This article argues that the public can and should be trusted with information about the food it consumes and that mandatory GE labeling is the right thing to do. First, the article lays out the strong public policy reasons for mandatory labeling, including an overview of the public’s well-founded concerns with these foods and the federal government’s historic failure to regulate them. Then, the article briefly explains why potential constitutional challenges should not succeed in presenting roadblocks to labeling laws.
The first argument in favor of labeling is that the U.S. public overwhelmingly desires to be an “informed citizenry” when it comes to GE foods. In a 2013 Huffington Post/YouGov poll, 82 percent of Americans said that GE foods should be labeled. Emily Swanson, GMO Poll Finds Huge Majority Say Foods Should Be Labeled, Huff Post Green (Mar. 4, 2013). In a 2010 poll, 93.1 percent of Americans said that GE foods should be labeled. Thomson Reuters, National Survey of Healthcare Consumers: Genetically Engineered Food 4 (Oct. 2010). These polls asked the question directly and did not attempt to hide the ball; there is no reason to suspect that the polls’ respondents did not likewise provide direct (and honest) answers. The American public is presumably not alone in this preference; the citizens of at least sixty other countries already enjoy the benefits of mandatory labeling. These countries include the members of the European Union, China, Russia, Japan, Australia, Saudi Arabia, Brazil, Turkey, Kazakhstan, and South Africa. See Center for Food Safety, International Labeling Laws.
The second, and most important, point is that labeling will enable Americans to make critical decisions about the foods they eat. It will enable them to avoid the health and other risks associated with these foods—and there are risks. The most basic risk stems from the fact that there are significant uncertainties about the health impacts of GE foods. Genetic engineering can cause unidentified disruptions in molecular structure and genetic composition, which may lead to unintended and unappreciated health consequences.
In a 2012 letter to the American Medical Association, Michael Hansen PhD, senior scientist for Consumers Union, provided some examples of these unintended—and previously unidentified—consequences. For instance, in a 2005 study in which a protein was inserted from beans into peas, the protein’s structure was somehow altered; mice eating the transgenic peas suffered from adverse immunological reactions to the inserted protein, as well as higher-than-normal reactions to natural proteins in the peas (and cooking the peas did not help). Letter from Michael Hansen to AMA 10-11 (Mar. 19, 2012) [hereinafter Hansen Letter]. As Hansen explained, this meant that though the implicated protein had the same amino acid sequence in both plants, it was safe in one and potentially harmful in another—even beyond the protein itself. Id. Another recent report from Belgium reviewed the molecular characterization of six transgenic plants and found that, but for one, the characterizations of the foods in the market were different from the characterizations originally reported by the company. Id. at 11. Further, the food safety standards body of the United Nations (UN) (the Codex Alimentarius Commission) has warned, “there is no definitive test that can be relied upon to predict allergic response in humans to a newly expressed protein.” U.N. Codex Alimentarius, CAC-GL 45-2003, Annex 1.
In addition to this uncertainty, there is a compelling string of animal studies showing effects such as intestinal damage, enlarged organ size, organ cellular abnormalities potentially indicating toxicity or compromised metabolic function, significant increased mortality for offspring of rats fed GE soy before and during pregnancy, and allergic-type reactions. See Jeffrey Smith, Genetic Roulette (2007) (summarizing many of these studies). These types of studies have been reported in scientific journals and by government and industry alike. Their authors include researchers from academic and medical institutions with expertise and PhDs in fields such as molecular biology and biomedical sciences. The concerns they raise are echoed in a statement endorsed by at least 800 scientists around the world calling for a moratorium on GE plants. Open Letter from World Scientists to All Governments Concerning GMOs. Further, a common thread among many of these studies is that important follow-up research—for instance, to determine whether the consumption of Bacillus thuringiensis (Bt) crops in fact impacts colon cancer, or whether the consumption of GE soy is contributing to the rise in soy allergies—has not been conducted, at least not publicly. The point is that these studies raise serious concerns and should be taken seriously.
FDA Nonregulation of GE Foods
One might assume that GE products are safe because they are on the market. In fact, in the realm of GE foods, any safety studies—if conducted—are conducted by the industry and need not be reviewed by the U.S. Food and Drug Administration (FDA). The FDA does not conduct or require safety testing of these foods, nor does it require industry to submit safety studies prior to marketing. Rather, through a 1992 draft policy statement, the FDA has largely assumed that GE components are “substantially similar” to their non-GE counterparts and has recommended nonmandatory testing protocols for companies to follow in ascertaining various effects. See 57 Fed. Reg. 22,984 (May 29, 1992). If harmful effects are found in these voluntary studies, the FDA encourages, but does not require, producers to consult with the agency. See id. Against this backdrop, it is worth noting that none of the GE crops currently on the market in the United States have actually undergone the minimum safety testing recommended by Codex Alimentarius, and even rigorous safety testing cannot account for each of the unintended consequences that GE foods may cause. See Hansen Letter at 3–5.
