INDEPENDENT REVIEW Assessment of the use of outputs from PIM-supported work on national SAMs and CGE models Agapi Somwaru March 2021 ii Contents Acknowledgments ........................................................................................................................................ iii Introduction ................................................................................................................................................... 1 Previous Evaluations ..................................................................................................................................... 2 Study Methods .............................................................................................................................................. 3 SAM Downloads ........................................................................................................................................... 3 Publications and Citations of Single-Country CGE Modeling Work ........................................................... 6 Online Surveys ............................................................................................................................................ 11 First Online Survey .............................................................................................................................. 11 Second Online Survey .......................................................................................................................... 14 Interviews: A Worldwide Bird’s Eye View ................................................................................................ 16 Summary and Conclusions ......................................................................................................................... 21 Appendix 1: Downloads of SAM Datasets, 2012–2019 ............................................................................. 23 Appendix 2: SAM Dataset Titles (with Digital Object Identifier) .............................................................. 24 Appendix 3: Selected Publications for Selected Countries using SAM/CGE Outputs, 2012–2019 ........... 29 Appendix 4: Complete Altmetric and Google Citation Analysis of All Sampled Publications (n = 269) .. 34 Appendix 5: First Online Survey ................................................................................................................ 42 Appendix 6: List of Key Informants for PIM-Supported Modeling ........................................................... 48 Appendix 7: Interviewed IFPRI Researchers Producing and Using IFPRI SAMs and CGE models ......... 50 References ................................................................................................................................................... 51 iii Acknowledgments This work was undertaken from March 2019 through November 2020. The review is independent; however, the assistance of IFPRI staff is gratefully acknowledged. The author would like to thank Frank Place, Rui Benfica, and Sherman Robinson for their valuable comments and suggestions. The author also thanks Indira Yerramareddy, Nilam Prasai, Melissa Skees, Erica Saito, Xinyuan Shang, Ghada Shields, Leigh Plummer, Ryan Miller, and other IFPRI staff for their valuable help with data/databases for this paper. Also gratefully acknowledged are all survey respondents for providing their time and insights to phone calls, electronic survey questionnaires, and emails. 1 Introduction This report evaluates the outputs from the CGIAR Research Program on Policies, Institutions, and Markets (PIM) on national social accounting matrices (SAMs)1 and single-country computable general equilibrium (CGE) models.2 The study aims to identify what policies, programs, strategies, and expenditure decisions were informed by SAMs and single-country CGE models. The report seeks to document what decisions were made based on the contribution of SAM/CGE models’ outputs that effectively shaped policy work or what changes were made because of CGIAR PIM– supported economywide modeling being available to decision makers. Based on the above, the report makes recommendations for CGIAR PIM–supported single economywide modeling work on how to effectively increase decision makers’ use of SAM/CGE outputs. Prior to 2002, staff of the International Food Policy Research Institute (IFPRI) worked primarily from the organization’s headquarters in Washington, D.C., mainly on global issues and/or selected policy issues common to many countries through multi-country research programs. After 2002, IFPRI began to decentralize its staff, placing researchers where they could engage more closely with individual countries on their own policy issues and in more interactive, demand-driven ways. The staff that were outposted mainly worked on regional issues or on individual-country projects that were case studies within the framework of multi-country research programs. The PIM- supported SAM/CGE modeling activities have augmented country groups’ and individual countries’ analytical capabilities, informed governments, and other stakeholders, and affected policy decision making. This report covers 2012 to 2019, a period that marks the beginning of PIM to the present. It adds to knowledge generated by previous reports on this subject. We begin by reviewing previous work evaluating IFPRI’s SAM and single-country CGE models and related work. Then we present the methods used to assess the usage of, and outcomes resulting from, PIM-supported economywide single-country SAM/CGE models. This includes the results from website hits, downloads of the SAMs, and citations searches of country CGE–related publications. This is followed by the outcomes of an electronic survey of major stakeholders as well as phone interviews of key contacts of users familiar with this PIM-supported modeling work. Also, the citations became a main source to identify contacts for a second electronic survey. Finally, we present some suggestions on single-country economywide SAMs/CGE modeling activities. 1 A social accounting matrix (SAM) can be defined as an organized matrix representation of all transactions and transfers between different production activities, factors of production, and institutions (households, corporate sector, and government) within the economy and with the rest of the world in a given year. A SAM is thus a comprehensive accounting framework capturing the full circular flow of income from production to factor incomes, household income to household consumption, and back to production (Wing 2004). 2 A CGE model is developed from microeconomic fundamentals and employed to illustrate (1) how a model may be calibrated using the economic data in a SAM, (2) how the resulting system of mathematical equations may be solved for the equilibrium values of economic variables, and (3) how perturbing this equilibrium by introducing a policy shock facilitates analysis of policies’ economywide impacts (Wing 2004). 2 Previous Evaluations As part of a process aimed at improving the effectiveness of its work, IFPRI systematically reviews the impact of its major research and related outreach programs. The first evaluation of SAM/CGE modeling activities took place in 2003. IFPRI’s director general commissioned a study to evaluate the impact of IFPRI research and related activities that used economywide models. The study aimed to evaluate activities involved in the development of SAM databases and economywide models for analyzing issues of trade and macro policy relevant to food policy in developing countries. Kym Anderson conducted this first assessment that reviews the impact of the single-country CGE modeling research and outreach program. The assessment report concluded that IFPRI had generated a very large number of economywide modeling outputs and produced and made publicly available numerous SAMs and economywide models, as well as methodologies associated with both. The modeling outputs were state-of-the-art, with some pushing the knowledge frontier. The report mentioned that the IFPRI SAM/CGE team was also engaged in substantial capacity strengthening via training workshops in many locations in Africa, Asia, Latin America, and the Middle East. The second evaluation of IFPRI SAM/CGE modeling activities was undertaken by a core team led by Chris Gerrard that conducted an evaluation of PIM global agricultural modeling that focused mostly on the 2011–2014 period. This was an evaluation of PIM-supported modeling work, and it acknowledged that some work began before PIM. The evaluation was commissioned by the Independent Evaluation Arrangement of CGIAR’s PIM (CGIAR-IEA 2015). The report is volume IV of the evaluation and focuses on agricultural modeling carried out in PIM. The following is a summary of the report’s findings on SAMs and economywide CGE models and their products. The second review focused on two PIM-supported activities: (1) “databases and tools for analyzing pro-poor growth and food security in Arab countries,” and (2) “case studies of country specific policies to promote agricultural transformation and poverty reduction in Africa.” Although countries were characterized by different income levels and the analysis was constrained by data scarcity and access (mainly in some Middle East and North Africa [MENA] countries), the report found that the country-level CGE modeling activities had responded to requests mostly from governments and international organizations. The single-country CGE models had many country- level policy applications, had built on IFPRI expertise that is widely recognized, and led to many requests from countries and organizations for analytical support. The report also stated that the successful record of this work had made IFPRI’s single-country economywide work in this area very demand-driven and highly sought. The quality of the leadership and staff was demonstrated by the quality and quantity of research outputs. The report suggested that improvements of these models might include disaggregating land and labor markets, incorporation of migration, greater use of micro-econometric estimations, and adding more emphasis on employment and natural resource use. The country-level CGE modeling teams made many contributions to improving national agricultural policy analysis training in several African countries in the context of the African 3 Union’s Comprehensive Africa Agriculture Development Programme (CAADP). According to the report, most of the projects in the MENA region, Yemen, Egypt, and Iraq were originated by requests from national governments that established data openness initiatives while seeking IFPRI’s involvement in analytical and policy work. Because of this successful record of work, IFPRI’s work in this area was highly sought. The most notable impact in this time period had been in Tanzania, where an analysis of the effects of reinstituting a maize export ban helped persuade the government not to do so. The report stated that capacity-strengthening work was undertaken in connection with the country-level CGE activities. The International Center for Agricultural Research in the Dry Areas (ICARDA) also requested support from IFPRI to analyze agricultural growth under different scenarios for its project in Iraq. Other work followed requests from the governments of Yemen and Egypt and from the World Food Programme (WFP) and the World Bank in the MENA region. In Africa, the requests came from national governments or local IFPRI offices, which in turn provided evidence to host governments. The types of policy analyses conducted varied in response to the specific requests but all used the PIM-supported IFPRI standard CGE model framework. The assessment of the PIM global agricultural modeling activities by the professional peers was very positive. The contributions of IFPRI were valuable, given the weak analytical capacities of developing countries in designing and evaluating agricultural policies. Study Methods