11.4 Agricultural biotechnology

This section focuses on the potential, risks and likely benefits of agricultural biotechnology to 2030. The benefits of agricultural biotechnology arise from its potentially large contribution to productivity gains and quality improvements. Productivity gains encompass essentially all factors of agricultural production: higher returns on land and livestock, labour and capital or simply lower input requirements per unit of outputs. This may mean higher crop and livestock yields, lower pesticide and fertilizer applications, less demanding production techniques, higher product quality, better storage and easier processing, or enhanced methods to monitor the health of plants and animals. Ultimately, higher productivity will result in lower prices for food and fibre, a benefit for all consumers but particularly important for the poor who spend a relatively large share of their incomes on food and fibre.

Higher productivity also holds the key in the fight against rural poverty. The underlying mechanisms of the productivity-poverty nexus have been discussed in Chapter 8. Biotechnology holds the promise of boosting productivity and thus raising rural incomes, in much the same way as the green revolution did in large parts of Asia during the 1960s to 1980s. It could kick-start a new virtuous cycle of productivity growth, increased output and revenues.

But there are also numerous risks and uncertainties associated with these new technologies that have given rise to a host of concerns and questions. The most important of these is whether and how developing countries can actually harness the potential of biotechnology to promote production and the productivity of the poor. This in turn raises other questions. Whether and to what extent are the needs of developing countries being taken into account in current research efforts? How fast and to what extent have GM crops been adopted by developing countries? Which crops took the lead? Are the products developed by and for developed countries suited to the economic and ecological environments of developing countries and to what extent will developing countries develop their own biotechnology applications? More specifically, will “orphan” crops such as millet or bananas, which often play a vital role in the livelihoods of the poor, receive sufficient attention by new research? Will farmers in developing countries be trained and equipped to reap the benefits of the new technologies? Will the proliferation of GM-based crops and livestock further weigh on biological diversity? How can consumer concerns about environmental safety and potential human health hazards be taken into account, at low costs and without unduly distorting international trade? The parameters that determine the answers to these questions are changing quickly and it is therefore impossible to provide definite answers, particularly in view of the long-term perspective of this study. Instead, the following section will discuss some of the factors that are likely to affect the development and adoption of these new technologies in the future.

11.4.1 What is agricultural biotechnology?

Many traditional forms of biotechnology continue to be used and adapted. Some biotechnologies, such as manipulating micro-organisms in fermentation to make bread, wine or fish paste, or applying rennin to make cheese, have been documented for millennia.

Modern biotechnology takes various forms. These include: (i) tissue culture, in which new plants are grown from individual cells or clusters of cells, often bypassing traditional cross-fertilization and seed production; (ii) marker-assisted selection (MAS), in which DNA segments are used to mark the presence of useful genes, which can then be transferred to future generations through traditional breeding using the markers to follow inheritance; (iii) genomics, which aims to describe and decipher the location and function of all genes of an organism; and (iv) genetic engineering, in which one or more genes are eliminated or transferred from one organism to another without sexual crossing. A GMO, also referred to as a living modified organism (LMO) or transgenic organism, means any living organism that possesses a novel combination of genetic material obtained through the use of modern biotechnology.¹

Marker-assisted selection. Traditionally, plant breeders have selected plants based on their visible or measurable traits (phenotype). But this process was often difficult, slow and thus financially costly. MAS helps shorten this process by directly identifying DNA segments (genes) that influence the expression of a particular trait. The markers are a string or sequence of nucleic acid that makes up a segment of DNA.² As more and more markers become known on a chromosome, it is possible to create a detailed map of the markers and corresponding genes that codify certain traits. Using detailed genetic maps and better knowledge of the molecular structure of a plant, it is possible to analyse even small bits of tissue from a newly germinated seedling. Once the tissue is analysed, it is possible to check whether the new seedling contains the specific trait.

These new techniques are also important because they are not stigmatized by the negative attributes associated with GMOs, which have resulted in growing concerns about the safety of these new products for consumers and the environment (see below). They have revolutionized conventional breeding and help accomplish significant genetic improvements across almost all crops and livestock. And, should consumers’ concerns vis-à-vis GMOs become more important, they could become the most crucial biotechnological application in the future.

Genomics. Genomics is the science of deciphering the sequence structure, the variation and the function of DNA in totality. More important than merely discovering and describing all genes of an organism is to describe the functions of the genes and the interactions between them. So-called functional genomics will help to discover the functionality of all genes, their functional diversity and the interactions between them. Functional genomics is expected to accelerate genetic improvement, the discovery of traits and to help solve intractable problems in crop production.

Recent progress in the mapping of the entire rice genome sequence, with the complete sequence expected to be delivered in 2004, represents a first important step towards understanding the overall architecture of the crop and provides valuable information for other techniques such as MAS or genetic transformation. But this would not yet include a full description of the biological functions of the various DNA sequences and their interactions, which would be a much more important step towards improved varieties. Many more years are likely to pass before all functions of all rice genes will be fully understood.

Genetically modified organisms. Current trends and applications. The first GMOs became commercially available in the mid-1990s. Since then, their importance has grown at an astounding pace. The number of GM varieties and species has increased rapidly and the area sown to GM crops has multiplied (some illustrative examples are given in Table 11.3). But the adoption across countries has been very uneven, with almost the entire expansion taking place in developed countries. Similarly, despite the rising variety of GM products available, commercial success has been concentrated on a few varieties or traits, notably herbicide-tolerant (Ht) maize and soybeans as well as Bacillus thuringiensis (Bt) cotton and maize. In 2001, Ht soybeans accounted for 63 percent of the area under transgenic crops, followed by Bt maize with a share of 19 percent (ISAAA, 2001).

