This item is only available as the following downloads:
THE VA LUE OF ORGANIC WINES FROM TUSCANY A HEDONIC PRICE ANALYSIS By LANE ALEXANDER ABRABEN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2014
2014 Lane Alexander Abraben
To my grandmother
4 ACKNOWLEDGMENTS First, I would like to thank Dr. Kelly Grogan, Dr. Zhifeng Gao and Dr. Evy Mettepenningen I owe them a sincere debt of gratitude for their assistance, knowledge, support, and input. They were patient and forgiving and were a pleasure to work with. I want to thank Dr. Jeffrey Burkhardt for allowing me to pursue this degree, for guiding me through it, and for giving me one of the most exciting opportunities I have ever been given. I also want to thank Jessica Herman for guiding me through the logistical n ightmare of graduate school and for being a great friend while she did it. Dr. Andrew Schmitz, Dr. William Bomberger, and Dr. Richard Kilmer taught the most fascinating and useful classes in my program and I enjoyed every minute of them. Dr. Guido Van Huylenbroeck Dr. Mieke Calus, Dr. Marijke D'Haese Dr. Joost Dessein Dr. Jeroen Buysse Martine De Witte, Wim Van Hauwaert and the rest of the University of Gent faculty and staff gave me an education that will serve me for the rest of my l ife and an experience that I will never forget. I also must thank a group of amazing friends. Frank Sortino, Elliott Doolittle, Jared North, Anthony Mullersman, Andrew Coffey, Daniel Horak, Cam Tu Nguyen, Ariel Guerrero, Chopper Hodes, Richard Schatz, Jos h Cocozza, Shane Runyon, Steven Smith, John Young, HK Wallace, Kim Love, and Kate Boulos were not directly involved in this project, but their involvement in everything else has given me t he encouragement to keep going. My fellow classmates Matt Smith, Ana Deville, Jan Smitka, Margitta Minah, Diane Isis Nyob Kristina Sobekova Brock Carpenter, Elizabeth Baker, and Misti Sharp filled my time in graduate school, both at home and abroad, with fun, adventure, and humor.
5 I also must thank Dr. Alice Martin for a ll of her help. Her guidance has been tremendously valuable and without it this project would likely still be incomplete. Another thanks goes to Mike McElwee. He has been a teacher, a mentor, and most importantly a friend who has been sharing his wisdom insight, and humor with me for nearly 15 years. I want to thank Arthur, Manny, Roberta, and Keith for always being proud of me regardless of whether or not I deserved it. The same can be said for my grandmother Selma, who did not live long enough to see the completion of this thesis but was My sincerest gratitude goes to my family. To Linsday and Logan, the two best siblings in the world whose support, encouragement, and love means more to me than I could ever put into words. I owe a huge thanks to my father, an incredibly talented man who taught me how to appreciate the highest points in life and how to learn from the lowest ones. He has passed on to me the values of hard work, passion, and perseverance and has encouraged me to always be my best. I want to thank my brilliant mother, the most selfless, caring, and supportive woman I have ever met. Words cannot express how much I appreciate her for everything she has done for me, for ever ything she has sacrificed, and for everything she has taught me. All of my accomplishments, both personal and academic, are thanks to her unrelenting support and love. She is a truly amazing woman and I can only hope to one day be as good of a parent as she has been.
6 And Christine, the love of my life, my best friend, and the most wonderful companion I could possibly ask for. She encourag es me and support s me in every way and fills every day of my life with joy and excitement This thesis and everything else is for her.
7 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INT RODUCTION ................................ ................................ ................................ .... 14 Organic Agriculture ................................ ................................ ................................ 14 Organic Wine Policy in the US ................................ ................................ ................ 15 Organic Wine Policy in the EU ................................ ................................ ................ 16 Trade of Organic Wines Between the US and EU ................................ .................. 18 Organic Labeling ................................ ................................ ................................ ..... 18 Italian Consumers of Organic Products ................................ ................................ .. 20 US Consumers of Organic Products ................................ ................................ ....... 22 Organic Price Premiums ................................ ................................ ......................... 26 2 REVIEW OF LITERATURE ................................ ................................ .................... 29 General Attitudes Towards Organic Wines ................................ ............................. 29 Wine Price Hedonic Models ................................ ................................ .................... 31 Hedonic Price Models of Italian Wines ................................ ................................ ... 34 Hedonic Price Models of Organic Wines ................................ ................................ 35 Criticism of the Hedonic Price Model for Wine ................................ ........................ 35 Information Provision ................................ ................................ .............................. 37 3 DATA AND METHODS ................................ ................................ ........................... 39 Dataset ................................ ................................ ................................ ................... 39 Summary Statistics ................................ ................................ ................................ 40 4 RESULTS ................................ ................................ ................................ ............... 44 Models with Individual Ratings and Average Rating ................................ ............... 44 Price Segmented Models ................................ ................................ ........................ 47 Price Models wi th Interaction Terms ................................ ................................ ....... 50 5 DISCUSSION AND CONCLUSIONS ................................ ................................ ...... 52
8 The Italian Market ................................ ................................ ................................ ... 52 The US Market ................................ ................................ ................................ ........ 53 Comparisons Between Markets and Implications ................................ ................... 54 Limitations of this Study ................................ ................................ .......................... 55 LIST OF REFERENCES ................................ ................................ ............................... 57 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 61
9 LIST OF TABLES Table page 3 1 Mean Price in Each Market and Mean Rating ................................ .................... 40 3 2 Vintage Frequency ................................ ................................ ............................. 41 3 3 Frequencies Grape Varietal, Appellation of Origin, and Organic Status .......... 42 4 1 Effects of wine characteristics on US price with average and individual rating scores ................................ ................................ ................................ ................. 44 4 2 Effects of wine characteristics on Italian price with average and individual rating scores ................................ ................................ ................................ ....... 45 4 3 Effects of wine characteristics on price, by price quartile, in the US market ....... 48 4 4 Effects of wine characteristics on price by price quartile, in the Italian market .. 49 4 5 Models of US and Italian Price Including Interaction Terms ............................... 51
10 LIST OF FIGURES Figure page 1 1 Certification Logos. A) USDA organic logo, B) EU organic logo, C) biodynamic logo certified by Demeter ................................ ................................ 20
11 LIST OF ABBREVIATIONS AIAB CT Italian Association for Organic Agriculture Cellar Tracker DOC Controlled Designation of O rigin DOCG EOWC EU FAO Controlled Designation of Origin Guaranteed European Organic Winemaking Carta European Union Food and Agriculture Organization GMO ICEA IFOAM IGT MIPAAF NOP ORWINE PPM RP SO 2 ST USDA WE WS WTP Genetically Modified Organism Environmental and Ethical Certification Institute International Federation of Organic Agriculture Movements Indicazione Geografica Tipica Italian Ministry of Food, Agriculture, and Forestry National Organic Program Organic Viticulture and Winemaking Parts per Million Robert Parker Sulfur Dioxide Steven Tanzer United States Department of Agriculture Wine Enthusiast Wine Spectator Willingness To Pay
12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Science THE VALUE OF ORGANIC WINES FROM TUSCANY A HEDONIC PRICE ANALYSIS By Lane Alexander Abraben May 2014 Chair: Kelly A. Grogan Major: Food and Resource Economics (FRE) Previous literature suggests that the price of wine is a function of its intrinsic attributes. Studies have examined the price of wines from various wine producing regions using the hedonic price model to determine the correlation between wine attributes and price. However few studies have examined the relationship between organic production or certification and price. This study examines the price premiums associated with organic production and organic certification for Tuscan red wines produced between 1999 and 2008. To do this we modeled current Italian and US pric es as a function of organic status, grape varietal, appellation of origin, vintage, and rating. This study finds that grape varietal and vintage tend to have the most significant relationsh ip with price. Organic status, when accounting for other relevant determinants of price does not have any significant association with price in the Italian market. However, wines produced organically but are not certified as organic do receive a price p remium in the US market. Vintage showed a significant and negative correlation with price for the 2006, 2007, and 2008 vintages
13 The data were divided into price quartiles to examine the differences in effects of the variables between price classes. This analysis found that appellation and grape varietal have significant and substantial correlations with price in the highest price quartile
14 CHAPTER 1 INTRODUCTION Organic Agriculture Organic agriculture has expanded considerably in the last decade. From 1999 to 2011, the global amount of agricultural land devoted to organic production increased from 11 millions hectares to 37 million hectares and the market for organic products has increased from $15.2 billion to $5 9 billion (IFOAM EU Group, 2013) Organic agriculture is a sustainable ecosystem management practice that eliminates the use of synthetic external inputs such as pesticides, fertilizers, genetically modified organisms (GMOs), preservatives, and other synt hetic additives Organic production makes use of agronomic, biological, and mechanical methods to reach similar results as conventional agriculture. The goal of organic agriculture is to maintain long term soil fertility, biologic al cycles, and biodiversi ty (Food and Agriculture Organization 19 99). In the European Union, 5.6 % of agricultural land is under organic farming practices, while in the United States, adoption of organic agriculture is much lower, with less than 1% of cropland and pasture certifi ed as organic (Eurostat, 2013; United States Department of Agriculture 2013). Organic viticulture, otherwise known as organic winemaking, became however the practice of orga nic winemaking has benefits that are not strictly consumer driven The organic viticulture movement has two main tenets: to increase microbial activity in the soil by avoiding the addition of synthetic sub stances to the soil and the general concern for ov erall environmental protection through red uced soil erosion and pollution.
