Labeling Genetically Modified Foods: How Do US Consumers Want to See It Done?
Mario F. Teisl, Luke Garner, Brian Roe, and Michael E. Vayda
University of Maine; The Ohio State University
If genetically modified food labeling occurs, policymakers will need to consider the form of the program: Should it be mandatory? What should be labeled? What should be on the label? Who should be in charge? We present results of a nationally representative survey of US adults aimed at answering these questions.
Key words: genetically engineered food, labeling, design, consumer preferences.The production and labeling of genetically modified (GM) foods has been controversial—debated at legislative and judicial levels and in the arena of public opinion. Polls have emphasized that a majority of consumers in the United States (US) desire GMFs to be labeled (Pew Initiative on Food and Biotechnology, 2001), and legislation has been entered at both the federal and state levels. For example, HR 3377 and S 2080—the "Genetically Engineered Food Right to Know Acts"—were introduced into the US House of Representatives and Senate, respectively. In addition, at least seven states have debated labeling and marketing requirements for GM foods (Pollack, 2001). Further, the current lack of harmonization of policies across countries also makes GM food labeling an international trade issue.
The debate surrounding GM food labeling has almost exclusively centered on the assumption that consumers' sole desire for information about GM foods is whether they are, in fact, genetically modified. Although this is an important primary question, the labeling issue is more complex than simply determining whether or not to label. If GM food labeling is to occur, policymakers need to consider the form of the labeling program: Should it be mandatory or voluntary? What foods should be labeled? What pieces of information should be placed on the label? Who should be in charge of monitoring compliance? To answer these questions, policy-makers have to balance the concerns and desires of a broad spectrum of consumers with the costs and complexities associated with supply chain management under alternative labeling programs. Here we present an overview of a recent survey of US adults aimed at understanding consumer preferences for specific attributes of alternative labeling programs.
During the summer of 2002, we administered a mail survey to a nationally representative sample of 5,462 US residents and an additional oversample (710 individuals) of Maine residents. The survey was administered with multiple mailings and with an incentive paid for returned completed surveys. A mail survey approach was used, because it allowed respondents to react to physical representations of potential GM labels and allowed us to manipulate these representations experimentally. An oversample of Maine residents was added to provide representative results for Maine state policymakers. (The Maine Agriculture and Forest Experiment Station provided some of the funding for this research.)
In total, 375 Maine residents and 2,012 US (non-Maine) residents responded to the survey for a response rate of 53% and 37%, respectively. For all analyses, the data are weighted to adjust for oversampling; responses for each state are weighted such that the proportion of respondents from that state is reflective of the state's proportion of the US adult population (as measured by the 2000 US Census) while maintaining the overall sample size. Specifically, weights are determined by the formula: State weight = (state adult population / US adult population) / (number of state respondents / total number of respondents).
Except for race, our survey respondents have characteristics similar to those of the US adult population (Table 1). The stated differences in race may be reflective of a true underlying difference and/or may reflect differences in the way the race questions are asked across the two surveys.
Table 1. Socioeconomic characteristics of respondents.
The mail survey instrument consisted of 44 questions in six sections.1 (Questions analyzed are referenced throughout the text—see the Appendix for actual question wording.) The content and wording of questions is based upon an analysis of issues raised in the labeling or consumer-perception literature (e.g., Boccaletti & Moro, 2000; Hallman & Metcalfe, 1995; Hoban, 1999; Huffman, Rousu, Shogren, & Tegene, 2003a; Rousu, Monchuk, Shogren, & Kosa, 2003; Roe, Teisl, Rong, & Levy, 2001; Teisl, 2003), state and federal policy needs, and previous focus group research (e.g., Teisl et al., 2002). Further, they are based upon conceptualizations of consumer reactions to labeling information as presented in Teisl, Bockstael, and Levy (2001) and Teisl and Roe (1998).
There are several statistical tests used that are differentiated by the type of data being tested. For responses to ratings questions (Tables 2, 3, 4, and 7) we use analysis of variance (in the form of pair-wise t-tests of means) to determine if the mean ratings are different across response categories. For responses to categorical choice questions (Tables 5 and 6), we use contingency tables to determine if the response distributions are different across response categories. All tests use a significance level of 10%.
We began by asking a series of questions to determine respondents' awareness and knowledge of current GM food production in the United States. We find 76% of respondents had heard of GM foods (Question 6), and a majority of these individuals had heard about GM corn and tomatoes. The high awareness of corn may be due to extensive news coverage in 2000 of the contamination of taco shells by GM corn not approved for human consumption. The high awareness of tomatoes may be because the Flavr Savr™ tomato was the first commercially available GM food in the United States (released in 1994).