Insect resistant traits. “Pest-protected” varieties were among the first GM crops to be developed, for the purpose of reducing production costs for farmers. Insect-resistant GMOs have been promoted both as a way of killing certain pests and of reducing the application of conventional synthetic insecticides. For more than 50 years, formulations of the toxin-producing bacteria Bacillus thuringiensis (Bt) have been applied by spraying, in the same way as conventional agricultural insecticides, to kill leaf-feeding insects. Studies on the safety of Bt for humans have not revealed any adverse effects on health.

In the late 1980s, scientists began to transfer the genes that produce the insect-killing toxins in bacteria Bacillus thuringiensis into crop plants. The intention was to ensure that all cells in these GMOs produced the toxin. Although no efforts were made to increase the growth rates or yield potential of the GM crops with these innovations, farmers have welcomed Bt crops because of the promise of better insect control and reduced costs. However, in the United States, the impact of Bt GMOs on crop yields and the number of conventional insecticide applications have varied widely by location and year. This is partly because of differences between the intended potential impact of the GM crops on target pests and their actual field performance. Some of these differences were a result of the uneven distribution of the toxin within the plants as they grew, some resulted from variations in target and non-target pest populations, and others were the result of toxins accumulating in plant-feeding insect pests, causing mortality of predators and parasites that ate those pests.

Herbicide-tolerant traits. The insertion of a herbicide-tolerant gene into a plant enables farmers to spray wide-spectrum herbicides on their fields killing all plants but GM ones. For that reason, the new GM seeds opened new markets for themselves and for herbicides. In fact, these crops contain a slightly modified growth-regulating enzyme that is immune to the effects of the active ingredient and allows it to be applied directly on the crops and kill all plants not possessing this gene.

Virus resistance. Virus-resistant genes have been introduced in tobacco, potatoes and tomatoes. The insertion of a resistance gene against potato leaf roll virus protects the potatoes from a virus usually transmitted through aphids. For that reason, it is expected that there will be a significant decrease in the amount of insecticide used. The introduction of a virus resistance gene in tobacco may offer similar benefits

Stacked traits. The so-called stacked traits embody a combination of properties introduced through GM technologies. The most important applications at present are combinations of herbicide tolerance (Ht) and insect resistance (Bt). A number of other combinations have already become commercially available, such as herbicide-resistant maize varieties with higher oil contents. In the future, the addition of more traits with specific value will be added with combinations of stacked traits that provide insect tolerance, herbicide resistance and various quality improvements such as high lysine and/or low phytate content, possibly even in conjunction with higher oil content.

GM farm animals and fish. While there was considerable growth in the development and commercialization of GM crops, GM livestock have largely remained outside commercial food production systems. At the experimental level, more than 50 different genes have been inserted into farm animals, but these efforts still require substantial skill and are not as routine as those for plants. Early research in the development of transgenic farm animals has also been accompanied by manifestations of perturbed physiology, including impaired reproductive performance. These experiences raised ethical problems of animal welfare.

So far, the prospect of foods from transgenic farm animals has not been well received by consumers. Surveys consistently show that the public accepts transgenic plants more easily than transgenic animals. Experimenting with and altering animals are less acceptable prospects and have broader implications. Various cultures and religions restrict or prohibit the consumption of certain foods derived from animals. The use of certain pharmaceutical products from transgenic animals, however, seems more acceptable to the public.

Highly successful research has been carried out on GM fish, but no GM fish have entered the market as yet (see Chapter 7). Most GM fish are aquaculture species that have received genes controlling the production of growth hormones, which raises the growth rate and yield of farmed fish. Ethical questions on the welfare and environmental impact of these GM fish have been raised, but it is also argued that GM fish share many attributes of conventionally selected alien fish species and genotypes, both of which are proven and accepted means of increasing production from the aquatic environment (FAO, 2000f).

How important are GMOs for agriculture? Current trends. The importance of GM crops has risen dramatically following the first endeavours of larger-scale commercialization in the mid-1990s. The six-year period from 1996 to 2001 witnessed a 30-fold increase in the global area grown with GM crops. With more than 52 million ha in the year 2001 (Table 11.4), the area planted with GM crops has reached a level that is twice the surface of the United Kingdom. At the same time, the number of countries growing GM crops has more than doubled.

This impressive growth notwithstanding, the annual increments in GM crop area have been levelling off both in absolute terms and in terms of percentage growth. This reflects to a large degree a saturation effect, as certain GM traits (soybeans) account already for a considerable share of the overall area. In addition, there was an actual decline in the area planted to GM canola, which is attributed to lower canola prices and the introduction of non-GM herbicide-tolerant varieties in Canada. But this slowdown also coincided with growing consumer concerns in developed countries that the new crops could jeopardize biosafety and pose a serious risk to human health. These fears have led to a growing pressure for legislation to label GM food, to increase the stringency of requirements for their approval and release or even to outright active resistance (Graff, Zilberman and Yarkin, 2000). These concerns were particularly forcefully voiced by consumers in developed countries. As a result, nearly all the additional area grown with GM crops came from developing countries, while area used for GM crops virtually stagnated in developed countries (Table 11.5). Canada’s GM area even declined, leaving the United States and Argentina as the principal growers of GM crops with an overall share of 91 percent.

Which GM crops are important? Soybeans, maize, canola and cotton represent almost 100 percent of area grown with GM crops globally in 2001 (Figure 11.5). Ht soybeans alone account for 58 percent of all GM crops. Ht soybeans are not only the most important transgenic crop but, after Bt cotton, also the most rapidly growing one. The rapid market penetration of these first GM crops is impressive, particularly when compared with the introduction of similar technologies, such as hybridized varieties of maize and sorghum. In 2001, GM soybeans accounted for 63 percent of all area under GM crops. GM varieties of maize, cotton and canola accounted for 19, 13 and 5 percent, respectively.

