EXCERPT: It is widely assumed that the rules of the World Trade Organization (WTO) establish a standard global framework for regulating trade in genetically modified (GM) crops and foods. But countries may have a much wider scope for making choices about GM foods than is often thought - and these choices can influence the science that will need to be done to inform those decisions.
------

Evaluating the acceptability of GM crops: the scope for autonomy in developing countries
By Erik Millstone
http://www.scidev.net/dossiers/index.cfm?fuseaction=policybrief&dossier=6&policy=55
SciDev, January 2005

SUMMARY

How much real scope do developing countries have in the way they trade or use genetically modified foods or crops? Despite assumptions about World Trade Organization (WTO) rules, member nations ”” including developing countries ”” have a significant degree of autonomy in choosing which GMOs to accept, and which to reject. That is because in general, they can decide for themselves whether scientific evidence on the safety of GM foods that is considered sufficient by other WTO member states will also suffice within their own borders.

INTRODUCTION

It is widely assumed that the rules of the World Trade Organization (WTO) establish a standard global framework for regulating trade in genetically modified (GM) crops and foods. But countries may have a much wider scope for making choices about GM foods than is often thought - and these choices can influence the science that will need to be done to inform those decisions.

This level of autonomy is important because it gives countries the opportunity to devise rules tailored to their particular priorities and conditions. At the same time, the choices they make will affect which risks are evaluated and controlled, as well as how much time, money, equipment, personnel and facilities is needed to implement and administer procedures for appraising risks and taking the subsequent decisions.

WHAT THE RULES SAY

The terms of the WTO treaty, and the wording of its ancillary agreements (such as the Agreement on Technical Barriers to Trade and the Sanitary and Phytosanitary Agreement ), suggest that they jointly provide the means for creating a single uniform global regime governing trade in agricultural commodities, including GM crops and foods. The SPS agreement, for example, stipulates that measures that are more restrictive than those recommended by the Codex Alimentarius Commission (the international body that sets global food safety standards) can only be imposed if they are based on sufficient scientific evidence and a scientific risk assessment.

Such conditions have been widely interpreted as meaning that once the risks of introducing a new GM crop or food have been subjected to a scientific risk assessment, and subsequently deemed acceptably safe by one WTO member state, that product should be acceptable to all member states. This is based on the assumption that science is universal, objective and reliable.

The rules do not stipulate, however, precisely which risks must (or can) be assessed, nor how much evidence is sufficient to justify a particular measure. They also fail to define the benchmarks against which the evidence should be judged. Indeed, both the WTO dispute panel and its Appellate Body have acknowledged that each WTO member state has the right to judge these issues autonomously.

What WTO rules require is that an individual member state, while able to choose from a wide range of testing requirements, must interpret these and any subsequent findings consistently, and apply its chosen criteria in ways that do not discriminate between domestic and imported products. As long as those conditions are satisfied, the WTO rules allow member states to exercise considerable autonomy.

OPTIONS IN RISK ASSESSMENT

There are a number of risk assessment methods and techniques, which vary in sophistication and capabilities. This range of methods seems likely to extend even further as a result of rapid scientific and technological changes now underway.

Governments of industrialised countries have applied a broad range of testing requirements to different categories of food and chemical products. These have varied according to the category of products, the country concerned, and the period of application. Most such countries, for example, require far more rigorous tests on pharmaceutical products than they do on food colourings or bleaching agents.

When GM foods were first introduced in the United States and United Kingdom in the mid-1990s, those countries assumed them to be acceptably safe as long as relatively coarse chemical analyses did not indicate substantial differences in levels of proteins, vitamins and minerals compared to those of their non-GM equivalents.

In the EU in the late 1990s, the approach was modified to require more refined and precise chemical analyses of a wider range of constituents which resulted in higher costs. The United Kingdom has also required, and conducted, farm-scale trials of the cultivation of GM crops.

