Genetic Modification Revisited: Speech to the NZ Institute of Agricultural and Horticultural Science Inc forum, 1 July 2009
NZIAHS convention at Lincoln University, Canterbury
Thank you for the opportunity to reassess our position on genetic modification in the light of the latest evidence.
I want to start today with basic principles; with the questions, "What is the purpose of Agricultural and Horticultural Science? And how does genetic modification fit with that purpose?"
I can see two related purposes for the science of food production.
First, from an international perspective, to produce enough quality food for the world’s growing population in a way that is reliable, affordable, protects the ecosystem services that sustain it and protects human health.
Second, from a New Zealand point of view, to enable our country to prosper through producing surplus food to trade and using our natural advantages of climate, soils, water, and knowledge to do so sustainably.
Does anyone have a problem with that statement of goals?
Right. In that case we need to look at the best ways to ensure food quantity, nutritional quality, safety; to sustain soil, water, biodiversity and crop genetic diversity; and to develop resilience to factors like accelerated climate change.
Second, we need to avoid creating market resistance or undercutting food production in less developed countries; and consider the role of intellectual property; the regulatory system; and opportunity costs.
On the way through I will assess the extent to which genetic modification can help us meet those goals.
Co-incidentally, a major international study investigating very similar questions was initiated at the 2002 World Summit on Sustainable Development, backed by the UN and the World Bank. Over three years some 400 authors systematically reviewed the science of food production and addressed the very questions I have just asked, so I will draw on its conclusions. It was peer reviewed twice and reported in 2008 as the International Assessment of Agricultural Knowledge, Science and Technology for Development.
That report asked whether we would have an agroecosystem in 2050 that could feed the world then and not exhaust itself in the process. What it found was that we do not have that system now, and we will not ever if we continue along present approaches both in biotechnology and policy. Summaries of the report are available here.
We have a growing population wanting to eat better despite limited cultivable land, water, oil and fertiliser. So, predictably, one promise of GM is to increase yields.
But no-one is starving today because there is a lack of food produced. In fact, even during the world’s most famous famines, those countries were exporting food! (Vandermeer and Perfecto, 2007)
People are malnourished because they lack the income to buy food or the land on which to grow it; they are starved of markets for their surplus production, starved by explicit cash subsidies to farmers in some countries and subtle subsidies such as artificially low fossil fuel costs and attendant mechanisation in others.
It is commonly held that intensive use of water, pesticides and fertilisers are needed to raise yields above what can be achieved with agroecological, including organic, agriculture, and that genetic modification can raise them further.
A number of studies have examined those claims. To quote from what is probably the largest meta-analysis on conventional and agroecological agriculture ever conducted:
"Model estimates indicate that organic methods could produce enough food on a global per capita basis to sustain the current human population and potentially an even larger population, without increasing the agricultural land base." (Badgley et al., 2007)
A recent empirical study completed by the UN found that in Kenya, conversion from conventional to organic production increased yields up to 179% and more than doubled yields in most of the 24 countries across Africa in which the study was conducted. (UNEP/UNCTAD 2008)
There are many organic techniques to increase yields by preserving soil moisture, improving fertility through rotations and natural fertilisers and controlling pests without poisons. A study in Africa by the International Water Management Institute and Earthscan found that in the same water stressed conditions, improving the yield of the low yield farms to 80% of the yield of high yield farms was enough to close the food gap. (Molden, 2007)
I have personally seen, some 20 years ago, a 60 acre plot in the McKenzie country, surrounded by bare earth and hieracium, support metre high vegetation and grazing by cattle without irrigation, just by growing the right plants in the right way.
The reputation organic methods have in some circles for poor yields is explained by the fact that comparisons are frequently made between established chemical based systems and land only recently converted to organics. (Badgley et al., 2007) It is typical of such conversions that yields initially drop and then return and surpass as the new methods bear fruit.
What, then, can GM crops add to this?
Almost all commercially released GM crops have been designed to increase resistance to either pests or herbicides. Any increase in yield would be a side effect resulting from easier management at massive scales, and not the purpose of the GM trait.
There is anecdotal evidence of both increased and lowered yields in GM crops in practice, and it is pretty obvious that it depends on environmental and management conditions and how the measurements are made. While the evidence has tended towards positive for cotton, it tends towards negative for soy beans and maize. A 2003 study published in Science found that:
"For insect resistant maize in the United States and herbicide tolerant soybeans in the US and Argentina average yield effects are negligible and in some cases even slightly negative." (Qaim & Zilberman, 2003)
The most recent and advanced study on cotton in the US state of Georgia has found lower profits and yields for farmers using GM varieties (Jost et al., 2008). These results would not justify the higher cost of seed and licensing, let alone potential market resistance.