Figure 11.5 GMO crops by country

Source: ISAAA (2001).

What is in the pipeline? Input-oriented technologies. The next important improvement is likely to result from a further market penetration of so-called stacked traits, combining the benefits from two or more genetic modifications. The first stacked traits of cotton and maize (Bt/Ht cotton and maize) have already been released, offering both herbicide tolerance and insect resistance. In parallel, herbicide tolerance and insect resistance are planned to be extended to other varieties, notably sugar beet, rice, potatoes and wheat, while new releases of virus-resistant varieties are expected for fruit, vegetables and wheat. Fungus-resistant crops are in the pipeline for fruit, vegetables, potatoes and wheat. In addition, efforts are being made to create new traits with greater tolerance to drought, moisture, soil acidity or extreme temperatures. Chinese researchers claim to have developed salt-tolerant varieties of rice, which could help mitigate water scarcity and allow land lost to salinization problems to be recovered. The potential to cultivate marginal land appears to be particularly interesting for poorer farmers who are often more dependent on these environments. However, the ability of poor farmers to pay for these new technologies may be much more limited. This suggests that both speed of research and speed of introduction in the field are likely to be less impressive than for the first generation of GM crops.

Output-oriented technologies. A shift in focus is expected with the transition from the first to the second generation of GM crops. The new generation of GM crops is expected to offer higher output and better quality of the produce. Many of these new traits have already been developed but have not yet been released on the market. They include a great variety of different crops, notably soybeans with higher and better protein content or crops with modified oils, fats and starches to improve processing and digestibility, such as high-stearate canola, low phytate or low phytic acid maize. Another promising application is cotton with built-in colours that would spare the need for chemical dyes.

First efforts have been made to develop crops that produce nutraceuticals or “functional foods”: medicines or food supplements directly within the plants. As these applications can provide immunity to disease or improve the health characteristics of traditional foods, they could become of critical importance for improving the nutritional status of the poor.

In the pipeline are also a number of non-food applications of GM technology. These include speciality oils (e.g. jet engine lubricants), biodegradable thermoplastics, hormones, “plantibodies” (e.g. human antibodies for treatment of infectious and auto-immune disease), vaccines or pharmaceuticals (e.g. anticancer drugs such as taxol) (Thomashow, 1999). Non-food applications that have already reached practical importance include a transgenic variety of Cynara cardunculus thistle, which is grown in Spain for electricity generation and GM poplars grown in France for paper production which demand less energy and produce less waste during processing.

The success of the second generation of GM crops will ultimately depend on their profitability at the farm level and their acceptance by consumers. Unlike the first-generation products, quality-focused products such as functional foods provide a higher perceived benefit for the consumer. This may increase the risk that consumers are willing to assume, suggesting a high market potential for the second generation of GM crops.

Specific products in the pipeline

Soybeans

• High oleic soybeans contain less saturated fat than that in conventional soybean oil. The oil produced is more stable and requires no hydrogenation for use in frying or spraying. For that reason, this variety has a “health” image.
• Soybeans with improved nutritional traits for animal feeding contain higher levels of two amino-acids (lysine and methionine) which will reduce the need for higher-cost protein meals in the preparation of feed mixes.
• High-sucrose soybeans have better taste and greater digestibility.

Rapeseed and canola

• High-lauric variety produces an oil containing 40 percent of lauric acid for chemical and cosmetic purposes.
• High-stearate variety produces oil high in stearic acid, solid at room temperature without hydrogenation. This oil could be used for baking, margarine and confectionery foods that cannot use liquid oils.

Maize

• Several researches, both conventional and biotechnological, aim to produce value-enhanced maize that will offer improved nutritional traits for livestock. Since grain is fed primarily as a source of energy, many of the new value-enhanced varieties aim to increase the content or availability of energy. But some new varieties will also include more protein and better amino-acid balances, which would reduce the need to buy supplemental feed ingredients.

Cotton

• Coloured cotton is already available on a niche market basis. This trait would reduce the need for chemical dyes. Fibre quality improvement, such as polyester-type traits, would make sturdier fabrics.
• Chinese researchers are breeding a new strain of cotton that includes rabbit keratin. Fibres of this cotton are longer and more resilient and they have an increased ability to maintain warmth. Research is also being carried out to develop wrinkle-resistant cotton and even fire-retardant qualities.

Prospects for the nearer term. Given the enormous speed of progress in generating and adopting new biotechnology applications, any longer-term outlook is necessarily speculative. Somewhat greater confidence, however, can be attached to forecasts of possible developments over the nearer term. The following short-term developments are discernible.

First, adoption rates for GM crops are likely to increase in developing countries. With the rapid adoption of Bt cotton in China, GM crops have made an important inroad into a potentially important market. China’s GM potential rests not only on the sheer size of its agriculture but also on the particular importance for China of soybeans, maize, and tobacco – crops in which GM traits have been introduced successfully elsewhere. Moreover, an approval of the respective GM traits in China may have important knock-on effects in other developing countries. The significant catch-up potential in some developing countries, however, masks limiting factors and constraints that prevail elsewhere. To the extent that the new GM products favour capital-intensive and labour-extensive environments, the incentives to adopt these technologies are limited in other developing countries.

Second, growth of the area under traditional GM crops such as Ht soybeans and Bt maize is likely to slow down. This is in part a reflection of the impressive growth in the past, which limits the remaining potential. GM soybeans, for instance, already account for two-thirds of soybean area worldwide and for an even larger share of developed countries’ soybean area. An expansion of GM soybeans must therefore come from an overall growth in soybean area rather than from a shift out of non-GM soybean production. Growth may also be curbed because of food safety and environmental concerns that have received particular attention in Europe.

