An Empirical Analysis of Producer Perceptions of Traceability in Organic Agriculture
Edmund M. Tavernier
Edouard Mafoua
Rutgers University
Paper presented at the 43rd Annual Meeting of the Southern Regional Science Association, Le Pavillion Hotel, New Orleans, LA March 11-13.
ABSTRACT: This study assesses producer perceptions of traceability issues in organic agriculture using survey data from the 2001 National Agricultural, Food and Public Policy Preference Survey. The survey provides food policy and socio-economic data that facilitates the examination of traceability issues in organic agriculture. The logistic regression models that are used to examine the relationship between traceability and organic agriculture are robust across models. The models indicate a decreased likelihood for federal government intervention on the part of organic producers for policies that improve traceability from consumer back to producer to improve food safety and tracking. However, the results suggest that organic producers with sales under $50,000 are more likely to indicate some willingness for government intervention. The analysis also shows that producers want food products made with biotechnology to be labeled if there is a scientifically determined difference in the product. While education may be an avenue to facilitate greater acceptance of traceability issues, the results indicate a negative relationship between the acceptance of traceability and farmers with a bachelor’s degree. This finding is surprising given the on-going debate surrounding biotech foods and traceability. Clearly the reluctance on the part of producers with bachelor’s degrees to embrace tracing could hinder such efforts if those producers assume leadership positions in their communities. Perhaps such results are a reflection of the confidence that those producers have in the quality of organic produce.
An Empirical Analysis of Producer Perceptions of Traceability in Organic Agriculture
Introduction
New challenges to the U.S. food supply have raised concerns over food safety. Two of the most important challenges are the increasing number of new food pathogens and the advent of genetically modified (GM) foods. To combat these challenges, mechanisms such as traceability have received increased attention in the literature.
Several definitions of traceability exist in the literature1. In this study traceability is defined as a process that requires documentation tracking foods from producers to consumers. This practice is commonly followed in organic farming through extensive documentation along the food supply chain. While it is unlikely that foods produced under such a system would be contaminated given the conditions of production in such a framework, the issue of traceability has been discussed in the context of genetically modified (GM) foods. For example, the European Commission has passed two new regulations to deal with the labeling and traceability of GM organisms2,3. The regulations address both food and animal feed and are intended to provide information to the consumer, to ensure transparency of GM ingredients in the food chain and to encourage the unblocking of an (unofficial) moratorium on GM crops4.
The issue of traceability has also been driven by food safety concerns such as scares like the human form of Bovine Spongiform Encephalopathy (mad cow disease), bioterrorism and contamination of the food supply. On Dec. 14, 2000, for example, Cargill Corp. announced a recall of 16.7 million pounds of turkey and chicken products because of contamination with listeria, a microbe that can cause high fever, severe headaches, neck stiffness, and sometimes death. More recently, Pilgrim’s Pride, the parent company of Wampler Foods announced a nationwide recall of 27 million pounds of turkey and chicken products after tests found listeria in the drain. This recent outbreak was blamed for at least 7 deaths and 39 illnesses in Pennsylvania, New Jersey and the surrounding states5. Traceability issues in other areas of the food supply chain are beginning to come under closer scrutiny from researchers.
Deasy6 argues that the dairy industry is under pressure to improve product security, implement efficient risk management and rapid response capabilities, and manage quality “from trough to table” to achieve full, verifiable traceability. Mousavi et al.7 state that the meat industry is seeking to establish reassurance on traceability and production techniques that may help promote confidence in the integrity and origin of their products. Liddell and Bailey8 demonstrate that the U.S. pork industry lags its major international competitors and major international customers in terms of developing programs for traceability, transparency, and assurance. In light of these concerns baby boomers are driving the demand for food products generated by alternative agriculture practices.
The National Research Council9 defines alternative agriculture as a system of food or fiber production that systematically pursues a (i) more thorough incorporation of natural processes such as nutrient cycling, nitrogen fixation, and beneficial pest-predator relationships into the agricultural production process, (ii) reduction in the use of off-farm inputs with the greatest potential to harm the environment or the health of farmers and consumers, (iii) productive use of the biological and genetic potential of plant and animal species, (iv) improvement in the match between cropping patterns and the productive potential and physical limitations of agricultural lands, and (v) profitable and efficient production with emphasis on improved farm management, prevention of animal disease, optimal integration of livestock and cropping enterprises, and conservation of soil, water, energy, and biological resources. These goals contain the same elements that form the framework for organic agriculture.
