Unstable Individual Preferences and Stable Aggregate Demand Wild

Unstable Individual Preferences and Stable Aggregate Demand: Wild and Farmed Fish in the French Market. Kyrre Rickertsen (Based on work with Frode Alfnes, Pierre Combris, Géraldine Enderli, Sylvie Issanchou, and Jason F. Shogren) Invited Presentation at the National Chung Cheng University, Chia-Yi, Taiwan, March 7, 2011 2111 2005

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Outline Presentation § Norwegian fish in the world market § Cod farming § Our research objectives § Survey § Sensory evaluation § Bidding – BDM § Results: Survey, sensory evaluation, BDM § Unstable individual preferences and stable market demand – Distributed paper – Preferences measured as hedonic scores and WTP 2 § Why unstable individual demand preferences? § Conclusions www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Wild Catch and Aquaculture 2007 PRC # 1 (>45 million tons) Norway #11 (3 million tons) (Source NSEC) 3 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Main Exporters Seafood, 2007 PRC #1 & Norway #2 (Source NSEC) 4 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Norwegian Export of Seafood 1988 – 2009 Growth in aquaculture (dark blue) Seafood exports US$ 7 billion (about 6% of Norwegian export value) (Source NSEC) 5 www. umb. no

NORWEGIAN UNIVERSITY OF LIFE SCIENCES School of Economics and Business Norwegian Export of Seafood in 2009 by Region 6 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Norwegian Export by Country France #1, Russia #2, and Denmark #3 (Source NSEC) 7 www. umb. no

NORWEGIAN UNIVERSITY OF LIFE SCIENCES School of Economics and Business Norwegian Export of Wild White Fish Cod Dominates 8 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Production of Farmed Salmon in Norway More than half the value of fish is farmed salmon 9 www. umb. no

NORWEGIAN UNIVERSITY OF LIFE SCIENCES School of Economics and Business Cod Farming http: //www. codfarmers. com/ www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Cod Farming - Minor product in the world - about 21, 000 tons in 2009 • Other farmed whitefish species include: – Tilapia (2. 8 mill tons) and pangasius (1 mill tons) • Other wild whitefish species include: – North-Atlantic cod (900, 000 tons), Pacific cod (400, 000 tons), Alaskan pollock (2. 8 million tons) • Norwegian farmed salmon (800, 000 tons) - Still in early development - 2009: Cod production passed 20, 000 tons - 1984: Salmon production passed 20, 000 tons • Can farmed cod production reach the levels of salmon production? – Depends on production costs and WTP in market www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Background Study § Our principal objectives: – Investigate the market potential (and WTP) for farmed cod in the European market – Determine how the expected growth in supply of farmed cod will affect the demand for farmed salmon and wild cod § Methods: – Sensory evaluations, experimental auctions, choice experiments, and surveys § Thanks to Sylvie Issanchou, Géraldine Enderli, Pierre Combris, Jason F. Shogren, Cod. Farmers, The Research Council of Norway § Work in progress. Any comments are very welcome www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Sample I § Three experiments at INRA’s laboratory in Dijon in December 07, and May 08, and December 08 § Dijon has about 150, 000 inhabitants and is representative for France in terms of income and socioeconomic composition – Fresh fish consumption is slightly below the mean for France but representative for non-costal regions § Participants were recruited from INRA’s consumer panel in Dijon § A total of 276 unique participants – 60 participated 3 times – 89 participated 2 times – 127 participated once 13 – Can say something about stability over time www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Sample II § 59% female § Mean age 47 years § 35% had not graduated from high school § 39% had finished studies at university level § 36% of households earned less than € 2, 000 per month § 26% of households earned more than € 3, 000 per month § Average household size 2. 5 § All the participants were eating fish at least once per month and buying fish at least every second month 14 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Sample III § Participant had to eat about 250 grams of fish so the experiments were conducted either at lunch or dinner time § Participants were paid € 25 to participate in a 90 minutes session § The participants evaluated five types of fish – Farmed salmon (SA) from Norway – Wild cod (WC) from North Atlantic area – Farmed cod (FC) from Norway – Wild monk (MO) – expensive substitute from North Atlantic area – Farmed pangasius (PA) – inexpensive substitute from Vietnam 15 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Each Experiment Consisted of Three Parts § Taste experiment – Fish served in rotational order – 10 points scale on computer – Uninformed tasting (e. g. , cod) in December 07 – Informed tasting (e. g. , farmed cod from Norway) in May 08 and December 08 § http: //www. eatmedaily. com/2009/05/stephen-colbert-eats-cat-food-video/ § Valuation experiment (BDM) – Training with orange juice or champagne – Bids for each type of fish (with tasting scores on screen) – Participants have the incentives to bid their true WTP § Survey 16 – Demographic, socioeconomic, attitudinal, and beliefs questions like “What is your best guess at the average market price for. . ” www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES 17 Available Data December 07 May 08 December 08 Informed tasting Bid 2 RC: fish, cut, price SC: fish, label, price, info scenarios Questionnaire 1 a Questionnaire 1 b +2 SA, FC, WC, MO, PA OJ, CH E 1, E 2, E 3 Uninformed tasting Bid 1 Information Bid 2 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Some Results from Questionnaire I Origin of Farmed and Wild Fish 1 = Very strongly disagree 10 = Very strongly agree I have a very positive view of fresh farmed fish from: France 7. 60 Countries of Northern Europe 7. 39 Countries of Southern Europe 5. 84 Other developed countries 4. 83 Third world countries 2. 78 I have a very positive view of fresh wild fish from: 18 The North Atlantic 8. 20 The Mediterranean 7. 09 The Pacific 6. 93 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES 19 Some Results from Questionnaire II Attitudes toward Fish and Fish Production 1 = Very strongly disagree 10 = Very strongly agree Farmed Wild Fish is healthy foods 5. 60 8. 23 Fish is safe to eat 5. 53 7. 13 I am concerned about environmental impact 7. 21 7. 96 I am concerned about the environmental sustainability 7. 45 7. 95 I am concerned about the welfare 6. 71 6. 96 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES 20 Some Results from Questionnaire III Attitudes toward Different Fish Types 1 = Very strongly disagree 10 = Very strongly agree Salmon Cod Monk Pangasius Tastes good 8. 11 7. 16 7. 63 3. 75 Good value for money 7. 19 6. 56 4. 59 5. 84 Easy to use in different dishes 8. 04 7. 03 6. 66 4. 57 Healthy food 8. 26 8. 01 8. 07 5. 01 Fat food 7. 31 4. 08 4. 54 4. 25 Safe to eat 6. 74 6. 73 6. 97 3. 85 Easy to prepare 8. 81 8. 33 6. 18 6. 76 Expensive 6. 38 6. 28 8. 71 3. 49 Whole family likes 7. 64 7. 44 7. 12 3. 91 Can be served on special occasions 7. 46 4. 72 8. 68 3. 04 Monday-to-Friday fish 6. 38 7. 55 2. 94 5. 55 www. umb. no