Third, there is a considerable growth potential for new GM applications in developed countries. Examples include GM fish varieties or GM crops for renewable energy. Other possibilities are GM-based nutraceuticals or GM applications for health and cosmetic applications. As these new applications are likely to produce much wider benefits than just cheaper food and feedstuffs, consumers in developed countries are also more likely to accept greater risks and thus to adopt these non-traditional applications at faster rates.

Prospects for the medium and longer term. Substantial progress has been made over the last five years, both in terms of theoretical advances and practical importance of biotechnology. These advances over such a short timespan make it impossible to identify specific products that are likely to dominate developments in biotechnology over the next 30 years. It is, however, easier to identify the overall parameters that are likely to affect future trends.

The overall direction of research and development is likely to be determined by economic incentives. Developments in prices of production factors and products are critical in the context of a 30-year outlook. These, in turn, will be crucially affected by future changes in the relative abundance/scarcity of production factors, notably land, capital and labour. In developed countries, costs of labour may increase relative to land, which would favour the further development of labour-saving technologies. In developing countries, by contrast, factor proportions may change in the opposite direction with increasingly abundant labour and increasingly scarce land. This would favour labour-intensive and land-saving technologies. The critical question in this regard is whether and to what extent the private sector will cater for diverging needs and to what extent investments from the public sector are needed to reconcile these needs.

Many of the currently available technologies have catered for land-intensive and labour-extensive environments. This is particularly noticeable for GM crops, where productivity gains are based on savings in input needs (labour, capital) even when output (yields) is stagnant or declining. This was one of the main factors that contributed to the high adoption rates for GM crops in developed countries. If today’s relative factor proportions are a guide to the future, the incentives to adopt these new technologies in developing countries are likely to be subdued. Moreover, private investors have little incentive to provide proprietary technologies where the chances of recouping investments in research and development are small. This suggests that the public sector will have to play a significant role in providing the technologies to cater for the specific needs of developing countries.

11.4.2 Why agricultural biotechnology matters to developing countries

The principal benefits. Productivity gains. Biotechnology has the potential to increase crop and livestock yields. The first generation of GM crops was largely input-oriented and provided the same or only marginally increased yield potential. The fact that some GM crops rendered higher yields in practice largely resulted from the effect that “built-in” inputs such as pesticides have reduced output losses that are typically caused by inappropriate or inadequate input applications. Moreover, the fact that GM technology embodies this expertise directly into the seeds is particularly important for environments where sophisticated production techniques are difficult to implement or where farmers do not command the management skills to apply inputs at the right time, sequence and amount. This suggests a much larger potential for GM crops (stacked traits of Bt/Ht cotton or Bt/Ht maize) in developing countries even for the first generation. The second generation of GM crops is expected to raise both the volume of output and the quality of the produce. These technologies are currently being tested but only a few traits are available in practice.

More, cheaper and better food. A second factor arises from the prospects for lower prices for better food. Higher productivity lowers production costs and will ultimately result in lower food prices. While this is not in itself a guarantor of improved food security or reduced poverty, more and better food at lower prices is particularly important for poor consumers. They would particularly benefit where GM products offer less expensive and nutrient-enriched food staples, which account for a large share of their food expenditure.

A higher capacity to feed a more populous world. The capacity that GM crops offer to produce more and better food is even more important when the future food needs of growing populations are considered. Much of the incremental food production in the future has to come from higher yields, yet the potential to raise actual yields through more traditional agronomic improvements such as earlier ploughing, scotch carts, higher fertilizer and pesticide applications is declining. A slowdown in yield growth has already been observed in some high-intensity systems in Asia, where the gap between yields attained by farmers and the economic maximum yield has narrowed noticeably.

The potential to save and improve resources or recoup marginalized land. A fourth factor is the potential to save resources or recoup marginalized land. Empirical studies suggest that the poor are cultivating the most marginal agronomic environments, and that they are more often dependent on these marginal growing environments than other groups.³ These marginal production environments are often characterized by drought or moisture stress, extreme temperatures, soil salinity or acidity. The potential to grow food in such environments is therefore doubly important in the fight against hunger and poverty: the potential to produce food where food is needed most helps ease the food problems of the poor directly. Moreover, to recoup land that was lost through environmental stress (e.g. soil salinity or acidity) could help contain further encroachments on areas with high environmental value or high sensitivity.

More and better non-food products. GM crops could also offer a more attractive way to produce non-food products. Plants with higher energy conversion and storage capacities could be bred and make a more meaningful contribution to alternative energy use in the future. If successfully implemented on a large scale, this could boost agriculture’s role as a carbon sink. Transgenic plants and animals could significantly extend the possibilities of various areas of technology and overcome some of the traditional constraints that medical research and applications face today.

The principal concerns. Notwithstanding the potentially large benefits of GM technology for developing countries, there are growing concerns that these new technologies are associated with significant costs, risks and problems.

Market concentration in the seed industry. Some concerns have arisen out of the significant market concentration in the seed industry. In 1998, 60 percent of the world market for seeds was controlled by just 35 companies. One company alone controlled over 80 percent of the market for GM cotton, 33 percent of the market for GM soybeans and 15 percent of the GM maize market (Then, 2000). This growing horizontal concentration is accompanied by an increasing vertical concentration between seed producers and agrochemical companies whereby larger agrochemical companies have been absorbing the few large seed companies that resulted from the horizontal consolidation process within the seed industry. The trend towards larger and more integrated operations (the so-called life science companies) was largely driven by the chemical industry. Chemical firms were looking for partners in the seed industry to protect the value of their intellectual property rights (IPR) in patented herbicides (Just and Hueth, 1999). The consolidation process between the agrochemical and the seed industry is currently being extended to a third stage, as the life science companies broaden their reach through strategic alliances with major trading companies such as Cargill or ADM. While this concentration process has offered new possibilities to reap scale effects and to overcome barriers in creating and commercializing GM products, it has also given rise to concerns that these non-competitive market structures may impose significant social and private costs (Phillips and Stovin, 2000). These are only now being considered.