The National Organic Standards Board defines organic agriculture as an “ecological production management system that promotes and enhances biodiversity, biological cycles, and soil biological activity. The system is based on minimal use of off-farm inputs and on management practices that restore, maintain and enhance ecological harmony”10. The National Organic Program (NOP) Final Rule issued in December 2000 after a decade of deliberations and was fully implemented on October 21, 2002. Those who grow or market organic products must comply with the rule. Under the USDA’s new NOP rules, all organically labeled products must conform to nationally set standards. Products carrying the “Organic” label must be produced without hormones, antibiotics, pesticides, synthetic fertilizers or genetic modification. The rule uses different terms to identify the levels of organic content.
For example, products carrying the “100% Organic” label cannot contain any non-organic ingredients while products with the “Organic” label must contain ingredients that are at least 95% organic. The “Made With Organic Ingredients” label must contain at least 70% organic ingredients and may not contain any added sulfites11. Govindasamy et al.12 find that the use of labels to identify products as certified organic has a positive effect on increasing sales from organic production and on the number of farmers being satisfied with their returns from organic production.
While labeling provides valuable information for consumers wishing to avoid non-organic products, it says little about the tracking mechanism of food products from producer to consumer. Whether such tracking is even necessary in an organic production food framework is an empirical question. Moreover, government agencies have introduced the science-based food safety program – Hazard Analysis and Critical Control Point (HACCP) to address new challenges to the food supply such as Escherichia coli and Salmonella enteritidis13. Further studies are also currently being undertaken to develop Organic HACCP recommendations for improved procedures for securing consumer oriented food quality of certified organic foods from “plough to plate”14. These studies will likely augment a Food and Agriculture Organization report that concludes organic farming potentially reduces the risk of E. coli infection15. Given the concerns about food safety, research examining traceability issues in organic agriculture on the producer side is clearly needed to inform the debate. This paper uses New Jersey producers as a case study to inform such debate.
The paper is organized as follows. In the next section, the organic foods market is examined in some detail. This section is followed by the data source and profile of agricultural producers used in the paper. Next, the results from the empirical analysis on traceability issues in organic agriculture are presented. This section is followed by a discussion of the results.
The Organic Foods Market
The organic foods market has experienced tremendous growth in recent years. Farm sales of organic produce in the U.S. have grown from $78 million in 1980 to about $7.8 billion in 200016,17. Retail sales of organic foods in the U.S. are projected to be $11 billion to $13 billion in 200318. The increase in sales and the continued expansion of sales of organic foods are driven in part by consumer concerns for food safety, animal welfare, and the environmental and ecological impact of food production and agro-processing19. These concerns have led to increasing investments in farm, agribusiness, and food marketing in alternatively produced products and suggest a responsiveness to the growing interest in organic foods20. This interest has led to a doubling of certified organic cropland in the US from 1992 to 1997 to 1.3 million acres17 and is projected to increase to 950,000 hectares (2.3 million acres) in 200318. The growing consumer demand has also led to the introduction of over 800 new organic products in the first half of 2000 (mainly desserts) and the introduction of new organic beverage products in 199921.
The increase availability of organic produce in supermarkets is likely to further spur the growth of organic foods. Consumers indicate that they are more likely to buy organic produce if they are readily available in such outlets22. This willingness has facilitated 20 percent growth in organic foods per year since 1990 and has increased the availability of organic products in nearly 20,000 natural food stores and 73 percent of all conventional grocery stores17. As the growth of organic foods continues to increase issues related to health will likely surface and traceability issues in organic foods may assume greater prominence. For example, the main criterion for organic acceptability of materials in organic management is that natural materials are allowed and synthetic materials are prohibited23. Thus meat, poultry and dairy labeled organic must come from creatures that are raised on organic grains and grasses, given access to the outdoors and fed no growth hormones or antibiotics.