Labeling, therefore, presents what should be a floor for risk avoidance and management. Consumers may avoid the risks by avoiding the foods, and the public health community may manage the risks by being able to identify and track GE consumption. Put differently, labeling can help to increase the chances of accurate results in what resembles a mass feeding study of GE foods on the American public and can help to ensure that citizens do not engage in that study unknowingly.
Environmental Harms and Other Concerns
Human health is not the only reason to label. Labeling will also enable consumers to avoid supporting an industry that clashes with their values and their visions for the world. Among other things, the increased use of GE crops has been directly linked to environmental harms, including increased pesticide use, increased herbicide use, development of herbicide-resistant “superweeds,” and depleted soils. See Antoniou et al., GMO Myths & Truths (2012) (referencing scientific studies) [hereinafter Antoniou]. There are also concerns that GE crop practices are jeopardizing biodiversity, for instance by wiping out the milkweed so crucial to the food supply of migrating monarch butterflies, whose population was down to its lowest level in two decades this year. Additionally, some religious groups are opposed to genetic engineering and do not wish to consume foods of GE descent. See Faith & GMOs. Further, unintended cross-contamination is a real threat. See e.g., Carey Gillam & Julie Ingwersen, U.S. Discovery of Rogue GMO Wheat Raises Concerns Over Controls, Reuters (May 31, 2013). It can jeopardize the status of organic and conventional farmers and exposes them to legal threats by the biotech industry for patent infringement. See, e.g., CBS News, Agricultural Giant Battles Small Farmers (Jan. 4, 2011). Finally, in some communities, GE practices have been linked to tragic social problems. Green America, Green American 15, 17 (Apr./May 2012) (discussing farmer suicides in India and birth defects in Argentina).
Debunking Other Antilabeling Claims
We do not need GE foods to feed the world. In 2008, the World Bank and the UN produced a report concluding that “agroecological farming”—not genetic engineering—was the key to food security. See Antoniou at 112. Agroecological farming focuses on practices such as soil and water conservation, minimal use of chemical pesticides and fertilizers, low energy inputs, and natural pest resistance and crop resilience. The United Nations Special Rapporteur on the Right to Food explained that agroecology “mimics nature not industrial processes” and that “yields went up 214% in 44 projects in 20 countries in sub-Saharan Africa using agroecological farming techniques over a period of 3 to 10 years . . . far more than any GM crop has ever done.” Id. And, eighteen African countries have objected to industry’s characterization of itself as their savior, stating:
We strongly object that the image of the poor and hungry from our countries is being used by giant multinational corporations to push a technology that is neither safe, environmentally friendly nor economically beneficial to us. We do not believe that such companies or gene technologies will help our farmers to produce the food that is needed in the 21st century. On the contrary, we think it will destroy the diversity, the local knowledge and the sustainable agricultural systems that our farmers have developed for millennia, and that it will thus undermine our capacity to feed ourselves.
Id., quoting Delegate Statement to UN Food & Agricultural Organization (1998).
Additionally, there are other, simpler ways to feed the world, including improved food distribution and the often overlooked efficiency of feeding plants directly to humans, rather than to animals first. Further, the “feed the world” argument is only relevant if labeling GE foods means that no one will want those foods and that they will inevitably disappear from the planet. (GE foods are rare on European shelves.) If this were so, it would suggest that the GE industry had not met its burden to prove to the jury of the American public that its products were safe.
It might also be tempting to think that voluntary labeling of non-GE products provides sufficient information to the public. While voluntary labeling does inform consumers of the non-GE status of some non-GE foods, not every non-GE food is labeled. Mandatory labeling is the only way to ensure that the public knows whether the foods it is consuming are genetically engineered. Also, as a matter of public policy, any “burden” of labeling should fall on the industries actually benefitting from the subject technology. In reality, studies have found that GE labeling would cost consumers at most $1.27 to $10.00 annually. See W. K. Jaeger, Economic Issues & Oregon Ballot Measure 27: Labeling of Genetically Modified Foods (2002); Joanna Shepherd-Bailey, Economic Assessment: Proposed California Right to Know Genetically Engineered Food Act (Prop 37) (2012).
Constitutionality of Labeling
Finally, there are the constitutional issues. When the Vermont legislature heard testimony on a GE labeling bill in spring 2012, an industry representative testified that if industry decided to sue, it would consider the “threat” posed by a labeling regime to be “imminent” even if the law hadn’t yet taken effect. An Act Relating to the Labeling of Food Produced with Genetic Engineering before the House Committee on Agriculture, 2011–2012 Leg. (Vt. 2012) (statement of Rachel Lattimore, Partner, Arent Fox Law Firm). Some in the state were concerned. In 1996, the state had lost its campaign to require labels on dairy products derived from cows treated with recombinant bovine growth hormone (rBGH). The Second Circuit ruled that the state could not require disclosure on the basis of “consumer curiosity” alone under the First Amendment. Int’l Dairy Foods Ass’n v. Amestoy, 92 F.3d 67, 73–74 (2d Cir. 1996).