Intellectual Property Rights (IPR). The impact of the application of IPR, the mechanisms for their enforcement and the excludability that is associated with them is another source of concern. In general, the excludability is of critical importance in encouraging private research in all sectors. Without it, innovators would not be able to recoup their investments, private research would languish, productivity gains would slow down and social welfare would suffer. Recognition of the importance of IPR has brought about a strengthening of legal protection for biotechnology processes and products and spurred on significant private investment in biotechnology. But the strengthening of IPR has also given rise to concern. First, the scope of intellectual property protection may be too wide, thereby choking off spillovers, follow-on innovations and diffusion. Second, IPR afford private companies the possibility of protecting the alteration of a single gene derived from freely accessible germplasm that has been generated by farmers and public research efforts over centuries. Developing countries in particular believe that they should be compensated for their contributions to existing genetic resources. The International Treaty on Plant Genetic Resources for Food and Agriculture (PGRFA), adopted in November 2001, addresses these concerns (Box 11.5). It could assume a pivotal role in facilitating access to plant genetic resources in the future and in safeguarding traditional indigenous contributions to the breeding process (farmers’ rights).

Biosecurity. A third area of public concern revolves around the risk that biotechnology applications in food and agriculture pose to human health and the environment. Consumers in all countries would like assurances that GM products reaching the market have been adequately tested, and that these products are being monitored to ensure safety and to identify problems as soon as they emerge. Because of the complexity of food products, research on the safety of GM foods is thought to be more difficult than carrying out studies on components such as pesticides, pharmaceuticals, industrial chemicals and food additives. Through the Codex Alimentarius Commission and other fora, countries discuss standards for GMOs and ways to ensure their safety. One approach, which is being used in assessing the risks of GMOs, derives from the concept of substantial equivalence.⁴ If the GMO-derived food is judged to be substantially equivalent to its traditional counterpart, then it is considered as safe as its conventional counterpart. If it is not, further tests are conducted.

Critics claim that only 1 percent of public research funds has been allocated to assess the risks associated with the introduction of GM technologies. It is suggested that the experience with traditional counterparts cannot be applied to products based on GM technology, as the substantial equivalence approach implies, and that the new technologies require a new risk assessment approach. Underestimating or ignoring the risks means that external costs associated with the technology are not fully accounted for and that the welfare gains of the new technology may be overstated. The recent accidental use for human food consumption of GM maize that contains a potentially allergen protein has reinforced such concerns.

Genes can end up in unexpected places. The artificially inserted genes might be passed on to other members of the same species, and perhaps to other species. Antibiotic-resistance genes are often inserted into GMOs as markers so that researchers can tell whether gene transfer has succeeded or not. These genes may be transferred to bacteria within the human body with yet unclear impacts. While this technique is now being replaced, other problems may remain. There is even a possibility that the gene for herbicide resistance may transfer to weeds, with potentially disastrous impacts for agriculture and food security.

Genes can mutate. It is still unclear what impact the artificially inserted gene has on the stability of the genome. There are claims that it may cause more unexpected mutations. While mutations could be neither new nor necessarily bad, GMOs may cause unexpected and undesirable instability.

“Sleeper” genes could accidentally be switched on. Organisms can contain genes that are not activated except under certain conditions, for example under the influence of pathogens or as a result of certain weather conditions. The “promoter” gene that is used to insert the new gene could activate “sleeper” genes, potentially in inappropriate circumstances.

Allergens can be transferred. Genes that cause allergies could be transferred into another species. The problem is twofold: it extends the range of potentially allergen products and creates uncertainty as to what products are potentially allergens. For example, an allergenic Brazil-nut gene was transferred into a transgenic soybean variety. It was found in testing, and the soybean was not released.

Sterility could be transferred. There is the theoretical risk that a dominant gene from a GURT plant could be passed on through cross-pollination to non-GURT plants, thereby reducing their fertility rate. However, this risk would be relatively small and, even if it were to happen, the inherited dominant non-germination gene would anyway be self-eliminating.

Controls over GM releases are inadequate. In 2000, a maize variety intended only for animal feed was accidentally used in products for human consumption. There is no evidence that this variety was dangerous to humans, but it could have been.

Animal welfare is at risk. There is evidence of abnormal physiology in some transgenic animals. Some effects on animals are unpredictable and could range from benign to distressful to dangerous.

Unintended effects on the resource base. Unusual traits in a plant could have unintended effects on the farming system. For example, a wheat variety capable of extracting more nutrients from the soil may exhaust the soil. Plants bred for land that has been made saline by unsuitable irrigation may enable a farmer to use even more brackish water, destroying the land completely.

Loss of biodiversity. GM plants could compete with traditional farmers’ varieties, causing loss of crops that have been bred for millennia to cope with local stresses. For example, the existence of traditional potato varieties in Latin America permitted a recovery from the catastrophic potato blight in Ireland in the 1840s. Today, traits from farmers’ varieties and wild relatives are often used to improve climate tolerance and disease resistance. GM crop varieties might also cross, and thus compete, with wild relatives of crops such as wheat and barley. This is especially risky in the developing world, where wild relatives may be found growing close to farmed crops.