Data Source and Producer Profile
The data for this study were obtained from the Agricultural, Food, and Public Policy Preference Survey conducted in May 200124. The study was undertaken to provide input in the 2002 Farm Bill process since The Federal Agricultural Improvement and Reform Act of 1996 that contains language authorizing federal farm programs, expired in 2002. The mail only survey was supported by the National Public Policy Education Committee comprised of faculty from land grant universities, Farm Foundation and statisticians from each state’s agricultural statistics service. Each state surveyed producers using a questionnaire composed of national questions, selected optional questions, and state-specific questions. The National Agricultural Statistics Service was responsible for the selection of the sample, printing and mailing of the questionnaires, and telephone follow up and data collection activities. The data collection effort consisted of first and second mailings of questionnaires. The questionnaires were sent to Kansas State University for data entry and aggregate analysis.
Twenty-seven states, including New Jersey, participated in the national survey. The New Jersey part of the national effort began in April 2001. During that month, a master survey consisting of questions to be asked of agricultural producers in all participating states was sent to Rutgers University from Kansas State University with the opportunity to include state specific questions. Questions were sought from the agricultural community and persons associated with that community such as Agricultural Agents and Agricultural Specialists. Their input was incorporated in the survey.
The New Jersey Agricultural Statistics Service chose a random sample of 631 farmers for participation in the survey. One hundred and forty-four of the surveys were returned giving a response rate of 23%. This rate matches the average response rate for the northeast and is above the national response rate of 20%. The participating states in the northeast include Maryland, New York, Pennsylvania and New Jersey.
Table 1 presents summary statistics for demographic and structural variables of the 102 observations used in the study. The demographic variables are education and sales, while the structural variables are organic agriculture, biotechnology, animal waste and traceability.
The demographic variables are defined as follows:
§ education represents the last year of school completed by the principal operator;
§ sales represent annual gross sales.
The structural variables are defined as follows:
§ organic agriculture is a priority area identified by farmers for research and extension education in the northeast. The analysis proceeds on the assumption that if farmers identify organic agriculture as a priority area for research and extension education they are more likely to be engaged in or are already engaged in alternative agricultural practices. This variable is used as a proxy for organic agriculture producers.
§ animal waste represents the provision of financial incentives by the federal government to facilitate the management of animal waste;
§ biotechnology represents the labeling of food products when there is a scientifically determined difference in the product; and
§ traceability represents the role of the federal government to increase efforts to improve traceability.
The demographic variable with the highest mean is the number of farmers indicating that their annual gross sales from farming in recent years, including government program benefits is less than $10,000. The mean value of that variable is 0.46, which, because of the construction of the variable may also be interpreted as a percentage. Thus 46% of farmers receive less than $10,000 from farm sales. Twenty-five percent and 9% of the producers have annual gross sales of $50,000 to $99,999 and $10,000 to $49,999, respectively.
A greater percentage of respondents have bachelor’s degree than any other education qualification. The mean of respondents with a bachelor’s degree is 0.29 compared to 0.20 of respondents indicating that they have an advance degree. Taken together, Table 1 suggests that 75% of respondents have some type of college education.
The traceability variable has a mean of 0.78 in the structural variables. Thus 78% of respondents would like the federal government to increase efforts to improve traceability from consumer back to producer in an effort to improve food safety and tracking. The high percentage suggests that transitioning to a traceability environment may not be a difficult task for organic producers. This finding is important because in the alternative agricultural framework, natural materials, in particular animal waste materials are the preferred source of materials used in organic agriculture. Perhaps it is not surprising then, that 66% of respondents would like the federal government to provide financial incentives to encourage the provision of the environmental benefits that the proper management of animal waste provides. Forty one percent of respondents believe that organic agriculture should be a priority area for research and extension education in the northeast.
Biotechnology has the highest mean in the structural variables. Ninety percent of respondents indicate that food products made with biotechnology should be labeled if there is a scientifically determined difference in the product. This figure is unusually high considering the anecdotal evidence that suggests that U.S. farmers are opposed to the labeling.
