by Ian Scoones, STEPS Centre co-director

A decade ago there was much hope – and even more hype – about the potentials of GM crops. GM crops were going to feed the world, solving issues of poverty and development, seemingly at a stroke. Technologies for dealing with drought or nutrient deficits were, it was claimed, in the pipeline. Even the pest-resistant Bt technologies that were already available offered the opportunity of reducing pesticide use and improving farmers’ incomes. GM crops were going to be of particular help, it was argued, to poorer farmers in the developing world, ushering in a new ‘gene revolution’ to succeed the ‘green revolution’ of previous decades.

A decade ago, of course, there were also those who predicted disaster and calamity – and still do. GM crops were going to result in all sorts of environmental and health catastrophes, and provide the basis for global domination of agriculture by a few large corporations. Just as the pro-GM lobby could be accused of excessive and unfounded hype, anti-GM campaigners often generated doomsday scenarios based on limited evidence.

But there were others offering a more balanced view too. The1999 Nuffield Council report argued:

“As GM crop research is organised at present, the following worst case scenario is all too likely: slow progress in those GM crops that enable poor countries to be self-sufficient in food; advances directed at crop quality or management rather than at drought tolerance or yield enhancement; emphasis on innovations that save labour-costs (for example, herbicide tolerance), rather than those which create productive employment; major yield-enhancing progress in developed countries to produce, or substitute for, GM crops now imported in conventional (non-GM) form from poor countries (Nuffield Council on Bioethics 1999: 66–67)”

So what has happened since? Have the hopes of the technology optimists been realised? Have the doomsday scenarios come to pass? In reality, of course, neither of the above: a more complex story has emerged. In some circumstances, some farmers have benefited from GM crop technologies, but not always; while others have had bad experiences or been by-passed altogether. Overall, the picture is decidedly mixed. Nevertheless, despite the accumulated experience and the collection of evidence from detailed impact studies, some wild claims are still being made and false expectations generated.

Certainly, GM crops have expanded rapidly in some parts of the world. The annual assessments by Clive James of the ISAAA (1), a non-profit organisation supported by the biotechnology industry as well as aid agencies and governments, portray a dramatic picture of GM crops sweeping the globe. However, although GM crops were planted in 25 countries in 2008, only eight of those countries planted more than a million hectares. In fact, about 98m hectares out of a global GM crop area of 125m hectares was grown in just three countries, overwhelmingly by large-scale farmers: the United States (62.5m hectares), Argentina (21m hectares) and Brazil (15m hectares). The GM crops commercialised to date are primarily insect-resistant Bt varieties of maize and cotton, and herbicide-tolerant varieties of soy.

A recent book by respected political scientist, Robert Paarlberg, called ‘Starved for Science: How biotechnology is being kept out of Africa’ (2) once again makes the case for GM crops as the core of a solution to agricultural development. The book argues that Africa, in particular, has been denied the benefits of GM crops because of the influence of European anti-GM campaigns, and that this amounts to a scandalous denial of vital, life-saving technology to poor people in Africa. Through a series of arguments, he claims the potentials of science-based agriculture, especially GM crops, and claims that inappropriate, precautionary biosafety regulation in particular is a major hurdle to the successful and widespread adoption of poverty-reducing technologies.

This book is not a fringe publication, produced by some unknown press. It is published by the university press at Harvard, is endorsed by both a Nobel prize winner (Norman Borlaug) and a former US president (Jimmy Carter), and has acknowledgements to a string of well-known academic researchers. Its arguments are deemed respectable by many commentators, and have been picked up by policymakers and lobby groups. For example, the indefatigable GM enthusiast Lord Taverne has argued recently that GM crops must be central to tackling high world food prices (3). Former UK Chief Scientist David King has also argued that GM crops must be behind a global response to challenges of climate change, allowing farmers to adapt to harsher conditions (4). In a similar vein industry lobby groups have recently argued that ‘tide is turning’ in favour of GM crops – in Europe and beyond – particularly as a result of the political recognition of the major challenges presented by the global food crisis. New efforts are afoot to make the case for a GM solution, especially in the vast potential market that is the developing world, through, for example, the industry-based Alliance for Abundant Food and Energy(5).