The study employed a variety of methods, including examining website visits, downloads of SAM datasets, and citations of IFPRI publications on the economywide CGE modeling work. We also used online surveys to obtain information from users of IFPRI single-country CGE models suggested by the IFPRI/CGE researchers. We conducted extended phone interviews with numerous development practitioners who have been identified as being knowledgeable on CGE modeling. Selected users of the models as identified by the CGE citation analysis were also surveyed to ascertain who is using the PIM-supported CGE models, how the models have been used, and what impact they have had on policies, programs, and investment/expenditure decisions in agriculture and the broader economy. SAM Downloads With PIM’s support, SAM datasets have been developed and/or updated and made available on Dataverse for download by the public. Dataverse, an open-source web application for archiving and sharing datasets, was developed and is managed by Harvard’s Institute for Quantitative Social Science, together with collaborators. Appendix 1 presents information on downloads of SAM datasets from 2012 to 2019 (through mid- December 2019). Over that period, SAM datasets were downloaded from Dataverse a total of 15,740 times. The pace of downloading increased over time (see Figure 1). The most frequently downloaded SAM dataset was the 2007 Social Accounting Matrix for China (968 times) followed by the Ethiopian SAM, which was download 693 times. The Ethiopian SAM dataset was developed through a collaboration involving the Ethiopian Development Research Institute, the http://www.iq.harvard.edu/product-development http://www.iq.harvard.edu/product-development 4 Institute of Development Studies, and IFPRI. The SAMs of Nigeria and South Africa were downloaded 595 and 563 times, respectively, followed by the downloads of Mexico’s (561 times) and Ghana’s (466) SAM datasets (Appendix 1). Universities constitute the majority (63%) of the 1,216 affiliations downloading SAMs, followed by government entities (9%), nongovernmental organizations (NGOs)/nonprofits/think tanks (9%), private entities (8%), and other international organizations (3%), while IFPRI and CGIAR accounted for 1% of the downloads (see Figure 2). Figure 1. Number of users downloading SAM datasets annually, 2012-2019* * Run through the middle of December 2019. Source: Dataverse guestbook. The wide range of institutions and the geographic spread of those downloading SAMs indicates the worldwide demand for PIM-supported SAMs. That includes faculty, staff, and students from a vast variety of universities. Downloading universities include, but are by no means limited to, the following: Agricultural Information Institute of Chinese Academy of Agricultural Sciences, Addis Ababa University, Arizona State University, Boston College, Bowen University, Cairo University, Cambridge, Carnegie Mellon, Clemson University, De La Salle University, Georgia Institute of Technology, Harvard Kennedy School of Government, Indian Institute of Management, London School of Economics, Norwegian University of Life Science, University of Reading, University of Istanbul, University of Alberta, University of Arizona, University of Hohenheim, University of Groningen, University Le Havre, University of Aberdeen, University of Alberta, University of Arkansas, University of Athens, University of Benin, University of Bologna, University of Toronto, University of Tokyo, University of Virginia, University of Warwick, Yale University, Zhejiang University, and University of Karachi. 245 951 1,295 1,667 1,711 2,815 3,240 3,108 0 500 1,000 1,500 2,000 2,500 3,000 3,500 2012 2013 2014 2015 2016 2017 2018 2019 5 Figure 2 Number of downloads by institutional affiliation of SAM user Source: Dataverse guestbook. A variety of international organizations and institutions downloaded SAMs, including the World Bank, the Food and Agriculture Organization of the United Nations (FAO), the Asian Development Bank, the International Fund for Agricultural Development (IFAD), Alliance for a Green Revolution in Africa (AGRA), and the International Monetary Fund (IMF). Others among nongovernment affiliations included the Bill & Melinda Gates Foundation, the Brookings Institution, the Overseas Development Institute (ODI), and Oxfam. The list of SAM downloads (Appendix 2) provides an in-depth and valuable understanding of the extensive use of SAMs for policy analyses. Appendix 2 includes the digital object identifiers (DOIs) that permanently identify an article or document and link to it on the web so readers can easily locate a document. The SAM downloads indicate a wide range of studies. Note that the national SAMs are built using reliable national accounts data of a country, along with data from the United Nations National Accounts Official Country Data database, the UN Comtrade database, and the World Bank World Development Indicators (WDI) databank. According to Sherman Robinson, research fellow (emeritus) at IFPRI and professor of economics (emeritus) at the University of Sussex, “IFPRI utilizing reliable National Accounts has developed SAMs for most of the developing countries in Africa with the exception of some countries, like Zimbabwe, due to lack of reliable data” (telephone interview). Unlike other sources, the PIM-supported IFPRI program of SAMs produces consistent, standardized, and reliable datasets for economywide analyses. Government 9% IFPRI and other CGIAR 1% NGO/ non- profit/ think tank 9% Other international 3% Private 8% University 63% Unknown 7% 6 PIM-supported SAM development and related datasets, such as climate data, land use categories, and so forth, constitute the most reliable repository of SAM datasets (GTAP 2019). The PIM- supported SAMs also draw on the Organization for Economic Co-operation and Development’s (OECD’s) Input-Output Tables (IOTs) database. The IOTs database was part of the Structural Analysis (STAN) databases project undertaken in the Economics Analysis and Statistics Division of the OECD Directorate for Science, Technology, and Industry (OECD 2016, 2018). The OECD IOTs database is the first publication of the OECD input-output tables that includes both reliable statistics and analyses that allows “new internationally comparable data for consistent industrial analysis at a detailed sectoral level.” Note that countries where IFPRI staff amounted to the primary or sole supplier of SAMs include South Africa, Mozambique, Malawi, Zambia, Uganda, Rwanda, Kenya, Tanzania, Ethiopia, Ghana, Nigeria, Rwanda, Senegal, Niger, and Mali in the Africa south of the Sahara (SSA) region. In Asia, PIM-supported SAM modeling activities include SAMs for Bangladesh, Pakistan, the Philippines, Nepal, Laos, Vietnam, Cambodia, Indonesia, and Myanmar. The SAMs of Egypt and Tunisia are reported in the MENA region. Most of the downloads reporting the use of IFPRI’s SAMs also report the use of IFPRI’s modeling methodologies, that is, standard multiplier analysis and country CGE modeling. For example, popular with many downloads was the IFPRI standard CGE model as set out by Löfgren, Harris, and Robinson (2002) in their paper A Standard Computable General Equilibrium (CGE) Model in GAMS. More recently, analytical work on value chain and country investment prioritization to inform development strategies has been informed by the Rural Investment and Policy Analysis (RIAPA) modeling framework (Benfica and Thurlow 2017) derived from the IFPRI standard model. In general, 78% of the articles that downloaded an IFPRI SAM also used an IFPRI CGE model specification, while the remaining 22% of the articles used SAM-based multipliers to analyze the effects of policies.3 Worldwide, users of the IFPRI SAM databases, SAM-based multiplier analysis, or CGE models had high praise for the significant contributions of James Thurlow, Channing Arndt, Xinshen Diao, and Clemens Breisinger in these areas of work. Publications and Citations of Single-Country CGE Modeling Work In evaluating the PIM-supported CGE modeling program, its publications are a key output to consider. Using an advanced search approach, a sample of 269 CGE publications was identified by assessing IFPRI’s publications repository from 2012 through the end of 2019. Those publications include 128 peer-reviewed and 141 non-peer-reviewed articles (Figure 3). Looking into both types of publications by year, in 2012 we have the largest number of both peer- and non-peer-reviewed publications (65 articles)—39 peer-reviewed and 26 non-peer-reviewed publications (Figure 4). The numbers for 2013 and 2014 are the second highest with 45 total for each year. Specifically, 2013 has 17 peer-reviewed and 28 non-peer-reviewed publications, while in 2014 the peer- and non-peer-reviewed publications are 16 and 29, respectively. Journal articles form the largest category within peer-reviewed publications, followed by book chapters (see 3 For a review on macro models and SAM-based multipliers, see Robinson (2006). http://ebrary.ifpri.org/utils/getfile/collection/p15738coll2/id/74845/filename/74846.pdf http://ebrary.ifpri.org/utils/getfile/collection/p15738coll2/id/74845/filename/74846.pdf 7 Figure 5). In 2012 there are 12 journal articles and 23 book chapters among the 39 overall peer- reviewed publications. In 2016 there are 7 peer-reviewed book chapters, 7 peer-reviewed journal articles, and 18 overall peer-reviewed publications. In 2013, there are 4 peer-reviewed book chapters, 12 peer-reviewed journal articles, and 17 overall peer-reviewed publications. When it comes to non-peer-reviewed publications the largest categories are working papers and discussion papers (Figure 6). The year 2014 has the largest number of non-peer-reviewed outputs (29) with 10 discussion papers and 7 working papers. In 2013 there are 10 discussion papers, 9 briefs, and 2 working papers/citations, while in 2012 there are 7 discussion papers and 7 working papers. Figure 3 Total number of peer-reviewed (PR) and non-peer-reviewed (non-PR) publications, 2012–2019 Source: IFPRI publications repository. Figure 4 Total number of peer-reviewed (PR) and non-peer-reviewed (non-PR) publications, 2012–2019 Source: IFPRI publications repository. 