The Cost of a GMO-Free Market Basket of Food in the United States
Barry K. Goodwin, Michele C. Marra, and Nicholas E. Piggott
North Carolina State University
We examine the consumer cost consequences of choosing GMO-free food over food that contains GMOs. Using text-mining algorithms applied to detailed product descriptions contained in a proprietary database of individual GMO and GMO-free foods at the retail level, we find that, when directly compared item by item, GMO-free food costs an average of 33% more than a comparable food item that is not GMO-free. When compared on a per-ounce basis, GMO-free foods cost an average of 73% more. Generalizing to the cost of a typical market basket of food consumed by American households, GMO-free food consumption would increase the average family food budget from $9,462 to $12,181 per year.
Key words: biotechnology, consumer food choice, GMO-free, market basket.Technological progress in the production of foods and fiber has led to unprecedented growth in the productivity of agriculture. The US Department of Agriculture (USDA) reports that total US agricultural output grew at an average annual rate of 1.49% between 1948 and 2011 while input use only grew at 0.07% per year (USDA Economic Research Service [ERS], 2015a). There are many reasons for this impressive growth, including improvements in cropping practices, input qualities, resource management, selective breeding, and other widespread innovations in production practices. One important innovation that many believe has increased productivity is the genetic modification of crops in order to achieve increased output, higher quality, or lower production costs. According to the USDA, 90% of corn, 93% of soybeans, and 90% of cotton planted in 2013 in the United States were genetically modified (USDA ERS, 2015b).
Technological improvements have increased the overall quality and variety of the US food supply while, at the same time, lowering overall food costs. The share of disposable personal income spent on food at home fell from 21.2% in 1930 to 5.7% in 2012 (USDA ERS, 2015c). However, these technological advances have not been viewed as positive by all consumers. In particular, despite scientific evidence to the contrary, skepticism and suspicion regarding the safety and quality of genetically modified organisms (GMOs) exists among many consumers, especially outside of the United States. This has led to efforts to legislate labeling of any foods containing GMOs (e.g., Proposition 37, which was defeated by California voters in 2012 and Vermont H.112, which was signed into law on May 8, 2014). To date, three state legislatures have passed mandatory GMO labeling laws and two states have passed laws against GMO labeling (Center for Food Safety, 2015). Similar labeling laws have existed in the EU for many years.
While there seems to be little lingering doubt about the yield increases brought about by GM crops with insect resistance, much less is known about the extent to which consumer concern about GMOs translates into price and food expenditure effects. A number of studies have evaluated consumers’ willingness to pay for GMO-free food. Studies by Huffman (2010), Bukenya and Wright (2007), and Tegene, Huffman, Rousu, and Shogren (2003) find that US consumers are willing to pay premiums ranging from 14% to 21% for food certified to be GMO-free. Lusk, Roosen, and Fox (2003) found that US consumers were willing to pay an additional $2.83 to $3.31 per pound for beef that was not fed GMO ingredients. They also found that analogous premiums in Europe ranged from $4.86 to $11.01. Research has also documented that the information that consumers have about GMO foods heavily influences their willingness to pay. For example, Lusk et al. (2003) found that a lack of knowledge about GMOs significantly increased a consumer’s stated willingness to pay for GMO-free foods. Such willingness-to-pay studies are also widely recognized to have a number of biases that can result in stated values far exceeding what consumers actually pay. The segregation and identity preservation needed to ensure food ingredients remain GMO-free from the farm gate to the retail store are also likely to be substantial. Such costs depend upon tolerance levels and the degree of regulation entailed.
Many retail outlets already offer foods that are certified to be GMO-free. The market share of GMO-free foods is modest, but some retailers are identifying such products in their in-house brands. For example, the Whole Foods supermarket chain recently announced a commitment to complete labeling of all foods containing GMO ingredients. Fernandez-Cornejo, Wechsler, Livingston, and Mitchell (2014) report that of the 7,637 new food products introduced between February 12, 2010 and February 11, 2011, 2.6% advertised that they were GMO-free and 8% advertised that they were organic.