15 means of increasing production. France is considered the being the first government to officially recognize organic farming (Robinson, 1 999). In the five top wine producing countries in Europe ( France, Ita ly, Spain, Germany, and Austria), the share of vineyard land under organic production is between 6.8% and 9.6%, whereas in the United States, only 3.0% of vineyard land is organic (IFOAM & FiBL, 2013). 1 Organic Wine Policy in the US The formation of organic wine policy in the United Sta tes began in 1990 with the passing of the National Organic Foods Act, which gave the United States Department of Agriculture (USDA) the task of establishing regulations for organic foods and products. The USDA then created the Nation al Organics Standards Board, an advisory board wh ich along with the Bureau of Alcohol, Tobacco, and Firearms, drafted the regulation concerning the production and labeling of organic wine, which ultimately prohibited finished products, such as wine, from being labeled as organ National Organic Program (NOP) was established with the goal of creating guidelines for the processing and labeling of organic products, including wine. They also established the list of allowed and prohibited substances in organi c products. Included in this list is sulfur dioxide a common additive in winemaking It is formed from the burning of elemental sulfur in the presence of oxygen. Its use dates back hundreds of years but today its most important properties for winemak ing are to 1 These figures reflect overall grape production and are not limited to vineyards producing only wine grapes.
16 prevent oxidation, to prevent bacterial inoculation, and to encourage quick and clean fermentation (Robinson 1999). Some people experience severe allergic reactions to sulfur dioxide. Consequently, i n the United States, all wines must carry t he phrase Other countries have similar compulsory labeling rules for sulfites, however in the European Union, this requirement does not exist. Some people believe that sulfites contribute to headaches experienced after d rinking wine. However, the headaches are likely caused by histamines and amines found in the wine (Jarisch & Wantke, 1996). In the United States, the NOP allow s for wines to be labeled as organic and carry the USDA organic seal, only if they are produced from organically grown grapes without the addition of sulfites. Total naturally occurring sulfite levels cannot exceed 20 cannot bear the USDA organic seal on the label (Orga nic Consumers Assoc i ation, 2014). Organic Wine Policy in the EU While many wines produced in the European U nion were made from organically grown grapes (which fell under European organic food regulation established in 1991), regulations did not address t he organic production of wine from those grapes until recently This prohibited wine producers from being able to use the European organic logo on the label. Wines producers were only permitted to label wines with the phr ase In Europe, m any private standards for organic wine production existed and organic wine producers, if certified, were allowed to use the logos of the private certifiers on their labels. However these private labels are not always easily
17 recognizable to cons umers and little is often known about their standards (discussed further below). Therefore a common EU wide regulation fo r organic winemaking was needed so that producers could take advantage of the more widely recognized official EU organic label. In 20 04, discussions to develop these regulations began with the Organic Viticulture and Wine Making (ORWINE) Project, which presented a set of recommendations for organic wine production. In 2009, the European Commission began work on the implementation of or ganic wine regulations, a process that was hampered by disagreements over sulfite use Because wines produced without the addition of sulfur dioxide are prone to oxidation, bacterial inoculation, and therefore spoilage, organic wine producers in the EU still wanted to be able to add the compound to keep their wines from declining in quality. These disagreements caused the discussions to stall within a few months. In 2011, the European Organic Winemaking Carta (EOWC) and the European wing of the Interna tional Federation of Organic Agriculture Movements (IFOAM EU) helped the European Commission resume discussions, leading to an agreement on the sulfite issue and ultimately approval of the fir st EU organic wine regulation, allowing winemakers to use the of ficial EU organic logo. The regulations went into effect on August 1 2012. They allowed for different total sulfite levels based on the style of wine. Prior to the passing of the new regulations maximum acceptable sulfite levels in conventional wines were set between 150 400 mg/L depending on the style of wine. The new regulations modestly reduced these levels for organic wines by 30 50 mg/L 2 or by roughly 7 35% In addition to wines produced after August 1, 2012, wines 2 For sulfur dioxide, t he conversion rate is conveniently 1:1 for ppm to mg/L, but US and EU regulations use ppm and mg/L, respectively.
1 8 produced before that dat e which had not yet been bottled could also make use of the EU label as long as they met the new standards (IFOAM EU Group 2013). Trade of Organic Wines Between the US and EU With widely varying definitions of what is considered organic in terms of sulfit e use in winemaking a new United States/European Union Organic Equivalence Arrangement was needed to address how organic wines produced in the US could be sold in the European Union and visa versa. This new arrangement went into effect on June 1 st 2012. ovided that the sulfite levels comply with the limits set by the U.S. Organic Labeling Organic labels signal to consumers that a product has met organic production standards, subsequently allow ing pro ducers to receive price premiums as some consumers are willing to pay more for organic products ( McCluskey & Loureiro 2000 ) In order for producers to use an official organic logo, an accredited certifying agent must certify that the production and processing methods comply with the ap propriate organic regulations. In the United States, products that are certified to be organic by one of the (Figure 1 1) on their labels (United States Department of Agriculture, 2000) In addition to this label, if producers want to include other logos signifying additional environmental or fair trade attributes of a product, they can seek out supplementary certification. Producers may seek out carbon neutral certification, biodynamic certification, sustainable production
19 certification, or a host of other certification programs that are region or industry specific. Standards for these programs often go above and beyond official regulations set b y the USDA or other government agencies and their respective logos can be used to show that a produc t meets even stricter standards. In the European Union, producers of organic products (excluding unpackaged produce and imports) are required to include the EU organic logo (Figure 1 1) on all certified organic products. European producers can opt to undergo further certification and include additional logos on products as well. Many country or industry specific certification schemes allow for producers to use eco labels to convey information about production practices when regulations for specific industries do not exist (as was the case for organic wine) or when those production practices meet stricter standards than those set by the European Commission (E uropean Counsil, 2007) Recognition and knowledge of eco labels is an important aspect of their effectiveness. Prior to 2012 and in the absence of any EU level organic wine regulation, many private organic certification schemes and logos filled the need for certification and labeling of orga nic products. A recent study examined consumer knowledge of private eco label schemes such as Demeter (a global organic and biodynamic certifier) and ECOCERT (an organic certification organization with a presence in over 80 countries) as well as umbrella organizations like Bio Suisse (a group of over 30 individual farmers associations ) (Janssen and Hamm, 2011 ). The study used focus groups from the Czech Republic, Denmark, Germany, the UK, and Italy. They found that Italian consumers were unaware or lacked knowledge of the prevalent ecolabel schemes in their country, with the EU logo and the Controllo Biologico (a private inspection and
20 certification organization) logo known by only some of the participants O ther p rivate organic certifier logos such as Demeter, the Environmental and Ethical Certification Institute (ICEA), and the Italian Association for Organic Agriculture (AIAB) were known by very few or none of the participants. 3 Furthermore, nearly all of the It alian participants were unaware of the differences between the various eco labels. Most participants had very little knowledge of the standards or control systems of the ecolabel schemes and many were confused about the definition of organic, organic stan dards, and how production was regulated. Consequently, Italian participants had no preference for one organic certification scheme over another. This study was done before the introduction of mandatory EU organic logos on organic products, so further res earch is needed to examine consumer knowledge of EU organic standards under the new regulations. A B C Figure 1 1 Certification Logos A) USDA organic logo, B) EU organic logo, C) biodynamic logo certified by Demeter Italian Consumers of Organic Products There are many types of consumers of organic products and they choose organic products to varying degrees and for a number of different reasons. In Italy, nearly 60% of the population regularly purchases organic products, either exclusively or i n addition to conventional products (Pelligrini & Farinello, 2009). Studies have shown 3 Quantitative terms were not included in this study.