Several influential groups of commentators have argued that, as molecular biological and biochemistry advance, it will become both possible and desirable to conduct more sophisticated analyses of GM foods, and to compare the results with those for non-GM equivalents, although this will incur higher costs (e.g. Kok and Kuiper, 2003, Spök et al, 2002). Those proposals have been endorsed by various European expert advisory committees, and may be introduced relatively soon (EFSA, 2004).

As shown in the table above, these evolving approaches to GM foods and crops stand in sharp contrast to those that currently apply to food additives and pesticides. These are subjected not just to chemical analyses, but also to testing in bacteria and cell cultures and live laboratory animals. While this approach provides much more data, it is also more costly. The testing for pharmaceutical products is the most comprehensive and expensive: in addition to all the rounds of testing to which food additives and pesticides are subjected, pharmaceutical drugs must also undergo clinical trials on humans.

EVOLVING APPROACHES TO RISK ASSESSMENT AND REGULATION

The United States ”” and until 2001 the EU as well ”” has chosen to make judgements about the acceptability of GM foods based on somewhat scant scientific evidence. They endeavoured to legitimise that approach by invoking the concept of "substantial equivalence". This concept was elaborated by the Organization for Economic Cooperation and Development (OECD) in a 1993 report, Safety Evaluation of Foods Derived by Modern Biotechnology: "
or foods and food components from organisms developed by the application of modern biotechnology, the most practical approach to the determination of safety is to consider whether they are substantially equivalent to analogous conventional food product(s), if such exist" (OECD, 1993, p 11).

Furthermore, the OECD proposed that: "if a new food or food component is found to be substantially equivalent to an existing food or food component, it can be treated in the same manner with respect to safety. No additional safety concerns would be expected" (OECD, 1993, p. 13).

When it was originally introduced, the concept of substantial equivalence was intended to be a relatively simple and cheap way of assessing the possible risks that GM foods might pose for human health. Instead of conducting lengthy and complex toxicological studies on laboratory animals or clinical trials with human volunteers, the idea was simply to analyse GM foodstuffs and their non-GM antecedents chemically to seek any significant chemical differences. The GM foods would be further tested if, but only if, any such differences emerged.

This approach took for granted that all non-GM foodstuffs that were already in use were unproblematically safe, despite a lack of baseline data on their composition.

The EU's 1997 Novel Food Regulation included a simplified procedure stipulating that when a novel food was considered to be substantially equivalent to an existing food, a company was required only to provide a justification for the claim of substantial equivalence, rather than a formal risk assessment. That procedure was used to authorise several GM foods for sale in the EU (Levidow & Murphy, 2002).

The interpretation and use of the concept of substantial equivalence by regulatory authorities was widely criticised, however (by the present author among others), for being not just simple, but simplistic (Millstone et al, 1999; Levidow and Murphy, 2002). Partly as a result of this criticism the EU almost entirely abandoned the concept of substantial equivalence in 2001, the only exception being the assessment of derivatives from GM foods that contained no detectable levels of genetic material or proteins, such as highly refined corn and soy oils.

The European Commission’s (EC) proposals in 2001 for a revised Novel Food Regulation explained that:

"
this proposal does not include a notification (simplified) procedure
for genetically modified foods which are substantially equivalent to existing foods. The use of this regulatory short-cut for so-called 'substantially equivalent' GM foods has been very controversial in the Community in recent years
and there is consensus at the international level
that whilst substantial equivalence is a key step in the safety assessment process of genetically modified foods, it is not a safety assessment in itself," (CEC, 2001).

Subsequently, the EC indicated that chemical analyses would no longer be considered a sufficient basis for making a decision about the safety of GM foods, but that such analyses should provide a starting point for a more sophisticated approach (EU, 2003, Recital 6). This change of policy illustrates the general point made in this briefing: namely that under the rules of the WTO, countries can decide for themselves whether scientific evidence on the safety of GM foods considered sufficient by other WTO member states will suffice within their own borders.