128 141 120 125 130 135 140 145 PR Non-PR 39 17 16 12 18 7 9 10 26 28 29 23 10 10 8 7 0 5 10 15 20 25 30 35 40 45 2012 2013 2014 2015 2016 2017 2018 2019 PR Non-PR 8 Figure 5 Total number of peer-reviewed publications by type, 2012–2019 Source: IFPRI publications repository. Figure 6 Total number of non-peer-reviewed publications by type, 2012–2019 Source: IFPRI publications repository. We analyze the citations of these papers in an effort to better understand how PIM-supported CGE modeling work has been used (Appendix 4). The 269 publications generated 3,328 citations. The average number of times a publication was cited in published articles over the period was 15.8. Among peer-reviewed papers the highest number of Google Scholar citations is 188 while for non- peer-reviewed papers the maximum number is 133 (Table 1). The citations include mostly peer- reviewed journal articles and book chapters. Using the so-called Altmetric score we found that peer-reviewed publications have higher scores than the non-peer-reviewed publications (Table 1). The Altmetric score for the period spanning 2012 through December 2019 is much higher for the peer-reviewed publications, averaging 10 with a 158 maximum score, than for the non-peer-reviewed publications, averaging 6 with a 0 5 10 15 20 25 30 35 40 45 2012 2013 2014 2015 2016 2017 2018 2019 Books Book chapters Synopsis Briefs Conference Papers Journal articles Reports 0 5 10 15 20 25 30 35 2012 2013 2014 2015 2016 2017 2018 2019 Working papers Synopsiss Research reports Reports Project papers Project Notes Proceedings Policy Working papers Policy Papers Journal articles Essays Discussion papers Data papers Conference Papers Briefs Book chapters 9 maximum score of 50. This indicates that peer-reviewed outputs reach an extensive audience and that it is a very effective way to communicate results when compared to non-peer-reviewed ones. Here are some examples of PIM-supported CGE articles: a journal article titled “Climate Change and Developing Country Growth: The Cases of Malawi, Mozambique, and Zambia” with the lead author from IFPRI has the highest Altmetric score (158); the paper titled “Climate Uncertainty and Economic Development: Evaluating the Case of Mozambique to 2050” has an Altmetric score of 66. Both papers are published by Springer in the journal Climate Change. Note that Altmetric references are used by a wide media outlet such as cable news, newspapers, and Internet sites worldwide. Besides Google Scholar citations and Altmetric scores, Table 1 reports Mendeley readership scores and policy document citations. The Mendeley readership score’s usefulness depends on the coverage and users from different disciplines, countries, academic status, age. Source: Author’s calculations from dataset of selected publications published by IFPRI and selected publications by IFPRI staff published in external sources. To gain a better understanding of the use of PIM-supported CGE modeling work we first present a detailed and representative list of CGE modeling work and publications used for selected Table 1: Altmetric score, Mendeley readers, Policy document citations and Google scholar citations for PIM-supported CGE publications Altmetric Score Mendeley Readers Policy Document Google Scholar Citations Citations All references (269) Mean 9 44 2 16 Median 3 28 1 6 Minimum 1 1 1 0 Maximum 158 335 6 188 Count 89 74 16 209 Peer-reviewed (128) Mean 10 45 2 24 Median 3 29 1 10 Minimum 1 1 1 0 Maximum 158 335 6 188 Count 60 68 13 101 Non Peer-reviewed (141) Mean 6 34 1 9 Median 3 19 1 4 Minimum 1 2 1 0 Maximum 50 87 2 133 Count 29 6 3 109 10 developing countries (Appendix 3), and then, using an advanced search approach, we provide the global Altmetric and citation numbers for all IFPRI CGE publications (Appendix 4). Publications listed in Appendix 3 are categorized by user, such as national and international institutions and government organizations. PIM-supported IFPRI researchers have worked closely with developing countries’ national research institutions, which constitute one of the categories. IFPRI researchers have conducted training in CGE development and CGE usage for economywide policy analyses. For example, PIM-supported IFPRI staff worked with Kenya’s Institute for Public Policy Research and Analysis and conducted targeted training courses with the Ministry of Finance and Planning in the development of CGE models and their use for understanding the economywide effects of different policies. Kenya SAMs were used by the International Livestock Research Institute for a range of studies. In Uganda, IFPRI has conducted training courses at the Economic Policy Research Centre (EPRC), the country’s leading institution in economics and development policy. This effort led to multiple EPRC publications, graduate student dissertations, consulting contracts, and peer-reviewed journal articles that used SAM/CGE models. More important, the PIM-supported CGE modeling has been associated with policymakers’ increased understanding of the various policy processes and the role of institutions in an economywide analysis. Most of the publications in Uganda fall in the international institutions and researchers (excluding IFPRI) category and cover topics like trade reform, the macroeconomics of health reform, agricultural development, and poverty reduction. For South Africa, peer-reviewed articles linked the energy sector to a detailed dynamic general equilibrium model. The study authored by Arndt, Davies et al. (2016) is an example of dynamic CGE modeling of the energy sector. For Tanzania, many peer-reviewed journal articles—on such topics as the effect of biofuels on economic development and the economic impact of different aid channels on dependency, absorptive capacity, and poverty reduction— have derived from IFPRI’s Tanzania SAM/CGE model. Finally, IFPRI’s Ghana SAM/CGE modeling generated a variety of publications. IFPRI’s Ghana SAM underpinned the World Bank’s MAMS (Maquette for MDG Simulations) CGE model, which has been used for multiple World Bank studies (Appendix 3). In a nutshell, this in-depth presentation of selected countries’ publications highlights the positive impact and potential ripple effect of PIM-supported CGE modeling work in developing countries. The Google Scholar citations and Altmetric scores show that the most cited studies cover a variety of subjects—such as climate change (112 Google Scholar citations), the role of various policy instruments in economic development (101 Google Scholar citations), economic development strategies and poverty, both regional and country specific (77 Google Scholar citations), and the impact of global food prices, trade, and commodities prices on developing countries in Africa, Asia, and South America4 (31 Google Scholar citations). The publications cover issues for specific countries (49%), regions, (33%) or the world (18%). 4 S. Morley and V. Piñeiro, A Regional Computable General Equilibrium Model for Honduras: Modeling Exogenous Shocks and Policy Alternatives, IFPRI Discussion Paper (Washington, DC: IFPRI, 2013). 11 Among the SAM/CGE publications in these topics, the ones with the most Google Scholar citations are “Climate and Southern Africa’s Water–Energy–Food Nexus” (188), “Introducing Carbon Taxes in South Africa” (112), “The Economywide Impacts and Risks of Malawi’s Farm Input Subsidy Program” (101), “Strategies and Priorities for African Agriculture” (96), and “Can Cities or Towns Drive African Development?” (77). Regarding the nature of the citations, 67% of the papers cited IFPRI CGE articles in the introduction section while 33% cited IFPRI CGE articles in the results and/or policy implications sections. We developed an email list of authors who cited the 269 PIM-supported CGE publications (3,328 publications). Those authors were then included in the second online survey we conducted to understand what outcomes may have resulted from the usage of PIM-supported CGE modeling work. Online Surveys We conducted two online surveys to identify outcomes attributable to the use of PIM-supported SAM/CGE modeling. The surveys were developed based on the Terms of Reference for this assignment and in extensive consultation with IFPRI and PIM staff. The first survey included respondents identified by the PIM-supported SAM/CGE modeling team. The second survey included the authors who cited IFPRI SAM/CGE publications/articles. First Online Survey The first survey consists of six parts. First, we ask the respondent questions about his or her professional affiliation and some other background information. Second, we ask about the respondent’s use of the SAM/CGE data and modeling tools, whether SAM/CGE data, products, and modeling have influenced their decision making, and what impacts may have resulted from the decision. The third section asks about the respondent’s awareness of international organizations using SAM/CGE data or products or modeling and how their decisions may have been affected by such work. The fourth and fifth sections are the same as the third but ask about national governments and NGOs, respectively. The final section asks for the respondent’s contact information for eventual follow-up. The survey was sent on August 8, 2019, via SurveyMonkey to 22 recipients identified by IFPRI staff who work on SAM/CGE modeling.5 One reminder message was sent on August 30, and the survey was closed on September 11, 2019. The response rate was very high (82%) because the survey’s participants are considered very knowledgeable as experts on the use of SAM/CGE modeling. Appendix 5 presents the survey questions, and Appendix 6 includes the list of the first survey participants as well as detailed information about their affiliations. 5 To comply with the General Data Protection Regulation (GDPR) requirements, all survey participants were contacted and asked whether they were willing to participate in our survey. 12 The survey helped provide insights on the uses of PIM-supported SAM/CGE modeling work. Most respondents (37%) were affiliated with a national government, followed by an international organization or development agency (17.6%) and a CGIAR center (17.6%). Twelve percent were from universities, while other academic research institutes or think tanks accounted for 11.8%. Only 5.9% were independent or from the private sector. The majority (53%) of the respondents were researchers/analysts, follow by managers/directors (23.5%), professors (17.7%), and independent consultants (5.9%). The survey indicates how respondents learned of the SAM/CGE modeling work. Most (29.4%) were directly contacted by the IFPRI SAM/CGE modeling team, 23.4% learned of it from a school or university course, and 17.6% through a publication that referenced IFPRI SAM/CGE modeling work. Other ways in which they learned about it included an IFPRI blog (11.8%), a conference or workshop (5.8%), a web search (5.9%), and other training course (5.8%). Regarding access of SAM data or CGE models, the majority received the modeling tools directly from the IFPRI Development Strategy and Governance Division modeling team through collaboration (29.4%). Other pathways included from a school/university course (23.5%), from published media that references CGE (17.7%), and through a web search (5.9%). The majority of respondents used SAM/CGE models for research papers; others used the modeling to prepare a report to the government or to upper management of the respondent’s organization, and almost half used the modeling to conduct a training course or to teach a class, or to prepare a report to donors. Note that individuals could select more than one option, implying that CGE models were used for multiple purposes. Most respondents (93.3%) used SAM/CGE models to evaluate the potential impact of policies, 66.7% used the modeling tools to better understand economic patterns or sectoral relationships in an economy, and 46.7% used them to better assess economic growth outcomes. It is interesting to note that 73.3% of respondents believe that the information coming out of the SAM/CGE analyses address the question at hand in a fully satisfactory manner. The topics or questions investigated by respondents using IFPRI SAM/CGE modeling work include growth strategies, economic linkages, climate change, and government investment options. Again, note that individuals could select more than one option, implying that, in some cases, CGE models were used for multiple purposes. Regarding the limitations of SAM databases and/or CGE models for addressing research questions, respondents (18%) stated as follows: • Lack of directly linking factor revenues by activities and factor incomes of households in the so-called IFPRI standard model • Lack of obtaining results of the distributional effects of policies across households’ categories and across space, which are both important dimensions for policymakers • Need of land and water resources allocation • Better allocation of income between factors of production—that is, labor, capital, and land • Documentation could be more transparent 13 All survey respondents have “hands-on” SAM/CGE modeling for their research work while 86.7% of respondents used SAM/CGE modeling analysis to influence decisions regarding policies, programs, strategies, directives, and/or expenditures. Here are some survey participants’ responses on the use of the data and CGE models in their research: • CGE analysis was used to design the macro framework of Ethiopia’s second five-year growth and transformation plan 2014/15–2019/20 (GTP II). • CGE model was used to design carbon tax options, climate change mitigation and adaption options, energy choices, and health insurance options in South Africa. • IFPRI’s economywide analysis for Uganda