21 that purchasers of organic products tend to be health conscious individuals with a positive attitude towards organics and a high awareness of the difference between org anic and conventional products. They are generally concerned about the impact of the production processes on the environment and believe that organic products are att itudes towards organic products are their concerns for personal health and concerns for the environmental impact of the production processes (de Magistris & Gracia 2008; Saba & Messina, 2003 Boccaletti & Nardella, 2000 ). These consumers tend to be betwe en 30 45 years of age and fall in a medium high income bracket (Crescimanno Ficani & Guccione 2001). Within the pool of reasons that people purchase organic products, personal health concerns tend to be th e most important (Chinnici , & Pecorino 2002; de Magistris & Gracia 2008 ). Surveys of Italian consumers have shown that some consumers are willing to pay a premium for organic products. Some organic consumers have indicated a maximum willingness to pay for organic products of 10% ( Boccaletti & Nardella, 2000) Other studies have shown that some consumers are willing to pay between 20% and 40% more for organic products ( Chinnici, & Pecorino 2002 ; Pelligrini & Farinello, 2009 ) In a survey of Sicilian consumers, an in person interview was used to collect socio demographic information about respondents and qualitative information about their purchasing behavior and perception of the quality and price of organic products ( Chinnici, & Pecorino 2002 ) They performed a c luster analysis of the data, which revealed that respondents were segmented into four clusters based on the
22 motivations driving their purchasing decisions: health conscious, curious, price conscious, or nostalgia. The group characterized by health consciou sness displayed the most deeply rooted consumption of organic products. This group had been purchasing organic products for two years or more at the time of the survey and expressed a willingness to pay 20 30% more for organic produce than conventional pr oduce A survey of Italian consumers examining attitudes, behaviors, and knowledge of organic food was conducted by the Italian Ministry of Food, Agriculture, and Forestry (MIPAAF) ( Pelligrini & Farinello, 2009 ) They found that that those consumers who r egularly purchase organic foods are willing to pay a premium of up to 40% for food produced with organic farming practices over conventional practices Consumers shopping at large grocery stores in Northern Italy were surveyed in a study assessing willingn ess to pay for pesticide free produce. The results showed that many consumers were concerned about the health risks associated with pesticide use in fruit and vegetable production; 89% of respondents indicated that they would be willing to pay more for pe sticide free produce. However, of those willing to pay more, only 36% indicated that they would be willing to pay a premium of more than 10% and only 12% of those willing to pay more would pay a premium of 20% or above ( Boccaletti & Nardella, 2000) US Con sumers of Organic Products Many studies have been conducted to identify the typical organic consumer in the United States and their willingness to pay for organic products (Govindasamy & Italia, 1999; Lin Smith, & Huang 2008; Nie & Zepeda, 2008 ; Li, Zepeda & Gould, 2007; Onyango Hallman & Bellows 2007) These studies, mostly conducted as consumer surveys, have shown that the price premiums consumers are willing to pay
23 for organic products range from 10% to 60% depending on the product and the cha racteristics of the respondents. In general, consumers of organic products are young (<30 years of age), are either of a lower income bracket or a higher than average income bracket (a bimodal distribution), are female, and have a college level of educati on or higher. Age appears to be a significant factor in the likelihood of purchasing organic products and willingness to pay a premium for organic products. Studies have found that there is a negative correlation between age and both likelihood to purchas e organic products and willingness to pay a premium for them Some studies have shown that females are more likely than males to purchase higher priced organic goods. No consistent relationship exists between education level and att itudes towards organic products (Onyango, Hallman, & Bellows, 2007; Lin, Smith, & Huang, 2008; Govindasamy & Italia, 1999). Consumer attitudes, beliefs, and knowledge are also significant predictors of likelihood to purchase organic products and higher willingness to pay. Tra its associated with higher WTP and higher likelihood of consuming organic products include higher concern for food safety, high value placed on healthiness of food, interest in cooking, specific dietary needs, higher knowledge of organic logos and processe s, and shopping at specialty stores and cooperatives (Govindasamy & Italia, 1999; Lin, Smith, & Huang, 2008; Nie & Zepeda, 2008; Li, Zepeda, & Gould, 2007; Onyango, Hallman, & Bellows, 2007). In a national telephone survey observing consumer awareness and knowledge of of GMO techniques, willingness to pay for organic labeled products, and importance of
24 various food characteristics, it was found that 44% of respondents were regular purchasers of organic foods ( Onyango, Hallman & Bellows 2007 ). In terms of likelihood of regularly purchasing organic foods, females were 8% more likely than males, those below the age of 33 were 7% more likely than older respondents, those with college degrees were 6% more likely than those without, and those who identified as liberal in their political views were 12% more likely than those who identified as conservative. Those who said that the naturalness of a product was important or extremely important were between 13% and 35% more likely than those who said naturalness was not important to regularly purchase organic foods. Respondents who said they followed a vegetarian or vegan diet were between 8% and 20% more likely to regularly purchase organic foods than those who did not follow on e of these diets. Those who placed a higher level of importance on US production of their food were 4% more likely to be regular purchasers. organic produce commands anywhere fr om a 15% to 60% price premium over conventional produce (Lin, Smith, & Huang, 2008) Higher income households were generally more likely to purchase higher priced produce than lower income households and those households with household heads below the age of 40 will generally purchase higher priced produce than those above 40 years old. This study found no significant relationship between education and willing ness to pay for organic products, however smaller households tend to purchase organic products mo re regularly and are willing to pay higher prices for them.
25 A cluster analysis of US food shoppers segmented respondents into 4 categories based on shopping methods, purchasing motives, environmental knowledge/concerns, socio demographic characteristics (N ie & Zepeda, 2008) It found that those shoppers who are the most regular purchasers of organic food s frequently shop at specialty stores, pay close attention to food labels, and place a high value on healthiness of food, food safety, freshness, and taste They have an interest in cooking and frequently do so, they tend to follow specific dietary needs for illness or fitness related concerns, and they place little importance on brand names or food convenience. They have a high level of organic food knowl edge and participate in environmentally friendly activities. Another national study of US food shoppers analyzed regularity of organic food purchases as a function of consumer characteristics such as incom e, demographics attitude and knowledge of organic products, buying behavior, and region using results Buying ( Li, Zepeda & Gould, 2007 ) The study found that demographic variables such as age, education, household size and gender had little impact on the regularity of organic food purchases. The characteristics of shoppers that had the largest impact on regularity of organic food purchases had to do with attitudes and knowledge about organic products. Specifically, a wareness and familiarity of the USDA organic logo was associated with a 16.7% increase in frequency of organic food purchases over those who were not familiar with the label. Those who believed that organic foods are more nutritious than conventional food s purchased organic foods with a 15.9% higher frequency than those who did not share that belief. Respondents who very much enjoy cooking purchase organic foods with a 9% higher frequency that those who do not enjoy
26 cooking. The study also found that con sumers who shop at cooperatives and health food stores made organic food purchases with 16.3% and 28.7% higher frequency than those who did not shop at those venues, respec tively. A survey of shoppers at grocery outlets in New Jersey was used to examine th e characteristics of consumers willing to pay a premium of 10% for organic products (Govindasamy & Italia, 1999). Smaller, higher earning households (>$70,000) with a female knowledgeable of alternative agriculture as the main shopper are the most likely to pay a premium for organic foods. Specifically, males were 12% less likely than females to pay the premium. Households earning less than $30,000 annually were 16% less likely than households in the $70,000+ income bracket to pay the premium and househo lds in the $30,000 to $49,000 income bracket were 26% less likely to pay the premium. Those under the age of 36, those between the ages of 36 and 50, and those between the ages of 51 and 65 were 52%, 38%, and 28% more likely to pay the price premium than those above 65 years old, respectively. Increased levels of education correspond to a decreased likelihood of paying the 10% premium, with those having attended college or graduate school being 18% less likely to pay the premium than those with a high sch ool level of education. Household size also showed a negative relationship with willingness to pay. Each additional household member decreased the likelihood of paying the premium by 8%. Organic Price Premiums In most cases, o rganic production is more co stly than conventional production. Greater labor inputs, lower production per unit of land, greater diversity of enterprises, the need for specialized equipment, limits to economies of scale, and incurrence of certification costs can all increase overall production costs. The higher costs are
27 passed on to consumers in the form of higher prices, so it is critical that consumers are able to identify organic products, are aware of the benefits associated with organic products, and are willing to pay a premiu m for them. From a production standpoint, organic conversion is being considered and carried out by a growing number of wine producers. A recent survey of Italian wineries found that 17% were in the process of converting to organ ic production, despite the higher costs associated with production and certification and the lower productiveness of organic methods (Castellini, Mauracher Procidano & Sacchi 2013 ). However, the ability of certain products to command price premiums is not entirely clear, and this is certainly the case with organic win es. S tudies have shown that wine consumers express positive attitudes towards organic wines and indicate that they would be willing to pay higher prices for them (Brugarolas Moll Bauz Mart nez Carrasco Martnez Poveda Rico Prez 2005; Forbes, Cohen Cullen Wratten & Fountain 2009; Remaud, Mueller Chvyl & Lockshin 2008) However, in one study of California wines in the US market (discussed in more detail in the review of literature ) it was found that that the indication of organic status on a wine bottle is linked to lower marke t prices, whereas organic certification alone without use of organic labeling results in price premiums (Delmas & Grant, 2010 ). This thesis makes use of the fact that prior to 2012, there were no official regulations concerning the labeling of organic wines in the European Union. Wines were being produced organically (according to independent certifiers) but organic winemakers were unable to advertise this in a way that consumers could easily
28 contain the EU organic logo on the label. Using a hedonic price model for Tuscan red wines, this study examines whether or not organic wine s were still able to command price premiums despite the lack of a uniform label. For the same bottles of wine, the relationship between price and organic status is estimated for both the US and EU wine markets. To the best of our knowledge, this is the f irst time that such a comparative study has been done. The results yield valuable insights for current organic wine producers, wine producers considering converting to organic growing practices, wine merchants, and policy makers. The following chapter dis cusses literature relevant to this study, including consumer attitudes towards organic products and specifically organic wines, other uses of hedonic price models to examine wine prices, and the validity of using hedonic price models to study wine. Chapte r three will discuss the data used for this study and the methods used to produce a hedonic price model for Tuscan wines. Chapter four will summarize the results, and chapter five will discuss conclusions drawn from the results as well as limitations to t his study and areas for future investigation.