To our knowledge, no existing research has considered the cost implications of adopting a totally GMO-free diet for a typical family. We attempt to fill this void by considering the composition of the typical US household’s food bill and the prominence of GMO ingredients across the diet. To this end, we utilize data from the US Department of Labor’s Bureau of Labor Statistics (BLS) on the market basket weights used in calculating the consumer price index (CPI) and the composition of the average household’s annual food bill that is reported in the Consumer Expenditure Survey. The CPI uses expenditure weights calculated from surveys of about 7,000 families per year and collects detailed purchase data for over 200 item categories. We use the latest market basket weights (2007-2008) and Consumer Expenditure Survey (2011) reported by the BLS.
We also consider consumption of the various food items that are likely to contain GMO ingredients, either directly or as an animal feed. This is made possible by the US Environmental Protection Agency’s (EPA) Food Commodity Intake Database (FCID; US EPA, 2014), which is comprised of data taken from the National Health and Nutrition Examination Survey (NHANES) of the Centers for Disease Control (CDC) and USDA, as well as the FCID recipe database, which derives the consumption of raw crop and livestock commodities from the dietary patterns reported in the NHANES. While the principal goal of the FCID is to monitor dietary exposure to pesticides, it also provides a detailed measure of the daily consumption of the base commodities (e.g., corn, beef, etc.) that are ingredients in the US food supply.
As we have noted, a variety of empirical studies have attempted to infer the price differences of existing GM commodities and GMO-free alternatives. Nearly all of this research has been done for broad commodity categories. For example, Barrows, Sexton, and Zilberman (2014) find that the adoption of GM corn lowered corn prices by 13% while adoption of GM soybeans lowered prices by 2%-65%. It is difficult to extend these aggregate price impacts through the marketing chain to infer how the cost of the typical grocery basket was impacted by GM crops. To reinforce these aggregate estimates, we adopt a unique approach that is empirical but largely anecdotal. We use proprietary grocery item pricing for conventional and certified GMO-free food items. In particular, we utilize market research data collected by Mintel from retail outlets over the preceding 12-month period for bakery, dairy, and snack food items. The Mintel data contain unit prices and detailed product descriptions. From this extensive database, we use a text-matching algorithm to identify comparable GM-containing food items for each GMO-free certified item (Levenshtein, 1966). We utilize text-mining algorithms applied to detailed product descriptions to identify GMO-free items as well as the unit size (in ounces) of the item.
Table 1 presents detailed price comparisons for food items certified to be GMO-free and comparable food items that do not have such a certification. The price comparisons are made on a per-unit (i.e., package) basis as well as on a price per ounce of product. The product descriptions do not contain unit sizes in every case and thus the per-ounce comparisons are missing in some cases. It is important to note that although our matching algorithms derive the closest product matches possible, the products being compared may nevertheless differ in ways that we do not observe and thus that are not accounted for in the price differences. By comparing a large number of similar products, we hope to diminish any biases that would reflect product differences other than the GMO-free certification. That said, we also evaluate price differences that are based upon the relevant literature surveyed above.
Table 1. Price differences between comparable GMO-free certified and non-certified food products.
Across all of the product matches, a GMO-free certification raises prices by an average of 33%. This is quite similar to the commodity price differences identified in the empirical literature. An interesting observation is that, for those products for which we are able to define unit size, GMO-free certified products seem to be packaged in smaller units. If one considers pricing on a per-ounce basis, the GMO-free certification adds 73% to price. The fact that unit sizes are absent in many cases may suggest that the 33% price difference is more reliable.
Before proceeding to an evaluation of how these price differences translate into household expenditure differences, it is interesting to consider the prevalence of GMO-containing food items in the typical consumer’s diet. We utilized the EPA’s FCID database to determine the typical US consumer’s intake of commodities in processed and prepared food items. Table 2 contains the results of this survey of consumption patterns.
Table 2. Average and median daily intake (grams) for selected food commodities (based upon the Food Commodity Intake Database, the NHANES/What We Eat in America Survey Data, and the US EPA Office of Pesticide Programs).