guided the choice of priorities in Uganda’s 2010– 2015 Agriculture Development Strategy and Investment Plan. When it comes to any visible or tangible impacts of SAM/CGE analyses, 92.3% of the respondents report a positive impact. The following are some examples of such impacts: • The analyses affected the strategies of trade agreements and growth perspectives. • Generated evidence-based effects of policies which were provided to government officials. • The CGE-based methods affected investment options in the Malawi country strategy. • The Statistics Department in Ghana now provides consistent economic data for users. • Used of carbon tax CGE modeling analysis with comprehensive recycling options contributed to stronger commitment to renewable energy. • Uganda’s Ministry of Finance, Planning, and Economic Development increased by 10% the national agriculture investment plan following evidence-based prioritization from the CGE model. Most respondents (76.7%) were aware of decision makers from an international, regional, or donor organization using IFPRI SAM data/CGE analyses or research outputs. The respondents were aware of CGE analysis resulting in a change in activities or decisions made by the organization regarding its policies or programs or expenditure decisions. International, regional, and/or donor organizations cited by survey respondents include the World Bank; the United Nations Development Programme; USAID; the Department for International Development; the European Commission; the West African Science Service Center on Climate Change and Adapted Land Use; the Institute for Economic Planning Uzbekistan; IFAD; FAO; AGRA; and the Institute of Statistical, Social, and Economic Research of the University of Ghana. Seventy percent of the national government respondents were aware of decision makers using or being exposed to IFPRI SAM data/CGE analyses or research outputs. The following is a partial list of national organizations: Ethiopia’s Office of the Prime Minister; Ethiopia’s National Planning Agency; Yemen’s Ministry of Planning and International Cooperation; the Egyptian Ministry of Finance (budget priorities); research staff of the Ministry of Finance and Ministry of Transport in Ghana; the South African National Treasury; the South African Department of Environmental Affairs; the South African National Planning Commission; the South African Department of Trade and Industry; the Rwandan government (national agriculture investment 14 plan), including the Ministry of Finance and Economic Planning and the Ministry of Agriculture; and the Ugandan government (national agriculture investment plan). Most respondents (90.9%) were aware of IFPRI SAM data/CGE analysis or products being used and resulting in a change in activities or decisions made by the national government regarding its policies, programs, strategies, or expenditure decisions, such as trade policies/agreements, crude oil and food subsidies, macroeconomic framework/analyses, decisions on tax revenue, and impact on different tax policies for analysis on poverty and investment options. Seventy five percent of the respondents acknowledge that decision makers from an nongovernmental organization, association, advocacy group, or civil society organization had used or were exposed to IFPRI SAM data/CGE analyses or research outputs. Examples include the Ethiopian Economic Association, the Union Tunisienne de l’Agriculture et de la Pêche, and the Trade & Industrial Policy Strategies group in South Africa. One example of a decision of the organization, association, or group affected by the SAM/CGE analyses is advocacy on foreign exchange rate policies. Second Online Survey For the second electronic online survey, we harvested 834 emails of authors of publications that cited PIM-supported CGE modeling outputs, but after requesting permission for their inclusion in the survey and two reminders, the list of those who agreed and participated was only 30. The second survey helped procure more insights on the uses of PIM-supported CGE modeling work by reaching out to those who cited IFPRI CGE model work in peer- and non-peer-reviewed publications. Most of the respondents (40.0%) were affiliated with a university, followed by other academic or research institution/think tank (26.7%), international organization or development agency (16.7%), and CGIAR entities (6.7%). The majority (53.3%) of the respondents were researchers/analysts/staff, followed by professors (33.4%), managers (3.3%), or independent consultants (3.4%). Regarding how the IFPRI CGE model/tool came to their attention, the participants indicated that CGE models/tools came to their attention through mostly (50.1%) a publication that references IFPRI’s CGE work and through collaboration in a project/study/program with IFPRI. Moreover, 80% used IFPRI SAM/CGE modeling analyses, such as articles, papers, reports, and modeling tools (e.g., datasets and/or run models). Regarding how the CGE modeling tools were accessed, 43% indicated that they downloaded the CGE model from a website. Seventeen percent received a SAM dataset from another person, and about 14% received a CGE model from another person. Only 13% received a SAM dataset or a CGE model from a training course. The responders indicated that the most common website source was the IFPRI e-brary. 15 When it comes to the ways in which the CGE modeling or CGE analyses have been valuable products to inform decision making, three-fourths of responders indicated that they used the analyses to prepare reports or publications on policy options; inform specific policymaking, such as prioritization of investments; inform decision makers on the impact of various program options; or demonstrate alternative growth paths to government officials. Regarding the value of CGE modeling products used to inform decision makers, 70% responded that they helped to better understand economic patterns or sectoral relationships in an economy, 20% to better assess economic growth outcomes, and 10% selected “other.” Examples of “other” include publication in highly ranked journals and a countrywide forest incentives program in Guatemala. Thirty-five percent of the respondents had used SAM/CGE modeling for their own work activities (e.g., research, projects, or advocacy) while 31.5% had their work used to inform policymaking decisions regarding policy options, programs/plans, or strategies. Regarding what activities or decisions were affected by SAM/CGE modeling and how SAM/CGE modeling outputs were used, here is a sample of responses: • CGE modeling used to affect decisions (beyond research): government planning and rural job opportunities (Ethiopia); options of agricultural reforms and resource allocation (Senegal 2019–2023); transformation strategies in Kenya; capital taxation in Tanzania; alternative financial options, like carbon tax, macroeconomic analysis, and accounting for climate information system in Ethiopia; evaluating the impact of climate policies in Chile • CGE modeling used for research activities: assessing the impact of free trade agreement; studying the effects of alternative policies for the Water Research Commission; developing gender-specific measures for the World Bank; evaluating different policy options and impacts of carbon tax, emission trade, and energy efficiency; identifying the socio- economic impacts of carbon tax and energy-efficiency policy; researching various scenarios for targeted income groups; evaluating trade barriers; analyzing the effects subsidies under alternative scenarios Regarding valuable impacts of SAM/CGE tools, responders revealed the usefulness of the tools especially in demonstrating to policymakers the macro- and microeconomic linkages of various sectors in a clear, coherent manner and capturing the complementarity and substitutability of various policies in an economy. Several respondents indicated that the CGE model is an effective tool that clearly shows the trade-offs of certain policies, evaluates detailed economywide benefits of interventions and/or of alternative policies, and provides estimates of impacts of a variety of adopted policy options. Some participants pointed out that the IFPRI CGE model serves as a valuable input to the Integrated Economic-Environmental Modeling Platform developed at the Inter-American Development Bank for medium- and long-run public policy and investment analysis. Participants also recognized the CGE model as a teaching instrument. Specifically, CGE models are used to teach undergraduate- and graduate-level applied trade policies. Responders also 16 indicated that CGE models are used for academic research and for publication in scientific journals. One responder indicated that the IFPRI CGE model offered a nice and elegant structure, a template to begin the modeling process. The responder stated that one can start with the IFPRI model and then incorporate features not available in it. Another responder stated that given the macroeconomic modeling nature, the IFPRI CGE is suitable to present macroeconomic variables to policymakers—for example, in the context of Ethiopia, and probably in many other countries that tend to focus more on macro variables such as what happens to GDP, domestic absorption, exports, imports, employment, and so on. Regarding how the SAM/CGE tool/dataset could be more useful, here are some suggestions from the survey. Several respondents indicated they would welcome sector disaggregation, especially at the national level; others suggested the model might account for gender or extend regional coverage that would allow for in-depth analyses. Others suggested that regular and/or more frequent data updating would make the model results more relevant. Besides data updating, others suggested that updates of the model equations’ specifications and adding new features of production technology that account for undesired outputs, such as emissions, to the standard IFPRI model and the current NEXUS SAM structure would allow for a better analysis, especially of developing countries. Some think that the main drawback of the standard IFPRI model is the lack of stock–flow relationships. Over the years, IFPRI has built SAMs for many countries, particularly in Africa. In doing so, it extended minimal effort in retaining stock variables as “satellite” accounts (employment, labor force, capital stock, land, etc.) that go with the flows in the SAM. Survey respondents said that the lack of incorporating stock variables remains a weakness of the IFPRI CGE model. The less (more) focus on stock (flow), then the less relevant the model becomes to specific contexts of each country. The model remains standard by lacking specificity to an individual country’s economic characteristics, but the inclusion of the stock–flow relationship would capture the essence of the country’s economic context. Finally, some suggested that setting up a so-called modeling support “desk” to handle economic and programming issues would be of great help to SAM/CGE users. Interviews: A Worldwide Bird’s Eye View From July through November 2019, we interviewed nine key researchers producing and using IFPRI SAMs and CGE models in an effort to better understand their usage and the outcomes resulting from SAM/CGE modeling (see Appendix 7 for the list of interviewees). The subsequent sections summarize the highlights of the interviews of key SAM/CGE modeling scientists. Through interviews and correspondence, we attempted to understand the impact that resulted from usage of IFPRI SAM/CGE modeling activities. 1. From Microcomputer Technology to Scenario Analysis: Hall-of-Fame View The CGE models through the time glass for use in food policy analysis in developing countries 17  In the 1980s and 1990s, CGE models were black boxes that prevented clear explanation of the results.  