FSP Synthesis Report IV: Building Locally Led Agricultural Policy Analysis Capacity: Lessons from Experience in Developing Countries


December 23, 2019 - <jayne@msu.edu>, <S.Babu@cgiar.org>, <boughton@msu.edu>, <Sheryl.Hendriks@up.ac.za>, <elimka23@gmail.com>, <Ferdi.Meyer@up.ac.za>, <staatz@msu.edu>, <lliverp@msu.edu>, <crawfor5@msu.edu>, <P.Dorosh@cgiar.org>, and <k.savadogo@cgiar.org>

Executive Summary

Agricultural policies affect almost everyone in the world, directly or indirectly. Improvements in agricultural policy analysis capacity can therefore significantly benefit society. This report synthesizes lessons learned from the Food SecurityPolicy Innovation Lab’s capacity strengthening efforts over the 2013–2018 period and from related capacity development initiatives conducted over the last four decades by the partner institutions in Africa and Asia.

“Capacity” is defined as the resources required for people and institutions to reach their objectives and achieve results in an efficient and sustainable manner. Capacity development is therefore the process of enhancing, improving, and unleashing such resources—in people, organizations, and systems. In this report, the term “locally led” refers to institutions registered in the host country, led by nationals of the host country, and often, but not necessarily, affiliated with a public university or government unit in the host country.

Why is the development of locally led agricultural policy analysis important? In principle, governments in developing countries could continue to receive policy analysis guidance from external research organizations, yet governments in most medium- and high-income countries rely on policy guidance from locally led research groups in their own countries. One lesson from capacity development initiatives is that governments are more likely to seriously listen to and engage with policy guidance provided by research units led by well-respected researchers in their own countries, who know the country, the culture, and the local politics surrounding agricultural policy issues. The impact of technical analysis and policy guidance cannot be divorced from policymakers’ trust and respect for the person/group providing it. For these reasons, well-functioning locally led policy analysis units play a crucial role in an effective policy environment, and they can raise the probability that policy analysis will contribute to policy impact. These points in no way downplay the importance of international research units—in fact, they must often play a crucial support role to raise the capacities of locally led policy analysis units to improve the quality of policy decisions in their countries.

Few governments or development partners want to fund research for its own sake. Most funders of agricultural policy analysis capacity are ultimately interested in policy impact, which involves much more than just developing local capacity for policy analysis. Therefore, this report takes a holistic approach to capacity development, with the aim of improving evidence-based policy guidance and policy impact.

Models of agricultural policy analysis units in developing countries
Institutional capacity for agricultural policy analysis and engagement has tended to follow one of the following three models: (i) individual technical advisors or research units embedded in government ministries; (ii) institutes affiliated with local universities; and/or (iii) independent policy institutes or think tanks. The report examines the performance of numerous examples of these three institutional forms.

For example, under the Food Security Policy Innovation Lab, the technical advisor model has been employed in theAgricultural Policy Analysis Unit of the Myanmar Ministry of Agriculture, Livestock and Irrigation. In recent years, the Bill & Melinda Gates Foundation has also utilized this model through the Tony Blair Institute. A more extensive application of this model has been to set up an entire policy or technical unit within a ministry or national research system, such as with the Bureau d’Analyses Macro-Economiques (BAME) in the Senegalese Agricultural ResearchInstitute, which is part of the Ministry of Agriculture. An example of university-based policy institutes is the Tegemeo Institute based at Egerton University in Kenya. Examples of independent agricultural policy analysis units include the Indaba Agricultural Policy Research Institute (IAPRI) in Zambia, the Bureau for Food and Agricultural Policy (BFAP) inSouth Africa, and the Bangladesh Institute of Development Studies (BIDS).

Key findings:

Six of the report’s many findings are highlighted here:

  1. The importance of the policy ecosystem. Capacity development has traditionally focused on improving individuals’ knowledge and skills. An important conclusion of this report is that strengthening locally led agricultural policy analysis organizations requires explicit attention to the external environment in which these units operate. This“policy ecosystem” determines the scope for locally led policy institutes to develop and thrive. This is especially the case in recent years as foundations and bilateral and multilateral funding organizations have created new organizations with inevitably overlapping mandates. Investments in a new externally funded policy research or advisory organization may unintentionally marginalize local policy units with similar mandates but with far fewer resources to carry them out. Donors need to carefully consider how their funding decisions may unintentionally affect the viability of other actors and functions in the system.

    For example, in the context of Africa, there are many more PhD-trained agricultural economists in 2019 than in 1980.The Association of African Agricultural Economists (AAAE) had 46 members in 2004, rising to over 470 members in2019 (AAAE Management Office, 2019). However, many of these African agricultural economists choose not to joinAfrican universities or policy analysis units due to perceived and/or real differences in workplace conditions compared with those of international organizations. Investments in individual capacity development do not necessarily improve the performance of locally led policy analysis units unless organizational and system-level issues are also addressed.One of the contributions of capacity development models emphasizing policy systems is the recognition that investing in the capacity of individuals alone may risk creating enclaves of better-educated nationals working, even in their own countries, for well-funded international organizations or new entities designed to respond to donor priorities. Such situations can restrict the development of locally led policy institutes.