We report both mean and median daily intakes, since it is not uncommon for certain segments of the population to avoid a specific commodity altogether (e.g., vegetarians will have no meat consumption). In the case of field corn, which is overwhelmingly comprised of GM corn, the typical individual consumes 76.5 grams of corn across a wide range of processed commodities. These commodities include ingredients such as corn flour, corn meal, corn syrup, and so forth. In the case of soybean products—a commodity that is nearly entirely GM—the average consumer has a daily intake of about 30.4 grams. The gram intake totals are of interest, but perhaps more enlightening is the broad range of food ingredients that contain the relevant agricultural commodities. In the case of meats, which use GM products as feed inputs, poultry is consumed the most, followed by beef products, and finally by pork. The median consumption of pork is quite low, reflecting the fact that a significant share of the US population does not consume pork products.
Though the quantitative measures of daily consumption of raw commodities is difficult to interpret in terms of the costs of a typical food basket, the FCID data do provide a detailed illustration of exactly how GM commodities are used in the US food supply. The breadth of food ingredients that contain GM commodities is impressive and serves to highlight the significant dietary changes that would be necessary to avoid consumption of GM commodities in the US diet.
Table 3 presents the latest average annual expenditures on food by US households. This is based on the 2011 Consumer Expenditure Survey of the BLS (US Department of Labor, 2014). The data are comprised of a survey of the population of 60.14 million US households. The average household consisted of 3.2 persons, 0.9 children, and had an annual gross income of $86,700. Total annual expenditures averaged $63,972, of which $8,315 went toward food. Of that amount, $4,944 was spent on food at home and $3,370 was spent on food away from home. Spending on broad categories of food items is included in Table 3.
Table 3. Expenditures on broad food categories for average household in 2011 (taken from the 2011 Consumer Expenditure Survey).
In order to decompose the total expenditures of $63,972 into specific food categories, we applied the 2007/08 CPI market basket weights to the total expenditures.1 Table 4 presents the detailed CPI weights and total expenditures on specific food categories. We have also included (where possible) the farm-to-retail value proportions. These are not used directly in our calculations but they do allow informal inferences regarding the degree to which farm price shocks might be reflected in retail prices. These data were collected from unpublished USDA sources (USDA ERS, 2014). We identify each food item category that is likely to have GMO ingredients. This includes cereal and bakery products, meat and poultry, dairy, beverages, prepared foods, and food away from home.
Table 4. Simulated impacts of a 33% price increase in foods containing GMOs on average household expenditures (based on the 2011 Consumer Expenditure Survey).
Of the total expenditures of $9,462 on food and beverages, $8,239 is spent on food items likely to contain GMO ingredients. We consider the impacts on the typical family’s food budget of a GMO-free diet. First, we consider the 33% price premium implied by our comparison of specific GMO-free certified food products and comparable conventional foods. In that case, the typical family’s food budget would increase from $9,462 to $12,181 each year. Of course, depending on household composition and consumption patterns, the price impacts could differ widely across households. We next consider the impacts of a modest 10% price premium for GMO-free food items. In that case, total expenditures would rise to $10,286 per year. In the case of a 20% price premium, which is the midpoint of the estimates reviewed by Barrows et al. (2014), total food expenditures would increase to $11,110. A 40% price premium would increase food expenditures to $12,758. Finally, if we apply the 73% price premium implied by our $/ounce price comparisons, total food expenditures would rise to $15,806.
Overall, our calculations suggest that the cost of a typical US family’s market basket of food would rise from 8% to 50% annually, depending on the impacts on retail prices from going to a GMO-free diet. To put this in perspective, consider a comparison of food spending in developed countries that regulate GMO ingredients to that of the United States. According to calculations presented by Civil Eats, the typical US family spends approximately 6.9% of its household budget on food at home as compared to 13.9% in France and 11.1% in Germany (Jones, 2011). Dietary differences beyond GMO regulation are likely reflected in these statistics, but it is likely that at least part of the budget differences reflect the higher costs associated with GMO-free foods.
In short, the budgetary implications of a GMO-free diet are substantial. GMO-free food items are shown to be more expensive than conventional alternatives. GMO ingredients play an important and ubiquitous role in the US food supply. Even small increases in the costs of these ingredients translate into significant impacts on the typical US household. Increased food costs would not only impact food consumption patterns but would also affect all classes of expenditures as limited income is redistributed across alternative consumption items.