But are all these arguments based on the type of ‘sound science’ that Paarlberg, Taverne, King and others are apparently so passionate about? Have GM crops had a positive impact on development – helping small-scale and subsistence farmers to improve farm productivity and climb out of poverty? Paarlberg’s (and others’) arguments look less compelling when more detailed scrutiny is applied. A good starting point for such an exploration is the iconic cases of ‘GM success’, which are repeatedly cited in publications such as ‘Starved for Science’. Key cases include the rapid expansion of GM cotton among ‘smallholders’ in India, China and South Africa. However, as Dominic Glover’s new STEPS Working Paper shows (published June 10), the story is complex and the impacts have been very mixed. Extravagant claims were made in the early years, but these have become more muted over time as the reality has hit home. As Glover’s paper shows, economic returns are highly variable, dependent on a range of factors. GM crops only perform well in good varieties, and it is these that have the largest effect. The start-up costs and technology fees sometimes put the GM seeds out of reach of poorer farmers, and those who are the major adopters tend to be relatively richer and with more land and other assets. And finally – and perhaps most critically – it is the institutional and policy environment that makes all the difference. Without support, credit and sustained backing, the new technologies very often fail.

These are not surprising conclusions. Indeed, they were predicted by many a decade ago, including by the highly prescient comments of the 1999 Nuffield report. Technologies do not exist in a vacuum; their success is not pre-determined. However, Florence Wambugu, the well-known Kenyan scientist argued that, with GM, “the technology is in the seed” (6). But technologies are always linked to social, economic and political contexts, and the assumption that nothing else matters beyond the technical fix is of course deeply flawed.

Much research has demonstrated this over the past decade, and the material contained in the STEPS Centre biotechnology archive (which launches on this website on June 10) illustrates this clearly. Study after study – from India, China, Africa, Latin America or Europe – shows how GM technologies must be examined in their socio-technical context, and that policies and politics matter. This applies just as much to the questions of intellectual property and access arrangements as it does to the regulation of safety, environmental impacts or effects on health. Contrary to Paarlberg’s claims, a precautionary approach makes much sense in the face of deep uncertainties, public concerns and contested politics. The elaboration of robust, transparent and accountable institutions, emerging from policy deliberations involving diverse stakeholders, is a critical feature of any new socio-technical landscape. Such institutions take time to develop and, just like the technologies themselves, cannot be simply transplanted from continent to continent. Just because the mid-west of the United States is making use of a particular technology in a particular institutional and policy context does not mean that an African country should adopt the entire package without proper discussion. That would be unsound science, and poor development practice – every bit as much as a blanket rejection of GM technologies would be.

The ‘pro-’ versus ‘anti-’ fundamentalisms of the GM debate have become so entrenched that discussions too often get stuck in an unhelpful impasse (7). But equally unhelpful is the deployment of weak argument and inadequate evidence in the name of so-called scientific debate. No-one is served by such rhetorical ploys, least of all ‘the poor’ who are co-opted, unknowingly, into the fruitless to and fro arguments. So how do we get beyond this impasse? What are the real lessons to be learned from the past decade? What are the realistic prospects of a pro-poor gene revolution? Five interlinked points are important:

1. GM is not the only biotech way. GM technologies are not the only biotechnological solution on offer. Marker-assisted selection and other genomic techniques, for example, offer important opportunities for enhancing conventional breeding through biotechnology. In many ways, in fact, transgenic-based genetic engineering looks rather old-fashioned – a blunt instrument for dealing with complex problems. Re-engineering for complex traits, such as drought tolerance or nutrient use efficiency – which involve multiple, interacting genes – is well-known to be a tough call. Despite the long-promised ‘pipeline’ technologies in these areas, results to date have been limited. A more holistic focus on varieties, genetics and environmental factors seems more likely to pay dividends. Whatever the genetic wizardry, the underlying background material and the wider environmental conditions really do matter. Solving complex problems in Africa through importing technologies ‘in the seed’ from elsewhere are doomed to failure. Investment in long-term, local, context-specific breeding and crop development programmes must not be forgotten in the rush to the attractive, well-funded, high-tech solution.
2. Abandon technology fundamentalisms. Technologies are never isolated from social, economic, political contexts. Agri-cultures – the many different ways farmers manage plants, their soils and the wider environment really matter. (8) Agriculture, as Paul Richards has described, is a performance, and one in which technology is only a bit-part, if vitally important, player (9) ‘Silver bullet’ perspectives can be grossly misleading: agriculture is a complex socio-technical system, where innovation must be managed across technological, institutional and policy frontiers. Beyond the need for new science – whether biotechnological or other kinds – such an approach requires the ability to assess technologies in context, and to create appropriate institutional and policy frameworks for their successful use – particularly in respect of ‘pro-poor’ goals. Today, the agricultural research establishment – in both national and international systems – is woefully lacking in this regard. With the rush to the technical fix, these other ‘soft skills’ are often forgotten.
3. Don’t expect too much from the corporate sector. Major biotechnology companies have little commercial interest in Africa. Why should they? They are accountable to their shareholders, not the rural poor of the global south. Their business models are focused on widespread adoption of standardised technologies on large farms, with high unit profits based on highly capitalised operations. Under such conditions, a focus on a select number of labour-saving, capital-intensive crop technologies, linked to an agribusiness centred on vertical integration and monopoly control makes much sense. It is thus no surprise that the prediction made by the Nuffield Council report a decade ago has come to pass: most of the GM crop technologies now available or in development are primarily for the large-scale commercial farmers of the rich world. The relative success of Bt cotton (and possibly maize) in developing countries perhaps gives a deceptive picture. The Bt trait is a simple, generic technology which could easily be backcrossed into local varieties to some effect (although not as much as is frequently claimed). A key challenge, however, is that proprietary technologies are critical to the agribusiness model. Some claim that patenting (or other forms of proprietary control) is essential for innovation and continued business viability. But such a model is rarely pro-poor. Only through publicly-based, open-source arrangements will poor farmers’ needs get a look in. Thus, it is not sensible to expect too much ‘pro-poor technology’ to emerge from the corporate sector, even if some spin-offs may be on offer through intellectual property-sharing agreements or licensing arrangements. The basic products, because of the mode of their design and delivery, are unlikely to offer much of a solution, unless they are substantially adapted to new settings.
4. Put farmers first. Who owns technologies and controls their development – whether GM or not – clearly matters a lot. In the end, it will be public research efforts that will make the difference to poverty reduction. As agricultural technology moves upstream, away from field settings to high-tech molecular biology labs and bio-secure units, access by farmers is even more restricted. As a result, a top-down, transfer-of-technology, pipeline model in public research systems is reinforced. But we know that this does not produce the best technologies. Involving farmers in priority-setting and upstream technology design – even at the level of high-tech laboratory research – is vital.(10) The best science results from asking the right questions about the right problems, and it is the potential users of technology who understand their own problems best.
5. Regulatory frameworks and wider policy issues are critical. Some see burdensome regulation as the big obstacle to the spread of GM technologies. They argue, for example, that Africa is being held back by precautionary regulations, and that in some way the ‘freedom to innovate’ (11) is being restricted. But such arguments miss fundamental points. In the face of deep uncertainty – indeed ignorance in many situations – a precautionary stance makes for sound policy. Building an appropriate regulatory infrastructure, which has wide buy-in, generates trust in regulatory institutions and is feasible to implement, is a major challenge in most developing-world settings. Every context is different, and requires a particular regulatory and policy response. Regulations need not be burdensome; indeed they should be facilitative, but they must be based on real, local evidence, and so have firm scientific underpinnings. Uniform, universalised regulations, pushed to the lowest common denominator by powerful interests, will not command respect. Equally important, though, they cannot be justified in the name of science. Making technologies work of the poor is inevitably a ‘slow race’, (12) but one which will result in more robust and effective governance mechanisms which ultimately benefit everyone.