In the late 1990s, with the clear statement of the theoretical underpinning of the CGE models, scenario analysis was established as a powerful tool to explain results. o Specification of the interaction linkages led to scenario analysis. o Interaction linkages utilizing applied policy economics led to “story” and scenario analysis. o Outcomes of scenario analysis (second-generation CGE) are used for in-depth, hands-on policy analyses around the world. Where IFPRI CGE modeling is used  IFPRI developed CGE single-country models for many developing countries in Africa to address development policy questions and support decision making.  The modeling has been used in Pakistan, China, India, and Cambodia. For example, in Pakistan models were used to analyze the Impacts of Climate Risks on Water and Agriculture in the Indus Basin (Yu et al. 2013) and evaluate Alternative Development Options.  Effects of NAFTA on U.S. and Mexico. This CGE study found that the U.S. and Mexico benefit from NAFTA, with much larger relative benefits for Mexico’s economy.  A CGE modeling study of 16 countries in South America demonstrated the effects on economic development of South American economies of various resource allocation decisions.  IFPRI CGE analysis is mostly used to analyze issues in agriculture and energy  IFPRI CGE modeling is widely used by central planning ministers of many developing countries in Africa, Asia, and South America.  Standard multiplier analysis is used in many studies. The most recent assessed the potential economic cost of the COVID-19 pandemic in Asia, including one on China's macroeconomy and agri-food system, that provided policy recommendations to stimulate economic growth and development.  IFPRI SAMs were used to build the OECD Input/Output Database (www.oecd.org/industry/ind/2673344.pdf).  Ministries of economics worldwide know policy and policy options, as well as their country and the needs of the country, but they o want clear knowledge of the linkages between various sectors in the country’s economy, o want clear technical outcomes, and o would like to avoid surprises. Nowadays, single-country CGE modeling offers exactly this framework. Second-generation CGE modeling http://www.oecd.org/industry/ind/2673344.pdf 18  CGE modeling utilizing scenario analysis allows policymakers to see the impacts of alternative policies in different time spans (e.g., in 2050 or 2070).  Scenario analysis enhances evidence-based policies for agricultural transformation and rural development in developing countries.  CGE modeling facilitates the use of economic modeling tools and is widely used by the African Growth and Development Policy Modeling Consortium. 2. China: A View from Asia IFPRI and the Chinese Academy of Agricultural Sciences (CAAS) facilitate and share evidence- based research findings using IFPRI CGE modeling in order to inform policymaking on agricultural development and the impact of specific policies, such as rural revitalization, on China’s agricultural economy (link). According to the CAAS director general, IFPRI’s and CAAS’s unique collaboration utilizing consistent SAM data-driven, evidence-based CGE modeling puts CAAS in a unique position to be able to inform policymakers on alternative options for China’s food security issues, on sustainable development, and on poverty reduction. Many Chinese institutions and government entities are using the IFPRI/CAAS CGE model results, especially projections for briefing policymakers. The PIM-supported CGE modeling allows China’s policymakers to use and share evidence-based model results on fostering agricultural development and modernization, reducing poverty, restoring the environment, and strengthening governance. IFPRI/CAAS CGE model results are also used for furnishing reports about China’s policy impacts to the Bill & Melinda Gates Foundation. 3. Egypt/Yemen/MENA: An In-Depth Look Using CGE Modeling A PIM-supported CGE modeling team using high computer hardware capacity and state-of-the- art software, and developing an innovative CGE approach, has established a unique capacity- building engagement in the region. These scenarios then serve as a results repository that can be used by any analyst to present and discuss the model outcomes with decision makers. In other words, the process alleviates the need of a CGE modeler or allows non-modelers to make great use of model results. This is an innovative PIM-supported activity for CGE modeling and can be considered an advanced CGE capacity building. In a nutshell, defining assumptions, running CGE simulations, building a database, and saving the results expedites the modeling process and reduces the need of a CGE modeler. The CGE scenario repository allows a broader and in-depth comparison of alternative scenarios. CGE experts incorporated their knowledge into a country’s CGE model and were in a position to develop a vast sequence of alternative scenarios. The process of using technology (both hardware and software) to build CGE models, saving the scenarios, and retrieving the results at any time automates the process. It can be called the second generation of CGE modeling. In this framework, the need for a CGE modeler is minimized as teaching or getting an expert requires time. The process provides a country’s CGE modelers more time to focus on interpreting the results rather than teaching CGE modeling. The accumulation of the results of many scenarios allows the creation of an online CGE modeling database. It is a new approach that calls for the creation of a repository or library of CGE modeling results and serves like an online CGE modeling cloud (according to the interviewee). https://en.cdf.org.cn/cdf2019en/index.htm 19 This process was used to analyze Egypt’s agricultural development strategies and its irrigation options. It was used to facilitate the process for prioritizing policies and investment in MENA countries by FAO. Also, IFAD used a similar process for the Near East and North Africa region. In particular, the Agricultural Investment for Data Analyzer (AIDA) toolkit was built on an economywide framework to analyze the allocation of investment in rural sectors. In Egypt, FAO is a facilitator of the process that evaluates policy and investment options and provides alternative allocation of Egypt’s budgets and turns to IFPRI for CGE modeling support. The database approach to CGE model results was used by Egypt’s Central Agency for Public Administration (knowledge-sharing initiative), by Yemen’s Central Planning Program (country program), and by Jordan. SAM databases are used to train economists who work for agencies of the national governments of North African countries such as central statistical offices and central planning entities. In addition to training and building SAMs, CGE training has been offered to research institutes in Tunisia, Egypt, and local universities yearly. Finally, new partnerships with the World Bank and IMF enable productive feedback and interactions with the IFPRI CGE modeling team. 4. An Outsider’s Evaluation of PIM-Supported Middle East CGE Modeling Key points for success are data and institutions. Conflicting views of team members on modeling and low-quality data are the main obstacles to successful economywide modeling, while innovation and the involvement of institutions in a clear approach to modeling produce great outcomes. Whereas a World Bank attempt to use an economywide model in Iraq failed, a collaboration between the World Bank and the IFPRI CGE modeling team in the Middle East (Egypt, Yemen, and Iraq) generated a very positive experience resulting in the first CGE model for Yemen. The PIM-supported CGE modeling effort, with updating data and an economywide modeling specification, was a positive contribution. According to the outsider observer, economywide modeling demands deep knowledge of a country’s politics while policy experiments call for policy expertise. In other words, policy experience is a very important factor and policy expertise matters the most to make the CGE economywide modeling meaningful. Moreover, economywide modeling done from an organization’s headquarters is problematic as it fails to be in touch with a specific country’s politics and policies. An appreciation of that is very crucial for modeling and especially for evidence-based economywide scenarios. A sensible methodology and a one-on-one dialogue with policy expertise allow an independent broad reach of a county’s policymakers with the modeling experts. Acknowledging data problems is important, as is seeking the best available data; also, modeling the economy of a country under study requires knowledge of the country’s economic behavior. A country’s government’s participation in articulating its goals is also very crucial; this is what happened in the case of recovery of Iraq’s economywide modeling, where the economywide modeling failed because of data constraints, lack of appropriate methodology, and inability of the modeling team to work and interact with the relevant local people. Running a model is one thing https://www.ifpri.org/publication/agricultural-investment-data-analyzer-aida 20 and requires technical knowledge; however, building and running a model needs to be coupled with thorough research of the country’s goals and policies. Modelers need to be in touch with a country’s government policies and politics. In sum, the availability of fast Internet access and rather inexpensive computers with storage capacity allows increased economywide country modeling with a vast number of scenario executions and storage. A repository of scenarios that allows for the storage and retrieval of a variety of scenarios (the second stage of CGE modeling) and minimizes the need of a modeler is especially beneficial for developing countries with weak analytical capacities for designing and evaluating policies. 5. Ghana: Increased Expectations and Valuable Analyses Training and training techniques have increased the appreciation of economywide modeling, such as, for example, by the Ghana Statistical Services. Increased interaction with the IFPRI CGE modeling team has led to research results being discussed and used by policymakers; however, sometimes communication with policymakers that lacks a two-way interaction falls short of positive results/impact. The successful CGE modeling work in Ghana—especially policy analysis of agricultural inputs— increased demand for economywide modeling in Uganda, Kenya, and Nigeria. Regarding that increased demand, see “How to Spend Uganda’s Expected Oil Revenues? A CGE Analysis of the Agricultural and Poverty Impacts of Spending Options” (Wiebelt et al. 2018). We also see increased demand for IFPRI CGE evidence-based analysis, such as, for example, by the National Agricultural Investment Plan (Eastern and Western Africa) in the context of CAADP. Ghana’s statistical agency has rebased national accounts and is already planning for the next round of SAM- building once the final revisions of the national accounts are completed. Several papers have used the updated Ghana SAM and CGE model: for example, see Ghana Statistical Service, Office of the President, Social Accounting Matrix (SAM) 2015, http://www2.statsghana.gov.gh/nada/index.php/catalog/95. An AGRA-funded study in collaboration with Ghana’s Ministry of Food and Agriculture aimed at monitoring and evaluating the input subsidy program using CGE analysis. An IFC-funded study on the impact of subsidy programs is relying heavily on IFPRI CGE modeling analysis. A SAM multiplier course based on Ghanaian data was conducted in 2009. Ten years later, people who were part of that course are using multiplier analysis in their own work. 6. Rwanda: Innovation in Strategy In response to a request by the Rwandan Ministry of Agriculture, the IFPRI CGE team provided a CGE analysis, based on scenario analysis, in support of Rwanda’s Strategic Plan for Agriculture Transformation phase 4 (PSTA 4). PSTA 4 is the agriculture sector strategic plan under Rwanda’s National Strategy for Transformation. http://www2.statsghana.gov.gh/nada/index.php/catalog/95 http://www.fonerwa.org/sites/default/files/Rwanda_Strategic_Plan_for_Agriculture_Transformation_2018.pdf 