The International Assessment found that yield gains were variable and unpredictable and declines too frequent to endorse claims of enhanced yields from commercial GM crops.
Food is not just bulk to fill the stomach, but nutrients to support health. The only GM crop ever designed to improve human nutrition the so-called golden rice is a technological solution for a policy problem. It raises some issues of principle that are typical of the GM debate vitamin A deficiency is normally addressed by eating green leafy vegetables which are easy to grow. Why engineer one grain to provide all nutritional requirements which it can’t do anyway when those considerable resources could have been devoted to arranging a more balanced diet in a locally resilient agroecosystem?
But there are practical issues too. Philanthropic agencies exploiting years of public sector research developed the rice, but not without stepping on the intellectual property claims of many different companies including the agrochemical giant Syngenta. In a symbolic gesture Syngenta donated its IP (but only in developing countries that would not otherwise recognise Syngenta’s claims, making it not seem so generous after all). But the interlocking IP ownership in the industry meant that just the gene transfer process required some 40 patented or proprietary processes or materials owned by over a dozen entities and the wider research needed 70 product or process patents held by 32 universities or corporations. This seriously slowed down the research and increased its costs. (Graff 2003; Spielman 2007)
GM, then, has not achieved anything much to improve nutrition. Moreover, it is unlikely to. . The technology is designed for, and encourages, large monocultures and intensive systems where soils often become depleted of micronutrients because fertiliser regimes are confined to NPK. It is certainly not designed for small mixed farms which use composts and a wide range of soil conditioners to maintain micronutrients. It has been estimated that more than half the micronutrients in intensively managed soils under large scale farming are not replaced by fertilisers (Zoebl, 2006) and what isn't there in the soil will not be there in the food.
Another aspect of food quality is safety, and it is here that arguments about GM have raged most fiercely.
It is often claimed that glyphosate or glufosinate tolerance enables farmers to avoid the use of more toxic herbicides in favour of less toxic ones. Yet because they are deliberately sprayed directly on the crop much greater quantities will be ingested. While they are less toxic than some herbicides the story is not that simple. In normal commercial formulations which also contain other ingredients they have been found to kill human umbilical, placental and embryonic cells. These toxic effects of commercial formulations have only recently come to light so we cannot rely on past studies to assure us of the safety of glyphosate herbicides, such as Roundup. They also have other effects on non-target organisms such as soil microbial communities:
"The toxicity exerted by glufosinate induced shifts in the microbial community structure with apparently long lasting significant effects. Changes in soil microbial populations can also affect soil functionality, thereby influencing nutrient turnover and the restoration process of the soil." (Pampulha et al., 2007)
If it is true that these herbicides are much more benign than their alternatives, then we should be very concerned that the evolution of multiply herbicide resistant weeds will deny us their use in future, leading to a return to the more toxic herbicides in non-GM crops.
Turning to Bt crops, we need to look at the effects of the various cry proteins in the food itself. Many people have been relaxed about possible health effects as the proteins are found naturally in soils with populations of Bacillus thuringiensis. Soil is ingested with food in various ways and inhaled as dust in quantities of tens to hundreds of mg. However to equal the dietary exposure from various commercial varieties of Bt corn a typical US consumer would have to eat between 14kg and 600 tonnes of soil. (reported from various sources in Heinemann, 2009 p153)
The claim is often made that GM foods are the most tested of any, and that no evidence of harm has ever been found. What is more, there is no obvious sign of human illness resulting from eating GM foods.
You don’t have to be an epidemiologist to know that science rarely finds what it doesn’t look for. Human ill health is rife everywhere. There are countless allergies and many metabolic disorders that are not well diagnosed but which undoubtedly have many causes. There is no way of knowing who has eaten GM foods and which ones, so no epidemiological work can be done. Nor could there ever be a matched control. On the question of whether GM food is making people sick, we simply have no idea, and few are trying to find out.
The next best proxy is animal feeding trials, of which there are alarmingly few, and most tests have been on the protein of interest, not the whole food. That assumes there is no possible risk from disruption of the genetic and physiological function of the plant, from the antibiotic markers that are used, or from multiple insertions of the target gene. It assumes the location of the gene is of no significance. If a protein is acutely toxic to humans such tests might discover that, but I take no comfort at all from claims of lack of harm when there have been so few tests of whole GM foods.