11.4.3 Who benefits and who bears risks and costs?

The principal problem: disjoint risks and benefits.As with any new technology, there are winners and losers associated with the use of GMOs and biotechnology. At country level, the costs and benefits accrue to different stakeholders and cause concerns about, or even the outright rejection of, the new technology. Addressing and reconciling these problems are part of the policy response of the respective country. A second, less common source of concern emerges when risks and benefits accrue to stakeholders in different countries. This either requires international policy coordination or leaves externalities unaddressed. The introduction of GM technology is associated with both dimensions of the problem, i.e. there are disjoint risk and benefits within and across countries. An analysis of these disjoint costs and benefits may help identify appropriate policies. It may also provide insights as to what directions the new technologies will take.

Rich versus poor countries. The risks and benefits associated with an innovation are the principal determinants for the degree of adoption by a country. The willingness to assume risks may therefore be disconnected from the extent and possibility of capturing the benefits of GM technologies. The disconnection of risks and benefits affects numerous stakeholders: consumers versus producers, developed versus developing countries and private companies versus public research institutions. For example, the benefits of GM maize – and thus the willingness of rich societies to assume the associated risks – may be too small to pursue the technology. The benefits of the same technology for poor societies may be large, but their ability to pay for it is too small to develop it. If left unaddressed, such disjoint interests can result in a situation that neglects the interests of the poor. Bridging these gaps calls for appropriate policy action and for international policy coordination.

High-value versus low-value goods. The willingness of a society to bear the risks of a new technology is positively related to the benefits drawn and expected from it. The benefit from cheaper food staples such as rice, maize or soybeans is likely to be small for rich consumers in the north and high for poor consumers in the south. Rich consumers are therefore unlikely to assume the same risks as poor consumers and, if the benefits are sufficiently small, they are rational to reject the new technology altogether.

Consumers in the developed countries accept the higher risks for functional food, medical applications or cosmetics as the (perceived) benefits from these applications exceed the risks that they carry. But given the small benefits from less expensive food staples for the same consumers, staples such as rice, wheat or coarse grains are likely to be most affected by a decline in research expenditure. Even more so are tropical staples such as some roots and tubers. But such food staples are of critical importance for consumers in developing countries, particularly for the rural poor.

What does it mean for the direction of research? The different perception and importance of risks and benefits in developed and developing countries could reduce the speed of progress and change the direction of GM research and development. In the north, which has the funds to afford research, consumers perceive the risks associated with these technologies to be high and their benefits to be small. Yet the greatest need and the greatest benefits are in the south, where the ability to pay for the development of the technologies is small. The allocation of risks and benefits could mean that the north may reduce investments in these technologies, as the consumers in the north are unwilling to accept their products, while the consumers in the south, in need of the products, are unable to pay for the technology.

Where the south can afford to develop and import the technology, this may adversely affect trade in the final products. GM-based food and fibre may be faced with growing non-tariff barriers, which reflect the lower willingness to accept risks in the north or an added form of protecting domestic agriculture. A smaller market volume for GM crops in turn may adversely affect the profitability of developing the new technology. A lack of market access may render research and development non-viable for the south.

What policies could help to reconcile risks and benefits? Appropriate policy action could help reconcile some of the conflicting interests of the various stakeholders.

Addressing the concerns of consumers in developed countries: more transparency. Part of the risk that makes consumers in developed countries reluctant to accept GM products is not actual risk but perceived risk. In part, this risk perception reflects a lack of transparency and calls for measures that help to maximize transparency for the consumer. Appropriate labelling is an important step towards higher transparency and thus towards lower risk. It would also help facilitate trade since labels help products to comply with international standards. However, excessive labelling requirements may result in “regulatory capture”, which in itself is detrimental to trade. Moreover, labelling of GM food requires segregation of GM from non-GM products and can generate substantial additional costs and incentives for fraud. The higher costs of segregation are a particular problem for poorer countries that often lack the necessary regulatory capacity.

The second part of the risk that consumers in developed countries face is real. It is a reflection of insufficient testing and premature releases of GM crops for field applications. This calls for better risk assessment procedures and commensurate rules and regulations to minimize the risks associated with applying GM technology.

Addressing the concerns of developing countries: lowering seed costs. A crucial factor that limits access to the new technologies in developing countries is the high cost of GM seeds. To the extent that high seed costs reflect the monopoly rents from a lack of competition in the life science industry, antitrust measures could provide a remedy. In addition, national and international research institutions themselves could charge royalties for the germplasm that has traditionally been provided free of charge to the private sector. These payments could then be used to promote targeted GM research for developing countries. This, of course, may reduce the overall incentive for the private sector to invest in crop research and could result in an overall loss in productivity.

Private-public partnerships in GM research. The public sector could explore a number of routes to collaborate with the private sector to target developing countries’ specific needs, such as poverty reduction, public goods provision and the capture of spillover effects. National and international research institutions could define specific breeding tasks, or the development of field-proven varieties with certain characteristics and put them out for competitive tender. Companies could use their patented germplasm as an input into work under such tenders, but the final product would have to be made available free of IPR charges or genetic technical use restrictions (Lipton, 1999).

11.5 Directions for agricultural research

Agricultural research has been crucial in meeting the challenge of increasing food production faster than population growth over the past 50 years. A main characteristic of research efforts was the focus on increasing productivity through a set of technologies in what has come to be known as the green revolution. The impressive global achievements mask considerable regional differences. Asia received most attention while sub-Saharan Africa was largely bypassed, as were, within many countries, the remote and poorest communities that did not have the resources, physical and financial, to capitalize on the potential of the new high-input demanding technologies (Conway, 1997). Also, the impacts of the new technologies on the environment (see Chapter 12) were largely ignored in the early years of the green revolution.

What has emerged most strongly in various reviews of the green revolution is that development of technologies by themselves is not enough. Relevant and sustainable technology innovation must be planned, developed, tested and delivered within a broad-based agricultural and rural development framework. Nor can productivity be the sole criterion to guide technology development; the potential implications of technologies for agro-ecological stability and for sustainability and equitability must be fully addressed at the “drawing board” stage. This has fundamental implications for the planning of future agricultural research strategies.

