above. In table A.1, Cooley and Kohl (2006) explain that expansion means “increasing the scope of operations of the organiza- The tool indicated in �gure A.3 from Cooley and Kohl (2006) tion. The way in which it expands has implications for the ca- provides a rapid assessment of the complexity or simplicity pacity of the organization. Replication is increasing the use of of the innovation, and thus a rough indicator of scalability. the innovation, but this is not done by the originating organiza- tion. Replication by government policy adoption appears to be FIGURE A.2: Which Form of Scaling? A Decision Guide a simple form, but if the innovation requires behavior changes and participation by the community, governments may lack capacities for such complex implementation. The other forms of replication have signi�cant implications for the originating organization’s management. Collaboration falls between the �rst two models and requires coordination and other forms of partnerships with one or more organization or network. FIGURE A.1: Organizational Roles in Scaling Up Source: Cooley and Kohl, 2006. TABLE A.1: Types of Scaling Up Expansion Growth Re-structuring or Decentralization Franchising Spin-off Replication Policy Adoption Grafting Diffusion and Spillover Mass Media Source: Cooley and Kohl, 2006. Collaboration Formal Partnerships, Joint Ventures, and Strategic Alliances Networks and Coalitions Source: Cooley and Kohl, 2006. TH INKING SY STEMATICA LLY A B OUT S C A LING UP A N N E X 2 — OV E RVIE W OF KEY L IT E RAT URE A N D CONC EPTS ON S C A LING UP 35 FIGURE A.3: Scalability Checklist Source: Cooley and Kohl, 2006. AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 36 A NNEX 2 — OVERVIEW OF KEY LITER ATURE A ND CONC EPTS ON S C A LING UP Hancock, J., 2003. Scaling-up the Impact of Good key points related to scaling up impacts. The checklist should Practices in Rural Development: A Working Paper to not be applied rigidly. The different parts of the checklist will be Support Implementation of the World Bank’s Rural used in different combinations and emphases depending on Development Strategy. Mimeo. Washington D.C.: the task: whether the task is (1) assessing the state of practice World Bank. of scaling up experiences to lay the ground for country strate- Hancock (2003) presents a “Checklist for Thinking System- gies, for instance; (2) developing a project, program, or policy; atically about Scaling-Up Impact� in table A.2. It is intended to or (3) evaluating a project, program, or policy. The checklist help development practitioners and their partners think through includes seven major items. TABLE A.2: Checklist for Thinking Systematically about Scaling Up Impact Source: Hancock, 2003. TH INKING SY STEMATICA LLY A B OUT S C A LING UP REFERENCES 37 REFERENCES Binswanger, H. P., and S. S. Aiyar. 2003. Scaling Up Community Royal Tropical Institute. 2005. Stakeholder-Driven Funding Mechanisms Driven Development Theoretical Underpinnings and Program for Agricultural Innovation: Case Studies from Sub-Saharan Africa. Design Implications. Mimeo. Washington, D.C.: World Bank. KIT Bulletin 373. Amsterdam. Binswanger, H. P., and T. Nguyen. 2004. Scaling Up Community Driven United Nations. 2010. The Millennium Development Goals Report. Development for Dummies. Washington, D.C.: World Bank. New York. Byerlee, Derek. 2000. Competitive Funding of Agricultural Research UNSCN et al. 2009. Scaling Up Nutrition: A Framework for Action. in the World Bank: Lessons and Challenges. RDV Working Paper. Available at www.unscn.org/�les/Announcements/Scaling_Up_ Washington, D.C.: World Bank. Nutrition-A_Framework_for_Action.pdf. Cooley, L., and R. Kohl. 2006. Scaling Up—From Vision to Large- World Bank. 2003. Reaching the Rural Poor: A Renewed Strategy for scale Change: A Management Framework for Practitioners. Rural Development. Washington, D.C.: World Bank. Washington, D.C.: Management Systems International. World Bank. 2004. “Reducing Poverty, Sustaining Growth: Scaling-Up Hancock, J., 2003. Scaling-Up the Impact of Good Practices in Rural Poverty Reduction—a Global Learning Process and Conference Development: A Working Paper to Support Implementation � Washington, D.C.: World Bank. in Shanghai. of the World Bank’s Rural Development Strategy. Mimeo. World Bank. 2004a. “Scaling Up a Community Driven HIV/AIDS Washington, D.C.: World Bank. Program in Malawi.� Social Development Note No. 96, October. Hartmann A., and J. Linn. 2008. Scaling Up: A Framework and Washington, D.C.: World Bank. Lessons for Development Effectiveness from Literature and World Bank. 2005. Agriculture and Achieving the Millennium Practice. Wolfensohn Center for Development, Working Paper 5. Development Goals. Washington, D.C. Washington, D.C.: Brookings Institution. World Bank. 2008. World Development Report 2008 on Agriculture Holcombe, Susan. 2011. Lessons from Practice: Assessing Scalability. for Development. Washington, D.C.: World Bank. Heller School for Social Policy and Management at Brandeis University, for the World Bank. World Bank. 2009. Agricultural Research and Competitive Grant Schemes: An IEG Performance Assessment of Four Projects in Horton, Susan, Meera Shekar, Christine McDonald, Ajay Mahal, Latin America. IEG Report No. 49149. Washington, D.C.: World and Jana Krystene Brooks. 2010. Scaling Up Nutrition: What Bank. Will It Cost? World Bank Directions in Development Series. Washington, D.C.: World Bank. World Bank. 2011. Mapping the Roads from Development Marketplace Agriculture and Rural Development Projects to Linn, Johannes F ., Artntraud Hartmann, Homi Kharas, Richard Kohl, Sustainable Practice. Washington, D.C.: World Bank. and Barbara Massler. 2010. Scaling Up the Fight against Rural Poverty: An Institutional Review of IFAD’s Approach. Brookings World Bank. 2011a. Republic of Peru Agricultural Research and Global Economy and Development Working Paper 43, October. Extension Project in Support of the Second Phase of the Washington, D.C. (For the International Fund for Agricultural Agricultural Research and Extension Program Implementation Development). Completion and Results Report. Washington, D.C.: World Bank. Mkhize, H. P. B., J. P. Regt, and S. Stephen, 2009. Scaling Up Local World Bank. 2012. Agricultural Innovation Systems: An Investment and Community Driven Development: A Real World Guide to Its Sourcebook. Washington, D.C.: World Bank. Theory and Practice. Washington, D.C.: World Bank. Rajalahti, Riikka, and Sara Farley. 2010. Designing and Implementing Agricultural Innovation Funds: Lessons from Competitive Research and Matching Grant Projects. Washington, D.C.: World Bank. AG R I C U LTUR E AND RURAL DE VE LOP ME NT D I S C U SS ION PA PER 53 Agriculture and Rural Development (ARD) 1818 H Street, NW Washington, D.C. 20433 USA ARD AGRICULTURE AND RURAL DEVELOPMENT Telephone: 202-477-1000 Internet: www.worldbank.org/ard