1 Note that modest differences in expenditures in broad categories arise from applying the CPI weights to the 2011 expenditures. For example, the weights imply total food expenditures of $8,791 as compared to the CES total of $8,315. We utilize the CPI weights because of the significantly greater detail that they provide.
Barrows, G., Sexton, S., & Zilberman, D. (2014). Agricultural biotechnology: The promise and prospects of genetically modified crops. Journal of Economic Perspectives, 28(1), 99-120.
Bukenya, J.O., & Wright, N.R (2007). Determinants of consumer attitudes and purchase intentions with regard to genetically modified tomatoes. Agribusiness, 23(1), 117-30.
Center for Food Safety. (2015). Issues: GE food labeling [website]. Available on the World Wide Web: http://www.centerforfoodsafety.org/issues/976/ge-food-labeling.
Fernandez-Cornejo, J., Wechsler, S., Livingston, M., & Mitchell, L. (2014, February). Genetically engineered crops in the United States (USDA ERS Economic Research Report No. 162). Washington, DC: US Department of Agriculture (USDA), Economic Research Service (ERS).
Huffman, W.E. (2010). Consumer acceptance of genetically modified foods: Traits, labels and diverse information (Staff General Research Papers). Ames: Iowa State University, Department of Economics.
Jones, N. (2011, March 29). Mapping global food spending (infographic): How much of our spending goes toward food? Civil Eats. Available on the World Wide Web: http://civileats.com/2011/03/29/mapping-global-food-spending-infographic/.
Levenshtein, V.I. (1966). Binary codes capable of correcting deletions, insertions, and reversals. Soviet Physics Doklady, 10(8), 707-710.
Lusk, J.L., Roosen, J., & Fox, J. (2003). Demand for beef with growth hormones and fed genetic corn. American Journal of Agricultural Economics, 85(1), 16-29.
Tegene, A., Huffman, W., Rousu, M., & Shogren, J. (2003, April). The effects of information on consumer demand for biotech foods: Evidence from experimental auctions (Technical Bulletin No. 1903). Washington, DC: USDA ERS.
US Department of Agriculture (USDA), Economic Research Service (ERS). (2014). Price spreads from farm to consumer [database]. Washington, DC: Author. Available on the World Wide Web: http://ers.usda.gov/data-products/price-spreads-from-farm-to-consumer.aspx#.U3DoM_ldXK0.
USDA ERS. (2015a). Agricultural productivity in the U.S. [database]. Washington, DC: Author. Available on the World Wide Web: http://www.ers.usda.gov/data-products/agricultural-productivity-in-the-us.aspx#28247.
USDA ERS. (2015b). Recent trends in GE adoption. Washington, DC: Author. Available on the World Wide Web: http://www.ers.usda.gov/data-products/adoption-of-genetically-engineered-crops-in-the-us/recent-trends-in-ge-adoption.aspx#.U2-igPldXK0.
USDA ERS. (2015c). Food expenditures [database]. Washington, DC: Author. Available on the World Wide Web: http://www.ers.usda.gov/data-products/food-expenditures.aspx#.U2-tR_ldXK0.
US Department of Labor, Bureau of Labor Statistics. (2014). Consumer expenditure survey: Expenditure tables [database]. Washington, DC: Author. Available on the World Wide Web: http://www.bls.gov/cex/csxstnd.htm.
US Environmental Protection Agency. (2014). Food commodity intake database: What we eat in America [database]. Available on the World Wide Web: http://fcid.foodrisk.org/.
The authors have worked as consultants to Monsanto, Syngenta, and Bayer Crop Science. This research was supported in part by a grant from Monsanto.
Suggested citation: Goodwin, B.K., Marra, M.C., & Piggott, N.E. (2016). The cost of a GMO-free market basket of food in the United States. AgBioForum, 19(1), 25-33. Available on the World Wide Web: http://www.agbioforum.org.
|© 2016 AgBioForum | Design and support provided by Express Academic Services | Contact ABF: firstname.lastname@example.org|