We next asked respondents to provide an estimate of the percent of food sold in the United States that is GM (Question 7). The Grocery Manufacturers of America estimates that at least 70% of the foods on grocery store shelves contain GM ingredients (Goldsbrough, 2000). However, about half of the respondents thought that less than 30% of the food supply has been genetically modified, and less than 10% thought the answer was 70% or more. Respondents seemed to have an imperfect appreciation for the amount of GM foods they had been eating.
We also asked respondents to provide a concern rating for eight different food production and processing techniques (Question 5). Respondents' number one food production concern was not the use of GM ingredients (Table 2); this general result parallels those from an annual survey administered by the International Food Information Council.2 In those surveys, respondents had relatively high concerns on a range of food safety issues (e.g., pathogens, chemicals, pesticides). Perversely, the four technologies of most concern do not trigger labeling, whereas techniques of lesser concern do require some sort of labeling.3 The apparent discrepancy between individuals' concerns about food production technologies and the labeling of foods using these technologies raises two questions. First, if government implements GM food labeling, why does it not impose (or at least advocate) a similar requirement for technologies of equal or higher concern? Second, given that labeled technologies are of lowest concern, one may question if labeling has the ability to reduce consumer concerns about food technologies.
Table 2. Respondents' ratings of concerns about food production and processing technologies.a
We next provided respondents a list of 32 potential benefits (Question 8) and concerns (Question 9) of using GM foods and asked them to rate each one on importance. Of the eight benefits rated as most important by our respondents, most accrue directly to consumers, and most of these relate to improvements in the food's health attributes (Table 3). Farmer benefits and nonhealth improvements in food attributes were less important to respondents. Currently approved GM foods primarily provide producer benefits with few benefits accruing directly to the consumer.4 However, one benefit of importance to both farmers and consumers is a reduction in pesticide use. Increased plantings of GM crops have led to an overall decrease in the amount of pesticides used and in a switch to less toxic pesticides (Fernandez-Cordejo & McBride, 2002). Genetically modified food producers may have similar success in marketing pesticide reduction as a benefit to consumers; however, the current lack of labeling may hamper their ability to do this.
Table 3. Average importance respondents place on potential benefits of genetically modified food.a
The eight concerns rated as most important by our respondents are related to risks directly impacting the consumer through perceived deteriorations in food safety or are related to potential negative environmental impacts (Table 4). An important component of all these concerns seems to be the uncertainty of long-term impacts. The high level of concern surrounding unknown long-term impacts is a consistent theme explaining consumers' negative reactions to new food technologies.5 For example, concerns about long-term health impacts seem to explain consumers' initial opposition to pasteurization (Huffman et al., 2003a) and microwave ovens (Devlin, 1998) and seems to be a factor in consumer acceptance of GM foods (Hoban, 1997). In general, consumers trust food scientists' abilities to determine the short-term safety of new food technologies but understand the limitations scientists face in determining long-term impacts (Levy & Derby, 2000). Interestingly, the level of technical knowledge about a new food technology does not seem to impact consumers' concerns; it is the lack of experience with the technology (Levy, 2001).
Table 4. Average importance respondents place on potential concerns of genetically modified food.a
If long-run uncertainties are the primary driver of consumer rejection of new food technologies, then a lack of labeling may actually prevent consumers' ability to develop direct or indirect experience with it; that is, they are not able to understand the extent to which they (directly), or other members of society (indirectly), consume GM foods without any apparent ill effects. For example, in focus group research (Teisl et al., 2002), when we told participants how much of the food supply is derived from GM crops, some found the information comforting; these participants seem to combine the prevalence of GM foods with the notion they had not heard or known of anyone getting sick as positive news.6
This may present an ironic twist to those who support labeling as a way of eliminating GM foods; labeling may after an initial short-run decline in sales actually lead to a broader long-run acceptance of the technology (if, in fact, producers are willing to ride out any short-term losses). This also presents a conundrum to producers; they may be trading off a potential short-run decline in sales against future sales growth. This is particularly true if consumers are not able to associate a technology with the provision of positive attributes. Research has indicated that some individuals will accept an increase in perceived risks if relevant benefits are bundled with the risk (Fischhoff & Fischhoff, 2001; Moon & Balasubramanian, 2001; Teisl, Roe, Vayda, & Ross, 2003).
Before asking respondents whether they want a GM food-labeling program, we first wanted to determine how much experience respondents had with GM labeling. Currently, producers voluntarily test and label their foods only to denote that they do not contain GM ingredients. Few respondents (12%) had seen such a label (Question 10); however, almost all respondents (85%) wanted GM foods to be labeled (Question 11). Of those who wanted GM foods labeled, we asked a question to determine how they wanted the program structured (Question 12); almost all respondents who wanted labeling wanted this labeling to be mandatory (Table 5), although respondents were split on whether they wanted all foods or just GM foods to display a label. Clearly, respondents were dissatisfied with the current approach to labeling.
Table 5. Respondents' desire for labeling genetically modified foods.