To meet the food security needs of an expanding global population in the decades ahead and to reduce poverty, there is a need to maintain and increase significantly agricultural productivity on land at present available across the developing world and at the same time to conserve the natural resource base. This will require (i) increasing productivity of the most important food crops both on the more fertile soils and on marginal lands; (ii) exploring possibilities for limiting the use of chemical inputs and substituting these inputs with biologically based inputs; (iii) more precise use of soil, water and nutrients in optimized integrated management systems; and (iv) increasing production efficiency and disease tolerance in livestock.

These challenges call for a comprehensive and complex research agenda that must integrate current advances in the molecular sciences, biotechnology and plant and pest ecology with a more fundamental understanding of plant and animal production in the context of optimizing soil, water and nutrient-use efficiencies and synergies. Effective exploitation of advances in information and communication technology will be necessary not only to facilitate the necessary interactions across this broad spectrum of scientific disciplines but also to document and integrate traditional wisdom and knowledge in the planning of the research agenda and to disseminate the research results more widely. This agenda calls for a three-dimensional research paradigm that integrates scientific investigation across genetics and biotechnology, ecology and natural resources and not least socio-economics to keep in focus the development environment that characterizes the livelihoods and food security of the poor.

The first dimension (genetics and biotechnology) has been discussed in the preceding section. The following section will focus on the other two dimensions.

Ecology and natural resources management research. A comprehensive understanding of the ecology of all life forms within the farming system is a prerequisite to the development of sustainable agriculture in the context of sound knowledge-driven use and management of the natural resources with which the farming system interacts. Recent advances in ecology have been described alongside molecular genetics as the second great revolution in modern biology. The past decade or so has seen the use of mathematical modelling, the articulation of comprehensive hypotheses and advances in experimental design in support of more precise laboratory and field experiments. These trends have transformed the study of ecology into a rapidly developing science (Begon, Harper and Townsend, 1990), which should lead to a better understanding of the complex dynamics that are at work within agricultural systems. Increasingly, natural resource management calls for closer collaboration between ecologists and agricultural scientists whether they are addressing technical (biotic or abiotic), social or economic dimensions of agricultural development. By definition, agro-ecosystem research embraces the “ecological and socio-economic system, comprising domesticated plants and/or animals and the people who husband them, intended for the purpose of producing food, fibre or other agricultural products” (Conway, 1987).

Within the biophysical boundaries of the ecosystem, ecological research has much to contribute at three levels: (i) at the level of the plant, its pests and predators; (ii) at the level of the plant and its competition from weeds; and (iii) at the level of the plant rooting system and the roles of beneficial and competing micro-organisms in the capture and utilization of soil nutrients. Research at the plant-pest-predator level is opening up new insights in pest control, in one measure through the genetic development of pest-resistant plant varieties and, at the other end, in the refinement of IPM.

At the level of plant-weed competition, the geneticist’s approach has been to develop herbicide-resistant crops, while the ecology-based approach has been to develop crop rotations and intercropping and high-density systems that minimize weed damage. Both approaches require still further research, particularly with greater focus on marginal lands and food crops that are mostly grown by the poor. More fundamentally, a better understanding of plant genetics and of plants’ relation to other competing plants may offer new insights to stable cropping systems. This would provide a more informed basis on which to design cropping systems and rotations that can sustain higher and more stable yields.

At the plant rooting levels, there is a need for much more fundamental research at the physical, biochemical/physiological and genetic bases of plant-micro-organism interactions and symbiosis (Cocking, 2001). Nutrient utilization and biological nitrogen fixation lie at the heart of this research, which can have enormous benefits for low-input agriculture and the poor. The benefits to sustainable agriculture and the environment are obvious in the reduction of dependence on chemically produced nitrogenous fertilizer and associated greenhouse gas emissions. Future ecosystem research must also address soil nutrient availability and utilization more comprehensively within alternative cropping systems in the context of NT/CA, integrated cropping and crop-livestock systems.

At the community and watershed levels, agro-ecology, ecosystem and natural resources management research must address the biophysical and socio-economic dimensions of resources use and their potential enhancement and/or depletion. In this context, CGIAR has recently articulated a comprehensive agenda of integrated natural resources management research that embraces the more important topics that need to be addressed (CGIAR, 2000):

• Water: model system flows (river-basin level) allocated across multiple users, with particular attention to onsite and offsite effects, develop recharge balance models for aquifers at risk of excessive drawdown.
• Forests: characterize the complexity of forest systems and the range of stakeholders who interact with them, and develop strategies to influence the global policy agenda.
• Fisheries: identify the types of farming systems and agro-ecologies into which integrated aquaculture-agriculture can be sustainably incorporated.
• Livestock: develop databases, models and methods for analysing livestock-based systems to help identify priorities for research and development interventions.
• Soils: develop soil erosion models for various multifunctional land use systems, and nutrient balance and flow models.
• Carbon stocks: document and model alternatives for above and below ground carbon stocks and relate changes in those stocks to global climate change impact.
• Pest and disease incidence: describe, define and track key insect pests using GIS and develop models that relate incidence to agroclimatic conditions.
• Biodiversity: project alternative scenarios of functional biodiversity under different land management systems.

Research priorities for the poor. Defining research priorities (and the development of stakeholder-specific research agendas) is becoming a complex and increasingly sophisticated process. It demands the interactive interpretation of information flows between the scientist and the end-user of research outputs. The criteria for making strategic choices among alternative research programmes vary depending on the stakeholder. In private-sector-funded research, the ultimate criterion is profit in one form or another. In academic research the goal is often loosely defined around scientific advancement and knowledge. And in national and international research institutions the broad objective is primarily the production of information and products for the “public”, usually termed “public goods”.