21 7. Impact on Economic Development Many innovative studies have established the interaction between policies and economic growth, poverty, and inequality using the IFPRI CGE and microsimulation modeling. Throughout Africa, Bangladesh, Pakistan, Peru, and Vietnam, research primarily focuses on evaluating the following:  Returns to public investments and policies  Rural and regional development strategies  Climate change and energy policy  Economic growth and structural transformation  Investments in agriculture and energy infrastructure  Development strategy and agri-food system transformation • Modules have been developed and documented for several country analyses (see Appendix 3 and Link). Summary and Conclusions This report provides evidence of the worldwide usage of PIM-supported SAM database and CGE modeling work. The country-level IFPRI CGE modeling expertise has responded to many requests from governments and international organizations globally; many countries have asked for analytical support. We summarize our key findings as follows. Of the SAM datasets downloaded, the SAM for China was the most downloaded, followed by those from Ethiopia, Nigeria, South Africa, Mexico, and Ghana. Most downloaders were affiliated with a university, followed by government, NGO or nonprofit/think tank, private entity, other international organization, and CGIAR. The wide range of institutions that have downloaded SAMs indicates the worldwide demand for PIM-supported SAMs. Faculty, staff, and students from a vast variety of universities downloaded the SAM datasets. A variety of international organizations and institutions downloaded SAMs, including the World Bank, FAO, the Asian Development Bank, IFAD, AGRA, and the IMF. Among nongovernment affiliations were the Bill & Melinda Gates Foundation, the Brookings Institution, ODI, and Oxfam. In evaluating the CGE modeling program, the CGE academic citations constitute a key output to consider. When we consider the Google Scholar citations for all publications (269), the highest number of citations is 188 and that was for evaluating climate change in CGE modeling. CGE publications by IFPRI staff amount to 223 publications when IFPRI is the publisher. Those include both peer-reviewed and non-peer-reviewed publications. When we consider IFPRI authors publishing in external sources, the number of publications is only 132 (including both peer- reviewed and non-peer-reviewed publications). Using an advanced search approach, a sample of 269 CGE publications was identified and 3,328 citations were generated. The citations include peer-reviewed journal articles and book chapters. https://ebrary.ifpri.org/digital/search/collection/p15738coll3!p15738coll11!p15738coll19!p15738coll17!p15738coll6/ 22 We conducted two online surveys to identify outcomes attributable to the use of PIM-supported SAM/CGE modeling. Most of the respondents to both surveys had hands-on experience with SAM/CGE modeling, and many used SAM/CGE modeling analysis. The respondents also used IFPRI CGE modeling for their own research work. The majority of the participants used CGE modeling for policy analysis, publications, and educational training. Our interviews of key researchers producing and using IFPRI CGE models afforded us a better understanding of the usage and evolution of CGE modeling work. The availability of fast Internet connections along with high-quality, rather inexpensive computers with greater storage capacity has allowed for increased economywide country modeling with a vast number of scenario executions and storage. A wide range of applications was reported for the SAMs and the CGE models, including assessing policy impacts and helping governments prioritize investments in development strategies. In sum, this evaluation resulted in a rich database of references produced by the SAM/CGE IFPRI team under PIM. It has produced a database of articles that have cited IFPRI CGE publications and a list of authors who cited IFPRI CGE publications. Such information might be entered into a contact database for various possible uses. Furthermore, it provided a comprehensive assessment of the PIM-supported work on the use of microcomputer technology/scenarios for effective country analysis to inform development policy. 23 Appendix 1: Downloads of SAM Datasets, 2012–2019 Downloads of SAM datasets from 2012 to 2019 Dataset Title 2012 2013 2014 2015 2016 2017 2018 2019* Cumulative** A 2006 Social Accounting Matrix for Rwanda 22 52 58 20 29 40 24 25 284 A 2003 Social Accounting Matrix for Kenya 5 18 20 41 48 47 38 38 259 A 2000 Social Accounting Matrix (SAM) for the Slovak Republic 9 5 3 29 3 18 25 26 144 2014 Social Accounting Matrix for Malawi 0 0 0 0 0 0 74 50 124 2012 Social Accounting Matrix for Mozambique 0 0 0 0 0 0 48 36 84 Rwanda Social Accounting Matrix (SAM) 2011 0 0 0 88 79 79 87 55 388 Namibia Social Accounting Matrix (SAM) 2007 0 0 0 59 45 49 30 32 215 A 2001 Social Accounting Matrix for Zambia 0 22 27 7 8 15 14 30 177 A 1992 Social Accounting Matrix (SAM) for Tanzania 11 11 21 0 4 19 12 18 124 Swaziland Social Accounting Matrix (SAM) 2007 0 0 0 18 27 34 39 52 170 Ghana Social Accounting Matrix 2005 10 23 27 71 81 103 89 55 466 Brazil Social Accounting Matrices (SAM); 1995-1996 -- Aggregated Version 0 8 0 8 7 30 30 22 105 A 1997 Social Accounting Matrix (SAM) for Egypt -- Aggregated Version 1 2 0 1 4 9 17 21 55 Ethiopia Social Accounting Matrix (SAM) 2005-06 0 0 0 74 182 222 105 110 693 Lesotho Social Accounting Matrix 2007 0 0 0 12 29 37 44 44 166 A 1993-94 Social Accounting Matrix (SAM) for Bangladesh 0 8 21 24 17 60 61 41 233 South African Social Accounting Matrices (SAM) for 1993 and 2000 1 10 4 34 14 32 40 31 209 A 2006 Social Accounting Matrix for Nigeria: Methodology and Results 24 29 36 54 80 115 102 100 595 A 2007 Social Accounting Matrix (SAM) Database for Zambia 0 0 58 37 48 58 44 43 288 A 1997 Social Accounting Matrix for Honduras 4 1 4 2 7 17 32 29 96 A 2000 Social Accounting Matrix for El Salvador 5 1 9 15 18 18 36 35 137 A 2007 Social Accounting Matrix (SAM) for Uganda 38 83 102 52 41 55 19 17 407 Mexico Social Accounting Matrix (SAM) 1996 3 4 2 6 3 20 17 35 91 Botswana Social Accounting Matrix (SAM) 2007 0 0 0 19 53 40 73 79 264 2013 Social Accounting Matrix for Kenya 0 0 0 0 0 0 133 84 217 A 2009 Social Accounting Matrix (SAM) Database for South Africa 0 0 79 113 86 102 98 85 563 A 1991 Social Accounting Matrix (SAM) For Zimbabwe 8 6 6 12 24 35 32 39 206 Social Accounting Matrices for Mozambique 1994 and 1995 3 8 19 7 15 8 16 19 95 A 2007 Social Accounting Matrix for Malawi 45 98 70 39 36 58 33 18 405 A 1995 Social Accounting Matrix (SAM) For Zambia 0 7 7 8 4 13 29 28 123 A 2002 Social Accounting Matrix (SAM) for Peru and sub-national matrices 0 5 13 46 12 38 22 31 172 2010/11 Social Accounting Matrix for Ethiopia 0 0 0 0 0 0 163 136 299 Morocco Social Accounting Matrix 1994 2 1 9 15 6 34 28 24 119 Iraq Social Accounting Matrix 2011 0 0 19 19 25 17 21 59 160 A 1997 Social Accounting Matrix (SAM) for Egypt (Disaggregated Version) 0 8 7 4 12 28 20 27 106 A 2007 Social Accounting Matrix for China 0 152 276 114 83 145 102 96 968 A 1995 Social Accounting Matrix (SAM) For Uruguay 0 2 0 3 2 4 24 27 85 A 1994 Social Accounting Matrix (SAM) for Peru 2 9 11 7 3 6 19 28 111 2012 Social Accounting Matrix (SAM) for Sudan 0 0 0 0 0 0 63 83 146 Tanzania Social Accounting Matrix (SAM) 2009 0 0 0 85 61 137 86 41 410 A 2008 Social Accounting Matrix for Mexico 19 120 101 66 36 90 73 56 561 A 2007-2008 Social Accounting Matrix for Pakistan 1 122 114 51 31 48 34 35 436 A 1998 Social Accounting Matrix (SAM) For Thailand 1 4 10 12 14 21 27 29 153 2013 Social Accounting Matrix for Uganda 0 0 0 0 0 0 106 84 190 A 1998 Social Accounting Matrix (SAM) for Paraguay 3 8 0 2 4 6 14 21 91 Bolivia Social Accounting Matrix 2012 0 0 0 90 66 81 91 47 375 Yemen Social Accounting Matrix 2012 0 0 15 23 37 32 22 36 165 2013 Social Accounting Matrix for Ghana 0 0 0 0 0 15 63 56 134 A 1995 Social Accounting Matrix (SAM) for Indonesia 0 2 13 15 34 65 56 55 240 Vietnam Social Accounting Matrix (SAM) 2007 0 0 0 19 41 54 61 56 231 A 1997 Social Accounting Matrix for Colombia 1 1 0 11 6 18 20 33 90 Brazil Social Accounting Matrices (SAM); 1995-1996 (Disaggregated Version) 0 7 0 25 17 20 23 35 127 Argentina Social Accounting Matrix 2000 3 8 6 12 19 39 36 46 171 Bolivia Social Accounting Matrix 1996 0 1 6 1 1 5 16 19 53 A 2001 Social Accounting Matrix for Kenya 0 4 2 9 22 15 20 23 95 A 1999 Social Accounting Matrix (SAM) for Uganda 2 8 5 9 5 13 14 30 151 Malawi Social Accounting Matrix (SAM) 1998 0 1 7 13 11 20 13 22 96 The 1996 and 1997 Social Accounting Matrices (SAM) for Vietnam 0 7 18 17 15 21 15 24 146 Chile Social Accounting Matrix 1996 5 3 6 4 1 21 12 28 82 2015 Social Accounting Matrix for Tanzania 0 0 0 0 0 0 205 113 318 Egypt Disaggregated Social Accounting Matrix 2010/11 0 0 0 0 19 115 73 90 297 Tunisia Social Accounting Matrix 2012 0 0 0 0 8 88 50 58 204 Mozambique Social Accounting Matrix 2007 0 0 0 46 40 64 31 25 206 South Africa: Social Accounting Matrix (SAM) 1993 1998 and 1999 5 29 47 32 15 92 35 82 397 A Regionally Disaggregated Social Accounting Matrix (SAM) for Mexico 2008 0 0 0 36 37 58 74 56 261 Poverty-Focused Social Accounting Matrices for Tanzania 1998-2001 9 47 35 30 34 64 44 74 440 Costa Rica Social Accounting Matrix 1997 3 6 2 1 2 8 23 26 71 *Run through the middle of December, 2019 ** Cumulative download is total download of dataset(s) including those that are not associated with any particular download years 24 Appendix 2: SAM Dataset Titles (with Digital Object Identifier) Thomas, Marcelle, and Romeo M. Bautista. 2008. A 1991 Social Accounting Matrix (SAM) for Zimbabwe. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/8QFXOE. International Food Policy Research Institute. 2001. A 1993–94 Social Accounting Matrix (SAM) for Bangladesh. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/M54V1B. International Food Policy Research Institute. 2003. Brazil, Social Accounting Matrices (SAM); 1995– 1996—Disaggregated Version. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/UKRSQB. International Food Policy Research Institute. 1999. Morocco, Social Accounting Matrix 1994. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/1Y5ITG. Arndt, Channing, Antonio Cruz, Henning Tarp Jensen, Sherman Robinson, and Finn Tarp. 1998. Social Accounting Matrices for Mozambique 1994 and 1995. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/SPXN3P. Chung-I Li, Jennifer. 2008. A 1998 Social Accounting Matrix (SAM) for Thailand. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/VXCJVV. International Food Policy Research Institute. 2002. A 1997 Social Accounting Matrix (SAM) for Egypt— Aggregated Version. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/NRDIL1. International Food Policy Research Institute. 2000. A 1997 Social Accounting Matrix (SAM) for Egypt— Disaggregated Version. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/DQGK1E. International Food Policy Research Institute. 2002. Mexico, Social Accounting Matrix (SAM), 1996. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/FYGKY2. Thurlow, James, and Dirk Ernst van Seventer. 2002. South Africa: Social Accounting Matrix (SAM), 1993, 1998, and 1999. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/JU6GVD. Danish Research Institute of Food Economics. 2002. The 1996 and 1997 Social Accounting Matrices (SAM) for Vietnam. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/LX6HCD. International Food Policy Research Institute. 1999. A 1995 Social Accounting Matrix (SAM) for Indonesia. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/0JZ8U7. International Food Policy Research Institute. 2003. Brazil, Social Accounting Matrices (SAM); 1995– 1996—Aggregated Version. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/KNCYAC. International Food Policy Research Institute. 2001. Malawi, Social Accounting Matrix (SAM), 1998. