"GM crops have higher yields, improved performances, and greatly reduce the use of agrochemicals. Farmers like them because they increase income." Lim Li Ching and Jonathan Matthews debunk these myths, documenting failures of GM crops around the world.
Thousands of controlled trials have shown significantly decreased yields with GM crops.
A study based on 8,200 trials of soya varieties in US universities in 1998  reports yield drags between top RR varieties and top conventional varieties averaging 6.7%. In some areas, best conventional varieties produced yields on average 10% higher than RR varieties sold by the same seed companies.
In May 2000, results of a two-year study by Nebraska Universitys Institute of Agriculture and Natural Resources showed RR soya yielded 6% less than their closest non-GM relatives and 11% less than high-yielding non-GM varieties . The yield penalty was attributed to the gene insertion process.
Similar yield drags have been reported since 1997
In 1997, the University of Purdue found that transgenic soya varieties yielded on average 12-20% less than unmodified varieties grown at the same locations .
Research published in 1998 by the University of Arkansas and Cyanamid revealed reduced profit levels and lower yields for GM soya and cotton compared with unmodified varieties .
The University of Wisconsin found GM soya yields from the 1998 harvest lower than non-modified varieties in over 80% of cases in trials across nine US states .
In Iowa, a 1999 survey of reported an average RR-soybean yield reduction of 4% in over 365 fields .
A review of 40 trials of soya varieties in the north central region of the US by in 1999 found a mean 4% yield drag in RR soya .
In the UK, reports of crop trials from the National Institute of Agricultural Botany show yields from GM winter oilseed rape and sugar beet 5-8% less than high-yielding conventional varieties .
In summary, yield losses, not yield gains, are more commonly associated with transgenic crops compared to best available conventionally-bred cultivars and hybrids .
Yield drag in soya is associated with problems in root development, nodulation and nitrogen fixation, particularly in drought or infertile conditions, as the bacterial symbiont responsible for nitrogen fixation is sensitive to both Roundup and drought . Furthermore, there is a metabolic cost to expressing herbicide-resistance or the Bt-endotoxin. For example, levels of proteins responsible for plant defence responses are depressed following Roundup application. Although these are eventually restored to normal, pathogens quickly infect the plants in sub-optimal growing conditions. This forces a diversion of energy to repair damage, resulting in an essentially irreversible tax on yields.
University of Minnesota economist Vernon W. Ruttan sums up: "Thus far, biotechnology has not raised the yield potential of crops" .
Yet, an indication of distorted perceptions was shown through an opinion poll of 800 farmers, most of whom (53%) chose RR varieties because of perceived higher yields than non-GM varieties. When actual data from their farms were analysed, exactly the opposite was found . "It is interesting to note... that increasing crop yields was cited by over half the farmers as the reason for planting GM soya, yet yields were actually lower".
Bt resistance and more pesticides
The other big claim for GM crops is reductions in pesticide use. In reality, herbicide tolerant and Bt-transgenic varieties of GM crops are trapping farmers into more reliance on pesticides.
Recently, hundreds of hectares of GM cotton fields in Bulukumba, South Sulawesi, were destroyed by pests. Officials said that there was "nothing to worry about", and a spokesperson from Monsanto (the GM Bollgard cotton seed supplier) asserted that "they are just larva which eat the leaves, but will not disrupt cotton production". But local farmers complained, pointing out that the supplier had claimed the cotton variety was resistant to all kinds of pests.
What happens when GM crops fail to deliver on their promise of pest resistance? Farmers in Australia are now being advised to spray additional insecticide on Monsantos GM Bt cotton, INGARD, "under conditions of reduced INGARD plant efficacy" . The latest official guidance  makes it clear that Bt cotton is in some circumstances failing to control the principal target pest it was introduced for, Helicoverpa armigera.
Even when GM crops express pest resistance, there is little evidence of reduced pesticide use. This is borne out by data on transgenic cotton - although to date one fourth of American cotton is produced with genetically engineered Bt varieties, no significant reductions in the overall use of insecticides were achieved . In fact, those insecticides that could be replaced by Bt cotton make up a minor proportion of the insecticides used.
Similarly, with Bt corn, there is no independent evidence of a reduction in overall pesticide applications despite industry claims. Nor is there economic advantage in using Bt corn except in areas with very high pest infestation. Insecticide use on US Bt corn has in fact slightly increased, with insecticide targeting European corn borer rising from about 4% of acres treated in 1995 to about 5% in 2000 .
Herbicide use shows a similar picture. Although the cultivation areas of herbicide-tolerant cotton in the US have doubled annually over the past few years, herbicide use has shown little reduction. More revealingly, the sales of total herbicides that can be used with GM cotton have risen drastically since the introduction of herbicide-tolerant cotton .
While the Roundup Ready soybean system simplifies weed management, it entails 2-5 times more herbicide use than other weed management systems . Tolerance to Roundup is emerging in several key weed species, contributing to increased chemical use. Unbiased field-level comparisons, drawing on official USDA data, show that RR soybeans require more herbicides than conventional soybeans, despite claims to the contrary [9,15]. In 1998, total herbicide use on RR soybeans was 30% greater on average than on conventional varieties in six US states .
