Pratt, Alejandro Nin, and Shenggen Fan (2010).  “R&D Investment in National and International Agricultural Research: An Ex-Ante Analysis of Productivity and Poverty Impact.”  IFPRI Discussion Paper 00986, May.

This study estimates the allocation of agricultural R&D investment necessary to maximize productivity growth and poverty reduction in developing countries.

Agricultural research and development spending varies significantly between developed and developing countries and among developing countries themselves.  Developed countries invest $2.36 in public R&D per $100 of agricultural output, while developing countries invest only $0.53.  Among developing countries, Asia spends 46 percent of all R&D in the group, Latin America 36 percent, and Africa 17 percent.  Individual countries are responsible for sizeable chunks of this investment: China generates 25 percent of total developing country agricultural R&D, Brazil 15 percent, and India 14 percent. 

A corollary of such stratified R&D investment is diverging agricultural productivity rates.  East Asian land productivity grew from $1,485 per hectare in 1992 to $2,129 in 2006.  Regional labor productivity grew from $510 per worker to $822 over the same period, and total factor productivity at nearly 3 percent per year.  The gap between East Asia and Sub-Saharan Africa steadily widened over this period.  Sub-Saharan African land productivity was 79 percent of that in East Asia in 1992, but by 2006, this figure had slid to 59 percent.

Differing agricultural R&D investment rates are important because the returns to such investment are very high.  The authors estimate that the average rate of return to national agricultural research systems is 60 percent, more lucrative than investment in education or roads.  They report that “The Asia and Pacific region has the highest ROR (78 percent); Africa has the lowest, but even the African RORs are high (49.6 percent).”  Interestingly, they find that rates of return from national agricultural research systems are significantly lower than those from international research systems such as the CGIAR group, indicating an opportunity for knowledge diffusion as well as the need for increased capacity building in national research systems.  CGIAR rates of return were 83 percent higher than their national counterparts in Africa, 72 percent higher in Asia and the Pacific, and 21 percent higher in Latin America.

In the final part of the paper, the authors use a transparent modelling and simulation framework to estimate the allocation of research investment necessary to maximize global agricultural productivity growth and poverty reduction.  They combine CGIAR and national spending as total investment that affects agricultural productivity in developing countries.  They evaluate four scenarios: business as usual agricultural R&D spending, 0.5 percent annual productivity growth, agricultural growth maximization, and poverty reduction maximization.  They estimate the impact of R&D spending on growth and poverty using elasticity values obtained from the literature.  They find that in order to maximize developing country agricultural output growth, R&D spending should be allocated to Southeast Asia and South Asia.  However, to maximize poverty reduction, spending should be allocated to Sub-Saharan Africa and South Asia.  Results were similar using a $1.25 per day and a $2 per day poverty line.  However, under the $2 per day scenario, relatively more funds would be invested in Asia rather than Sub-Saharan Africa under both the growth maximization and poverty minimization options.