29 CHAPTER 2 REVIEW OF LITERATURE General Attitudes Towards Organic Wines willingness to pay for organic wines ( Brugarolas Moll Bauz Martnez Carrasco Martnez Poveda & Rico Prez 2005 ; Forbes, Cohen Cullen Wratten & Fountain 2009 ; Remaud, Mueller Chvyl & Lockshin 2008 ). The literature has shown that attitudes concerning organic wines are generally positive, with some consume rs indicating that they would be willing to pay a premium for them, however these results differ between regions and consumer segments. Consumer knowledge and recognition of organic schemes and logos is also a barrier to organic wine acceptance. Using a r ecruitment questionnaire, the ORWINE Project conducted a focus group expectations of organic wine. This study found that participants generally had the belief that the ab sence of chemical additives and pesticides resulted in organic wine being healthier than non organic wine. However, many participants also believed that organic wine did not taste as good as non organic wine This belief could be as a result of the absen ce of organic wine in specialized wine stores, which might give the impression that premium organic wines are scarce or non existent. Participants also made suggestions regarding labeling of organic wines. One suggestion was that labels contain a list of a dditives that are absent in the wine making process, thereby Participants also made suggestions for strategies to regulate additive use, which included general prohibition of any additives that might be detrimental to human health,
30 the general prohibition of additives that affect taste and authenticity of the wine, and lower maximum levels of additives (Stolz & Schmid, 2008). A contingent valuation study exploring Spanish wi pay for organic wine asked respondents how much more they would be willing to pay for an organic wine than a conventional wine with comparable characteristics. The authors found that respondents were willing to pay an average price premium of 17%. A lifestyle based cluster analysis of the same data revealed a price premium range of 12 2 5 % ( Brugarolas Moll Bauz Martnez Carrasco Martnez Poveda Rico Prez 2005 ). analyzed attitudes regardin g environmentally sustainable wines us ing a structured questionnaire The authors found that 75% of respondents would prefer to drink wines made using environmentally sustainable practices and 72% indicated an intention to purchase a sustainable wine ove r a conventionally grown wine. The study also found that 93% of respondents indicated a desire to see labels containing information about the environmental qualities of the wine Among respondents, 73% indicated that they would be willing to pay more for an environmentally sustainable wine, with roughly two thirds saying they would be willing to pay a price premium of 5 11% ( Forbes, Cohen Cullen Wratten & Fountain 2009 ). A study of how Australian consumers value organic production practices relative to other wine traits including price, region of origin, and environmental claims (environmentally responsible and carbon neutral) used a discrete choice experiment and f ou n d that the organic attribute had the lowest level of importance, accounting for only 3% of the importance for wine choice. Price carried the highest importance with
31 65% while the region of origin and environmental claim carr ied 17% and 14% of the importance, respectively. A latent class model which segmented the respondents into groups characterized by price sensitivity, concern for the environment, preference for organics, and regional preference d id not produce results that varied greatly from the overall analysis in term s of relative organic importance The class that was characterized by environmental concern and preference for organics gave the organic claim attribute a 9% level of importance. The authors conclude that only a small segment of wine consumers (14%) in t he Australian wine market are willing to pay a 22% price premium for organic wines ( Remaud, Mueller Chvyl & Lockshin 2008 ) Wine Price Hedonic Models It is possible to determine the relative effects of wine characteristics on the price of individual win es using hedonic price analysis. This form of analysis examines the price of a product as a function of its endogenous characteristics or attributes in a perfectly competitive market (Rosen, 1974). A common use of hedonic pricing is the estimation of the price of a home as a function of its attributes (square footage, number of rooms, neighborhood characteristics, etc.). Hedonic price analysis has been used to (Oczkowski, 1994 ; Combris, Lecocq &Visser 1997; Angulo, Gil, Gracia, & Sanchez 2000; Bombrun & Sumner, 2003; Benfratello, Piacenza & Sacchetto 2004; Costanigro, McCluskey & Mittelhammer 2007; Brentari & Levaggi, 2011). However, very few studies consider the effects of organic status o n price (Delmas & Grant, 2008 ). Most hedonic price studies for wine have used data concerning wines from one country or one particular region and have used, in most cases, objective independent variables. These variables can be derived from the label itself, such as region of origin,
32 grape varietal, and vintage. Objective characteristics have been shown to be better predictors of price than sensory characteristics such as taste, aroma, tannic acidity, and body (Lecocq & Visser, 2 006; Brentari & Levaggi, 2011 ; Benfratello, Piacenza & Sacchetto 2004 ; Combris, Lecocq & Visser 1997 ). Wine scores, such as those reported by Wine Spectator, Wine Advocate, or other wine rating sources, are also included in many hedonic price studies, as these scores are commonly indicated in retail settings H owever, the impact of these scores on price varies among different studies (Bombrun & Sumner, 2003; Costanigro, McCluskey & Mittelhammer 2007 ) A critical part of hedonic framework is the exclusion of exogenous variables in the model. Wine raters weigh all the sensory characteristics of a wine to determine the rating it is given, so rating is a function of those qualities and therefore cannot be considered an endogenous characteristic S ome studies have examined rating as a dependent variable to determine the effect of wine characteristics on rating scores. A study estimating a hedonic price model and a rating model for 519 Bordeaux wines found that sensory characteristics are respons ible for the majority of variance in rating but had little impact on price. This study included both objective and subjective characteristics as variables and found that the objective characteristics made up the majority of the significant price determina nts. However, results from the rating model showed the sensory characteristics made up the majority of the significant rating determinants ( Combris, Lecocq &Visser 1997 ). Oczkowski (1994) found that, in the Australian wine market, grape varietal has t he largest impact on price, with an increase in price of 55% for sparkling wine, and an increase of roughly 25% for sweet white wines, Pinot Noir, and Vintage Port compared
33 to the average price of wines used in the study However, this appears to be a res ult that is specific to Australian wines; Angulo et al. (2000) found in their hedonic price model of Spanish wines, that region of origin and vintage were the two most influential in determining price among the variables they tested. Grape varietal held n o significant effect on price. Bombrun and Sumner (2003) found in their hedonic model of Californian wine prices that 72% of the price variation could be explained by objective characteristics, with the most influential being app ellation and vintage. They too found that grape varietal was generally less important than other variables. A study examining the role of quality and reputation in determining the price of wines from Bordeaux developed five price models each using different combinations of info rmation available to consumers (Landon & Smith, 1997) The study found that a model employing individual winery reputation and collective reputation as variables was the best at determining price. The individual winery reputation variable was represented by classification derived from long term Wine Spectator quality ratings as well as a classification of Bordeaux producers by the wine rater Robert Parker. The collective reputation variables include regional appellation designations in addition to indust ry determined quality classifications. Bordeaux quality classifications group producers into classes from First Growth (highest quality) to Fifth Growth (lowest quality). Producers are also designated into region specific quality classes including St. Em ilion Premier Grand Cru Class, St. Emilion Cru Class, Graves Cru Class, Cru Grand Bourgeois, and Cru Bourgeois. This study concludes that winery and collective reputation are the leading price determinants in the market for Bordeaux wines.
34 Costanigro, McC luskey & Mittelhammer (2007) examined California and Washington red wines using a hedonic price model and found that the effects of attributes var y across price categories. In their analysis wines were segmented into commercial, semi premium, premium, a nd ultra premium groups designated by price In addition to a pooled price model, they performed individual analyses of each of these categories. The pooled model showed that regional appellations command price premiums over wines labeled only with the s tate of origin. The segmented models showed that the p rice effect of rating was highest in the ultra premium category and decreased in magnitude with each lower price segment The importance of grape varietal varied across segments as well. Merlots comm anded the highest price premium in the commercial segment while Cabernet Sauvignon and Pinot Noir commanded the highest price premiums in the ultra premium segment. This study shows that the relative importance of wine attributes is not consistent across all price segments so it can therefore be important to examine each segment individually. Hedonic Price Models of Italian Wines This thesis focuses on Italian wines, which make up the second largest share of wines produced behind France (FAO, 2014). The following studies speci fically analyze Italian wines. Findings of these studies suggest that objective label characteristics, particularly winery and regional reputation, play a large role in determining the price of wine in Italy. Brentari and Levaggi (2 011) examined Italian wines segmented by their distr ibution channels and found that, in a model only including label characteristics 78% of total price variance was explained by the variables In a separate model including only sensory characteristics only 35% of price variance was explained by the variables
35 In a region specific study of Barolo and Barbaresco wines from the Piedmont region of Italy, winery and region reputation were found to be more influential price determinants than the sensory char acteristics, which play a smaller role in driving market prices ( Benfratello, Piacenza & Sacchetto 2004 ). Hedonic Price Models of Organic Wines Some winemakers produce organically certified wine but choose not to put an organic certification label on the bottle, while others both certify and label their wine as organic. A study of Californian wines finds that these two strategies, while sequential and closely related, have nearly opposite price effects (Delmas & Grant, 2008) Their analysis shows tha t wines produced by wineries that have undergone organic certification but do not advertise their certification on their labels command a 13% price premium over conventional, non organic wines. The inclusion of organic certification information on wine la bels has an opposite price effect, reducing price by 20% below conventional, non organic wine. The authors believe that organic certification can improve wineries reputations through access to trade organizations and that certification results i n higher q uality wines This theory is support by a second analysis of determinants of wine quality, which showed that rating scores increased with certification. They attribute the counterintuitive result pertaining to labeling to the general lack of kn owledge about organic wines and organic wine labels and a belief that organic wines are of inferior quality Criticism of the Hedonic Price Model for Wine The application of the hedonic price model for wine research is not without criticism. Unwin (1999) examined previous studies employing hedonic regression and suggested that the model be abandoned in future wine price studies for a number of
36 reasons. The paper discusses the issues of multicollinearity between independent variables, the treatment of wine ratings as an independent variable instead of being considered an endogenous variable, and the wine market being treated as purely competitive Unwin also argues that the research is bounded by the availability of the data and that the unavailability and subsequent disregard of other potentially significant variables does not coincide with theoretical foundations. Furthermore, he argue s that consumers do not have access to or fully understand the explanatory variables used in the model and it is therefo re a poor predictor of consumer behavior. problem that is not specific to wine research, but is in fact a problem that spans many fields of statistical research and can be accoun ted for through the proper use of analytical tools. He discusses the methods that can be used to deal with multivariate models with endogenous variables, such as a hierarchal regression or instrumental variable procedures and shows that there is evidence that pure competition, or a similar market state, is not an unreasonable assumption for the wine market. Oczkowski (1994) found that the state of the wine market in Australia could be considered close to one of perfect competition. Thrane also believes t hat data driven research is also not a trait that is specific to wine research but instead is a common problem in general hedonic price modeling. He also makes the argument that consumer knowledge and the relevance of these types of models have less to do with the methods being employed and more to do with the lack of communication between researchers and wine consumers and that consumer behavior cannot be predicted with models based on supply oriented explanatory variables.