Analysis thus shows that RR soybean systems are not likely to reduce herbicide use or reliance. Claims otherwise are based on incomplete information or analytically flawed comparisons that do not tell the whole story . And as for RR corn, USDA data suggest that in 2000, the average RR corn acre was treated with about 30% more herbicide than the average non-GM corn acre .
Worryingly, research from the University of Alberta has revealed the rapid creation of multiple herbicide resistant canola plants in Canada as a result of pollen flow over significant distances . Cross-hybridizations occurred between a glyphosate-resistant variety and either glufosinate- or imidazolinone-resistant varieties. The evidence pointed to resistant gene movement via pollen flow from one field to another. Unusually, this occurred rapidly and multiple times, such that, through random crossing, certain plants showed triple resistance . One of the triple-resistant plants was found over 550 m from the pollen sources, greatly exceeding the 100-m buffer mandated for seed producers.
The greater expense of GM seeds and increased herbicide costs can already hit farmers pockets. Add to these the costs of yield drag and technology fees, and it is bad news for profitability. For example, the added costs for soya producers can be more than 12% of gross income per acre .
The Leopold Center for Sustainable Agriculture, Iowa State University, interviewed 800 Iowa farmers in 1998 to determine if growing GM crops was more profitable . Random surveys of 62 continuous cornfields, 315 rotated cornfields, and 365 soya fields concluded that the difference in profitability was non-significant for both crops. Thus, the farmers who raised GM crops did not gain any competitive edge.
The first farm-level economic analysis of Bt corn, in demonstrating less net profit, lower corn prices, and lost corn exports, questions whether planting GM corn is worth the cost . From 1996-2001, American farmers paid at least $659 million in price premiums to plant Bt corn, while boosting their harvest by only 276 million bushels - worth $567 million in economic gain. The bottom line for farmers is a net loss of $92 million - about $1.31 per acre. Furthermore, the US has foregone about 350 million bushels of corn export sales to the European Union since 1996/97 because the EU doesn't want GMOs. This is thus part of a triple negative for farmers - lost corn exports, lower corn prices and less net profit from Bt corn.
Furthermore, while transgenic cotton varieties may make pest control easier, they are not always worth the added expense when it comes to yield and fibre quality. Research by the University of Arkansas shows that many conventionals are the highest yielding varieties . Comparing the economics of a Bollgard/Roundup Ready variety with a conventional variety, "in a year when insect pressure was low the farmer spent about $10 an acre less for insect control with the conventional variety than he did with the more expensive stacked gene variety".
And can we put a price tag on the environment? Research points to the popularity of GM crops with many North American farmers because of their "convenience". A University of Nebraska report shows that farmers are using the technology to needlessly destroy weeds to get a "weed-free" field . The study demonstrates not only reduced profits, but also destruction of biodiversity.
Lessons from the South
We would do well to draw on the experiences of farmers in the South. The viability of non-GM alternatives has been demonstrated in a review of 208 projects/initiatives from 52 countries, adopted by 8.98 million farmers on 29 million hectares of land in Asia, Africa, and Latin America . Using a range of sustainable agriculture technologies - none of which involved GM farmers have achieved yield increases of 50-100% for rainfed agriculture, and 5-10% for irrigated crops.
Low-tech innovations by Southern farmers have boosted production . For example, in East Africa, corn faces two major pests stem borer and Striga, a parasitic plant. By planting a local weed (napier grass) that the stem borer prefers, pests are lured away from the corn into a honey trap the grass produces a sticky substance that kills stem borer larvae. By planting another weed, Desmodium, between rows of corn, Striga wont grow, as it is adverse to Desmodium. Pesticides are replaced by natural predators, and fertilisers by natural dung, crop wastes and plants that fix nitrogen from the air.
Further, going organic, entailing a restriction in the use of synthetic fertilisers and pesticides while excluding GM technology, could be more beneficial for the economies of developing countries. The FAO has recently urged poor nations to boost exports of organic produce to take advantage of booming markets in developed countries .
Sustainable agriculture and organic farming are not a panacea. They however offer alternative approaches to GM technology that have been demonstrated to provide increased yields and more income, while remaining environmentally friendly. No myths about this.
Benbrook, C.M. (1999) Evidence of the magnitude and consequences of the Roundup Ready soybean yield drag from university-based varietal trials in 1998, Ag BioTech InfoNet Technical Paper Number 1, www.biotech-info.net/RR_yield_drag_98.pdf
University of Nebraska (2000) Research shows Roundup Ready soybeans yield less, IANR News Service,www.biotech-info.net/Roundup_soybeans_yield_less.html
Benbrook, C.M. (2001) Troubled times amid commercial success for Roundup Ready soybeans: glyphosate efficacy is slipping and unstable transgene expression erodes plant defenses and yields, Ag BioTech InfoNet Technical Paper Number 4,www.biotech-info.net/troubledtimes.html
'Economist: Biotech has not made impact yet', Farm Progress, 21 November 2000.
See Thalmann, P. & V. Kung (2000) No reduction of pesticides use with genetically engineered cotton, for WWF International, www.biotech-info.net/WWF_inter_update.pdf; and Thalmann, P. & V. Kung (2000) Transgenic cotton: Are there benefits for conservation? A case study of GMOs in agriculture, with special emphasis on freshwater, www.panda.org/resources/publications/water/cotton/transgenic.html
Benbrook, C.M. (2001) Do GM crops mean less pesticide use? Pesticide Outlook, October 2001.