Consumers’ Willingness to Pay for Genetically Modified Foods with Product‐Enhancing Nutritional Attributes
通过实验拍卖发现，美国消费者愿意为通过基因内技术增强营养的蔬菜支付溢价，且信息环境显著影响其支付意愿。
The commercial successes of genetic modification (GM) during the past decade in the United States and Canada have been in feed, fiber, and oil crops but not primarily in food crops, with the exception of refined vegetable oils.1 Early GM traits (e.g., herbicide tolerance, insect resistance) were obtained by transferring genes across species, largely from soil bacteria. This transgenic nature of genetically modified organisms (GMOs) has been one dimension of consumer resistance to GM—raising biodiversity, environmental, ethical, and safety concerns—and has been a factor in the larger controversy surrounding GM (for reviews of the GM debate, see 6, 16). The global debate over GM encompasses a diverse set of interested parties who have disseminated information into the public domain, spanning the spectrum from Greenpeace calling GMOs "Frankenfoods" to the biotechnology industry suggesting that GMOs are "foods to feed the world" (see e.g., 7, 4). This conflicted information environment has a direct impact on consumers' perspectives and valuations for GM foods (12 and has played a role in the implementation of diverging adoption, labeling, and trade policies internationally. While to date commercially available GM crops have been transgenic in nature, recently bioengineering breakthroughs have occurred using intragenics, where genes are moved long distances within species and without antibiotic markers. For example, the potato is the fourth leading source of calories worldwide (UN 15) and has a diverse, but very difficult to manipulate, genome. However, it can be manipulated using intragenic methods to move traits from primitive to commercial varieties. That is, genomic and metabolic pathway discoveries can be rapidly introduced into established commercial varieties to fast-track the breeding processes for new potato varieties. An additional advantage of these methods is that they do not use antibiotic markers to identify the location of inserted genes.2 These are all proffered as reasons for a low regulatory hurdle for intragenic foods. A second neoteric development tied to intragenic breakthroughs is a renewed interest by some bioengineering companies to develop GM food crops with "product-enhancing consumer attributes," or traits that directly benefit consumers. With the exception of the short-lived marketing attempts in the mid-90s of the "Flavr-Savr tomato" and a "high solids tomato" produced by Zeneca, commercially successful GM crops in the United States have possessed input traits (traits that reduce either the cost of production or the variance in the cost of production to farmers) and hence have benefited consumers only to the extent that they have lowered food prices.3 With new intragenic GM techniques, it is feasible to dramatically enhance product attributes such as antioxidant and vitamin content in horticultural crops, thus developing new foods with attributes of direct value to consumers. New research has the potential to differentiate consumers' willingness to pay (WTP) for food products containing enhanced nutrients due to intragenic and transgenic GM methods. The objective of this article is to assess consumers' WTP for new intragenic fresh vegetables with product-enhanced antioxidant and vitamin C levels. Individuals from a random set of telephone numbers in two metropolitan areas were contacted by an independent survey group in 2007 to obtain their agreement to participate, and participants came to a central location—a laboratory or classroom. Their WTP was obtained in a unique series of multiple-round random nth-price experimental auctions with randomized label and information treatments. The article is organized as follows. In the following section, an overview of the conducted experimental auction is provided. The next section presents a summary of the collected data and analysis. In the final section, some conclusions are presented. Data on consumer attitudes toward GM food labels with and without enhanced nutritional attribute statements were obtained from a series of laboratory experiments conducted in the spring of 2007. Participants for the study were solicited from the general public by an independent marketing organization in two different cities (Des Moines, Iowa and Harrisburg, Pennsylvania) in order to obtain a representative sample. Each experimental session consisted of four primary steps. After completing a series of consent forms and receiving financial compensation in the amount of $45, the ninety-eight participants were trained in the experimental auction method to be utilized: the random nth-price auction (13. Training consisted of instructions, examples, a two-round practice auction, and a postpractice quiz on the nth-price auction mechanism. After gaining familiarity with the nth-price auction, participants were provided with one of four randomly assigned information treatments: (a) an agribusiness (pro-biotech) perspective on GM consisting primarily of positive statements about it, (b) an environmental (anti-biotech) perspective on GM consisting primarily of negative statements about it, (c) both the pro- and anti-biotech perspectives, and (d) the pro- and anti-biotech perspectives with verifiable factual information on GM from independent sources. Each perspective was limited to a single standard page size and organized in a common fashion. A key reference point in the experiments was a fifth information "treatment" that was empty or did not contain any information. After participants took a few minutes to digest the information treatment assigned to them, the auction began. In each of the four rounds of bidding, three products were offered for sale: one pound of broccoli, one pound of beefsteak tomatoes, and five pounds of russet potatoes—each in plain packaging. Products in each round bore a label indicating the type of commodity and the product weight. We refer to a label treatment consisting of only these statements as a "plain label." Two additional labeling statements were injected into two of the rounds: "intragenic GM" and "transgenic GM," both with the additional statement: "Enhanced levels of antioxidants and vitamin C."4 The ordering of food labels across sessions was randomized. After completion of all bidding rounds, the binding round was drawn, bids were posted and ranked on a whiteboard in the front of the lab (no bids were posted prior to this point), the random n was drawn to determine the clearing price, and winners were identified. All participants were then asked to complete a short exit questionnaire. Nonwinners were told that they were free to leave, and winners were told to go to an adjacent room to complete their purchases, exchanging money for goods. Given the incomplete regulatory status of the intragenic foods, we were unable to obtain the product-enhanced GM fresh vegetables to deliver to winners. As an alternative, winners were given plain labeled food products, similar to procedures followed by others in similar