37 Information Provision Organic w labels before 2012 has left a gap in the provision of information on Italian wine labels prior to that year. Wine consumers tend to exhibit a high demand for information about their pr oducts, and their purchasing decisions are shaped by the provision of this information (Dimara & Skuras, 2005 Boatto, Defrancesco, & Trestini, 2011 ). This suggests that wine labels should provide pertinent information to consumers in order to effectively differentiate products. Information provision, particularly on the topic of environmental attributes, is an important aspect of consumer preference s and their purchasing decisions and is influent ial in increasing willingness to pay for organic products ( Leire & Thidell, 2005, Johnston Wessells Donath & Asche 2001, Loureiro, 2003 ; Loureiro & Lotade, 2005, Didier & Lucie, 2008). Purchase decisions by wine consumers in particular are largely guid ed by label information (Thomas & Pickering, 2003). Common findings among previous hedonic price studies (discussed above in this chapter) are that objective characteristics such as region of origin and vintage (year of production) play a large role in d etermining the price of wine. Winery reputation is also a highly important determinant of wine price, particularly in Italy. In general, consumers are open to the idea of organic wines, but perceptions of organic wines are still somewhat negative. While many consumers have indicated that they are willing to pay a premium for organic wines, others have expressed a belief that organic wines are of a lesser quality than conventionally produced wines. The provision of information is a critical aspect of the marketing of organics products and wine consumers have
38 expressed a desire to see more informative labeling on wine labels regarding production practices and additives. The inability of Italian producers of organic wines to label their products with widel y known environmental signals may have left them unable to take advantage of the potential benefits of organic status. However, as Delmas and Grant (2008) found, it is possible that wine producers c ould still have reap ed the benefits of organic certificat ion without the use of the organic label. To examine if organic wine producers in Italy we re commanding price premiums despite their inability to use the EU organic logo, we have performed a hedonic price analysis of Tuscan red wine s including a variable for organic status Using data for Tuscan red wines, this study examines wines from a prominent Italian wine growing region that, to our knowledge, has never been studied using th e hedonic pricing method. Our study also contributes to the limited literatu re on price effects of organic wine certification and it has important managerial implications for wine producers considering converting to organic production and seeking org anic certification.
39 CHAPTER 3 DATA AND METHODS Dataset The dataset for our study consist s of 631 individual premium red wines from 60 wineries in the Tuscany region of Italy. The dataset creation process started with a list of wines available through a premium wine wholesaler called WineTip. The list only had data on winery name, wine name/vintage, general region of origin (Tuscany, Piedmont, etc), and wholesale price. This list was missing many vintages for each wine, and did not contain retail price values, which were not only more relevant to the study, but were also avai lable for the wines added to the dataset. Wholesale price was therefore replaced with retail price. Variables added to the dataset include rating, organic status, and current retail price in both US and Italian markets. From the original list, wines were added and removed based on availability of the necessary variables To obtain the current price data, grape varietal, classification/ap pellation of origin, and rating for wines added to the original list, we utilized the wine reference website Vinopedia .com. This site lists all online retailers offering each wine and lists the individual prices for each retailer in addition to the characteristics listed. For our dataset, we used the average current price for each wine, separately examining US retailers and Italian retailers, to give us an average current retail price for both markets. When varietal, appellation of origin, or rating score data were not available through Vinopedia.com, other sources such as Cellartracker.com and individu al producer webs ites were used. Organic status was obtained by referencing producer websites and by searching through a variety of organic certification databases including ICEA, Demeter, and AIAB.
40 Summary Statistics Within the dataset, median wine price in the Italian 42.50 ($ 58.33 ) and median US price is $ 82.75 ( 60.30 ) 1 Using current price has the advantage of representing actual market price paid by consumers instead of release price, which is the manufacturer s suggested retail price. O nly 446 of the wines sampled have a US price in addition to the Italian price Multiple ratings sources were included in the International Wine Cellar, the Wine Enthusiast, and the Cellar Track er average user score (Table 3 1) 2 Table 3 1. Mean Price in Each Market and Mean Rating Variable N Mean Std. Dev. Minimum Maximum Italian Price ( 631 59.41 59.48 8.00 630.00 US Price ($) 446 112.21 103.74 18.00 972.00 Rating Average Cellar Tracker Wine Enthusiast Steven Tanzer Wine Spectator Robert Parker 631 509 274 280 554 533 91.97 91.25 92.38 91.29 91.61 92.26 2.54 1.85 2.88 2.21 2.82 2.55 83.00 82.00 80.00 86.00 83.00 85.00 98 .00 99.00 100.00 97.00 100.00 100.00 Vintages range from 1999 to 2008. The years 2000, 2002, and 2008 are the least represented, with 7.8%, 3.5% and 7.8% of the data, respectively. The other vintages are more or less evenly distributed with each representing between 9 14% of 1 Exchange rates calculated using current rate of 1 euro = 1.37 US dollars. 2 Rating sc ores are on a scale of 50 100. All wine raters follow a similar format: 96 100 is an extraordinary wine; 90 95 is an outstanding wine of superior quality; 80 89 is a good or very good wine with above average qualities; 75 79 is a mediocre or average wine with some flaws; 70 74 is a below average wine; and below 70 is a very poor wine.
41 the data (Table 3 2) The vintage represents the year of harvest. Wines are typically released into the market 1 2 years after harvest. Table 3 2. Vintage Frequency Variable Frequency Percent Cumulative Freq. Cumulative Percent 1999 57 9.03 57 9.03 2000 49 7.77 106 16.80 2001 66 10.46 172 27.26 2002 22 3.49 194 30.74 2003 71 11.25 265 42.00 2004 83 13.15 348 55.15 2005 68 10.78 416 65.93 2006 87 13.79 503 79.71 2007 79 12.52 582 92.23 2008 49 7.77 631 100.00 Many grape varietals are r epresented in these data, with s angiovese comprising 48.5% of the wines, other single varitals (including cabernet sauvignon, cabernet franc, merlot, and syrah) comprising 17.1%, with the remaining 34.4% being made up of various blends. Regarding appellation of origin, 34.6 % of wines are classified as DOCG (controlled designation of origin guaranteed), 10.9% are classified as DOC ( controlled designation of origin ), and 54.5% are classified as IGT ( Indicazione Geografica Tipica ). 3 Uncertified o rganic wines made up 17.4 % of t he data and certified organic wines made up 10.6%, with the balance being conventional wines making up 72.0% of the data (Table 3 3). 3 The Italian appellation classification system is meant to signify levels of quality that exceed the quality of wines sold as vino da tavola meaning table wine. DOC repre sents quality wines with controlled production standards. It was considered to be awarded quite liberally and therefore did not enjoy much credibility. DOCG represents the highest quality appellations and is designated to only the finest Italian wines to identify and reward them for their superior quality. These appellations are not only controlled by response to the much criticized DOC/DOCG designations to represent high quality wines that did not follow the same strict production standards, particularly standards concerning allowable blending or grape after wines are designated as IGT (Robinson, 1999).
42 Table 3 3 Frequencies Grape Varietal, Appellation of Origin, and Organic Status Variable Frequency Percent Cumulative Freq. Cumulative % Varietal Sangiovese 306 48.49 306 48.49 Blend 217 34.39 523 82.88 Other single varietal 108 17.12 631 100.00 Appellation DOCG 218 34.55 218 34.55 DOC 69 10.94 287 45.48 IGT 344 54.52 631 100.00 Organic status Conventional 454 71.95 454 71.95 Uncertified Certified 110 67 17.43 10.62 564 631 89.38 100.00 Using SAS statistical software and following the traditional hedonic model we have modeled both current Italian price and current US price as a function of vintage, grape varietal, appellation of origin, rating, and organic status Price and rating were treated as continuous variables, while grape varietal, appellation of orig in, organic status and vintage were treated as discrete variables. By including a model of current US price in addition to the current Italian price, we are able to examine the differences in effects between the two markets. This allow s us to determine if US consumers of Italian wines value organic wines differently than Italian consumers of the same wines. In addition to a unified model where all 631 wines are included in the regression with the above listed variables a model containing interaction terms between organic status and varietal, vintage and appellation is estimated, allowing us to determine if the effect of the organic status depends on any of the other variables. Quartile price segment models are also e stimated to examine the effects of wine characteristics in different price categories. The first quartile represents the lowest priced wines in the dataset and the wine prices raise with each next quartile.
43 ull hypothesis of no heteroscedasticity is rejected in most models To account for heteroscedasticity, the models are estimated using a Heteroscedasticity Consistent Covarian ce Matrix The full regression specification to model the price of wine i in market j is as follows: For the regression specification including interaction variables, the fol low ing terms are added:
44 CHAPTER 4 RESULTS Models with Individual Ratings and Average Rating We first determine whether the overall model should include each individual rating score or just the average score. The results for the US price and Italian price models are shown in Table 4 1 and Table 4 2, respectively where a different rating is used in each model R s quared values for these regressions indicate that the models using average rating scores provide the best fit for both US and Italian price. Models using Cella r Tracker (CT) scor es yield the lowest R squared values for both prices, while the models using Wine Enthusiast (WE), Steven Tanzer (ST), Wine Spectator (WS), and Robert Parker (RP) explain roughly the same level of variance in log price for both markets Tab le 4 1 Effects of wine characteristics on US price with average and individual rating scores ln( US Price) Average CT WE ST WS RP Constant 65.66** 0.9 0 44.08** 23.25** 42.97** 53.87** ln( Rating ) 15.53** 0.77* 10.73** 6.16** 10.49** 12.9 0 ** Uncertified Org. 0.15** 0.17** 0.25** 0.1 0 0.2 0 ** 0.03 Certified Organic 0.05 0.04 0.04 0.11 0.09 0.03 Varietal Blend 0.19** 0.26** 0.34** 0.33** 0.22** 0.34** Varietal Single 0.46** 0.7 0 ** 0.6 0 ** 0.87** 0.56** 0.75** Appellation DOC 0.03 0.16 0.04 0.14 0.14 0.05 Appellation IGT 0.13* 0.06 0.3 0 ** 0.16 0.09 0.22** Vintage 2000 0.06 0 .00 0.11 0.08 0.12 0.01 Vintage 2001 0.02 0.16 0.26 0.08 0.03 0.01 Vintage 2002 0.26 0.06 0.17 0.15 0.32* 0.18 Vintage 2003 0.06 0.12 0.16 0.23 0.14 0 .00 Vintage 2004 0.12 0.1 0 0.06 0.09 0.1 0 0.11 Vintage 2005 0.14 0.26** 0.28 0.05 0.26** 0.07 Vintage 2006 0.34** 0.11 0.18 0.24 0.28** 0.34** Vintage 2007 0.42** 0.3** 0.4 0 ** 0.4 0 ** 0.34** 0.47** Vintage 2008 0.43** 0.43** 0.46** 0.8 0 ** 0.34** 0.48** R Squared 0.52 0.31 0.47 0.44 0.47 0.51 N 444 381 210 221 398 391 Notes: and ** indicate significance at the 10% and 5% level, respectively.