These risks and others, which have never been satisfactorily addressed in published peer reviewed studies, are explored in a new online, free-to-the-public resource for citizens, regulators, journalists and scientists called the Biosafety Assessment Tool (https://bat.genok.org/bat/). The five year international cooperation was led by Norway and the University of Canterbury.
Only recently have long term and reproductive animal tests been published using the whole food. I find it a terrible indictment on the industry, sufficient to destroy all credibility in their ethics, that they have pushed these crops at farmers and these foods on the market for over a decade without doing basic safety testing. It is also an indictment on the regulatory authorities who have allowed them to.
Pryme and Lembcke in 2003 could find only nine animal feeding studies of whole GM foods and of these, the industry studies found no health effects but the independent studies found significant effects that merited further study. Yet no further research had been done to confirm or refute them. There is always a problem funding safety tests that are not required by authorities and not in the interests of the owners of the IP. It can even lead to difficulties in obtaining accurately characterised samples of GMOs for independent testing.
Since then, and in fact just in the last two years we have four more studies which indicate all is not well.
The 2007 reanalysis by Seralini and others of Monsanto’s data on Bt corn MON 863 found toxic effects on the livers and kidneys of rats. Seralini’s work was supported by a local Environmental Science and Research (ESR) internal evaluation ignored by the New Zealand Food Safety Authority. (http://www.sustainabilitynz.org/news_item.asp?sID=190). It matters who funds the safety research.
Kilic and Akay last year in Turkey found granular degeneration of the livers of rats fed Bt corn over three generations.
Last July Austrian researchers studied the effects of a Bt corn combining MON 810 and NK603 on breeding mice and found effects on the kidneys and "time related negative reproductive effects of the GM maize". (Velimirov et al 2008)
Finally, just last November Italian scientists concluded from their study that:
"the consumption of MON810 maize ”¦induced alterations in intestinal and peripheral immune response of weaning and old mice." (Finamore et al, 2008)
This is not a picture of thorough testing of GM foods with no evidence of harm. It is not surprising that so many European countries have banned BT corn MON 810 Austria, France, Luxembourg, Greece, Hungary and most recently Germany.
Factors like yield, accessibility and affordability of seeds and technology, reliability, and resilience are sometimes put together in a measure of food security for the poor. FAO statistics, measuring two different levels of undernourishment in the US showed that food security has not improved in the decade of commercialisation of GM foods. In Argentina and Paraguay, where over 60% of arable land is in GM, food security has decreased since the mid-nineties when GM crops were first widely commercialised.
The International Assessment saw the sustainability of agricultural systems as a key challenge to food security. Its co-chair Dr. Hans Herren said:
"Agriculture is at the centre of the multiple looming crises of water, soil degradation, energy costs, biodiversity loss, climate change, population growth, dwindling natural resources and increasing inequities."
With regard to biodiversity, there have been too few long term, properly controlled studies of the effects of GM crops to make much comment. However, Bt corn was found in a 2007 study to increase mortality and reduce growth of ecologically important stream insects. (Rosi-Marshall, 2007)
Water is the key limiting factor in much of the world and even in parts of NZ, I recently visited farms in Hawkes Bay which has just come out of its third serious drought in 3 years. Many of those farms had no grass at all on their hills. One of the great promises of GM technology for the last 25 years has been to develop drought and salt tolerant plants, so I looked for evidence of how they were getting on. It seems that in the US alone more than 1,000 applications have been received to field test such plants but not one has ever been commercialised (Heinemann, p 64). The contribution of GM to overcoming adverse environmental conditions is, after 25 years, zero.
In the meantime, the World Bank has noted significant yield gains through disease resistance and drought tolerance from selective breeding despite too little funding in comparison to GM. New maize varieties and hybrids are yielding 20 percent more on average under drought conditions. Significant gains have also been made in breeding wheat for drought and heat-stressed environments, and rice that survives flooding. Such advances will be especially important in adapting to climate change (WorldBank, 2007)
If we get our priorities straight, we could do this better and faster.
Sustaining agricultural production into the long term future requires access to a wide range of germplasm both across and within species. Changing climatic conditions will require selective breeding to adapt. GM crops are not available for selective breeding and when they are grown in large monocultures they crowd out other varieties. GM patents are a major obstacle to seed saving and local adaptive breeding. Just five companies control more than 95% of gene transfer patents.
