Agricultural Biotechnology: Overregulated and Underappreciated
The pursuit of an integrated action plan, including regulatory reform, will help the United States and the world reap enormous benefits that now are thwarted.
The application of recombinant DNA technology, or gene splicing, to agriculture and food production, once highly touted as having huge public health and commercial potential, has been paradoxically disappointing. Although the gains in scientific knowledge have been stunning, commercial returns from two decades of R&D have been meager. Although the cultivation of recombinant DNA-modified crops, first introduced in 1995, now exceeds 100 million acres, and such crops are grown by 7 million farmers in 18 countries, their total cultivation remains but a small fraction of what is possible. Moreover, fully 99 percent of the crops are grown in only six countries—the United States, Argentina, Canada, Brazil, China, and South Africa—and virtually all the worldwide acreage is devoted to only four commodity crops: soybeans, corn, cotton, and canola.
Attempts to expand “agbiotech” to additional crops, genetic traits, and countries have met resistance from the public, activists, and governments. The costs in time and money to negotiate regulatory hurdles make it uneconomical to apply molecular biotechnology to any but the most widely grown crops. Even in the best of circumstances—that is, where no bans or moratoriums are in place and products are able to reach the market—R&D costs are prohibitive. In the United States, for example, the costs of performing a field trial of a recombinant plant are 10 to 20 times that of the same trial with a virtually identical plant that was crafted with conventional techniques, and regulatory expenditures to commercialize a plant can run tens of millions dollars more than for a conventionally modified crop. In other words, regulation imposes a huge punitive tax on a superior technology.
Singled out for scrutiny
At the heart of the problem is the fact that during the past two decades, regulators in the United States and many other countries have created a series of rules specific for products made with recombinant DNA technology. Regulatory policy has consistently treated this technology as though it were inherently risky and in need of unique, intensive oversight and control. This has happened despite the fact that a broad scientific consensus holds that agbiotech is merely an extension, or refinement, of less precise and less predictable technologies that have long been used for similar purposes, and the products of which are generally exempt from case-by-case review. All of the grains, fruits, and vegetables grown commercially in North America, Europe, and elsewhere (with the exception of wild berries and wild mushrooms) come from plants that have been genetically improved by one technique or another. Many of these “classical” techniques for crop improvement, such as wide-cross hybridization and mutation breeding, entail gross and uncharacterized modifications of the genomes of established crop plants and commonly introduce entirely new genes, proteins, secondary metabolites, and other compounds into the food supply.
Nevertheless, regulations in the United States and abroad, which apply only to the products of gene splicing, have hugely inflated R&D costs and made it difficult to apply the technology to many classes of agricultural products, especially ones with low profit potential, such as noncommodity crops and varieties grown by subsistence farmers. This is unfortunate, because the introduced traits often increase productivity far beyond what is possible with classical methods of genetic modification. Furthermore, many of the recombinant traits that have been introduced commercially are beneficial to the environment. These traits include the ability to grow with lower amounts of agricultural chemicals, water, and fuel, and under conditions that promote the kind of no-till farming that inhibits soil erosion. Society as a whole would have been far better off if, instead of implementing regulation specific to the new biotechnology, governments had approached the products of gene splicing in the same way in which they regulate similar products—pharmaceuticals, pesticides, and new plant varieties—made with older, less precise, and less predictable techniques.
But activist groups whose members appear to fear technological progress and loathe big agribusiness companies have egged on regulators, who need little encouragement to expand their empires and budgets. The activists understand that overregulation advances their antibiotechnology agenda by making research, development, and commercialization prohibitively expensive and by raising the barriers to innovation.
Curiously, instead of steadfastly demanding scientifically sound, risk-based regulation, some corporations have risked their own long-term best interests, as well as those of consumers, by lobbying for excessive and discriminatory government regulation in order to gain short-term advantages. From the earliest stages of the agbiotech industry, those firms hoped that superfluous regulation would act as a type of government stamp of approval for their products, and they knew that the time and expense engendered by overregulation would also act as a barrier to market entry by smaller competitors. Those companies, which include Monsanto, DuPont-owned Pioneer Hi-Bred, and Ciba-Geigy (now reorganized as Syngenta), still seem not to understand the ripple effect of overly restrictive regulations that are based on, and reinforce, the false premise that there is something uniquely worrisome and risky about the use of recombinant DNA techniques.
The consequences of this unwise, unwarranted regulatory policy are not subtle. Consider, for example, a recent decision by Harvest Plus, an alliance of public-sector and charitable organizations devoted to producing and disseminating staple crops rich in such micronutrients as iron, zinc, and vitamin A. According to its director, the group has decided that although it will continue to investigate the potential for biotechnology to raise the level of nutrients in target crops above what can be accomplished with conventional breeding, “there is no plan for Harvest Plus to disseminate [gene-spliced] crops, because of the high and difficult-to-predict costs of meeting regulatory requirements in countries where laws are already in place, and because many countries as yet do not have regulatory structures.” And in May 2004, Monsanto announced that it was shelving plans to sell a recombinant DNA-modified wheat variety, attributing the decision to changed market conditions. However, that decision was forced on the company by the reluctance of farmers to plant the variety and of food processors to use it as an ingredient: factors that are directly related to the discriminatory overregulation of the new biotechnology in important export markets. Monsanto also announced in May that it has abandoned plans to introduce its recombinant canola into Australia, after concerns about exportability led several Australian states to ban commercial planting and, in some cases, even field trials.