45 Table 4 2 Effects of wine characteristics on Italian price with average and individual rating scores ln(Italian Price) Average CT WE ST WS RP Constant 68.31** 0.88 50.38** 29.52** 43.08** 50.69** ln( Rating ) 15.98** 0.65 11.97** 7.41** 10.39** 12.07** Uncertified Org. 0 .00 0.1 0 0.1 0 0.11 0.07 0.14** Certified Organic 0.16** 0.2 0 ** 0.06 0.29** 0.16** 0.25** Varietal Blend 0.11** 0.21** 0.36** 0.24** 0.12** 0.26** Varietal Single 0.43** 0.78** 0.59** 0.91** 0.56** 0.72** Appellation DOC 0.09 0.15 0.07 0.13 0.2 0 ** 0.05 Appellation IGT 0.04 0.02 0.27** 0.05 0.01 0.14** Vintage 2000 0.13 0.02 0.36** 0.18 0.12 0.12 Vintage 2001 0.04 0.01 0.22 0.3 0 0.04 0.07 Vintage 2002 0.36** 0.04 0.2 0 0.07 0.24 0.3 0 ** Vintage 2003 0.04 0.17* 0.35** 0.19 0.09 0.05 Vintage 2004 0.17** 0.09 0.02 0.02 0.14 0.18* Vintage 2005 0.12 0.3 0 ** 0.06 0.06 0.24** 0.09 Vintage 2006 0.32** 0.19* 0.07 0.22* 0.29** 0.29** Vintage 2007 0.42** 0.38** 0.28** 0.41** 0.36** 0.44** Vintage 2008 0.38** 0.43** 0.36** 0.69** 0.29** 0.46** R Squared 0.50 0.29 0.48 0.44 0.44 0.46 N 629 509 274 279 554 533 Notes: and ** indicate significance at the 10% and 5% level, respectively. Because average rating models yield the best fit in both markets and using individual scores may introduce bias of individual rating agents the models with average rating scores will be used for our analysis. Results from these models show that rating has a positive and significant correlation with bot h US and Italian price, with a 1% increase in rating associated with a 15.53 % and 15.98 % increase in price respectively. In the US market, uncertified organic wines command a 15% higher price than conventional wines, however in the Italian market, there is no significant impact. Interestingly, the results are nearly opposite for certified organic wines. In the US market, certified organic has no significant impact on price, but in t he Italian market, certifie d organic is associated with a 16% lower price than conventional wines.
46 There is no significant price difference between DOC and DOCG designated wines in either market, however, in the US market, IGT designated wine receive s a 13% lower price than DOCG win es. In the both the US and Italian market s the price of single varietals is higher than the base variety of sangiovese by 46 % and 43% respectively Prices of b lended varietals are also significantly higher than sangiovese in both market s, commanding 19% and 11% price premiums in the US and Italy, respectively The i mpacts of vintage on price are consistently and significantly negative for the three most recent vintages in both markets. In the US market, wines from these vintages receive 34%, 42%, and 43 % lower prices than the base category of 1999 for the 2006, 2007 and 2008 vintages respectively. In the Italian market, these prices are 32%, 42%, and 38% lower than the base category for the same vintages, respectively. This is not surprising due to t he perception that wines improve with age and therefore i t would be expected that recent vintages would have lower prices than older, aged wines However, what is surprising is that in the Italian market, wines from 2002 receive a significant premium of 3 6% above the base year, despite 2002 being one of the worst years for Tuscan wines in terms of quality. The 2002 vintage is the least represented year in the dataset, with only 22 wines from this year. The limited data for this year could explain these a typical results. In a bad year, only the best wines would be saved for 10 years, which could be resulting in some selection bias. Wines from 2004 are shown to receive a 17% lower price than wines from the base year, despite 2004 being considered an excell ent year for the region (Parker, 2013 ) However, this result is less
47 surprising given that the base year of 1999 was also considered to be an excellent year for the region and wines from 1999 have aged five more years than 2004 wines. Price Segmented Models To determine if the price effects of the variables include d in this study varied between less and more expensive wines, we also estimated segmented price models with the data split into quartiles. Table 4 3 and Table 4 4 show the results of US and Italian price segment ed models, respectively. The last column in each table represents the full price model in each market for comparison. Price Segmented Model in the US Market Results from th e segmented price models show that the effects of wine characteristics do vary between price categories. In the US market (Table 4 3) the price of uncertified organic wines is 7% higher than conventional wines in the second quartile but is not significantly different in the other price segments. Certified organic wines in the first quartile received a price that is 13% lower than conventional wines. Certified organic has no significant impact in the oth er segments. W e see that an increase in rating still has a positive association in all price segments except the third quartile. The impact of a 1% increase in rating corresponds to a 2.74% and 2.59% increase in price in the first and secon d quartiles, r espectively. T he premium in the 4 th quartile is much greater at 12.63%. Wines made from the single varietals command significantly higher prices in the second and fourth quartile with premiums of 7% and 62% above sangiovese respectively, but in the third quartile the price is 11% lower than sangiovese Blended varietals also command significantly higher prices in the second and fourth quartiles, and a significantly lower price in the third quartile.
48 The wines with DOC designation command significantly lower prices in all segments except the third. The prices of the DOC wines are 24%, 12%, and 46% lower than DOCG wines in the first, second, and fourth segments, respectively. IGT wines have significantly lower prices t han DOCG in the second and fourth price segments with 6% and 47% lower prices, respectively. Effects of vintage throughout the four segments are scattered and mostly insignificant, with the exception of the fourth segment, where prices of wines from 2001 2004, 2006, 2007 and 2008 are significantly lower than the base year (24%, 28%, 43%, 41%, and 27%, respectively). Table 4 3 Effects of wine characteristics on price by price quartile, in the US market ln(US Price) Q1 Q2 Q3 Q4 Full Constant 8.5 0 7.44** 1.22 49.71** 65.66** ln( Rating ) 2.74** 2.59** 0.74 12.19** 15.53** Uncertified Organic 0.03 0.07* 0.03 0.07 0.15** Certified Organic 0.13* 0.05 0.01 0.1 0.05 Varietal Blend 0.08 0.07** 0.08** 0.25** 0.19** Varietal Single 0.02 0.07** 0.11** 0.62** 0.46** Appellation DOC 0.24** 0.12** 0.01 0.46** 0.03 Appellation IGT 0.03 0.06** 0.03 0.47** 0.13* Vintage 2000 0.18 0.01 0.05 0.05 0.06 Vintage 2001 0.09 0.03 0.04 0.24* 0.02 Vintage 2002 0.12 0.05 0.04 0.23 0.26 Vintage 2003 0.02 0 .00 0.02 0.04 0.06 Vintage 2004 0.06 0.06 0.08 0.28** 0.12 Vintage 2005 0.06 0.03 0.08 0.18 0.14 Vintage 2006 0.04 0.06 0.06 0.43** 0.34** Vintage 2007 0.14 0.08* 0 .00 0.41** 0.42** Vintage 2008 0.35** 0.09 0.04 0.27* 0.43** R Squared 0.48 0.26 0.29 0.54 0.52 N 111 112 106 115 444 Notes: and ** indicate significance at the 10% and 5% level, respectively. Price Segmented Model in the Italian Market Our models of Italian wine prices segmented into quartiles (Table 4 4) yield sim ilarly varying results. Uncertified organic wines receive a 23% lower price than conventional wines in the first quartile but receive no significant difference in the other three segments. On the other hand, certified organic wines receive a 7% higher price
49 and a 26% lower price than conventional wines in the second and fourth quartiles, respectively. Table 4 4. Effects of wine characteristics on price, by price quartile, in the Italian market ln(Italian Price) Q1 Q2 Q3 Q4 F ull Constant 32.06** 0.33 5.7* 51.31** 68.31** ln(Rating) 7.82** 0.71* 0.39 12.36** 15.98** Uncertified Organic 0.23** 0.01 0.01 0.07 0.00 Certified Organic 0.01 0.07** 0.01 0.26* 0.16** Varietal Blend 0.02 0.05** 0.04 0.17** 0.11** Varietal Single 0.29** 0.05* 0.05 0.57** 0.43** Appellation DOC 0.01 0.03 0.07* 0.23** 0.09 Appellation IGT 0.00 0.04* 0.06* 0.36** 0.04 Vintage 2000 0.25** 0.01 0.01 0.21 0.13 Vintage 2001 0.08 0.04 0.04 0.05 0.04 Vintage 2002 0.36** 0.00 0.05 0.28 0.36** Vintage 2003 0.05 0.01 0.01 0.11 0.04 Vintage 2004 0.09 0.02 0.04 0.08 0.17** Vintage 2005 0.01 0.05 0.01 0.04 0.12 Vintage 2006 0.06 0.04 0.04 0.26** 0.32** Vintage 2007 0.10 0.05 0.02 0.20 0.42** Vintage 2008 0.06 0.01 0.06 0.23 0.38** R Squared 0.45 0.22 0.15 0.46 0.50 N 134 175 161 159 629 Notes: and ** indicate significance at the 10% and 5% level, respectively. Rating has a significant positive association in the first, second and fou rth quartiles, corresponding to a 7.82%, 0.71%, and 12.36% increase in price with each percentage increase in rating Like the US market, the largest imp act is in the fourth quartile. Blended varietals receive 5% lower prices than sangiovese wines in the second quartile and 17% higher prices in the fourth quartile. Wines made from single varietals command significantly higher prices in all segments except the third, with premiums of 29%, 5%, and 57% over sangiovese wines in the first, second and fourth segments, respectively. Appellation of origin seems to have significant impacts for the higher priced wines. DOC and IGT wines command 23% and 36% lower prices than DOCG wines in the fourth quartile. DOC wines actually receive a 7% premium ov er DOCG wines in the
50 third quartile, while IGT wines receive 4% and 6% lower prices in the second and third quartiles, respectively. Effects of vintage in the Italian segmented price models are largely insignificant. Price Models with Interaction Terms The price models with interaction terms yield similar results to t hose with out interaction terms. Table 4 5 shows the US and Italian price models with interaction terms in the second and third column, and US and Italian price models without interaction te rms in the fourth and fifth columns, respectively. Variable coefficients for the non interacted variables are nearly identical in significance level but are slightly different in value Another notable difference between the Italian models is that the ce rtified organic coefficient loses its significance in the interacted model, suggesting that the single varietal variable is driving the certified organic result in the non interacted model. Aside from this interaction, the impact of organic status on pric e generally does not depend on other wine characteristics. The interaction terms show that the 2008 vintage affects the impact of the uncertified organic status in both markets and the impact of the certified organic status on the Italian price. We also see that the interaction between IGT designation and the certified organic status in the US market has a significant positive effect on price. Other n otable differences between the US models are that the effects of all significant variables are greater in the interact ed model except for vintage which has lesser impacts in the interacted model. Also, the value of the coefficients for varietal increase and the value of coefficients for vintage decrease in the interacted model.