Most of you will know the story of Percy Schmeiser. He had selectively bred his canola to thrive in his local conditions and used RoundUp at the end of the season and to control spread of his plants. When his crop was inadvertently contaminated by RoundUp Ready canola he lost the use of glyphosate and all his saved seed and his breeding programme. This was a heavy enough loss without the additional crippling financial losses arising from prosecution and conviction of unauthorised growing of patented seed despite his being the victim of adventitious contamination.. GM is incompatible with seed saving and with the development of local varieties to suit local conditions, which is an important option in achieving sustainability.
So the record and the prospects of GM crops increasing food yields, security, safety or sustainability are rather poor. But let’s face it: that’s not the motivation of most people involved with them anyway. In New Zealand we do not grow crops to feed the hungry they can’t afford to pay enough for them. We grow to feed the discriminating, high end of the market that can pay good prices. In that case the first lesson is to grow what the market wants. Our markets are consumer led we should have learned that lesson from fat lambs in the sixties.
The EU and Japan are primary markets for our food exports and both have strong consumer resistance to GM foods. While their governments may allow tolerances of minor contamination at the border, the consumer and therefore those who supply them don’t. Even a barely measurable GM contaminant in a corn pizza topping led to its rejection in Japan. If such rejections become more frequent it will damage our brand overall.
While I would argue that our so-called “clean green” brand is largely undeserved, it is certainly an important marketing image and has been valued at around a billion dollars. The benefits of GM organisms of any kind would have to be overwhelming for us to even contemplate damaging that image. And the benefits are far from overwhelming.
Overseas experience is that contamination of non-GM crops is inevitable if GM crops are grown nearby. It isn’t just the pollen drift it’s the mix ups in the seed store or in the field, the contamination of harvesting machinery, and the brand contamination of it being known that we grow GM crops.
Back in the late nineties partnerships between apple and kiwifruit growers and research institutes were developing GM varieties in containment. A surge of consumer resistance led to firm statements from those industries that they would discontinue all such developments. Both industries have done well, especially their organic variants, and have never looked back.
Our regulatory system was set up with the intention that NZ would become a GM grower and it is often described as the strictest system in the world. Anecdotally, that seems to be true for the import and management of insignificant materials for lab work that pose no real risk, but sadly astray when it comes to outdoor use of GM.
None of the recommendations of the royal commission of enquiry into GM on measures to ensure non GM growers could continue to grow their crops without contamination have been implemented, eight years after the report. The approach of the authorities is that when GM crops are ready to be released a 1% or more allowable contamination standard will be set for other crops. It is simply unethical for one set of growers to damage the market opportunities of others in this way.
Monitoring and enforcement of conditions on field tests has been derisory. Every time there is a breach brassicas with large holes in the containment netting in the eighties; tamarillos with negligible containment in the nineties, so bad it was commented on by the royal commission; salmon in outdoor tanks with screens open enough for eggs to pass through, discharging into a natural river system; brassicas that are allowed to flower contrary to their consent conditions just this year, and no penalties when a member of the public discovered the breach each time we are told procedures have been thoroughly reviewed and tightened and each time it happens again. This is not a system that can be relied on to manage a release, even if there were benefits from such a release.
The consenting process is similarly slack. The High Court has overturned ERMA’s decision to grant consent for Agresearch’s application to put virtually any animal gene into virtually any mammal with strong comments that such a blanket permit was contrary to the basis of our law, which is a rigorous case by case analysis.
If there is money to be made from GM, it is not made by the growers, but by those who hold the IP. This will not, on the whole, be New Zealanders. The global seed market is controlled by some five transnational firms. The patents for GM products and processes are even more closely held by these same companies. I would guess that all the field tests that have occurred in New Zealand have required payment or licensing with promise of future payment, to a number of overseas patent holders. Where is the long term benefit to New Zealand in this?
Finally, science funding is limited. Money poured down the black hole of GM crops, as it has been in this country now for well over a decade, is money with a high opportunity cost. It is not available to better understand our soils and their microorganisms; to better understand adaptation strategies to climate change in our particular conditions; to better understand human nutrition and its relationship to soils and agricultural techniques; to better understand crop protection without pesticides and soil fertility without urea, to mention just a few. The focus on GM has diverted research organisations and students and teachers away from other kinds of science which might have yielded more benefit.