Establishing research priorities that specifically address the needs of the poor or multidimensional goals such as food security, sustainability, conservation of biodiversity and natural resources, becomes increasingly difficult as the distance (socio-economic as well as spatial) between the research planners and the target beneficiaries widens. In essence, bridging this gap is the challenge of the new research agenda, not only at the research planning stage but also at the interpretation and field-testing stage of the technologies and other research outputs.

In determining research priorities, it is also vitally important to understand how new technologies may influence the lives and livelihoods of the poor. New technologies can lead to increased productivity on the family farm, resulting in increased food consumption and family incomes. Such technologies can result in lower food prices, benefiting a wide range of the urban and rural poor. Growth in agricultural output generally leads to increased employment opportunities for the landless and the poor in both the rural and urban non-farm economy. Perhaps not emphasized adequately in research programmes is that new technologies can lead to the production of food crops rich in specific micronutrients that are often deficient in the diets of the poor.

Research to underpin a new technology revolution with greater focus on the poor must put special emphasis on those crop varieties and livestock breeds that are specifically adapted to local ecosystems and that were largely ignored throughout the green revolution. These include crops such as cassava and the minor root crops, bananas, groundnuts, millet, some oilcrops, sorghum and sweet potatoes. Indigenous breeds of cattle, sheep, goats, pigs and poultry and locally adapted fish species must also receive much greater priority. A particular focus in the new research agenda should be on plant tolerance to drought, salinity and low soil fertility as nearly half of the world’s poor live in dryland regions with fragile soils and irregular rainfall (Lipton, 1999).

Research modalities and dialogue. Research that addresses only one component in the development chain, for example crop yield potential, will not result in an equitable or sustainable increase in food production. New research efforts should address a minimum of four critical questions at key salient points along the research continuum, from the conception/planning stage to the stage of application of the outputs by the targeted beneficiary. The key questions are: (i) whether the technology will lead to higher productivity across all farms, soil types and regions; (ii) how the technology will affect the seasonal stability of production; (iii) how the technology will impinge on the sustainability of the targeted farming system; and (iv) what are the sectors that will benefit most (or lose out) as a result of the widespread adoption of the technology. It is comparatively easy to tailor these questions to specific research programmes depending on the nature of the research to be undertaken, but it is much more difficult to arrive at well-supported answers, in particular at the research planning stage. This research challenge needs scientists from a range of disciplines and from different agencies, both public and private, to engage in close collaboration, not only among themselves but also with the intended beneficiaries – the farmers – either directly or through the extension services.

Effective dialogue among all scientists and extension workers in this research development continuum also calls for a new information-sharing mechanism that embraces transparent interactive dialogue and easily accessible information. Modern information and communication technology can provide the vehicle for this information sharing and dialogue, opening up the possibility of a global knowledge system through which the sharing of global knowledge on all emerging technologies relating to food and agriculture can be effectively realized (Alberts, 1999). This in turn will lead to the strengthening of the research process at all stages.

National government and international donor support for research has declined significantly over the past decade, despite compelling evidence on very high rates of return to investment in agricultural research and in particular in genetic improvement programmes for crops and livestock. This is particularly worrying at a time when there is a widely shared consensus on the absolute need and importance of strengthening agricultural research. While more and more funds go into biotechnology research, the other areas mentioned above are trailing behind. This is especially true for research focusing on marginal areas and crops. The private sector can and must contribute more than just funding. As outlined above, its expertise, technologies and products are essential to the development and growth of tropical agriculture based on rapidly advancing biotechnologies and genetically engineered products. It is argued by some that incentives (e.g. in the form of tax concessions) should be offered to induce private sector participation. It is also argued that collaborative partnerships with private sector companies or foundations in well-articulated and mutually beneficial agreements can mobilize the required cooperation and make significant contributions to agricultural research in the developing countries.

To conclude, the scientific community bears a responsibility to address ethical concerns. On the one hand, it must ensure that the technologies and products of research do not adversely affect food safety or risk damage to the environment. In this context, timely and transparent communication of relevant research findings and their interpretation (risk analysis) to all pertinent audiences including the general public is required. On the other hand, scientists, together with public servants, politicians and private sector leaders, bear a more profound humanitarian responsibility to ensure that all people can realize their most fundamental right – the right to food.

¹ This definition of LMO is taken from the Cartagena Protocol on Biosafety, Article 3(g). In Article 3(i), “modern biotechnology" is defined as “the application of: a. In vitro nucleic acid techniques, including recombinant deoxyribonucleic acid (DNA) and direct injection of nucleic acid into cells or organelles, or b. Fusion of cells beyond the taxonomic family, that overcome natural physiological reproductive or recombination barriers and that are not techniques used in traditional breeding and selection”.
² Of particular importance are the so-called quantitative trait loci that represent economically significant expressions of traits such as higher yields, improved quality or better resistance to diseases or various forms of abiotic stress.
³ In fact, these environments are available to the poor because only the poor are willing to accept the low factor returns (wages) for the inputs they can provide (low-skill labour).
⁴ Substantial equivalence acknowledges that the goal of the assessment is not to establish absolute safety but to consider whether the GM food is as safe as its traditional counterpart, where such a counterpart exists. It is generally agreed that such an assessment requires an integrated and stepwise, case-by-case approach. Factors taken into account when comparing a GM food with its conventional counterpart include: (i) identity, source and composition; (ii) effects of processing and cooking; (iii) the transformation process, the DNA itself and protein expression products of the introduced DNA, and effects on function; and (iv) potential toxicity, potential allergenicity and possible secondary effects; potential intake and dietary impact of the introduction of the GM food.