51 Some interactions were droppe d from the model due to insufficient variation in the variables being interacted. Table 4 5 Models of US and Italian Price Including Interaction Terms ln(Price) US, Int Italian, Int US Italian Constant 65.72** 69.48** 65.66** 68.31** ln( Rating ) 15.52** 16.23** 15.53** 15.98** Uncertified Organic 0.49** 0.25 0.15** 0.00 Certified Organic 0.22 0.14 0.05 0.16** Varietal Blend 0.25** 0.14** 0.19** 0.11** Varietal Single 0.59** 0.56** 0.46** 0.43** Appellation DOC 0.01 0.07 0.03 0.09 Appellation IGT 0.14** 0.04 0.13* 0.04 Vintage 2000 0.06 0.14 0.06 0.13 Vintage 2001 0.00 0.04 0.02 0.04 Vintage 2002 0.31 0.36** 0.26 0.36** Vintage 2003 0.03 0.04 0.06 0.04 Vintage 2004 0.13 0.18* 0.12 0.17** Vintage 2005 0.06 0.09 0.14 0.12 Vintage 2006 0.30** 0.32** 0.34** 0.32** Vintage 2007 0.31** 0.36** 0.42** 0.42** Vintage 2008 0.27** 0.22** 0.43** 0.38** Uncertified Org.*Varietal Blend 0.14 0.23 0.52 0.50 Uncertified Org.*Varietal Single 0.18 0.45 Certified*Varietal Blend 0.01 0.07 Certified*Varietal Single 0.62** 0.55** Uncertified Org.*Appellation IGT 0.28 0.02 Certified*Appellation DOC 0.16 0.01 Certified*Appellation IGT 0.28* 0.05 Uncertified Org.*Vintage 2000 0.01 0.09 Uncertified Org.*Vintage 2001 0.14 0.04 Uncertified Org.*Vintage 2002 0.14 0.01 Uncertified Org.*Vintage 2003 0.46 0.14 Uncertified Org.*Vintage 2004 0.09 0.04 Uncertified Org.*Vintage 2005 0.25 0.16 Uncertified Org.*Vintage 2006 0.11 0.07 Uncertified Org.*Vintage 2007 0.31 0.28 Uncertified Org.*Vintage 2008 0.45* 0.44* Certified*Vintage 2000 0.12 0.10 Certified*Vintage 2001 0.30 Certified*Vintage 2002 0.59** Certified*Vintage 2003 0.26 0.10 Certified*Vintage 2004 0.17 0.05 Certified*Vintage 2005 0.32 0.02 Certified*Vintage 2006 0.18 0.16 Certified*Vintage 2007 0.36 0.02 Certified*Vintage 2008 0.43 0.29* R Squared 0.57 0.53 0.52 0.50 Adjusted R Squared 0.53 0.50 0.51 0.49 N 444 629 444 629 Notes: and ** indicate significance at the 10% and 5% level, respectively.
52 CHAPTER 5 DISCUSSION AND CONCLUSIONS In order to determine the effects of organic growing practices and organic certification on the Italian and US price of Tuscan red wines, we have esti mated a hedonic price model as a function of rating, varietal, appellation of origin, vintage and organic status Our models have produced some very interest ing results. The Italian Market The results concerning organic status suggest that whether a wine is organic or not seems to have little clear and consistent association with price. When not accounting for the effects of interactions in our model, the Ita lian market model shows that certified organic wines receive a 16% lower price than conventional wines. However the model accounting for interaction effects shows no significant difference in price between conventional, uncertified, or certified wines suggesting that the effects of the certified organic variable depend on varietal and vintage specifically single varietals and the 2002 and 2008 vintage s The model shows that, when accounting for interaction effects, varietal and vintage have the most significant association with price in this market. Blended varietals and single varietals have 14% and 56% higher prices than sangiovese respectively. This association between varietal and price is particularly strong for the top price quartile where ap pellation also shows a significant association with price However, appellation is not significantly associated with price in the interacted model. Vintage specifically the more recent vintages, has a significantly ne gative association with price which is to be expected, as wines that are older tend to be valued higher by consumers.
53 The US Market In the US mar ket model not accounting for interaction effects, uncertified organic wines comm and 15% higher prices than conventional wines yet certified organic wines do not command any pre mium over conventional wines. However, when accounting for interaction effects, uncertified organic wines command a 49% higher price than conventional wines while certified organic wines still do not command any signifi cant price premium. Drawing comparisons to the results of Delmas and Grant (2011), we see somewhat similar results in terms of organic status. Italian wines produced organically and sold in the US market receive significant price premiums over convention al wines as long as they are not certified organic and therefore not labeled as such. It is somewhat difficult however to draw direct comparisons between the results of this study and those of Delmas and Grant, as their study went further to treat organic labeling as a separate variable altogether Varietal and vintage both have significant correlation with price in the US market as well. Both single varietals and blended varietals command significantly higher prices than sangiovese wines, with price prem iums of 59% and 25%, respectively. Like the Italian models show, the association between varietal and price is greatest in the top price quartile. While appellation shows no significant price association in the interacted model, it does appear that appellation is significantly associated with price in the top price quartile in the US market as well. More recent vintages ( 2006 2008 ) receive significantly lower prices than the base year of 1999 again suggesting that older wines are more highly value d by consumers.
54 Comparisons Between Markets and Implications Organic status seems to have little correlation with price in the Italian market, whereas in the US market, uncertified organic wines command significantly higher prices. This should be of considerable interest to producers. Producers who sell most of their wines in domestically (in the Italian market) are not commanding the price premiums that would help offset the cost of conv erting to organic production or the higher cost of certifying organic productio n. A possible cause of this could be lack of consumer knowledge about organic practices. If consumers are unaware of the potential benefits of organic production, or are not i nformed that a wine is produced organically, they do not have the necessary information needed to differentiate organic wines from conventional wines. Organic wine producers would likely benefit from including information about the organic production prac tices on the wine label if they want to be able to command price premiums for their wines. T hose producers who sell most of their wines in the US market could certainly benefit from converting to organic production practices but shou ld probably forgo cer tification since it is a costly process and does not appear to have any significant relationship with price. Organic wine producers might also benefit from forming or joining some sort of organic wine trade organization that would enable them to collective ly educate consumers about organic wine. Regardless of the market in which producers sell their wines, a common finding is that varietal is significantly associated with price premiums. Wines made from single varietals or wines th at are blended varietals tend to receive higher prices than wines made from sangiovese
55 The appellation of origin seems to have a significant price association for ultra premium wines (wines in the top price quartile) but is of less importance for the lower priced wines. Associations between vintage and price show that younger wines (2006 2008) receive significantly lower prices in both markets. The wine rating in almost all models, was positively and significantly correlated with higher wine prices, which is not an une xpected result. However what is interesting is the association of rating and price when examining the data segmented into price quartiles. We found that in both the US and Italian markets, a 1% increase in rating corresponded to a much higher price premi um in the fourth (most expensive) quartile than in the three lesser priced quartiles. Rating was significantly and positively correlated with price in the lo west quartile as well but not to the same degree Limitations of this Study Our study is certainl y not without limitations. Our price data for both US and Italian markets came from online retailers. These prices may not accurately reflect the prices consumers are paying for wine in supermarkets or specialty wine stores, where most wine purchases are made. Furthermore, the mean price for both the US and Italian markets is considerably high suggesting that many of the wines in our dataset collectable wines or wines p urchased for special occasions. The prices of these wines might be driven by winery name and reputation rather than the other label characteristics. The fact that consumers would ultimately have to pay shipping and packaging costs in addition to the pric es used in our study further suggests that the
56 wines we have chosen for our dataset have additional value for consumers that are is not being captured by our models. Most wines were available through a number of online retailers at different prices, as ret ailers have different markups on products. To remove some of the potential bias introduc ed by individual retailers, we used the average of those prices in our dataset. However some wines were only available through one or two reta ilers. In these cases, retailer bias could not be averaged out. It is also worth noting that the organic status of a wine could also be considered a production variable instead of a label characteristic, particularly in the case of wine produced in the EU prior to organic lab eling regulations. It might therefore be appropriate to include the organic variable in a price model that accounts for other production level characteristics, such as vineyard size, cases produced, number of employees, winery age, etc.