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/YHXS8K. Thurlow, James, and Peter Wobst. 2008. Poverty-Focused Social Accounting Matrices for Tanzania, 1998–2001. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/HNCAP1. 25 Appendix 2, continued International Food Policy Research Institute. 2007. A 1994 Social Accounting Matrix (SAM) for Peru. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/PALCAR. Laens, Silvia. 2008. A 1995 Social Accounting Matrix (SAM) for Uruguay. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/FAZWO3. University of Kiel. 2005. A 1995 Social Accounting Matrix (SAM) for Zambia. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/9RMBRA. International Food Policy Research Institute. 2005. A 1997 Social Accounting Matrix for Colombia. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/EFGLUB. International Food Policy Research Institute. 2005. A 1997 Social Accounting Matrix for Honduras. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/KKPK0E. Molinas, José R., and César Cabello. 2007. A 1998 Social Accounting Matrix (SAM) for Paraguay. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/K1ML0U. Dorosh, Paul, and Moataz El-Said. 2008. A 1999 Social Accounting Matrix (SAM) for Uganda. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/OFB6QR. Koronczi, Karol. 2008. A 2000 Social Accounting Matrix (SAM) for the Slovak Republic. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/PCJ4PF. International Food Policy Research Institute. 2005. A 2000 Social Accounting Matrix for El Salvador. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/M86MMS. International Food Policy Research Institute. 2005. A 2001 Social Accounting Matrix for Kenya. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/PH2CCG. Thurlow, James, David Evans, and Sherman Robinson. 2008. A 2001 Social Accounting Matrix for Zambia. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/XKJ2HZ. International Food Policy Research Institute. 2000. Argentina, Social Accounting Matrix 2000. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/RJOYV7. International Food Policy Research Institute. 2005. Bolivia, Social Accounting Matrix, 1996. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/AKSGF1. International Food Policy Research Institute. 2005. Chile, Social Accounting Matrix, 1996. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/TFAGMI. International Food Policy Research Institute. 2005. Costa Rica, Social Accounting Matrix, 1997. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/ZWDQSH. 26 Appendix 2, continued Thurlow, James. 2008. South African Social Accounting Matrices (SAM) for 1993 and 2000. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/94ZDKL. International Food Policy Research Institute. 2005. A 2001 Social Accounting Matrix for Kenya. Washington, DC: IFPRI [dataset]. https://doi.org/10.7910/DVN/PH2CCG. Breisinger, Clemens, James Thurlow, and Magnus Duncan. 2007. Ghana, Social Accounting Matrix, 2005. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/IEONLO. Wobst, Peter. 2008. A 1992 Social Accounting Matrix (SAM) for Tanzania. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/I7XAUV. Nwafor, Manson, Xinshen Diao, and Vida Alpuerto. 2010. A 2006 Social Accounting Matrix for Nigeria: Methodology and Results. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/LHXP97. Nin-Pratt, Alejandro, James Thurlow, and Samuel Morley. 2011. A 2002 National Social Accounting Matrix (SAM) for Peru and Sub-national Matrices for the Coastal and Sierra/Selva Regions. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/BUURDE. Diao, Xinshen. 2012. A 2006 Social Accounting Matrix for Rwanda. Washington, DC: International Food Policy Research Institute [dataset]. http://hdl.handle.net/1902.1/18224. Douillet, Mathilde, Karl Pauw, and James Thurlow. 2012. A 2007 Social Accounting Matrix for Malawi. Washington, DC: International Food Policy Research Institute [dataset]. http://hdl.handle.net/1902.1/18578. Thurlow, James. 2012. A 2007 Social Accounting Matrix for Uganda. Washington, DC: International Food Policy Research Institute [dataset]. http://hdl.handle.net/1902.1/18662. Debowicz, Dario, Paul A. Dorosh, Sherman Robinson, and Syed Hamza Haider. 2012. A 2007–2008 Social Accounting Matrix for Pakistan. Washington, DC: International Food Policy Research Institute [dataset]. http://hdl.handle.net/1902.1/19361. Debowicz, Dario, and Jennifer Golan. 2012. A 2008 Social Accounting Matrix for Mexico. Washington, DC: International Food Policy Research Institute [dataset]. http://hdl.handle.net/1902.1/19037. Zhang, Yumei, and Xinshen Diao. 2013. A 2007 Social Accounting Matrix for China. Washington, DC: International Food Policy Research Institute [dataset]. https://doi.org/10.7910/DVN/LGZ3VV. Zambia Institute for Policy Analysis and Research (ZIPAR); International Food Policy Research Institute. 2014. A 2007 Social Accounting Matrix (SAM) Database for Zambia. Lusaka, Zambia: ZIPAR; Washington, DC: IFPRI. http://dx.doi.org/10.7910/DVN/24702. Human Sciences Research Council (HSRC); International Food Policy Research Institute. 2014. A 2009 Social Accounting Matrix (SAM) Database for South Africa. Pretoria, South Africa: HSRC; Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/24774. 27 Appendix 2, continued International Food Policy Research Institute. 2014. Botswana Social Accounting Matrix (SAM), 2007. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28526. Ethiopian Development Research Institute (EDRI); Institute of Development Studies (IDS); International Food Policy Research Institute. 2014. Ethiopia Social Accounting Matrix (SAM), 2005–06. Addis Ababa: EDRI; Brighton and Hove, East Sussex, UK: IDS; Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28527. International Food Policy Research Institute. 2014. Iraq Social Accounting Matrix, 2011. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/26587. International Food Policy Research Institute. 2014. Lesotho Social Acccounting Matrix (SAM), 2007. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28529. International Food Policy Research Institute. 2014. Mozambique Social Accounting Matrix (SAM) 2007. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28556. International Food Policy Research Institute. 2014. Namibia Social Accounting Matrix (SAM), 2007. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28531. International Food Policy Research Institute. 2014. Rwanda Social Accounting Matrix (SAM), 2011. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28532. International Food Policy Research Institute. 2014. Swaziland Social Accounting Matrix (SAM), 2007. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28538. International Food Policy Research Institute. 2014. Tanzania Social Accounting Matrix (SAM), 2009. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28540. University of Copenhagen; The World Bank; Central Institute for Economic Management (CIEM); International Food Policy Research Institute. 2014. Vietnam Social Accounting Matrix (SAM), 2007. Copenhagen: University of Copenhagen; Hanoi: CIEM; Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28541. International Food Policy Research Institute; Ministry of Planning and International Cooperation, Government of Yemen; Kiel Institute for the World Economy. 2014. Yemen Social Accounting Matrix, 2012. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/26496. International Food Policy Research Institute; Institute for Advanced Development Studies; Kiel Institute for the World Economy. 2015. An Agriculture-Focused, Regionally Disaggregated Social Accounting Matrix (SAM) for Mexico, 2008. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/28446. International Food Policy Research Institute; Inter-American Development Bank; Institute for Advanced Development Studies; Kiel Institute for the World Economy. 2015. Bolivia Social Accounting Matrix, 2012. Washington, DC: IFPRI. http://dx.doi.org/10.7910/DVN/29015. International Food Policy Research Institute; Central Agency for Public Mobilization and Statistics. 2016. Egypt Disaggregated Social Accounting Matrix, 2010/112. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/DH37H9. 28 Appendix 2, continued International Food Policy Research Institute; Institut Supérieur Agronomique de Chott-Mariem. 2016. Tunisia Social Accounting Matrix, 2012. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/E0O4MM. Economic and Policy Analysis Unit; International Food Policy Research Institute. 2017. 2010/11 Social Accounting Matrix for Ethiopia. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/G84XIB. Department of Labor, Auckland, New Zealand; International Food Policy Research Institute. 2017. 2012 Social Accounting Matrix for Mozambique. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/BAMNFN. Ghana Statistical Services; Institute for Statistical, Social and Economic Research; International Food Policy Research Institute. 2017. 2013 Social Accounting Matrix for Ghana. Washington, DC: IFPRI [dataset] http://dx.doi.org/10.7910/DVN/YVZ8KR. International Food Policy Research Institute. 2017. 2013 Social Accounting Matrix for Kenya. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/KYIRMV. International Food Policy Research Institute. 2017. 2013 Social Accounting Matrix for Uganda. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/XDNIGO. International Food Policy Research Institute. 2017. 2014 Social Accounting Matrix for Malawi. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/REUCQR. International Food Policy Research Institute. 2017. 2015 Social Accounting Matrix for Tanzania. Washington, DC: IFPRI [dataset]. http://dx.doi.org/10.7910/DVN/PPXXD9. International Agricultural Trade and Development Group, Humboldt-Universität zu Berlin; Department of Agricultural Economics, Khartoum University; Department of Agricultural Economics, Sudan University of Science and Technology. 2018. 2012 Social Accounting Matrix (SAM) for Sudan. Washington, DC: International Food Policy Research Institute [dataset]. http://dx.doi.org/10.7910/DVN/DO3MSH. 29 Appendix 3: Selected Publications for Selected Countries using SAM/CGE Outputs, 2012–2019 1. Uganda International Institutions and Researchers (excluding IFPRI) 1. Matovu, J. M. 2012. “Trade Reforms and Horizontal Inequalities: The Case of Uganda.” European Journal of Development Research 24, 753–776. http://www.palgrave-journals.com/ejdr/journal/v24/n5/abs/ejdr201235a.html 2. Kabajulizi, J., R. D. Smith, and M. R. Keogh-Brown. 2013. Macroeconomic Implications of Health Sector Reforms in Uganda: A Computable General Equilibrium Analysis. London: Department of Global Health and Development, London School of Hygiene and Tropical Medicine. http://ecomod.net/system/files/Conference%20paper_EcoMod2013_Judith.pdf 3. Estache, A., J-F. Perrault, and L. Savard. 2012. “The Impact of Infrastructure Spending in Sub-Saharan Africa: A CGE Modeling Approach.” Economics Research International 875287: 1–18. http://www.hindawi.com/journals/ecri/2012/875287/ 4. McArthur, J. W., and J. D. Sachs. 2013. A General Equilibrium Model for Analyzing African Rural Subsistence Economies and an African Green Revolution. Africa Growth Initiative Working Paper 12. Washington, DC: Brookings Institution. 5. Boysen, O., and A. Matthews. 2012. Impact of EU Common Agricultural Policy Reform on Uganda. Overseas Development Institute. 6. Kyalimpa, F. D. 2014. Prospects for Economic Growth and Poverty Reduction in Uganda a Computable General Equilibrium (CGE) Analysis. Dundee, UK: School of Business, University of Dundee. http://discovery.dundee.ac.uk/portal/files/4161190/Kyalimpa_phd_2014.pdf 7. Smith, R. D., and M. R. Keogh-Brown. 2013. “Macroeconomic Impact of Pandemic Influenza and Associated Policies in Thailand, South Africa, and Uganda.” Influenza and Other Respiratory Viruses 7 (2): 64–71. http://onlinelibrary.wiley.com/doi/10.1111/irv.12083/pdf 8. Robichaud, V., L. Tiberti, and H. Maisonnave. 