Crop Biotechnology: Benefits, Risks and Ownership
Gordon Conway; President,
Speech delivered March 28, 2000 in Edinburgh, Scotland at: GM Food Safety: Facts, Uncertainties, And Assessment; The Organization For Economic Co-Operation And Development (OECD) Edinburgh Conference on the Scientific and Health Aspects of Genetically Modified Foods
Let me begin with three observations:
First, I wish to stress that the Rockefeller Foundation's interest in crop biotechnology is solely driven by our desire to help feed the hungry of the world over the next twenty years. We believe biotechnology has an important role to play in meeting this objective. On current evidence, we assess the potential benefits to the developing countries as greatly exceeding the likely risks. If, however, new information causes this equation to be reversed, we would need to rethink this component of our food security program.
Second, I believe it is important to recognize that biotechnology comprises a wide variety of techniques. Some of these, for example forms of tissue culture and marker-aided selection, exploit the advances of modern cellular and molecular biology to improve the process of traditional plant breeding. Many of the current successes in producing new varieties for developing countries have been the result of the application of these methods rather than of the particular technique of genetic engineering.
Third, the biotechnology debate is partly about benefits and risks, and the balance between them. In that sense it lies in the political arena. Scientists and other experts can provide evidence - some theoretical, some based on experiments and experience- of the likely benefits and hazards and the probability of their occurrence. This may lead to a call for more tests, or greater regulation, or a more precautionary approach but, in the end, politicians need to decide, after consulting with all the stakeholders, what each country's policy should be.
It is not, however, a simple question of benefits and risks, since the debate over crop biotechnology raises at least seven overarching and interacting issues1:
In this paper I will deal with only some of these issues, beginning with my (admittedly imperfect) perception of the benefits and risks, and then briefly mentioning the critical questions of ownership and access.
Worldwide nearly 40 million hectares of transgenic crops were grown in 1999. This comprised over 40 different transgenic crops, but the largest acreages were in cotton, corn, soybean and rapeseed (In the USA 55% of all cotton, 50% of soybeans and 33% of corn were transgenic varieties)2.
These early applications have been relatively straightforward manipulations of single genes, for example:
The initial benefits in the US were envisaged to be the lowering of production costs for American farmers, primarily through better pest and weed control, and a reduction in pesticide use (indeed this was one of the reasons why the agro-chemical companies got in on the act; they foresaw a declining market for pesticides), with accompanying environmental improvement. On balance these benefits have been realized - although performance has been patchy. 3
Bt corn was primarily introduced to control the European corn borer, which causes losses in some years (e.g., 1997) of over $1billion, and an average reduction of yields of about 4.72 bushels per acre. The Bt gene results in very effective control, but the pest infestation is naturally highly variable, both from year to year and from place to place (the last major infestation was in 1997. In areas where Bt corn is widely grown the corn borer populations fell in 1999 to the lowest levels ever recorded). Some farmers in some years have benefited in terms of higher yields (for example, in Iowa Bt corn growers gained a 11.7 Bu/acre yield advantage in 1997) and a profit over the cost of the seed, others have seen little return and with declining prices may have suffered losses. With the collapse of the cornborer population there is little incentive for farmers to buy Bt seed for 2000.
About half of corn growers spray with prophylactic insecticides against the cornborer. These sprays (not needed with Bt corn) can cause outbreaks of spider mites and other secondary pests. A further benefit of cornborer control is the reduction in pathogens. The cornborer carries the fungus Fusarium and there is also a close correlation between cornborer infestation and a high incidence of carcinogenic mycotoxins. (European Union insistence on importing non Bt corn may result in a higher level of contamination by carcinogenic aflatoxin)4.