Other companies have explicitly acknowledged giving up plans to work on certain agbiotech applications because of excessive regulations. After receiving tentative approval in spring 2004 from the British government for commercial cultivation of a recombinant maize variety, Bayer CropScience decided not to sell it because the imposition of additional regulatory hurdles would delay commercialization for several more years. And in June 2004, Bayer followed Monsanto’s lead in suspending plans to commercialize its gene-spliced canola in Australia until its state governments “provide clear and consistent guidelines for a path forward.”
Another manifestation of the unfavorable and costly regulatory milieu is the sharp decline in efforts to apply recombinant DNA technology to fruits and vegetables, the markets for which are minuscule compared to crops such as corn and soybeans. Consequently, the number of field trials in the United States involving gene-spliced horticulture crops plunged from approximately 120 in 1999 to about 20 in 2003.
Setting matters aright
The public policy miasma that exists today is severe, worsening, and seemingly intractable, but it was by no means inevitable. In fact, it was wholly unnecessary. From the advent of the first recombinant DNA-modified microorganisms and plants a quarter century ago, the path to rational policy was not at all obscure. The use of molecular techniques for genetic modification is no more than the most recent step on a continuum that includes the application of far less precise and predictable techniques for genetic improvement. It is the combination of phenotype and use that determines the risk of agricultural plants, not the process or breeding techniques used to develop them. Conventional risk analysis, supplemented with assessments specific to the new biotechnology in those very rare instances where they were needed, could easily have been adapted to craft regulation that was risk-based and scientifically defensible. Instead, most governments defined the scope of biosafety regulations to capture all recombinant organisms but practically none developed with classical methods.
In January 2004, the U.S. Department of Agriculture (USDA) announced that it would begin a formal reassessment of its regulations for gene-spliced plants. One area for investigation will include the feasibility of exempting “low-risk” organisms from the permitting requirements, leading some observers to hope that much needed reform may be on the horizon. However, regulatory reform must include more than a simple carve-out for narrowly defined classes of low-risk recombinant organisms.
An absolutely essential feature of genuine reform must be the replacement of process-oriented regulatory triggers with risk-based approaches. Just because recombinant DNA techniques are involved does not mean that a field trial or commercial product should be subjected to case-by-case review. In fact, the introduction of a risk-based approach to regulation is hardly a stretch; it would merely represent conformity to the federal government’s official policy, articulated in a 1992 announcement from the White House Office of Science and Technology Policy, which calls for “a risk-based, scientifically sound approach to the oversight of planned introductions of biotechnology products into the environment that focuses on the characteristics of the . . . product and the environment into which it is being introduced, not the process by which the product is created.”
One such regulatory approach has already been proposed by academics. It is, ironically, based on the well-established model of the USDA’s own plant quarantine regulations for nonrecombinant organisms. Almost a decade ago, the Stanford University Project on Regulation of Agricultural Introductions crafted a widely applicable regulatory model for the field testing of any organism, whatever the method employed in its construction. It is a refinement of the “yes or no” approach of national quarantine systems, including the USDA’s Plant Pest Act regulations; under these older regimens, a plant that a researcher might wish to introduce into the field is either on the proscribed list of plant pests, and therefore requires a permit, or it is exempt.
The Stanford model takes a similar, though more stratified, approach to field trials of plants, and it is based on the ability of experts to assign organisms to one of several risk categories. It closely resembles the approach taken in the federal government’s handbook on laboratory safety, which specifies the procedures and equipment that are appropriate for research with microorganisms, including the most dangerous pathogens known. Panels of scientists had stratified these microorganisms into risk categories, and the higher the risk, the more stringent the procedures and isolation requirements. In a pilot program, the Stanford agricultural project did essentially the same thing for plants to be tested in the field: A group of scientists from five nations evaluated and, based on certain risk-related characteristics, stratified a number of crops into various risk categories. Importantly, assignment to one or another risk category had nothing to do with the use of a particular process for modification or even whether the plant was modified at all. Rather, stratification depended solely on the intrinsic properties of a cultivar, such as potential for weediness, invasiveness, and outcrossing with valuable local varieties.
What are the practical implications of an organism being assigned to a given risk category? The higher the risk, the more intrusive the regulators’ involvement. The spectrum of regulatory requirements could encompass complete exemption; a simple “postcard notification” to a regulatory authority (without prior approval required); premarket review of only the first introduction of a particular gene or trait into a given crop species; case-by-case review of all products in the category; or even prohibition (as is the case currently for experiments with foot-and-mouth disease virus in the United States).
Under such a system, some currently unregulated field trials of organisms modified with older techniques would likely become subject to regulatory review, whereas many recombinant organisms that now require case-by-case review would be regulated less stringently. This new approach would offer greater protection and, by decreasing research costs and reducing unpredictability for low-risk organisms, encourage more R&D, especially on noncommodity crops.