Using peer-reviewed scientific evidence as recent as 2008 and 2009, I have revisited my earlier conclusions that GM food crops will not help us to feed the world in the long term; that they will not improve food security; that they are not proven to be safe for human or animal consumption; that they offer no assistance in protecting ecosystem services or achieving sustainability; that they will not please our markets or lead to greater access; that any profits will be made mainly by overseas holders of the IP; and that our regulatory system is not up to the task of managing them. I have revisited these conclusions carefully, and I see no reason to change them.
Genetic science has given us wonderful diagnostic techniques, marker assisted breeding, and laboratory manufacture of medicines by microorganisms. There is more to be done in those fields. But the best way I can describe Genetic Engineering of crops and animals is as a clever and expensive technology looking desperately for a use a smart solution looking for a problem. I give the last word to the Assessment report:
"An increase and strengthening of agricultural knowledge, science and technology (AKST) towards agroecological sciences will contribute to addressing environmental issues while maintaining and increasing productivity." (Global Summary, Key Finding 7, page 10)
* Badgley, C., Moghtader, J., Quintero, E., Zakem, E., Chappell, M.J., Avilés-VÃ¡zquez, K., Samulon, A., and Perfecto, I. (2007). Organic agriculture and the global food supply. Ren Ag Food Sys 22, 86-108.
* Fernandez-Cornejo, J., and Caswell, M. (2006). The First Decade of Genetically Engineered Crops in the United States. In Economic Information Bulletin (U.S. Dept. of Agriculture, Economic Research Service), pp. 36.
* Finamore et al, (2008) Intestinal and Peripheral Immune Response to MON 810 Maize Ingestion in Weaning and Old Mice J. Agr. Food Chem. 56, 11533-11539.
* Graff, GD et al, (2003) The public-private structure of intellectual property ownership in agricultural biotechnology. Nat. Biotechnol. 21, 989-995
* Heinemann, JA, 2009 Hope not Hype TWN
* Jost, P., Shurley, D., Culpepper, S., Roberts, P., Nichols, R., Reeves, J., and Anthony, S. (2008). Economic comparison of transgenic and nontransgenic cotton production systems in Georgia. Agron J 100, 42 51.
* Kilic, A and Akay, MT (2008) A three-generation study with genetically modified Bt corn in rats: Biochemical and histopathological investigation. Food Chem. Toxicol. 46, 1164-1170.
* Molden, D., ed. (2007). Water for food. Water for life: A comprehensive assessment of water management in agriculture. (Colombo and London, Int. Wat. Mgmt. Inst. & Earthscan).
* Pampulha, M.E., Ferreira, M.A.S.S., and A. Oliveira, A. (2007). Effects of a phosphinothricin based herbicide on selected groups of soil microorganisms. Journal of Basic Microbiology 47, 325-331.
* Pryme, IF and Lembke, R (2003) In vivo studies of possible health consequences of genetically modified food and feed - with particular regard to ingredients consisting of genetically modified plant materials. Nut. Health 17, 1-8
* Qaim, M and Zilberman, D, (2003) Yield Effects of Genetically Modified Crops in Developing Countries. Science 299, 900-902
* Rosi-Marshall et al (2007) Toxins in transgenic crop by-products may affect headwater stream ecosystems. Proc. Natl. Acad. Sci. USA 104 16204-16208.
* Seralini, G-E, et al (2007) New analysis of a rat feeding study with a genetically modified maize reveals signs of hepatorenal toxicity. Arch. Environ. Contam. Toxicol. DOI: 10,1007/s00244-006-0149-5, 596-602.
* Spielman, DJ (2007) Pro-poor agricultural biotechnology: can the international research system deliver the goods? Food Policy 32, 189-204
* UNEP/UNCTAD (2008) Organic Agriculture and Food Security in Africa
* Vandermeer, J., and Perfecto, I. (2007). The agricultural matrix and a future paradigm for conservation. Con Biol 21, 274 277.
* Velimirov, et al (2008) Biological effects of transgenic maize NK603xMON810 fed in long term reproduction studies in mice. Bundesministerium fur Gesundheit, Familie und Jugend, Sektion IV
* WorldBank (2007). World Development Report 2008: Agriculture for Development (Washington, D.C., World Bank).
* Zoebl, D. (2006). Is water productivity a useful concept in agricultural water management? Agricultural Water Management 84, 265-273.