Bt cotton has also proven very successful against a range of serious cotton pests (pink and cotton bollworms and tobacco budworm). Benefits have been higher yields and reduced pesticide applications (the aggregate gain for the USA was $92 million in 1998; in Alabama Bt cotton has stopped cotton growers from going out of the business, because of the heavy resistance there of pests to insecticides). In some locations the return of birds, butterflies and other wildlife has been documented. Most dramatic has been the effect in China where millions of acres are now under Bt cotton. Yields are up, pesticide applications down from an average of 12 to 3 per season and, reportedly, there is a significant reduction in human poisoning from pesticides.
Glyphosate ('Roundup') is a herbicide that controls a broad spectrum of weeds and is easy to use. 'Roundup Ready' crops are resistant to the herbicide. It replaces cocktails of more selective herbicides (in Illinois farmers using no-till systems are replacing 6-8 conventional sprays with 2 'Roundup' sprays); however, the cost of the alternatives has fallen sharply in response to the introduction of the 'Roundup Ready' varieties. While many farmers see reductions in cultivation costs - depending on their farm system and mix of weeds - this has not been universal. Glyphosate is a broad spectrum herbicide, but is considered relatively safe because it breaks down rapidly in the soil. Under no-till systems this permits rotations with other crops and promotes soil conservation.
In practice it is difficult to draw a distinct line between 'traditional' plant breeding techniques (through which we have been 'redesigning nature' for thousands of years) and genetic engineering5. But the capacity of genetic engineering to move genes across genera and families, and between animals and plants, may give rise to unanticipated interactions within the genome with unknown effects. It is a new technology with which we have had limited experience. While we gain experience we need to move cautiously. As a general rule, we should probably be more cautious the greater the phylogenetic difference involved in the gene transfer.
The most serious environmental risk is the likelihood of transgenes escaping from cultivated crops into wild relatives (or contaminating organic varieties on nearby farms). This is a justified concern. Genes from existing inorganic crops can and do pass to organic crops, and vice versa, and genes from both transfer to wild relatives. Even self-pollinated crops, such as rice, will cross with wild rices. Such risks may be higher in developing countries; wild relatives are often common and cultivated land is far more mixed with uncultivated land than in the industrialized countries.
Much attention has been paid to the distances move by pollen, but the key questions are:
Only extensive, well-designed and monitored field tests will give us the answers.
Another potential hazard arises from plants containing genes from viral pathogens that confer resistance to these same pathogens. Expressing the viral genes in plants somehow disrupts the virus infection process. But, exchange of these genes with other viral pathogens may be possible, creating entirely new virus strains with unknown properties.
A third significant risk - the potential for pests to evolve resistance to the toxins produced by Bt genes - is well known, as are some of the counter-strategies. One answer is to employ refuges of non-Bt crop plants. Another uses two or more toxin genes each with a different molecular target. But experience indicates that we need to anticipate the eventual breakdown of control. Introduction of Bt into a wide range of crops implies a much higher selection pressure than from spraying the insecticide on a single crop. Organic farmers, who use Bacillus thuringiensis as an insecticidal spray, would be disadvantaged if resistance became widespread. We need to carefully monitor insect populations for resistance and alternative strategies continuously need to be developed.
In general, Bt is regarded as a safe insecticide, since it mostly kills pest caterpillars leaving beneficial insects unharmed (and for this reason is often used as a key pesticide in integrated pest management programs). Field experience with Bt corn confirms this.6 However, in a recent, much-publicized laboratory experiment, pollen from Bt corn was shown to kill caterpillars of the Monarch butterfly.7 It was not a surprising result: the Bt toxin's characteristic is that it kills chewing insects such as caterpillars. Subsequent investigations in the field show that pollen concentrations fall away very rapidly from the edge of the cornfield. Beyond three meters the pollen load on neighboring milkweed plants - the natural food of the Monarch caterpillar - will be safe for the caterpillars. In the not too distant future, the problem may be solved by using gene promoters in the genome that prevent the expression of Bt in the pollen. Recent research has also shown that the Bt toxin remains active and persistent in the soil, binding to clays and humic acids.8 These cases demonstrate the importance of more detailed monitoring of the effects of Bt crops on the environment, and the need to weigh the downsides against those of conventional insecticide treatment.
