The Areal et al study has significant differences from the Heinemann et al study. Heinemann et al measure up to 50 years of data; Areal et al measure only 10 years (1999-2009) and only one or a few years in each individual study. In our opinion, short-term studies provide little reliable indication either of consistency in the data or sustainability of yield. Short-term studies between newly introduced “GM seed” vs. conventional also often lack the necessary detail to determine if the comparison is to germplasm and management (e.g., use of fertilizer and pest control) of equal quality differing only in the GM trait. It is very well established that the best germplasm is often only available with a GM trait. The trait may have no contribution to yield (Gurian-Sherman, 2009, Heinemann et al., 2013). This is especially important in developing countries where conventional seed may be recycled to save costs.
If claimed yield gains are taken at face value, then average yield increases were 0.25t/ha in developed countries. If verified, that yield gain would still be less than the average difference of 0.35 t/ha in maize yields between Western Europe (9.2 t/ha) and Canada + USA (8.8 t/ha) during the 10 year period 2001-2011, making GM as a yield enhancing strategy in developed countries less effective than Western European non-GM approaches.
Moreover, Areal et al found no clear contribution from the GM trait to yield in developing countries. They say: “The results show that there was no difference between Bt and HT traits relative to non-GM crops in terms of agronomic performance when analysis was conducted for developing countries” p. 20.
It is therefore not possible from the Areal et al meta-analysis to conclude that GM is the reason for any detections of increased yield.
The study is silent on the scale of the different studies and how many hectares were involved in each; how many hectares were part of repetitive experiments that would provide confidence in reproducibility on location; and on different agronomic practices on the GM vs the non-GM plots. This information is critical for assessing the generality of the claims.
Areal et al draw from studies of unverified accuracy and reliability, such as reports from the industry selling the relevant seed [e.g., Barwale et al (2004) AgBioForum: J. Agrobiotechnol. Man. Econ 7, 96-100 2]. Unfortunately, using surveys of performance and popularity of proprietary seeds from the industry that sells them has the unavoidable effect of undermining claims of objectivity.
The Areal et al meta-analysis also excluded significant research that had conclusions contrary to their overall findings. For example, the peer-reviewed 4 year study by Jost et al (2008) which found that farmers of conventional cotton in the US state of Georgia were on average financially better off (Jost et al., 2008). Interestingly, it was because profitability correlated with yields and conventional on average had higher yields. We believe the omission of this study was unfortunate both because of its findings and because of its length.
In reporting ‘gross revenue’, the Areal et al paper implies but does not establish who derives the benefit. It would be a mistake to over-interpret the claim as farmer profit. For example, Barwale et al only report on estimated differential value per hectare, they do not demonstrate that the benefit was captured by the farmer, locally or even by the country.
It is therefore not possible from the Areal et al meta-analysis for us to determine if the sampling of studies was comprehensive or representative and that all the underlying data was reliable.
Despite the differences in methodology and information sources, the studies have some similarity in their conclusions. Neither study found a yield benefit in GM-led agroecosystems over conventional in developed countries. Heinemann et al found yield gains were linked to breeding, not GM. Areal et al also find that GM inputs are more expensive than conventional, a finding consistent with Heinemann et al.
Areal et al find some evidence of a correlation between use of GM crops and increased yield or ‘gross revenue’. The authors carefully and responsibly disclose that they cannot distinguish between GM contributing to yield gains or other factors, namely the skewed adopter effect. They say that it “is worth noting that although it cannot be discerned whether the advantages of cultivating GM crops were due to the technology itself or to farmers’ managerial skills (GM adopter effect)” p. 27, or available seed. And likely any short-term increase they measured will recede in the future because as they say: “the GM adopter effect is expected to diminish as the technology advances” p. 27.
Therefore, any yield or economic gain, if actual, is unlikely to be sustainable as predicted by Heinemann et al.
Economic gains were attributed to higher yields in developing countries. “It is concluded that in general GM crops performed better economically than non-GM crops, mainly due to the greater yields obtained by GM crops, despite production costs usually being higher for the GM variety” p. 27. However, yield gains were attributed to lack of pest management in the plots with conventional crops, not to the crop being GM. Thus most economic advantages are not derived from GM per se, but from lack of availability of pest management options in conventional plots.
This again underlines what both Heinemann et al and others have said (Pretty, 2001): it is important to compare GM/conventional agriculture to best practice, not to one another to determine what kind of agriculture is both most productive and sustainable. Moreover, in making GM to non-GM comparisons, the control must be in all other ways equal to the GM test plot.
Conclusions
Claims of yield increases correlated with the use of GM can be explained by the confounding variable that conventional controls were poor controls as they lacked any comparable attempt to limit the effects of pests and weeds. As Areal et al say: “The absolute yield difference between developing and developed countries for HT crops may be due to the relatively low efficacy of conventional treatments in developing countries prior to the adoption of GM crops” p. 30. The claims are built on a mixture of data sources, many with competing financial interests, which have not been vetted. We question the choice of studies and whether they are representative. Finally, the data is highly skewed toward one location, India, and one crop, Bt cotton, measured in 1-year increments over the short period between 2002-2007.
Additional and important sustainability factors were not addressed in the Areal et al study. Such as: “It is worth noting that the present paper does not cover any socio-economical implications of an increased dependency of farmers on multi-national companies controlling the GM seed market. In such cases, multi-national companies would have the power to increase input prices, which would raise GM production costs, hence gross margin differences between GM and conventional crops may change” p. 31.
- Gurian-Sherman, D. (2009). Failure to Yield. Union of Concerned Scientists.
- Heinemann, J. A., Massaro, M., Coray, D. S., Agapito-Tenfen, S. Z. and Wen, J. D. (2013). Sustainability and innovation in staple crop production in the US Midwest. Int. J. Ag. Sustain. in press.
- Jost, P., Shurley, D., Culpepper, S., Roberts, P., Nichols, R., Reeves, J. and Anthony, S. (2008). Economic comparison of transgenic and nontransgenic cotton production systems in Georgia. Agron. J. 100, 42–51.
- Pretty, J. (2001). The rapid emergence of genetic modification in world agriculture: contested risks and benefits. Environ. Conserv. 28, 248-262.
1 Economic and agronomic impact of commercialized GM crops: a meta-analysis F. J. AREAL, L. RIESGO AND E. RODRÍGUEZ-CEREZO Journal of Agricultural Science (2013), 151, 7–33.
2 From paper: “All figures given in the table are based on a survey conducted by Mahnyco (sic)…”.