To determine who respondents wanted to administer GM labeling, we presented them with a list of organizations and asked them to choose which one organization they would prefer (Question 13). In terms of general organizational groups, most individuals chose either a government agency or an independent organization (Table 6); this result is similar to Huffman, Rousu, Shogren and Tegene (2003b). Of those choosing a government agency, most individuals wanted either the US Food and Drug Administration or the US Department of Agriculture to administer labeling. Respondents' familiarity with these two agencies, and their positive evaluation of their traditional handling of food labeling, may explain respondents' strong preferences for these agencies. Independent organizations, as a group, were chosen by almost 10% of respondents. However, of the four independent organizations, few chose the two that are currently the largest independent certifiers of GM-free foods (the Identity Preservation Program and Genetic ID).
Given their health-related concerns, it is not surprising that many respondents preferred health organizations. However, of the health-related groups, the American Medical Association has stated they find no scientific justification for the general labeling of genetically modified foods (American Medical Association, 2000), and the other three have apparently not taken any official position on the issue. In focus groups (Teisl et al., 2002) we found that some people did not want health organizations to administer a labeling program, because many of these organizations target only specific diseases or conditions (e.g., heart disease), whereas GM foods could impact the whole body. Thus, it is not surprising that the two health organizations focusing on specific diseases were chosen by few respondents.
Table 6. Percent of respondents preferring specific groups to oversee labeling program.a
It may be initially surprising that relatively few respondents wanted environmental groups to administer GM labeling, because at least two of these groups (the Organic Consumers Association and Greenpeace) have strongly supported this type of labeling. Respondents may have made a clear distinction between an organization's ability to promote advocacy and their ability to administer a labeling program.
The bottom line is that consumer preferences for who should administer a labeling program for GM foods seem directly at odds with the current reality. Most consumers desired a federal agency to administer this program—apparently because of the relatively high level of trust people place in these agencies—yet these agencies have been reluctant to take on this task. At the same time, the groups most active in promoting GM labeling, or those currently involved with GM labeling, garner little public support. Given that the ability of a labeling program to communicate to consumers depends, at least in part, on the degree to which consumers trust the organization in charge of the program (Slovic, 1993; Blaine & Powell, 2001), the government's reluctance to administer such a program may be inadvertently hindering consumer acceptance of GM foods.
To help determine the information respondents want to see on a GM label, we asked them to rate the importance of seven potential pieces of information that could be placed on such a label (Question 15). Given respondents' concerns about potential health risks, it is not surprising that they placed a high degree of importance on knowing about any warnings associated with the genetic modification (Table 7). Respondents were likely to desire contact (a phone number or website) and certifier information, because it would allow for a simpler, more credible label7 while allowing individuals a venue to pursue more information. Respondents would also like information about a GM food's benefits.
Respondents' desires that GM food labels provide benefit and risk information implies that a simple yes/no approach to labeling may be of limited usefulness. The use of biotechnology in food production can have multidimensional effects on product quality; some consumers may want to know about some or all of these changes. A simple GM label may actually be harmful to individuals who are concerned about GM content but willing to accept it if the genetic modification provides benefits; the harm comes from a simple GM label causing these individuals to avoid a food they would otherwise consume. Although the context is different, this general result is consistent with findings by Rousu et al. (2003), where they suggest that GM labeling of low-nicotine cigarettes may confuse some consumers, leading to inappropriate purchase decisions and reductions in welfare.
Table 7. Importance ratings of potential information pieces for a genetically modified food label.a
Policies allowing consumers to make purchase decisions match personal preferences are inherently desirable, whether the attributes concern end-use characteristics or process attributes—as long as these policies are not too costly. The results indicate that most respondents desire a labeling program for GM foods. However, this finding is based upon responses to a question that did not present the costs or benefits of instituting a labeling program. A decision to impose labels should recognize both its benefits and costs. Thus, the research here does not conclude that a labeling program should be instituted; rather, the findings can provide guidance on how a labeling program should look if it is determined that such a program is warranted. In terms of the GM debate thus far, one thing is relatively clear: A simple GM food label will be of limited usefulness to consumers because it would only allow consumers to differentiate GM food products from non-GM food products. Given that genetic engineering can produce a wide variety of outcomes (both positive and negative), simple GM food labels are not likely to allow most consumers to differentiate products in the manner they most desire and could actually hurt those consumers who would be willing to accept GM content when there are GM-linked benefits.8
1 Note there are several questions (Questions 16, 17, and 22) that ask respondents to react to specific labels; the information presented in these labels was experimentally manipulated and varied across respondents. The analyses of these data are beyond the scope of this paper and, as a result, we will not fully describe these data; interested readers can contact the first author.