The most publicized health risk is that GM crops carrying antibiotic genes used as markers may generate antibiotic resistance in livestock or humans. This has not been demonstrated experimentally, however the mechanism for such transfer exists. Alternative selection markers are now available and should be used.
There is also concern that transgenes may increase allergies, through the inadvertent transfer of allergenic proteins (as occurred in the transfer of a gene from Brazil nut to soybean). Allergies can be tested for, where known, but there may be surprises, especially where genes are transferred across large phylogenetic distances.
Other fears have less scientific basis. There is no convincing evidence to date that the process of gene transfer confers a health risk. But there clearly need to be thorough, well-conducted experiments, open to public scrutiny, to further test this proposition.
There is also no a priori reason why ingesting pieces of transgenic DNA is likely to be hazardous, any more than the large quantities of DNA from numerous sources ingested every day in normal diets. But long-term assessments need to be urgently put in place.
This summary assessment goes some way to explaining European (and growing American) concerns. European consumers gain no benefit from the GM food currently available (nor, for that matter do American consumers); the benefits virtually all go to the American farmer and, of course to the biotechnology companies. And, although in some situations there are clear environmental benefits, these are largely discounted in the public debate. On the other hand, there are risks, some real some imaginary. There is also concern over the domination by a few multinationals of the food chain and a fear that consumer safety and choice is being compromised.
In the future, we are likely to see much greater attention to output traits, i.e. qualities in the product of value to consumers. This second generation of biotechnology products is already underway and the new transgenic plants are visible in company and university laboratories and greenhouses. They include crops with better flavor and appearance, greater shelf life, improved nutritive value and so on. Some crops will be grown as sources of plastics, others to provide pharmaceuticals. Whether this will change public perceptions remains to be seen.
When we turn to the developing countries the evidence, so far, suggests a very different balance of benefits and risks.
The Green Revolution was one of the great technological success stories of the second half of the 20th century. Food production in the developing countries kept pace with population growth, with both more than doubling over the past forty years. Yet it was in some respects flawed.9 Many did not benefit. Today, over 800 million people, equivalent to 15% of the world's population, get less than 2000 calories per day and live a life of permanent or intermittent hunger and are chronically undernourished. Most of the hungry are women and young children.
Lack of proteins, vitamins, minerals and other micronutrients in the diet is also widespread. About 100 million children suffer from vitamin A deficiency. They are more likely to develop infections and the severity of the infection is likely to be greater. Each year half a million go blind and some 2 million die as a result. Iron deficiency is also common. About 400 million women of childbearing age (15-49 years old) are afflicted by anemia caused by iron deficiency. As a result they tend to produce stillborn or underweight children and are more likely to die in childbirth.
And if nothing new is done, the numbers of poor and hungry will grow. Most populations in the developing world are still increasing rapidly. By the year 2020, twenty years from now, there will be an additional 1.5 billion mouths to feed. What is the likelihood they will be fed?
The prognosis is not good. There is widespread evidence of declines in the rates of yield growth (Figure 1). At the same time investment by donor countries in agricultural research is falling rapidly (Figure 2). This is part of a general pattern of reduced public aid to the developing countries.
Some argue argue that lack of food is simply a problem of unequal distribution. If poor people were not poor they could buy the food they need. This is true, but oversimplistic and not very helpful. There are no signs the world is about to engage in a massive redistribution of wealth. Food aid on the scale required would be costly in both economic and environmental terms for the industrialized countries and would inhibit local farmers from producing for the market. And the practical reality is that the majority of the poor live in rural areas. The only way they can increase their incomes is through agricultural and natural resource development, which means greater productivity.
