Social Impacts: Poverty and Hunger, 2010-2012

The goal of this study was to assess how technical change in agriculture may have differential effects on indicators of well being, including poverty levels, hunger and food security, and nutrition. SPIA funded proposals from CIMMYT, WorldFish, and IRRI in June 2011 to document impacts on poverty and other indicators of well being. In addition to these three, SPIA commissioned a paper from IFPRI to assess CGIAR research impacts in Ethiopia on the same indicators. At the July 29th 2014 workshop in Minnesota ahead of the AAEA meetings, SPIA will bring together some of the authors of those case studies, some of the reviewers of those studies, and other agricultural/development economists to talk about what we have learnt, the methodological and data measurement challenges, and what CGIAR could do better to improve future efforts to document poverty impacts.


The goal of this study was to assess how technical change in agriculture may have differential effects on indicators of well being, including poverty levels, hunger and food security, and nutrition. There have been a number of advances in empirical economic work over the last ten years that can be brought to bear on this complex technology-poverty-food security issue. These innovations include a significant growth in the use of experimental and non-experimental methods for causal inference in development economics; advances in both the amount of household data available and the econometric techniques for analyzing these data; new spatial maps of poverty at sub-national levels; and a range of applications of general equilibrium models under different scenarios. The project is broadly following the steps outlined in this 2010 briefing note.


SPIA hired two consultants (Alain de Janvry and Elisabeth Sadoulet) to take stock of current approaches and outline future options. Their task was to evaluate the recent advances in data availability and analytical techniques in terms of their application to ex post assessment of impacts of agricultural research on poverty as measured by income poverty.


A brainstorming workshop titled “New approaches to documenting linkages between agricultural research, poverty and hunger” was held in December 2010 at IFPRI (Washington DC). A number of experts on poverty and nutrition in relation to agricultural research and development attended and click Events page for more information.


SPIA organised a two-stage call for proposals that ran between January and April 2011, with $500,000 funding from USAID for grants, for collaborations between the CGIAR centers and advanced research institutions/universities. Three proposals were funded and run between between June 2011 and mid-2013:

(i) Measuring the poverty and food security impacts of improved maize in Africa: A combined econometric and micro-economywide modeling approach
CIMMYT / Michigan State University / UMB-Norway / University of Zambia / University of Malawi

This study will measure the ex post impact of modern maize on poverty and food security in Malawi and Zambia, linking non-experimental econometric methods with economywide modelling. The average impact on adopters will be estimated with panel data methods to control for observed and unobserved heterogeneity as well as instrumental variables and treatment effects models (difference-in-differences with matching, propensity score matching) to control for endogeneity and observed selection bias. Endogeneity of access to subsidies and simultaneity of seed and fertilizer choices will be addressed.The econometric analysis generates technology coefficients for different household groups that will feed into micro-economywide models to capture spillover and local general equilibrium effects.

(ii) Improved methods for estimating technology adoption and impact: The case of integrated aquaculture-agriculture in Bangladesh
World Fish Center / Oregon State University

This study will estimate adoption, poverty reduction and household nutrition impacts from the promotion of integrated aquaculture-agriculture technologies in Bangladesh. Several research and dissemination projects developed and promoted IAA. With-without and before-after surveys showed broad adoption among project participants with significant economic and nutrition benefits. These surveys will be complemented by a new survey administered to the originally sampled participant and non-participant households. The project will use a novel simulation-based approach to impact assessment, the Tradeoff Analysis Model for Multi-Dimensional Impact Assessment (TOA-MD). TOA-MD is designed to simulate what would be observed if it were possible to conduct a controlled experiment to measure the effects of farms adopting a new production technology for all farms in Bangladesh where IAA is technically feasible. The analysis will use data from control and treatment groups to simulate adoption and impacts on poverty and nutrition in the target population, and will test various hypotheses related to adoption and impact.

(iii) Assessing the poverty and food security impacts of IRRI contributions to modern varietal replacement in Bangladesh, India, Indonesia, and the Philippines during 1990-2010
IRRI / University of Missouri-Columbia

The study will involve the compilation of time-series subnationally disaggregated survey-based data on the adoption of modern rice varieties so as to identify the dynamics of varietal replacement, and the traits conferred by the replacement process during 1990-2010. IRRI sourced conferred traits will be subsequently identified, and the consequent input use and output effects of contributed host-plant resistance, yield potential maintenance, abiotic stress tolerance, and improvements to milling recovery rates will be estimated via an array of methods by season and production environment. Effects on international and local rice prices, rice consumption, and food insecurity will be modeled via a subnationally disaggregated global structural econometric rice trade model based on comparisons between a historical baseline and an estimated counterfactual. Price responses and input-output effects will be used to calculate farmer revenue implications for poor adopters and nonadopters, as well as expenditure effects for poor consumers.

(iv) CGIAR-wide impacts in Ethiopia: A SPIA-commissioned country-level study

SPIA have directly commissioned a team led by IFPRI scientist Xinshen Diao, to coordinate input from all CGIAR Centers with a history of research in Ethiopia, namely IFPRI, CIMMYT, ILRI, CIP, ICARDA, and CIAT, in collaboration with the Ethiopian Agricultural Research Institute and the Ethiopian Development Research Institute.

The study, entitled, "Assessing the impacts of food staples research on income growth, poverty reduction and household nutrition in Ethiopia, 1995-2010" will use a country-level macroeconomic model to assess the impact of the adoption of research-led technologies, based on the best available evidence regarding the adoption and productivity impacts of commodity-specific technologies, natural resource mangement technologies and policy research. The study will run to mid-2013.

Related documents:

  • SPIA Briefing Note by Derek Byerlee, James Stevenson, Tim Kelley, Mywish Maredia, and Ross Conner, May 2010 (PDF)
  • Scoping study - Recent Advances in Impact Analysis Methods for Ex-post Impact Assessments of Agricultural Technology: Options for the CGIAR by Alain de Janvry, Andrew Dustan and Elisabeth Sadoulet, October 2010 (PDF)
  • Call for proposals, January 2011 (PDF)
  • WorldFish Impact Assessment report, case study (ii) above, March 2015 (PDF)
  • IRRI Impact Assessment report, case study (iii) above, June 2015 (PDF)

Recent blog entries

Recent publications

illustrative photo