Under the First Amendment, there are basically two tests that a court might apply to commercial speech regulation. The first generally applies to restrictions on speech and involves intermediate-type scrutiny that requires a “substantial” state interest. Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n of N.Y., 447 U.S. 557, 564, 566 (1980). The second generally applies to disclosure requirements and involves a rational basis-type test. Zauderer v. Office of Disciplinary Counsel of the Supreme Court of Ohio, 471 U.S. 626, 650–52 (1985). In Int’l Dairy, the court applied the Central Hudson test even though it was assessing a disclosure requirement and not a restriction on speech. Because the court found Vermont’s statute was based only on “consumer curiosity” and Vermont did not have said “substantial interest,” Vermont could not require labels on rBGH-derived products. 92 F.3d at 74.
In making this determination, the court noted that though the citizens of Vermont had concerns about the safety of rBGH-derived milk, the state had not “adopted” those concerns as rationale for its statute. The court continued that, even if the state had adopted those concerns, the concerns would need to be reasonable. The court said that there was “no scientific evidence from which an objective observer could conclude that [rBGH] has any impact at all on dairy products.” Id. at 73. It also noted that the FDA had determined there were “no human safety or health concerns” with rBGH products and that “neither consumers nor scientists” could distinguish rBGH milk from non-rBGH milk. Id.
None of these “consumer curiosity” factors are present in the GE context. First, there is scientific evidence—a small fraction of it discussed above—from which objective observers (including scientific experts) can conclude that genetic engineering has an impact on food products. Second, the FDA has not actually determined that GE foods are safe for human consumption. Unlike the “thorough review” that FDA says it conducted in support of its Final Rule approving the use of rBGH, FDA has not conducted a thorough review of GE foods and has issued only a draft policy statement that lacks the force of law. Also, the draft GE policy statement, unlike the rBGH Final Rule, discusses numerous risks associated with the product at issue (e.g., allergenicity and toxicity). Third, unlike in the rBGH case, a quick online search will reveal numerous ways in which scientists can distinguish GE foods from non-GE foods. See e.g., GMO Testing (last visited July 30, 2013). Finally, as far as the state itself being concerned about GE foods, the current version of Vermont’s new GE bill provides an example that reflects many weeks of testimony and makes concerns explicit in its purpose and findings sections. See H.112, 2013–2014 Leg. (Vt. 2013).
In addition, the Second Circuit has boxed the Int’l Dairy case into a corner through subsequent decisions. Several times, it has held that the Zauderer test should apply to disclosure requirements and has limited Central Hudson’s applicability to situations where the state can provide no better interest than “consumer curiosity.” E.g., N.Y. State Restaurant Ass’n v. N.Y. City Bd. of Health, 556 F.3d 114, 134 (2d Cir 2009). Further, other circuits have endorsed Zauderer as the proper test to apply to disclosure requirements. Like the Second Circuit, they have ruled that Zauderer should apply to disclosures even when the state’s interest is not in “preventing consumer deception” (as it was in Zauderer). Pharmaceutical Care Mgmt. Ass’n v. Rowe, 429 F.3d 294, 310 n.8 (1st Cir. 2005); Discount Tobacco City & Lottery, Inc. v. United States, 674 F.3d 509, 556 (6th Cir. 2012). In any case, preventing consumer deception would be a logical reason to require labels on GE foods (otherwise, a consumer might assume that a nonlabeled food was not GE).
The other constitutional issues that tend to arise in state-level GE labeling discussions are preemption and the dormant commerce clause (DCC). Like the First Amendment, neither should be an obstacle. The federal Food, Drug, & Cosmetic Act has an express preemption clause as well as a savings clause, so that only listed categories of labeling are preempted. 21 U.S.C. § 343-1; Pub. L. No. 101-535, § 6(c)(1). Properly drafted, GE labeling requirements fall under none of those categories. Under the DCC, a statute that is not discriminatory need only show that any burden on interstate commerce is not “clearly excessive” in relation to the local benefit. Pike v. Bruce Church, Inc., 397 U.S. 137, 142 (1970). A GE labeling statute would not be “discriminatory” because it would not favor in-state products over out-of state products: all products would need to be labeled, regardless of where they were produced. Then, the statute should survive the Pike test because any “burden” on interstate commerce (e.g., that companies would have to change their labels) would be outweighed by the local benefits. See, e.g., Nat’l Elec. Mfrs. Ass’n v. Sorrell, 272 F.3d 104, 108–13 (2001) (upholding under DCC Vermont’s requirement for labels on mercury products).
Given the numerous benefits of GE labeling and the legal support for GE labeling laws, governments should not be skittish about requiring companies to tell when they use GE technology in creating the foods we eat. In turn, the companies should be happy to tell. As Jerry Greenfield (of Ben & Jerry’s) put it when testifying before Vermont’s legislature: “I think food producers and companies ought to be proud to say what’s in, or not in, their products. They should be screaming it from the rooftops. If you don’t feel great about what’s in your product, get out of the business.” Andrew Stein, In Statehouse, GE Labeling Bill Is Praised & Panned, VTDigger.org (Feb. 15, 2013).