I believe we need a new revolution - a Doubly Green Revolution, that repeats the successes of the old but in a manner that is environmentally friendly and much more equitable. This is going to take the application of modern ecology in such areas as integrated pest management and the development of sustainable agricultural systems. It is also going to need much greater participation in the development process by farmers themselves. But I also believe it is going to need the application of modern biotechnology - to help raise yield ceilings, to produce crops resistant to drought, salinity, pests and diseases, and to produce new crop products of greater nutritional value.11
In response to this challenge The Rockefeller Foundation, has funded, over the past 15 years, some $100 million of plant biotechnology research and trained over 400 developing country scientists. At several locations in Asia there is now a critical mass of talent applying the new tools of biotechnology to rice improvement.
Chinese scientists, with funding from the Rockefeller Foundation, have produced a new rice variety using anther culture (a form of tissue culture), called La Fen Rockefeller, that is providing farmers in the Shanghai region with 15 - 25% increases in yield. Scientists in West Africa have also used anther culture to cross the high yielding Asian rices with the traditional African rices. The result is a new plant type that looks like African rice during its early stages of growth (in particular, it is able to shade out weeds) but becomes more like Asian rice as it reaches maturity, resulting in higher yields with few inputs.
Progress is also being made in the production of transgenic cereals for the developing countries. The most significant achievement has been the introduction of genes that produce beta-carotene - the precursor of vitamin A - in the rice grain. Beta-carotene is present in the leaves of the rice plant, but conventional plant breeding has been unable to put it into the grain. Dr. Ingo Potrykus of the Swiss Institute of Plant Sciences in Zurich, who carried out the work with Rockefeller funding, transferred one bacterial gene and two daffodil genes.12 The transgenic rice grain has a light golden-yellow color and contains sufficient beta-carotene to meet human vitamin A requirements from rice alone. He has also added a gene from the French bean to rice that increase its iron content over threefold.
We are now embarking on a program to support biotechnology research in Africa, and in particular to fund the development of crops which withstand the stresses - drought, salinity, pests, diseases and weeds - that face poor farmers on relatively marginal lands.
The only way to gain a better assessment of benefits and risks is to conduct relevant and appropriate trials and tests that are independently monitored, and available for public scrutiny.
For example, in the case of the vitamin A rice we need to ask a series of critical questions. What levels of intake are needed for children to obtain sufficient vitamin A from a diet based on rice? Is there a danger of excessive intake? Is there a risk associated with excess intakes of beta-carotene or vitamin A?13 Are there any other potential health hazards? Could an allergy have been transferred (daffodils are responsible for 'daffodil pickers rash').
We also need to know what effects the beta-carotene might have on the environment, for example will it affect the insects that suck the grain? If beta-carotene gets into the grain of wild rices will it persist and will it have any significant effect?
There may be other concerns. Some may feel it is better, despite the cost and logistics, to provide children with vitamin A tablets. Others may believe it is desirable for children to get their vitamins A from homegrown leafy vegetables. I tend to agree with them, but realistically it is going to be many years before the poor farmers in the seasonally arid regions can grow vegetables throughout the year. Moreover the critical challenge is to improve the nutritive value of the rice gruel on which babies are typically weaned in many Asian countries. This is when the lack of vitamin A is so important.
The developed countries are clearly better equipped to assess such hazards. A crucial need is to provide the necessary training in biosafety for the developing countries.
More important than the potential hazards, at least to my mind, is the question of who benefits from biotechnology. In the industrialized countries the new life science companies (notably the big six multinationals, who through a process of mergers are now able to combine crop biotechnology with agrochemical and seed production, - AgrEvo, Dow, Dupont, Monsanto, Novartis and Astra-Zeneca) have dominated the application of biotechnology to agriculture.