While the percentage of respondents answering in the affirmative may be high for certain categories, the percentage that prefers a certain action within that category may be low. For example, although 78% of respondents indicate that the federal government should increase efforts to improve traceability from consumer back to producer to improve food safety and tracking, a one-to-one mapping from organic agriculture to traceability suggests that only 26% of respondents who indicate a preference for organic agriculture also indicate such preference for traceability (Table 2). In the sales category, most respondents who indicate a preference for alternative agriculture have annual sales of less than $10,000. The respondents were not asked whether they were organic food producers. However, in a recent survey most organic producers in Pennsylvania, New York and New Jersey described themselves as small farmers12. In that regard there appears to be an inverse relationship between farm size as measured by sales and organic agriculture; i.e. the greater the sales from farming the less likely are respondents to indicate a preference for alternative agriculture.
Results of Empirical Analysis
The results of the empirical analysis are presented in this section. The analysis seeks to test the hypothesis that there exists a relationship between organic agriculture producers and traceability. The dependent variable is traceability. The variables used in the empirical analysis are defined in Table 3. The results are obtained using the logit methodology and are presented in Table 4. In addition to the results, Table 4 also presents goodness of fit measures such as the Chi-square test statistic, the Mc Fadden R2 test statistic, and the percent of successful predictions.
The Chi-square test statistic tests the null hypothesis that the coefficients of all the independent variables equal zero. The null hypothesis is rejected at the significant level 0.0001 and better, indicating that the model has significant explanatory power. The Mc Fadden R2 values are 0.18 and 0.32 and also indicate an extremely good fit for the estimated models25. This statistic is generally low in binary dependent variable models estimated with cross-section data26. The percent of correct predictions is also very high. This measure predicts whether or not an event will occur given a set of explanatory variables27. The models correctly predict 83.3% to 87.3% of the responses.
The variables included in the models are chosen for their theoretical and heuristic relationship reported in previous studies. For example, the organic acceptability of materials requires the use of non-synthetic materials such as animal waste. Thus the animal waste variable captures the relationship between traceability and that form of soil enhancement.
Model 1 has a significance level of 0.0001, the Mc Fadden R2 is 0.18 and the model correctly predicts 83.3% of the responses. The estimated coefficients are all significant at 5% level or better. There is a negative relationship between traceability and organic agriculture and a positive relationship between traceability and waste and biotech. In the case of organic agriculture, the results suggest that producers who view organic agriculture as a high priority area for research and extension are 26 % less likely to want the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking.
With respect to biotechnology, the results of this research suggest that producers who want food products made with biotechnology to be labeled if there is a scientifically determined difference in the product are 32% more likely to want the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking.
With respect to the animal waste variable, the results suggest that producers who want the federal government to provide financial incentives to encourage the provision of the environmental benefits afforded by the management of animal waste are 18% more likely to want the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking. This result appears contradictory when one considers the negative relationship between organic agriculture and traceability. However, producers, while cognizant of the absence of pesticides on organic produce may be concerned about possible pathogens in animal waste.
Model 2 incorporates the structural variables with socio-economic and demographic characteristics. The model has a significance level of 0.00001 the Mc Fadden R2 is 0.31 and the model correctly predicts 87.3% of the responses. The structural variables are robust across models 1 and 2, and there are statistically significant coefficients in the sales and education categories.
The results suggest that there exists a positive relationship between farmers whose gross sales are less than $10,000 and farmers whose gross sales are between $10,000 and $49,999, and traceability. The analysis indicates that farmers with gross sales of less than $10,000 are 19% more likely to indicate that the federal government should increase efforts to improve traceability from consumer back to producer to improve food safety and tracking. Farmers whose gross sales are between $10,000 and $49,999 are 28% more likely to indicate that the federal government should increase efforts to improve traceability from consumer back to producer to improve food safety and tracking. The variable that measures farmers with sales between $50,000 and $99,999 is not significant.
In the education category, the results suggest that producers with a bachelor’s degree are 28% less likely to want the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking. This finding may reflect a level of confidence, regarding the absence of chemicals in organic produce, on the part of such producers.
The results from model 2 seem to suggest that farm size, in terms of sales, matters in ones perception regarding whether the federal government should increase efforts to improve traceability. However, the model does not address whether “organic” producers with small farms are more likely to indicate a preference for traceability. To address that issue, interaction variables were created between the sales and organic agriculture variables. The results of that analysis are presented in model 3.