Although the EU has abandoned its previous approach, it has not yet decided on a replacement. As we saw above, a wide range of alternatives are available. Recent indications suggest that, in addition to chemical analyses, European regulators will also require at least some data from live animal studies, although it is not yet clear whether those studies will be short-term, longer-term, or lifetime feeding studies, or even multi-generational lifetime feeding studies.

One approach suggested recently by Kok and Kuiper (2003) is to replace the relatively crude comparative chemical analyses with a more sophisticated and finely grained analytical screen, while retaining the comparisons with non-GM antecedents. In the United States, meanwhile, the authorities continue to consider data from chemical analyses to be sufficient in most cases.

THE EVOLUTION OF SCIENTIFIC IDEAS AND TECHNOLOGICAL TOOLS

Until the mid-1990s, genetics and molecular biology were dominated by a set of assumptions generally presumed to be unproblematic. One was that there was a one-to-one correspondence between genes and proteins, in other words that each gene directs the expression of a single protein. Indeed a new scientific discipline called 'genomics' evolved to study the functions of individual genes in an organism’s genome.

In the late 1990s, some startling new findings threw doubt on this paradigm. Compelling evidence emerged indicating that, across all species, the number of proteins considerably exceeded the number of genes. That anomaly suggests that the expression of proteins may be controlled by combinations of genes, rather than just by individual genes. If this turns out to be true, then to understand the consequences of eating some foodstuffs, it may not be sufficient just to know the sequence of amino acids that characterise individual genes; instead, genomics would need to be supplemented with what has been called 'proteomics', the study of proteins and their interactions.

In practice, any effects from consuming GM foodstuffs may also depend not only on the genes and the proteins, but also on the ways those proteins are metabolised in the body ”” that is, on what is now being termed 'metabolomics'. All these developments have led some to suggest that a less crude approach to assessing the risks of a GM food could be provided by not only analysing for levels of proteins, vitamins, minerals and anti-nutrients, but by using a more elaborate analytic approach involving genomics, proteomics and metabolomics.

According to some interpretations, such an approach might require toxicological and immunological assessments, but only if proteomic and metabolomic novelties emerge, or if such novelties seemed potentially problematic, for example if they were entirely unfamiliar or structurally similar to toxic or allergenic compounds.

Some commentators have argued that the testing requirements for GM foods should be even more exacting than this. They point out that GM foods may be consumed in large quantities on a daily basis, even though pharmaceuticals ”” which are subjected to far more stringent tests ”” are taken in comparatively tiny amounts. Thus testing on a par with that for pharmaceuticals might be appropriate. No country has yet adopted this option. But there is nothing in the rules of the WTO, or for that matter of the EU, that would prevent individual countries, or groups of countries, from doing so.

ASSESSING THE ENVIRONMENTAL RISKS FROM CULTIVATING GM CROPS

Most of the above discussion has focused on testing for food safety. A more complex but essentially analogous set of issues and considerations applies to assessing the environmental consequences of cultivating GM crops.

In principle, one might assume that the consequences of eating GM food would be similar in California, Caracas and Calcutta, at least if all these populations are equally healthy and well-nourished. But the consequences of cultivating a GM crop depend heavily on context: the ecology and farming practices of Andhra Pradesh are very different from those of, say, Zambia. Thus it is reasonable to anticipate that many equally valid scientific assessments of the environmental impacts of cultivating GM crops will emerge.

Furthermore, experience from Europe indicates that different countries may choose different benchmarks against which to evaluate the environmental impacts of GM crops. For example, whereas the UK government’s risk assessors have chosen to compare the environmental risks of cultivating GM crops with those of 'conventional' high-technology farming, their Austrian counterparts have chosen the risks of organic farming, which is widely practised in the country, as their comparative benchmark.