So far the focus of these companies has been on developed country markets where potential sales are large, patents are well protected and the risks are lower. Most of the GM crops being currently grown in the developing countries are cash crops, for example Bt cotton in China. There is less interest in small farmer food crops because the returns are deemed to be low. Nevertheless, this situation could change. The major biotechnology companies are buying up local seed companies. More important they have embarked on an aggressive policy of identifying and patenting potentially useful genes, many of which may be of value to poor farmers (in the case of the Vitamin A rice, Dr. Potrykus had to use some 24 patented technologies)
Of particular concern in the developing countries have been proposals to introduce terminator genes. These would ensure that the seeds harvested from genetically engineered crops are sterile and hence could not be kept for next season's sowing. At present some 80% of developing country farmers keep their seed from year to year. In my speech to the board of Monsanto in June of last year I listed seven steps they should take to improve the their public accountability and responsibility. One of these was that they should disavow the use of the terminator technology, and I was pleased that in October they agreed to do this. However, there remain the other six steps that still have to be addressed (including the abandonment of antibiotic markers and the donation of technologies such as used in the production of Vitamin A rice).
Until recently plant breeders have relied on the Plant Variety Protection (PVP) system to protect their rights. Under this system breeders gain protection for their varieties, but farmers can save the seed and other breeders can use the varieties to produce new varieties. In the US this has been largely replaced by a system of patents - applied to genes, pieces of the genome, technologies of gene transfer, and existing and new varieties. It is justified in terms of protecting intellectual property rights and stimulating innovation. For the developing countries, however, it is likely to move potentially beneficial innovations out of the reach of the poor.
Part of the answer is for the developing countries to adopt a mixed approach. On the one hand, they could encourage the for-profit sector to develop and market high quality, locally adapted, premium seeds (especially hybrid seed, which farmers can, if they wish, keep for the next season's crop although the yields are likely to be lower) for the commercial and semi-commercial farmers. Protection would be through a modified PVP system. On the other, they would encourage a strong public sector seed system that serves poorer farmers. This would provide an economic incentive for private sector research, innovation, and marketing, and help ensure that the public sector had access to new technologies. Over time, more and more farmers from the semi-commercial sector should be able to buy seeds on a regular basis. And, hopefully, many farmers would move from being really poor to the semi-commercial or commercial categories.
A key part of such an approach would be the stimulation of public-private partnerships whereby genomic information and technologies are made available to public plant breeders. For example, it is hoped that the patent holders will donate the technologies used in producing vitamin A rice.
There can be no doubt that the agricultural biotechnology industry is going through a very difficult phase. As an independent commentator has put it "GM foods have become a lightning rod for a host of modern concerns; skepticism about the regulatory process, increasing anxieties about food, hostility to high intensity agriculture, patenting of seeds and consolidation of companies into a handful of world-wide giants."
The opposition in Europe and the effective ban there on imports of GM foods is a major blow to US farmers and the industry (for example, the value of soybean exports to Europe is about $1.5 billion). Nevertheless, the latest figures suggest that US farmer confidence in GM crops is holding up, better than was expected. There has been no decline in the placement of orders of GM soybean and next year's cotton acreage is expected to remain at about 50% GM. There has been a drop in sales of Bt corn seed. This seems to be partly a response to uncertainty about the markets, but is largely a function of the disappearance of the European cornborer. If the pest continues to be absent next season there will be no advantage in growing Bt corn. However, sales of seed are only a partial indicator of what will be sown. Farmers are hedging their bets.
In contrast, most people seem to accept the numerous genetically engineered vaccines and other pharmaceuticals. Some 100 products are in common use - including insulin, hepatitis vaccine and various products for cardio-vascular disease (in several instances they replace blood derived products where the risk of contamination by disease organisms was high). It is highly likely that when we make the breakthrough and produce an effective HIV/AIDS vaccine it will be genetically engineered. Few are likely to object. Genetically engineered Vitamin A rice could bring a similar order of benefits, and with the probability of few risks.
Finally, I wish to note that although I have separated the equation of benefits and risks and the question of ownership, there is a significant linkage. The assessment of risks cannot be left to the private sector. Unless there is a clear public involvement in the nature and progress of biotechnology, including public investment, greater regulation and improved public understanding, the risks are likely not to be properly assessed and addressed, and the benefits will go to the rich rather than the poor.