Model 3 has a significance level of 0.00001, the Mc Fadden R2 is 0.32 and the model correctly predicts 86.3% of the responses. The structural variables are robust across the three models and the education variable is robust across Models 2 and 3. The results suggest that “organic producers” with annual sales under $10,000 are 27% more likely to want the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking. However “organic producers” with sales between $10,000 and $49,999 are 39% more likely to indicate such a preference.
Discussion
Logistic regression models are used to examine the relationship between traceability and organic agriculture. The coefficient of organic agriculture is robust and significant at the 1% level across the models and indicates that organic producers are less likely to want the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking. This result may be a reflection of the belief that organic food products are healthier than products grown from conventional agriculture. Sixty-six percent of consumers in a Hartman Group survey cited health as a “top motivator” for purchasing organic products28. However, the USDA organic label does not imply that organic foods are safer or better than other kinds of food.
The results suggest that organic producers with sales under $50,000 are more likely to indicate willingness for such a preference and illustrate the importance of examining organic producers of different farm size categories when addressing issues of traceability. This finding is particularly important because the arguments on both sides of the traceability issue have been driven primarily by the food safety concerns of consumers and have largely ignored the perceptions of producers. The perceptions are important for they provide information that may help guide an education campaign on organic agriculture.
In addition to organic agriculture, the analysis provides producer perceptions on the relationship between animal waste management, biotechnology, sales and education on traceability. The results suggest that producers who want the federal government to provide financial incentives to foster the environmental benefits of animal waste are more likely to want the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking. This finding is robust and significant at the 5% level across the models. Whether this finding is a recognition on the part of producers that animal waste may contain deadly pathogens is not clear. However, the analysis indicates that producers who want the federal government to provide financial incentives to foster the environmental benefits of animal waste are 18% more likely to want the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking.
The labeling of foods containing biotechnology has generated significant debate. Michael J. Phillips, executive director for food and agriculture at the Biotechnology Industry Organization argues that the labeling of biotech foods is not only unnecessary, it would actually confuse consumers who might think that the process of biotechnology might in of itself have an impact on food safety29. Oregon’s Measure 27 that would require the labeling of all foods and food additives that have been genetically engineered was defeated by over 70% of Oregon’s voters this past November30. While the debate continues, the uncertainty surrounding issues of biotechnology are perhaps captured in the willingness of producers to have biotech products tracked. Such a position seems to converge with that of consumers who have expressed some willingness for such a preference31. The analysis indicates that producers who want food products made with biotechnology to be labeled if there is a scientifically determined difference in the product are more likely to want the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking. This finding may help insulate producers from the charge of indifference regarding issues of biotechnology by consumers.
While organic producers with sales under $50,000 are more likely to indicate a willingness for the federal government to increase efforts to improve traceability from consumer back to producer to improve food safety and tracking, organic producers overall, indicate a differing preference. The results do lend support to earlier findings that organic producers are likely to be small farmers and may signal a divergence between small and large farmers, that is farmers with sales in excess of $50,000, with respect to traceability and organic agriculture. This divergence may create increased difficulty for greater acceptability on traceability issues in organic agriculture if large farms that invariably have the most resources are “not on board” the move to such efforts.
One avenue that could facilitate greater acceptability of traceability issues is education. However, the results indicate a negative relationship between traceability and farmers with a bachelor’s degree. This finding is surprising given the on-going debate surrounding biotech foods and traceability. Clearly the reluctance on the part of producers with bachelor’s degrees to embrace tracking could hinder such efforts if those producers assume leadership positions in their communities. Perhaps the results are a reflection of the confidence that such producers have in the quality of organic produce. Whether that food producing sector continues to enjoy that status and its significant growth may hinge on continuing to assure consumers of the wholesomeness of organic foods. Traceability affords one avenue to achieve that goal and the results suggest that small organic producers are willing to embrace the concept.
The authors are, respectively, Associate Professor, Department of Agricultural Food and Resource Economics, and Post-Doctoral Associate, Food Policy Institute Cook College, Rutgers University, 55 Dudley Road, New Brunswick, NJ 08901.