The Stanford model also offers regulatory bodies a highly adaptable, scientific approach to the oversight of plants, microorganisms, and other organisms, whether they are naturally occurring or “non-coevolved” organisms or have been genetically improved by either old or new techniques. The outlook for the new biotechnology applied to agriculture, especially as it would benefit the developing world, would be far better if governments and international organizations expended effort on perfecting such a model instead of clinging to unscientific, palpably flawed regulatory regimes. It is this course that the USDA should pursue as it reevaluates its current policies.
At the same time as the U.S. government begins to rationalize public policy at home, it must stand up to the other countries and organizations that are responsible for unscientific, debilitating regulations abroad and internationally. U.S. representatives to international bodies such as the Codex Alimentarius Commission, the United Nations’ agency that sets food-safety standards, must be directed to support rational science-based policies and to work to dismantle politically motivated unscientific restrictions. All science and economic attachés in every U.S. embassy and consulate around the world should have biotechnology policy indelibly inscribed on their diplomatic agendas. Moreover, the U.S. government should make United Nations agencies and other international bodies that implement, collude, or cooperate in any way with unscientific policies ineligible to receive funding or other assistance from the United States. Flagrantly unscientific regulation should be made the “third rail” of U.S. domestic and foreign policy.
Uncompromising? Aggressive? Yes, but so is the virtual annihilation of entire areas of R&D; the trampling of individual and corporate freedom; the disuse of a critical, superior technology; and the disruption of free trade.
Strategies for action
Rehabilitating agbiotech will be a long row to hoe. In order to move ahead, several concrete strategies can help to reverse the deteriorating state of public policy toward agricultural biotechnology.
First, individual scientists should participate more in the public dialogue on policy issues. Perhaps surprisingly, few scientists have demanded that policy be rational; instead, most have insisted only on transparency or predictability, even if that delivers only the predictability of research delays and unnecessary expense. Others have been seduced by the myth that just a little excess regulation will assuage public anxiety and neutralize activists’ alarmist messages. Although defenders of excessive regulation have made those claims for decades, the public and activists remain unappeased and technology continues to be shackled.
Scientists are especially well qualified to expose unscientific arguments and should do so in every possible way and forum, including writing scientific and popular articles, agreeing to be interviewed by journalists, and serving on advisory panels at government agencies. Scientists with mainstream views have a particular obligation to debunk the claims of their few rogue colleagues, whose declarations that the sky is falling receive far too much attention.
Second, groups of scientists—professional associations, faculties, academies, and journal editorial boards—should do much more to point out the flaws in current and proposed policies. For example, scientific societies could include symposia on public policy in their conferences and offer to advise government bodies and the news media.
Third, reporters and their editors can do a great deal to explain policy issues related to science. But in the interest of “balance,” the news media often give equal weight to all of the views on an issue, even if some of them have been discredited. All viewpoints are not created equal, and not every issue has “two sides.” Journalists need to distinguish between honest disagreement among experts, on the one hand, and unsubstantiated extremism or propaganda, on the other. They also must be conscious of recombinant DNA technology’s place in the context of overall crop genetic improvement. When writing about the possible risks and benefits of gene-spliced herbicide-tolerant plants, for example, it is appropriate to note that herbicide-tolerant plants have been produced for decades with classical breeding techniques.
Fourth, biotechnology companies should eschew short-term advantage and actively oppose unscientific discriminatory regulations that set dangerous precedents. Companies that passively, sometimes eagerly, accept government oversight triggered simply by the use of recombinant DNA techniques, regardless of the risk of the product, ultimately will find themselves the victims of the law of unintended consequences.
Fifth, venture capitalists, consumer groups, patient groups, philanthropists, and others who help to bring scientific discoveries to the marketplace or who benefit from them need to increase their informational activities and advocacy for reform. Their actions could include educational campaigns and support for organizations such as professional associations and think tanks that advocate rational science-based public policy.
Finally, governments should no longer assume primary responsibility for regulation. Nongovernmental agencies already accredit hospitals, allocate organs for transplantation, and certify the quality of consumer products ranging from seeds to medical devices. Moreover, in order to avoid civil legal liability for damages real or alleged, the practitioners of agricultural biotechnology already face strong incentives to adhere to sound practices. Direct government oversight may be appropriate for products with high-risk characteristics, but government need not insinuate itself into every aspect of R&D with recombinant DNA-modified organisms.
The stunted growth of agricultural biotechnology worldwide stands as one of the great societal tragedies of the past quarter century. The nation and the world must find more rational and efficient ways to guarantee the public’s safety while encouraging new discoveries. Science shows the path, and society’s leaders must take us there.
Henry I. Miller (firstname.lastname@example.org) is a research fellow at Stanford University’s Hoover Institution. Gregory Conko is the director of food safety policy at the Competitive Enterprise Institute. This article is derived from their book The Frankenfood Myth: How Protest and Politics Threaten the Biotech Revolution (Praeger Publishers, 2004).