  2. Recognize and support diverse kinds of policy analysis needs. Policy institutes can produce three types of analysis:(1) demand-led analysis that responds to government requests; (2) analysis that is considered important to some stakeholders but not necessarily welcomed by governments (e.g., analysis of the impacts of government-imposed trade bans); and (3) “discovery” analysis that is not demanded by governments or stakeholders, but which creates new insights that influence future policy discussions and decisions. There are many examples of “researcher-led”discovery analysis leading to an improved understanding that subsequently influenced policies, such as IFPRI-led research on gender power relations, which has shown policymakers that gender relations affect not only household resource allocation but also the pace of a country’s agricultural productivity growth. Another example is MSU-led research demonstrating how dietary change in Africa is altering employment in agrifood systems. Especially in light of rapid demographic and economic transformation in Africa, this third category of policy analysis can help African governments anticipate future opportunities and emerging challenges, so that they can proactively rather than reactively respond to them.A policy ecosystem that produces growth-enhancing, equitable policy decisions relies on all three types of analysis.Attempts to achieve quick wins or capture low-hanging fruit frequently rely on an evidence base created by research investments made five to ten years earlier. If support for the third type of research dries up, the evidence base to inform demand-led policy debates in the future may be lost.

  3. The importance of building credibility and demand for evidence among stakeholders. Developing strong local demand for the work carried out by agricultural policy analysis organizations is crucial for their sustainability. This means that to generate quality knowledge products, policy analysis units might need to be more proactive, building a constituency of support from a wide range of public and private stakeholders, including civil society. It also means that policy institutes must understand that their mandates include actively working with governments and other stakeholders to translate research findings into policy discussions and ultimately policy impact.

  4. Evidence-based policy analysis requires data generation. Agricultural policy institutes cannot produce evidence-based analysis without data. In recent years, the Bill & Melinda Gates Foundation has invested significantly in generating open-access data on farmer and consumer behavior through the Living Standards Measurement Surveys–Integrated Surveys on Agriculture (LSMS-ISA). This investment in data has made it possible for local researchers to provide evidence-based analysis to policymakers and to achieve a greater degree of consensus among researchers around particular policy issues. However, the number of African countries with nationally representative datasets is remarkably few. The LSMS-ISA surveys discussed above are carried out in only 8 of sub-Saharan Africa’s 51 countries. It is therefore difficult to provide up-to-date evidence-based analysis to guide policy decisions on specific policy issues in most African countries. Creating an evidence base to guide policy discussions is a major priority for strengthening the effectiveness of locally led policy units. Ideally, data collection activities should be based on close collaboration between local policy institutes (which are likely to be in the best position to understand existing data gaps in the county) and government statistical offices, with clear protocols on how the data will be collected, supervised, processed, released, and utilized.

  5. Appropriate policy analysis models depend on objectives and timeframe for impact. All three models of policy analysis—embedded advisors or units in government ministries, university-affiliated policy institutes, and independent policy think tanks—can be effective. The effectiveness of any particular model depends on objectives and timeframe for impact. The common denominator of success in any institutional capacity development model is effective leadership—setting a conducive internal culture that incentivizes individuals to bring out their best and to support the institute’s objectives. This can be achieved in any model, but it is especially important to the policy institute model because the range of activities that a policy institute can fulfill in a country’s policy ecosystem is broader than what is usually possible in the embedded technical advisory model. For funding organizations aiming to improve the policy-enabling environment over the long run and for sustainable development, clearly the locally led policy institute—affiliated with a credible national organization—is likely to be the most effective model.

  6. The importance of leadership and management. The case studies examined in the full report demonstrate that leadership is critical for effective utilization of policy research institutions. In some cases, policy institutes received strong support from government and local stakeholders in their early years, but leadership in the university and institute were unable to raise demand for the institute’s services and eventually core funding. A particular challenge for independent policy analysis institutes is that unless the institute succeeds in sustainably raising core funds for its operations, it often turns to local consulting activities to sustain its members. The core activities of the project—those established by government, core donors, or the institute’s board of directors—can become neglected as staff pursue independent consulting arrangements to augment their salaries during periods when institutional funding is inadequate to fully support the staff. Also, because all three organizational models depend on external funding, research agendas can become driven by outside priorities.

Effective leadership of locally led policy institutes requires an effective balance between externally facing and internally facing responsibilities. Externally facing responsibilities include managing relations with governments, donors, and other stakeholders, thereby creating demand for the institute’s activities; effectively prioritizing the various “asks” by stakeholders for the institute’s work; and effectively managing relations with the institute’s board of directors. Internally facing responsibilities include providing incentives to attract and retain quality research and administrative staff, and creating a conducive work environment that rewards good performance. For policy institutes affiliated with local universities, enlightened university leadership is also crucial for ensuring the effectiveness and sustainability of the policy research institute.


Today, many more nationals in Africa and other developing regions possess strong agrifood systems policy analysis skills as compared to 25 years ago. Many have been educated in world-renowned institutions, possess valuable technical skills, and can operate effectively in their countries given their superior knowledge of local culture and their connections with centers of local power. Greater support for locally led institutional policy analysis capacity can ensure that this greater supply of well-trained individuals translates into more effective and sustainable local policy analysis and policy impacts.

[To read the complete report click on the "download file" button]



Accessibility Questions:

For questions about accessibility and/or if you need additional accommodations for a specific document, please send an email to ANR Communications & Marketing at anrcommunications@anr.msu.edu.