At odds with a vast body of economic evidence reporting exceptionally high rates of return to investments in agricultural research and development (R&D), growth in public R&D spending for food and agriculture has slowed in numerous, especially rich, countries worldwide. The observed R&D spending behavior is consistent with a determination that the reported rates of return are perceived as implausible by policy makers. We examine this notion by scrutinizing 2,242 investment evaluations reported in 372 separate studies from 1958 to 2011. We find that the internal rate of return (IRR) is the predominant summary measure of investment performance used in the literature despite methodological criticisms dating back more than a half century. The reported IRRs imply rates of return that are implausibly high. We investigate the reasons for these implausibly high estimates by analytically comparing the IRR to the modified internal rate of return (MIRR). The MIRR addresses several methodological concerns with using the IRR, has the intuitive interpretation as the annual compounding interest rate paid by an investment, and is directly related to the benefit–cost ratio. To obtain more credible rate of return estimates, we then develop a novel method for recalibrating previously reported IRR estimates using the MIRR when there is limited information on an investment's stream of benefits and costs. Our recalibrated estimates of the rate of return are more modest (median of 9.8% versus 39% per year); however, they are still substantial enough to question the current scaling back of public agricultural R&D spending in many countries.

Research-enabled growth in agricultural productivity is pivotal to sub-Saharan Africa’s overall economic growth prospects. Yet, investments in research and development (R&D) targeted to many national food and agricultural economies throughout Africa are fragile and faltering. To gain insight into what could be driving this trend, this article updates, summarizes and reassesses the published evidence on the returns to African agricultural R&D. Based on a compilation of 113 studies published between 1975 and 2014 spanning 25 countries, the reported internal rates of return (IRRs) to food and agricultural research conducted in or of direct consequence for sub-Saharan Africa averaged 42.3%py. In addition to the 376 IRR estimates, the corresponding 129 benefit-cost ratios (BCRs) averaged 30.1. Most (96.5%) of the returns-to-research evaluations are of publicly performed R&D, and the majority (87.6%) of the studies were published in the period 1990–2009. The large dispersion in the reported IRRs and BCRs makes it difficult to discern meaningful patterns in the evidence. Moreover, the distribution of IRRs is heavily (positively) skewed, such that the median value (35.0%py) is well below the mean, like it is for research done elsewhere in the world (mean 62.4%py; median 38.0%py). Around 78.5% of the evaluations relate to the commodity-specific consequences of agricultural research, while 5.5% report on the returns to an “all agriculture” aggregate. The weight of commodity-specific evaluation evidence is not especially congruent with the composition of agricultural production throughout Africa, nor, to the best that can be determined, the commodity orientation of public African agricultural R&D.

No abstract

The CGIAR is a unique and highly successful institutional arrangement for funding and conducting multilateral agricultural R&D. Established in 1971, the CGIAR has spent about $60 billion in present value terms mainly on R&D for staple food crops to support the world's poor. In this study, we provide a quantitative assessment of the past payoffs to CGIAR research. We compiled a comprehensive set of all known studies reporting estimates of returns to CGIAR research and to public research undertaken by low-and middle-income countries. We transformed these rate of return estimates into standardized benefit-cost ratios (BCRs), and we conducted a statistical meta-analysis. As a robustness check, we also undertook two complementary assessments. These comprised a selective review of nine ("billion-dollar") studies reporting large-scale benefits to CGIAR investments and an approximation approach based on an attribution of total factor productivity growth. The results are remarkably similar across the methods. All the evidence points to an overall BCR on the order of 10:1 across the portfolio for past CGIAR research investments.

No abstract