2014. Impact of Increased Public Education Spending on Growth and Poverty in Uganda: An Integrated Micro-Macro Approach. Working paper 2014-01. UNICEF, Rome. https://ideas.repec.org/p/lvl/mpiacr/2014-01.html National Research Institutes IFPRI conducted a training course at EPRC in Kampala in 2008 and this led to multiple EPRC publications, graduate dissertations, consulting contracts (see below), and peer-reviewed journal articles. 9. Shinyekwa, I., and J. Mawejje. 2013. Macroeconomic and Sectoral Effects of the EAC Regional Integration on Uganda: A Recursive Computable General Equilibrium Analysis. Research Series 105. Kampala, Uganda: Economic Policy Research Unit, Makere University. http://ageconsearch.umn.edu/bitstream/159672/2/series105.pdf http://www.palgrave-journals.com/ejdr/journal/v24/n5/abs/ejdr201235a.html http://ecomod.net/system/files/Conference%20paper_EcoMod2013_Judith.pdf http://www.hindawi.com/journals/ecri/2012/875287/ http://discovery.dundee.ac.uk/portal/files/4161190/Kyalimpa_phd_2014.pdf http://onlinelibrary.wiley.com/doi/10.1111/irv.12083/pdf https://ideas.repec.org/p/lvl/mpiacr/2014-01.html http://ageconsearch.umn.edu/bitstream/159672/2/series105.pdf 30 Government and International Organizations IFPRI’s Uganda SAM underpins the World Bank’s MAMS CGE Model (in DEC-PG) and has been used for multiple World Bank studies. After the first publication, only the MAMS Uganda model is referred to, but it implies IFPRI’s Uganda SAM. 10. Rosetti, N., and M. Kakande. 2012. Modeling the Contributions of Reduced Gender Inequality to GDP Growth and Poverty Reduction. Budget Monitoring and Accountability Unit, Ministry of Finance, Planning and Economic Development, Government of Uganda. 11. Matovu, J. M., E. Twimukye, A. Musisi, and S. Levine. 2013. “Uganda.” In M. V. Sanchez and R. Vos. Financing Human Development in Africa, Asia and the Middle East. The United Nations Series on Development. London: Bloomsbury Academic. http://www.bloomsbury.com/us/financing-human-development-in-africa-asia-and-the-middle-east- 9781780932200 IFPRI Researchers 12. Dorosh, P., and J. Thurlow. 2014. “Can Cities or Towns Drive African Development? Economywide Analysis for Ethiopia and Uganda.” World Development 63 (10): 113–123. http://www.sciencedirect.com/science/article/pii/S0305750X1300226X 13. Dorosh, P., and J. Thurlow. 2012. “Agglomeration, Growth, and Regional Equity: An Analysis of Agriculture- versus Urban-Led Development in Uganda.” Journal of African Economies 21 (1): 94–123. http://jae.oxfordjournals.org/content/21/1/94 14. Benin, S., J. Thurlow, X. Diao, A. Kebba, and N. Ofwono. 2012. “Uganda.” In X. Diao, J. Thurlow, S. Benin, and S. Fan, eds., Strategies and Priorities for African Agriculture: Economywide Perspectives from Country Studies. Washington, DC: IFPRI. ISBN: 978-0-89629-195-9. http://www.ifpri.org/publication/strategies- and-priorities-african-agriculture 15. Van Campenhout, B., K. Pauw, and N. Minot. 2013. The Impact of Food Price Shocks in Uganda: First- Order versus Long-Run Effects. IFPRI Discussion Paper 1284. http://books.google.com/books?hl=en&lr=&id=ja32AQAAQBAJ&oi=fnd&pg=PP5&dq=cge+model+*uga nda*&ots=Npj9nCwilF&sig=AsJIwpTPdRwvhpEEm8SwJ7JQbNY#v=onepage&q=cge%20model%20*ug anda*&f=false 2. Tanzania International Institutions and Researchers (excluding IFPRI) Many peer-reviewed journal articles have been derived from others from IFPRI’s Tanzania SAMs, especially prior to 2008. This list is restricted to those since 2010. 1. Bezabih, M., M. Chambwera, and J. Stage. 2014. “Climate Change and Total Factor Productivity in the Tanzanian Economy.” Climate Policy 11 (6): 1289–1302. http://www.tandfonline.com/doi/pdf/10.1080/14693062.2011.579300 2. Estache, A., J-F. Perrault, and L. Savard. 2012. “The Impact of Infrastructure Spending in Sub-Saharan Africa: A CGE Modeling Approach.” Economics Research International 875287: 1–18. http://www.hindawi.com/journals/ecri/2012/875287/ http://www.bloomsbury.com/us/financing-human-development-in-africa-asia-and-the-middle-east-9781780932200 http://www.bloomsbury.com/us/financing-human-development-in-africa-asia-and-the-middle-east-9781780932200 http://www.sciencedirect.com/science/article/pii/S0305750X1300226X http://jae.oxfordjournals.org/content/21/1/94 http://www.ifpri.org/publication/strategies-and-priorities-african-agriculture http://www.ifpri.org/publication/strategies-and-priorities-african-agriculture http://books.google.com/books?hl=en&lr=&id=ja32AQAAQBAJ&oi=fnd&pg=PP5&dq=cge+model+*uganda*&ots=Npj9nCwilF&sig=AsJIwpTPdRwvhpEEm8SwJ7JQbNY#v=onepage&q=cge%20model%20*uganda*&f=false http://books.google.com/books?hl=en&lr=&id=ja32AQAAQBAJ&oi=fnd&pg=PP5&dq=cge+model+*uganda*&ots=Npj9nCwilF&sig=AsJIwpTPdRwvhpEEm8SwJ7JQbNY#v=onepage&q=cge%20model%20*uganda*&f=false http://books.google.com/books?hl=en&lr=&id=ja32AQAAQBAJ&oi=fnd&pg=PP5&dq=cge+model+*uganda*&ots=Npj9nCwilF&sig=AsJIwpTPdRwvhpEEm8SwJ7JQbNY#v=onepage&q=cge%20model%20*uganda*&f=false http://www.tandfonline.com/doi/pdf/10.1080/14693062.2011.579300 http://www.hindawi.com/journals/ecri/2012/875287/ 31 3. Branca, G., E. Felix, I. Maltsoglou, L. E. Rincón, and J. Thuelow. 2014. Producing Biofuels in Low-Income Countries: An Integrated Climate and Economic Assessment for Tanzania. Working Paper 2014-018, UNU-WIDER, Helsinki, Finland. http://www.wider.unu.edu/publications/working-papers/2014/en_GB/wp2014-018/ 4. Arndt, C. 2013. Impacts of World Fuel and Agricultural Price Changes: An Economywide Analysis of Tanzania. Seattle: University of Washington. http://faculty.washington.edu/bdillon2/opafs_docs/Arndt.pdf 5. Schürenberg-Frosch, H. Modeling the Effects of Aid-Financed Education Programmes on Sectoral Production and Income Distribution—A CGE Application to Tanzania. University of Duisburg-Essen, Germany. 6. Estrades, C. 2013. Guide to Microsimulations Linked to CGE Models: How to Introduce Analysis of Poverty and Income Distribution in CGE-Based Studies Version 2. AGRODEP Technical Note TN 09. http://www.agrodep.org/sites/default/files/AGRODEP_TN09.pdf National Research Institutes IFPRI conducted SAM training in Tanzania in 2005, but the target audience was the national statistics agency (NBS). Unlike in Uganda, IFPRI never provided training to a national research institute in Tanzania. 7. Kaliba, A. R., E. R. Mbiha, J. M. Nkuba, and P. M. Kingu. Economic Effects of Different Aid Channels on Dependency, Absorptive Capacity, and Poverty Reduction in Tanzania. https://www.gtap.agecon.purdue.edu/resources/download/4054.pdf IFPRI Researchers 8. Arndt, C., K. Pauw, and J. Thurlow. 2012. “Biofuels and Economic Development: A Computable General Equilibrium Analysis for Tanzania.” Energy Economics 34 (6): 1922–1930. http://www.sciencedirect.com/science/article/pii/S0140988312001648 9. Arndt, C., W. Farmer, K. Strzepek, and J. Thurlow. 2012. “Climate Change, Agriculture, and Food Security in Tanzania.” Review of Development Economics 16 (3): 378–393. http://onlinelibrary.wiley.com/doi/10.1111/j.1467-9361.2012.00669.x/abstract 10. Pauw, K., and J. Thurlow. 2012. “The Role of Agricultural Growth in Reducing Poverty and Hunger: The Case of Tanzania.” In S. Fan and R. Pandya-Lorch, eds., Reshaping Agriculture for Nutrition and Health. Washington, DC: IFPRI. http://www.ifpri.org/publication/reshaping-agriculture-nutrition-and-health 11. Diao, X., A. Kennedy, A. Mabiso, and A. Pradesha. 2013. Economywide Impact of Maize Export Bans on Agricultural Growth and Household Welfare in Tanzania. Discussion Paper 1287. Washington, DC: IFPRI. 3. Kenya International Institutions and Researchers (excluding IFPRI) IFPRI’s Kenya and Ethiopia SAMs were used by the International Livestock Research Institute for a range of studies, especially before Karl Rich departed for Sweden in the late-2000s. The first article on gender disparities written by a member of the IFPRI SAM team, Bernadette Wanjala, was published in 2009. http://www.wider.unu.edu/publications/working-papers/2014/en_GB/wp2014-018/ http://faculty.washington.edu/bdillon2/opafs_docs/Arndt.pdf http://www.agrodep.org/sites/default/files/AGRODEP_TN09.pdf https://www.gtap.agecon.purdue.edu/resources/download/4054.pdf http://www.sciencedirect.com/science/article/pii/S0140988312001648 http://onlinelibrary.wiley.com/doi/10.1111/j.1467-9361.2012.00669.x/abstract http://www.ifpri.org/publication/reshaping-agriculture-nutrition-and-health 32 Estache, A., J-F. Perrault, and L. Savard. 2012. “The Impact of Infrastructure Spending in Sub-Saharan Africa: A CGE Modeling Approach.” Economics Research International 875287: 1–18. http://www.hindawi.com/journals/ecri/2012/875287/ National Research Institutes IFPRI worked closely with the main national research institute and conducted targeted training courses with the Ministry of Finance and Planning. Government and International Organizations IFPRI’s Kenya SAM underpins the World Bank’s MAMS CGE Model (in DEC-PG) and has been used for multiple World Bank studies. IFPRI Researchers 1. Thurlow, J., J. Kiringai, and M. Gautam. 2012. “Kenya.” In X. Diao, J. Thurlow, S. Benin, and S. Fan, eds., Strategies and Priorities for African Agriculture: Economywide Perspectives from Country Studies. Washington, DC: IFPRI. ISBN: 978-0-89629-195-9. http://www.ifpri.org/publication/strategies-and-priorities-african-agriculture 2. Mabiso, A., K. Pauw, and S. Benin. 2012. Agricultural Growth and Poverty Reduction in Kenya: Technical Analysis for the Agricultural Sectoral Development Strategy (ASDS)—Medium Term Investment Plan (MTIP). ReSAKSS Working Paper 35. http://www.resakss.org/sites/default/files/pdfs/agricultural-growth-and-poverty-reduction-in-kenya- 50989.pdf 4. Ghana International Institutions and Researchers (excluding IFPRI) 1. Arndt, C., F. Asante, and J. Thurlow. 2014. Implications of Climate Change for Ghana’s Economy. WIDER Working Paper 2014/020. Helsinki: UNU-WIDER. http://www.econstor.eu/handle/10419/96301 2. Adam, M. A. 2014. Oil Boom, Fiscal Policy, and Economic Development: A Computable General Equilibrium Analysis of the Role of Alternative Fiscal Rules in Ghana’s Emerging Petroleum Economy. University of Dundee. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.613589#sthash.o7VMR18e.dpuf 3. van Dijk, M., M. Kuiper, and L. Shutes. 2014. “What Does the Future Hold for the Poor in Ghana? An Assessment of the Impact of Climate Change.” Paper prepared for the 17th Annual Conference on Global Economic Analysis, June 18–20, Dakar, Senegal. https://www.gtap.agecon.purdue.edu/resources/download/6916.pdf 4. Cooke, E. F. A., S. Hague, J. Cockburn, A-R El Lahga, and L. Tiberti. 2014. Estimating the Impact on Poverty of Ghana’s Fuel Subsidy Reform and a Mitigating Response. Working Paper 2014-02. Rome: UNICEF. National Research Institutes 5. Obeng, C. K. 2014. Impact of Import Liberalisation on Poverty: A Dynamic Computable General Equilibrium and Microsimulation Analysis for Ghana. University of Cape Coast, Ghana. http://mpra.ub.uni-muenchen.de/58182/1/MPRA_paper_58182.pdf http://www.hindawi.com/journals/ecri/2012/875287/ http://www.ifpri.org/publication/strategies-and-priorities-african-agriculture http://www.resakss.org/sites/default/files/pdfs/agricultural-growth-and-poverty-reduction-in-kenya-50989.pdf http://www.resakss.org/sites/default/files/pdfs/agricultural-growth-and-poverty-reduction-in-kenya-50989.pdf http://www.econstor.eu/handle/10419/96301 http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.613589#sthash.o7VMR18e.dpuf https://www.gtap.agecon.purdue.edu/resources/download/6916.pdf http://mpra.ub.uni-muenchen.de/58182/1/MPRA_paper_58182.pdf 33 6. Amporfu, E., D. Sakyi, and P. B. Frimpong. 2014. Demographic Dividend of Ghana: The National Transfer Accounts Approach. Kumasi, Ghana: Department of Economics, KNUST. http://www.ntaccounts.org/doc/repository/Amporfu%202014.pdf Government and International Organizations IFPRI’s Ghana SAM underpins the World Bank’s MAMS CGE Model (in DEC-PG) and has been used for multiple World Bank studies. IFPRI Researchers 7. Breisinger, C., X. Dia