KEY ISSUES AND IMPLICATIONS FOR DEVELOPING COUNTRIES

Although WTO rules require that standards set for products and processes that are internationally traded should be consistent and non-discriminatory, each WTO member state has considerable autonomy when deciding what to accept and what to reject (Millstone & van Zwanenberg, 2003). This applies, at least in principle, to developing countries just as much as any others.

However, the range of choices open to developing countries is limited in part by their capacity to establish, finance and administer a risk assessment and regulatory system. As we have seen, different approaches to testing differ hugely in cost. The costs of conducting the tests typically fall on the companies requesting consent to market their products. But the administration that reviews and interprets the resulting data will also incur costs.

The more complex, thorough and exacting the tests, and the more extensive the resulting data, the greater the cost of assembling the expertise required to analyse and interpret those data. As a result, developing countries may need help with amassing enough trained personnel for this task. But there may also be considerable scope for collaboration with other countries in this position.

That, at least, would be the desirable scenario. In reality, the United States and the EU are economically and politically powerful bodies that exercise considerable influence over their smaller and weaker trading partners. Developing countries may find that they face strong pressure to design and introduce risk assessment regimes that conform with one or the other of those systems.

But they should remember that there is nothing in either the science, or the WTO's rules, that compels developing countries to adopt the standards or approaches to risk assessment and/or evaluation embraced by industrialised countries.

REFERENCES:

{1} CEC (2001) Commission of the European Communities, Proposal for a regulation on genetically modified food and feed (COM 2001 - 425 final), http://europa.eu.int/comm/food/fs/biotech/biotech08_en.pdf, also at http://www.foodstandards.gov.uk/consultations/gmfoodfeed.htm www.foodstandards.gov.uk/consultations/gmfoodfeed.htm

{2} EFSA (2004), Draft Guidance Document for the Risk Assessment of Genetically Modified Plants and Derived Food and Feed, April 2004, prepared by the Scientific Panel on Genetically Modified Organisms of the European Food Safety Authority (available at http://www.efsa.eu.int/consultation/372/consultation_guidance_gmo_01_en1.pdf )

{3} EU (2003) regulation (EC) No 1829/2003 of the European Parliament and of the Council on genetically modified food and feed http://europa.eu.int/eur-lex/pri/en/oj/dat/2003/l_268/l_26820031018en00010023.pdf

{4} Kok, E.J. and Kuiper, H. A. (2003) ‘Comparative safety assessment for biotech crops’, Trends in Biotechnology, volume 21, issue 10, October 2003, pp. 439-444

{5} Levidow, L. and Murphy, J. (2002) The Decline of Substantial Equivalence: How Civil Society Demoted a Risky Concept. Paper for Conference on Science and Citizenship in a Global Context: Challenges from New Technologies, at the Institute of Development Studies, University of Sussex, 12 13 December (available at www.ids.ac.uk/ids/env/ biotechpaperrev1Peter1.pdf )

{6} Millstone, E., Brunner, E. & Mayer, S. (1999) 'Beyond "substantial equivalence"', Nature, volume 401, number 7, pp. 525-526

{7} Millstone, E. & van Zwanenberg, P. (2003) 'Food and agricultural biotechnology policy: How much autonomy can developing countries exercise?' Development Policy Review, 2003, volume 21, number 5-6, pp. 655-667. OECD (1993) Safety Evaluation of Foods Derived by Modern Biotechnology: Concepts and Principles, OECD, Paris. Both available from http://www.wto.org/english/docs_e/legal_e/legal_e.htm#tbt

{8} Spök, A. Hofer, H. Valenta, R. Kienzl-Plochberger, K. Lehner, P, Gaugitsch H. (2002) Toxikologie und Allergologie Von GVO-Produkten, Austrian Federal Environment Agency, Vienna. Available from http://www.umweltbundesamt.at/publikationsdetail.html?&pub_id=1066

*Erik Millstone leads the Environment and Energy Programme at Science and Technology Policy Research (SPRU), the University of Sussex, UK.