57 LIST OF REFERENCES Angulo, A. M., Gil, J. M., Gracia, A., & Snchez, M. (2000). Hedonic prices for spanish red quality wine. British Food Journal, 102 (7), 481 493. Benfratello, L., Piacenza, M., & Sacchetto S. (2009). Taste or reputation: What drives market prices in the wine industry? estimation of a hedonic model for italian premium wines. Applied Economics, 41 (17), 2197 2209. Boatto, V., Defrancesco, E., & Trestini, S. (2011). The price premium for wine quality signals: does retailers' information provision matter?. British Food Journal 113 (5), 669 679. Boccaletti, S., & Nardella, M. (2000). Consumer willingness to pay for pesticide free fresh fruit and vegetables in italy. The International Food and A gribusiness Management Review, 3 (3), 297 310. Bombrun, H., & Sumner, D. A. (2003). What determines the price of wine. The Value of Grape Characteristics and Wine Quality Assessments.AIC Issues Brief, 18 Brentari, E., & Levaggi, R. (2010). Hedonic price f or the italian red wine: A panel analysis. Available at SSRN 1440092, Brugarolas Moll Bauz, M., Martnez Carrasco, L., Martnez Poveda, A., & Rico Prez, M. (2005). Determination of the surplus that consumers are willing to pay for an organic wine. Span ish Journal of Agricultural Research, 3 (1), 43 51. Castellini, A., Mauracher, C., Procidano, I., & Sacchi, G. (2014). Italian market of organic wine: A survey on production system characteristics and marketing strategies. Paper presented at the 140th Semi nar, December 13 15, 2013, Perugia, Italy, (163348) the consumers of organic products. British Food Journal, 104 (3/4/5), 187 199. Combris, P., Lecocq, S., & Visser, M. (1997). Estimation of a hedonic price equation for Bordeaux wine: does quality matter?. The Economic Journal 107 (441), 390 402. Costanigro, M., McCluskey, J. J., & Mittelhammer, R. C. (2007). Segmenting the wine market based on price: Hedonic regression when different prices mean different products. Journal of Agricultural Economics, 58 (3), 454 466. Crescimanno, M., Ficani, G., & Guccione, G. (2002). The production and marketing of organic wine in sicily. British Food Journal, 104 (3/4/5), 274 286. de Magistris, T., & Gracia, A. (2008). The decision to buy organic food products in southern italy. British Food Journal, 110 (9), 929 947.
58 Delmas, M. A., & Grant, L. E. (2014). Eco labeling strategies and price premium: The wine industry puzzle. Bus iness & Society, 53 (1), 6 44. doi:10.1177/0007650310362254 Didier, T., & Lucie, S. (2008). Measuring consumer's willingness to pay for organic and fair trade products. International Journal of Consumer Studies, 32 (5), 479 490. Dimara, E., & Skuras, D. (2 005). Consumer demand for informative labeling of quality food and drink products: A european union case study. Journal of Consumer Marketing, 22 (2), 90 100. European Counsil. European Commision, (2007). Council regulation (EC) N o 834/20 07 of 28 J une 2007 on organic production and labelling of organic produc ts and repealing regulation (EEC) N o 2092/91 (834/2007). Retrieved from website: http://eur lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:32007R0834 :EN:HTML Eurostat. (2014). Table food_in_porg1. Ce rtified organic crop area by crops products [Table]. Retrieved from http://appsso.eurostat.ec.europa.eu/nui/submitViewTableAction.do Food and Agriculture Organization. (1999, January). Organic agriculture. Committee on agriculture, Rome, Italy. Retrieved f rom http://www.fao.org/docrep/meeting/X0075e.htm Forbes, S. L., Cohen, D. A., Cullen, R., Wratten, S. D., & Fountain, J. (2009). Consumer attitudes regarding environmentally sustainable wine: An exploratory study of the new zealand marketplace. Journal of Cleaner Production, 17 (13), 1195 1199. Govindasamy, R., & Italia, J. (1999). Predicting willingness to pay a premium for organically grown fresh produce. Journal of Food Distribution Research 30 44 53. IFOAM EU Group. IFOAM EU Group, (2013). Eu rules for organic wine production. Retrieved from JCBGAM printing solutions website: http://ifoam eu.org/sites/default/files/page/files/ifoameu_reg_wine_dossier_201307.pdf Janssen, M., & Hamm, U. (2012). Product labelling in the market for organic food: Consumer preferences and willingness to pay for different organic certification logos. Food Quality and Preference, 25 (1), 9 22. doi: http://dx.doi.org/10.1016/j.foodqual.2011.12.004 Jarisch, R., & Wantke, F. (1996). Wine and headache. International archiv es of allergy and immunology, 110(1), 7 12. Johnston, R. J., Wessells, C. R., Donath, H., & Asche, F. (2001). Measuring consumer preferences for ecolabeled seafood: An international comparison. Journal of Agricultural & Resource Economics, 26 (1)
59 Landon, S ., & Smith, C. E. (1997). The use of quality and reputation indicators by consumers: The case of bordeaux wine. Journal of Consumer Policy, 20 (3), 289 323. Lecocq, S., & Visser, M. (2006). What determines wine prices: Objective vs. sensory characteristics Journal of Wine Economics, 1 (01), 42 56. Leire, C., & Thidell, . (2005). Product related environmental information to guide consumer purchases a review and analysis of research on perceptions, understanding and use among nordic consumers. Journal of C leaner Production, 13 (10), 1061 1070. Li, J., Zepeda, L., & Gould, B. W. (2007). The Demand for Organic Food in the US: An Empirical Assessment. Journal of Food Distribution Research 38 (3). Lin, B. H., Smith, T. A., & Huang, C. L. (2008). Organic premium s of US fresh produce. Renewable agriculture and food systems 23 (03), 208 216. Loureiro, M. L. (2003). Rethinking new wines: Implications of local and environmentally friendly labels. Food Policy, 28 (5), 547 560. Loureiro, M. L., & Lotade, J. (2005). Do fair trade and eco labels in coffee wake up the consumer conscience? Ecological Economics, 53 (1), 129 138. Loureiro, M. L., & McCluskey, J. J. (2000). Consumer preferences and willingness to pay for food labeling: a discussion of empirical studies. Journa l of Food Distribution Research 34 (3), 95 102. Nie, C., & Zepeda, L. (2011). Lifestyle segmentation of US food shoppers to examine organic and local food consumption. Appetite 57 (1), 28 37. Oczkowski, E. (1994). a hedonic price function for australian premium table wine*. Australian Journal of Agricultural and Resource Economics, 38 (1), 93 110. Onyango, B. M., Hallman, W. K., & Bellows, A. C. (2007). Purchasing organic food in US food systems: A study of attitudes and practice. British Food Journal 1 09 (5), 399 411. Organic Consumers Association. (2014). Clearing up confusion about organic wine. Retrieved from http://www.organicconsumers.org/Organic/OrganicWine.cfm Organic Trade Association. (2012, July 11). US EU organic equivalence arrangement. Retri eved from http://www.ota.com/GlobalMarkets/US EU Organic Equivalence Arrangement.html
60 Parker, R. (2013, December 31). The wine advocate vintage chart (2012 1970) Retrieved from http://www.erobertparker.com/newsearch/vintagechart1.aspx Pellegrini, G., & Farinello, F. (2009). Organic consumers and new lifestyles: An italian country survey on consumption patterns. British Food Journal, 111 (9), 948 974. Remaud, H., Mueller, S., Chvyl, P., & Lockshin, L. (2008). Do Australian Wine Consumers Value Organic Win e?, Robinson, J. (1999). The oxford companion to wine. (2nd ed., pp. 673 674). New York, NY: Oxford University Press. Rosen, S. (1974). Hedonic prices and implicit markets: Product differentiation in pure competition. The Journal of Political Economy, 3 4 55. Saba, A., & Messina, F. (2003). Attitudes towards organic foods and risk/benefit perception associated with pesticides. Food Quality and Preference, 14 (8), 637 645. Stolz, H., & Schmid, O. (2008). Consumer attitudes and expectations of organic wine Thomas, A., & Pickering, G. (2003). The importance of wine label information. International Journal of Wine Marketing, 15 (2), 58 74. Thrane, C. (2004). In defence of the price hedonic model in wine research. Journal of Wine Research, 15 (2), 123 134. U nited States Department of Agriculture. USDA, Electronic Research Service. (2013). Organic Production. Retrieved from website: http://www.ers.usda.gov/data products/organic production/documentation.aspx United States Department of Agriculture. USDA National Organic Program. (2000). National organic program (Organic foods production act provisions). Retrieved from website: http://www.ecfr.gov/cgi bin/text idx?c=ecfr&sid=3f34f4c22f9aa8e6d9864cc2683cea02&tpl=/ecfrbrowse/Title07/7c fr205_main_02.tpl Unwi n, T. (1999). Hedonic price indexes and the qualities of wines. Journal of Wine Research, 10 (2), 95 104.
61 BIOGRAPHICAL SKETCH Lane Abraben was born, raised, and educated in Gainesville, Fl orida. He received a Bachelor of Science in B usine ss A dmin istration and a Master of Science in Food and Resource Economics from the University of Florida. He has also received a Master of Science in Rural Development from the University of Gent in Belgium. His interest in organic food production and his extracu rricular interest in wine converged to become the basis of his thesis on price premiums for organic wines.