2 Results available at http://www.ific.org/research/loader.cfm?url=/commonspot/security/getfile.cfm&PageID=1491.
3 Not all irradiated food needs to be labeled, only those foods nutritionally altered by the irradiation.
4 This is also generally true for nonfood GM crops such as cotton. However, cigarettes made from tobacco genetically modified to reduce nicotine are currently available in some states.
5 This hypothesis is consistent with Slovic (1987) who indicates a major factor impacting a consumer's evaluation of a new technology is the degree to which risks are "unknown"—that is, risks that are not observable or evident, have effects that are delayed, or are not definitively known to science (Marks, 2001).
6 Others in the groups reacted negatively; they found the idea that they had been eating GM foods in ignorance as disturbing—apparently, a lack of labeling led these individuals to feel deceived.
8 An anonymous reviewer points out that if a similar study was done in Europe and found similar results, it would provide evidence that Europe's GM-labeling policy is suboptimal. Boccaletti and Moro's (2000) findings seem to support this idea; they found that some Italian consumers would be willing to consume GM foods if the benefits of the genetic modification were presented.
American Medical Association. (2000). Genetically modified crops and foods (Report 10 of the Council on Scientific Affairs [I-00]). Available on the World Wide Web: http://www.ama-assn.org/ama/pub/article/2036-4030.html.
Blaine, K., & Powell, D. (2001). Communication of food-related risks. AgBioForum, 4(3&4), 179-185. Available on the World Wide Web: http://www.agbioforum.org.
Boccaletti, S., & Moro, D. (2000) Consumer willingness-to-pay for GM food products in Italy. AgBioForum, 3(4), 259-267. Available on the World Wide Web: http://www.agbioforum.org.
Fernandez-Cornejo, J., & McBride, W.D. (2002). Adoption of bioengineered crops (Agricultural Economic Report No. 810). Washington, DC: United States Department of Agriculture Economic Research Service.
Fischhoff, B., & Fischhoff, I. (2001). Publics' opinions about biotechnologies. AgBioForum, 4(3&4), 155-162. Available on the World Wide Web: http://www.agbioforum.org.
Goldsbrough, P. (2000). Quiz yourself about foods made from genetically modified crops. West Lafayette, IN: Purdue University. Available on the World Wide Web: http://www.purdue.edu/UNS/html4ever/0007.Goldsbrough.foodquiz.html.
Huffman, W.E., Rousu, M., Shogren, J.F., & Tegene, A. (2003a). Consumers' resistance to genetically modified foods in high income countries: The role of information in an uncertain environment (D. Gale Johnson Lecture—Topic II). University of Chicago. April 25.
Marks, L. (2001). Communicating about agrobiotechnology. AgBioForum, 4(3&4), 152-154. Available on the World Wide Web: http://www.agbioforum.org.
Moon, W., & Balasubramanian, S.K. (2001). Public perceptions and willingness-to-pay a premium for non-GM foods in the US And UK. AgBioForum, 4(3&4), 221-231. Available on the World Wide Web: http://www.agbioforum.org.
Roe, B., Teisl, M.F., Rong, H., & Levy, A.S. (2001). Characteristics of successful labeling policies: experimental evidence from price and environmental disclosure for deregulated electricity services. Journal of Consumer Affairs, 35(1), 1-26.
Rousu, M., Monchuk, D.C., Shogren, J.F., & Kosa, K.M. (2003). Consumer perceptions of labels and the willingness to pay for "second generation" genetically modified products (with an evaluation of the FDA's proposed rules on labeling). Research Triangle Park, NC: RTI International.
Teisl, M.F., Halverson, L., O'Brien, K., Roe, B., Ross, N., & Vayda, M. (2002). Focus group reactions to genetically modified food labels. AgBioForum, 5(1), 6-9. Available on the World Wide Web: http://www.agbioforum.org.
Teisl, M.F., Roe, B., Vayda, M., & Ross, N. (2003, June-July). Willingness to pay for genetically modified foods with bundled health and environmental attributes. Paper presented at the 7nd ICABR International Conference on Public Goods and Public Policy for Agricultural Biotechnology, Ravello, Italy.
This project was financed in part by a United States Department of Agriculture National Research Initiative Grant (#2001-35400-10548) and the Maine Agricultural and Forest Experiment Station. Although the research described in this article has been funded in part by the USDA, it has not been subject to the Agency's required peer and policy review and therefore does not necessarily reflect the views of USDA; no official endorsement should be inferred. All errors and omissions are the sole responsibility of the authors.
Suggested citation: Teisl, M.F., Garner, L., Roe, B., & Vayda, M.E. (2003). Labeling genetically modified foods: How do US consumers want to see it done? AgBioForum, 6(1&2), 48-54. Available on the World Wide Web: http://www.agbioforum.org.
In This ArticleTable 1
|© 2003 AgBioForum | Design and support provided by Express Academic Services | Contact ABF: firstname.lastname@example.org|