circumstances, such as 1, 14, and 2. Receptiveness to the experiments was positive, and no complaint from participants was received. Table 1 presents the percent difference in bid prices across information treatments, averaged over individuals and commodities (broccoli, tomato, and potato), and between intragenic and transgenic labels with enhanced nutritional attributes and a plain label conventional alternative. Several key results emerge from the bid price data. Averaged over all information treatments, consumers are willing to pay a premium for both intragenic (25%) and transgenic (5%) labels with enhanced vitamin C and antioxidant content relative to a conventional plain label alternative. This indicates that despite consumers' well-documented perceptions of GM foods as being weakly inferior to non-GM foods (see 9 for a review and meta-analysis), the positive attribute of enhanced nutrition mitigates this negative valuation, resulting in a willingness to pay a premium for products with a GM label. This is the first evidence that U.S. consumers are willing to pay a premium for an intragenic GM labeled food product relative to a conventional alternative. Moreover, this result opens the door to the possibility that the food industry in the United States may have an incentive to voluntarily label an intragenic GM food product as intragenic GM. While the bid price data in table 1 shows that GM foods with enhanced nutrition have the potential to capture a position in the food market, we can shed more light on bidding behavior. First, consistent with the findings of 8 and 12, information injected into the experiments affects bidding behavior. Second, consider the mean difference in bid prices for a product with an intragenic enhanced-nutrition GM label compared with the bid prices for a similar product with a plain food label. Participants who did not receive any GM information were willing to pay a premium of 31%. The premium jumped to 63% when participants received pro-biotech information. When they received anti-biotech information, the GM label was discounted 12%. When participants received pro- and anti-biotech information or all three types of information, the premium was about 19%. Second, the bid prices for a product with a transgenic enhanced-nutrients GM label relative to those for a similar product with a plain food label were all somewhat lower than for the intragenic food label—premiums were lower and discounts larger. To reveal additional information about participants' bidding behavior, we summarize outcomes on bid price rankings in tables 2 and 3. Table 2 displays the percentage of individuals who, based upon their bid prices, preferred the intragenic GM label with enhanced nutrition over the conventional plain labeled product (i.e., those for which BidIntra>BidPlain), those who preferred the plain labeled product (BidIntra<BidPlain), and those who were indifferent between the two (BidIntra=BidPlain). Table 3 presents similar results for the transgenic GM label with enhanced nutrition. Across all information treatments (the first line of table 2), approximately half (48%) of the sample preferred intragenic GM labeled commodities to plain label alternatives, 20% preferred the plain label alternative, and 32% were indifferent. This indicates that across the information treatments, approximately 80% of the sample viewed the intragenic GM label with enhanced nutrition as weakly superior to a conventional label. When participants received the pro- and anti-biotech perspectives and third party information, half of the sample preferred the intragenic GM label, with the remaining half of the sample being fairly evenly divided between preferring the plain label (27%) and being indifferent between the two (23%). This contrasts significantly with the findings for relative preferences between the transgenic GM label with enhanced nutrition and the plain label (table 3), where only about a quarter (29%) preferred the transgenic GM label and 50% preferred the plain label. Finally, we provide some information about how participants' attributes affected their preferences for GM versus a plain labeled product. Two ordered probit models are estimated (one for intragenic GM and one for transgenic GM). For each model, the dependent variable was coded as counts denoting which label was preferred or whether the individual was indifferent. For brevity, in lieu of presenting coefficient estimates which are difficult to interpret (5), table 4 reports the marginal effects, predicted probabilities, and significance levels from the two models of consumers' preference relationships.5 Individuals who indicated that they frequently read food labels were less likely to prefer either the intragenic (13.8%) or transgenic (9.9%) labels and more likely to be indifferent or prefer the plain labeled products. Conversely, individuals who engaged in regular physical exercise were more likely to prefer the intragenic and transgenic labels (23.1% and 15.7%, respectively) and less likely to be indifferent or prefer the plain labeled product. Individuals who entered the experiments with a positive opinion of GM were less likely to prefer intragenic GM and more likely to be indifferent or prefer the plain labeled product, but were more likely to prefer transgenic GM and less likely to be indifferent or to prefer the plain labeled product. While the controversy over benefits, costs, and hazards of genetically modified foods continues to unfold in the arenas of global politics and public information campaigns, the advancements in intragenic bioengineering present a new piece to the puzzle. Overall, we find in our experiments that consumers do value enhanced nutrition (antioxidants and vitamin C) obtained through GM. But consumers are more accepting of foods with enhanced nutrition obtained through intragenics compared with transgenics. Most notably, we find that consumers are willing to pay more for intragenic labels with enhanced nutrition compared with conventional plain labels. This opens the door for voluntary private sector labeling of GM foods with enhanced nutrition. These results pose a dilemma for individuals and groups that have historically taken a position of staunch opposition to GM. While intragenics may present a more palatable form of bioengineering compared with transgenics, our laboratory experiments indicate that there is the potential for an even greater crowding out of non-GM foods. Although our findings present a somewhat positive picture for the potential of intragenics to obtain a foothold in the food market, they also suggest that information injected into the public domain will continue to play an important role in determining consumer acceptance. Our findings reveal that the information available to consumers when making purchase decisions has a significant effect on relative valuations for GM and non-GM labels. While pro-biotechnology information disseminated by agribusiness in isolation has significant positive effects on consumer valuations for GM labels, this effect is reduced when anti-biotechnology information is simultaneously injected into the market.