More CGIAR agitprop

Two recent CG op-eds in the Telegraph (Shakuntala Haraksingh Thilsted and Ismahane Elouafi, 23 October 2023; and Ismahane Elouafi, 22 November 2023) commit errors that will mislead the public about historical and potential gains to investments in the CGIAR, errors that explain why the CGIAR finds it so difficult to restore its credibility as a scientific partnership. Thilsted is director of nutrition, health, and food security for the CGIAR and Elouafi is the incoming Executive Managing Director of the CGIAR, so they ought to know better.

The October piece (“The world is precariously dependent on just a handful of staple food crops—we must diversify”, henceforth [ST&IE]) misrepresents the global food situation and proposes “solutions” that have been tried for decades, have failed for various reasons, and whose failures will persist for basic environmental and economic reasons.

[ST&IE] contend that ‘the war in Ukraine has had such a significant impact on global food security because the world is precarious dependent on only a handful of staple food crops” and propose “a range of alternative solutions … to improve the quality of diets and nutrition, while also reducing global dependency on a few staple crops and the countries that dominate their export”.  

[ST&IE] continue to argue that global food supplies would be “more resilient” (“resilient” meaning, in this context, “less variable in quantity and in price”) if they were “diversified beyond the big three—rice, maize, and wheat”. To achieve greater resilience, [ST&SIE] propose “more investment and support” in such “now neglected” crops as pearl millet (P. americanum, a cereal), quinoa (C. quinoa, a member of the amaranth family), cassava (M. esculenta, a root), cowpea (V. unguiculata, a legume), teff (E. tef, a cereal), and pigeon pea (C. cajun, a legume). Though [ST&IE] do not mention sorghum (S. bicolor, a cereal), it is like pearl millet in production environments and research history and is a logical candidate for the diversification efforts suggested by [ST&IE].

There are several defects in the “now neglected” crops reasoning. The most obvious is that these crops have not been neglected in research. There has been international research on cassava at CIAT (HQ in Colombia) and at IITA (HQ in Nigeria) for more than 55 years and national research in Brazil (the genetic home of M. Esculenta) and in many Asian countries for longer periods. The Indian national programs in sorghum and millet began soon after WW2 and have been successful. The success of many years of ICRISAT pigeonpea work in India has not been replicated elsewhere, notably in sub-Saharan Africa where there has been a crying need for more grain legume production. The narrow genetic and production bases of teff (genetic origin in highland Ethiopia) and quinoa (the Andes) have made them specialty crops because of the difficulty of increasing their yields outside their areas of origin.

A second defect is that the market model implied by [ST&IE]— quantity shocks to maize and wheat production (resulting from war in 2022 or from, say, climate change) lead to higher prices for the big three. Such higher prices would in theory induce increased production and consumption of the “now neglected” crops (call them the NNC). This implicit market has failed because of low supply response in, and weak consumer demand for, the NNC. The low supply response is, as I have said, largely owing to the narrow production adaptability of the NNC (except for cassava). The demand issue is simply that people will not eat more millet and sorghum or cassava as incomes rise, or as relative prices fall. Even as Indian yields of millet and sorghum grew (partly as the result of public-private research on hybrid cultivars) the areas under those crops in India did not increase because of competition from more profitable rainfed crops (like cotton) or from irrigated staples such as wheat and rice. Higher incomes only induce cassava consumption as pig feed. Tef and quinoa are novelty crops in the rich countries, but because of their low adaptability and weak supply response, will never contribute much to global food supply.

A third defect is misleading claims about the scale of results from CGIAR work. While it is true that CGIAR (and national program researchers) have improved yield, adaptability, and nutritional content of orange-fleshed sweet potato (OFSP), there is no published evidence of adoption 1 of OFSP cultivars. It is, moreover, true that bean (Phaseolus) cultivars can « contribute to healthier diets » but where is the evidence that support for CGIAR research, past or present, has advanced that contribution? Even in Latin American countries with long traditions in Phaseolus, there has not been general growth of production or consumption despite decades of research.

The November article (« Science our best bet to fight and beat food insecurity », henceforth [IE]) makes one wonder why the CGIAR spends money on communications when its CEO signs pieces stuffed with simple errors.

Elouafi writes « Global agricultural productivity has declined by nearly 21 percent in the past 60 years ». This remarkable claim refers to a 2021 paper (Ortiz-Bobea et al 2) which estimated the response of agricultural production to weather. [IE]’s presentation of Ortiz-Bobea implies that if an index of global agricultural productivity had been 100 in 1960, then climate change had reduced it to 79 in 2020. This is not what Ortiz-Bobea et al found.  They found a 21% decline in total factor productivity (TFP) over what it would have been in 2020, ceteris paribus, NOT from what it was in 1961 (Fig 3a and 3b, p. 309 in Ortiz-Bobea). So, if TFP was 100 in 1961 and rose to about 210 in 2020 in the absence of climate effects, then climate change would have lowered it to something like (1-.21)*210 in 2020, not to something like (1-0.21)*100 in 2020.

[IE] goes on to confound the well-established achievements of rice, wheat, and maize breeding, in the CGIAR Centers and in many national public and private collaborations, with aspirational efforts in « climate resilient crops ».  Such aspirational efforts are focused either on niche crops, such as date palm, grass pea (L. sativa), and lentils, with limited global potential (or on wheat, of which I spoke above). Such aspirational efforts might work or might not work—evidence to date is that they have not had global effects on quantities or prices-- but their potential success bears no relation to the request for US$4 billion in new money. If this work had such large expected benefits, then why is it so underfunded today in One CGIAR?

Elouafi next contends that research on carbon sequestration in agriculture can « provide smallholder farmers with supplemental income ». This is divorced from reality for technical reasons (any viable method of significant C sequestration on small farms would require substantial fallowing of crop land or reforestation), economic reasons (the unit price of sequestered C price would have to be improbably high to make up for the low physical sequestration potential), and incentive reasons (we knew from current C sequestration schemes that monitoring costs are heavy).

The following paragraph in “Science our best bet” “suggests some confusion about the history of agricultural technology since WW2.

« Nobel Laureate Dr Norman Borlaug and Professor M.S. Swaminathan, for example, led the creation of a high-yielding rice variety that drove vital increases in food production in the second half of the 20^th century – averting a global famine and saving a billion lives. « 

Borlaug worked on wheat, not rice. The late Swaminathan was a DG of ICAR (the Indian national program) and of IRRI (the world’s leading rice research center) but there was a long history of rice (and wheat) work before the CGIAR began to be incorporated in the 1960s. The rice research did not generate one « high-yielding rice variety » but families of widely adaptable materials. Talk of global famine being averted and a billion lives saved is unscientific because the counterfactual can never be established. The claim of a reduction in infant mortality of 30% “across the developing world” is misleading as an inspection of the source for the claim of 30% will show.3

Both Telegraph pieces make unsupported claims of actual and potential benefits from CGIAR research and are of course timed to attract attention at COP-28. Such claims do no one any good, except perhaps for the the bureaucrats hired by One CGIAR to make them.