NORWEGIAN UNIVERSITY OF LIFE SCIENCES School of Economics and Business Sensory Evaluation 21 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES Joint Distribution of the Hedonic Scores in the Three Experiments 10 = “like very much” and 0 = “do not like at all” 22 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Mean Hedonic Scores § Corresponds well with Dec 07 Uninf Salmon 7. 90 7. 65 7. 47 7. 43 Farmed Cod 7. 06 6. 21 5. 65 6. 29 Wild Cod 7. 06 5. 66 6. 42 6. 47 Monk 7. 43 6. 39 7. 19 7. 03 Pangasius 3. 49 3. 76 3. 12 3. 21 § Taste scores stable for salmon but change for other fishes § Changes may be due to: – Seasonal quality differences/seasonal effects in taste • FC – Uniformed vs. informed tasting 23 Questionnaire questionnaire • WC, MO, PA www. umb. no May 08 Dec 08

NORFISH 2010 -2013 NORWEGIAN UNIVERSITY OF LIFE SCIENCES Difference in Hedonic Scores FC – WC, Dec. 2008 Segments Are Important 24 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES BDM (1964) - Bids § Explained the BDM § Trained on OJ or champagne § Participants stated their maximum willingness to pay for each of the five fish types § Binding fish type was randomly drawn § (Individual) random prices drawn for selected fish § If the bid was above the random price participant had to buy fish at the randomly drawn price 25 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES The Market Demand Curves, i. e. , Joint Distribution of Bids in the Three Experiments 26 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Mean Bids in € per kg and Percentage Zero Bids December 07 Uninformed Informed May 08 Informed December 08 Informed Bid % Salmon 10. 30 11 10. 22 10 10. 24 11 10. 06 12 Farmed Cod 7. 11 29 6. 80 31 7. 43 26 8. 32 20 Wild Cod 6. 85 37 6. 83 36 9. 13 19 9. 17 14 Monk 9. 59 29 9. 64 30 11. 38 18 10. 88 17 Pangasius 3. 74 60 3. 10 65 2. 25 72 2. 64 66 27 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Summary Changes in Mean Bids and Zeros § About 10% zero bids for salmon § About two-thirds zero bids for pangasius § Uniformed vs. informed in December 07 – FC and PA reduced § Informed bids December 07 vs. bids in 08 – Increased substantially for FC, WC, and MO – Stable for SA – Reduced substantially for PA – Can these changes be explained? 28 • Later www. umb. no

NORWEGIAN UNIVERSITY OF LIFE SCIENCES SCHOOL OF ECONOMICS AND BUSINESS The Bids for Wild Cod Shifted Substantially 29 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES 30 Ranking of Bids and Hedonic Scores Both types of wild fish moved one step up in the rankings in 08 1 2 3 4 5 Bids: December 07, uninf. SA MO FC WC PA Bids: December 07, informed SA MO WC FC PA Bids: May 08 MO SA WC FC PA Bids. December 08 MO SA WC FC PA HS: December 07, uninf. SA MO FC WC PA HS: May 08 SA MO WC FC PA HS: December 08 SA MO WC FC PA www. umb. no

NORFISH 2010 -2013 NORWEGIAN UNIVERSITY OF LIFE SCIENCES Difference in WTP: FC – WC, December 2008 Segments Are Important 31 www. umb. no

NORWEGIAN UNIVERSITY OF LIFE SCIENCES School of Economics and Business The Distributed Paper 32 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Unstable Individual Bids and Stable Market Demand § Hedonic scores and bids are measures the strength of individual preferences § We have two research questions in this paper: – Question 1 Are individual bids and hedonic scores stable? – Question 2: Will unstable individual bids (hedonic scores) result in stable market demand (stable market taste)? • Aggregation will wash out individual specific effects such as variety seeking behavior • Aggregation will wash out effects of random anchors such as exposition to advertising § Problem with material: 33 – How does it relate with existing literature? Discussed later! www. umb. no

NORWEGIAN UNIVERSITY OF LIFE SCIENCES Stability of Bids – Neoclassical Theory § Theory: Consumers’ choices are based on the maximization of a stable utility function subject to a budget constraint Bid § Stable individual preferences will, ceteris paribus, result in stable bids in experimental auctions conducted at different times S p. A § Stable individual bids is a sufficient but not a necessary condition for stable aggregate demand curves D A § Becker (1962), theoretical paper 34 – Showed how stable aggregate demand curves could be derived from random choices within the budget set www. umb. no 1 Quantity

NORWEGIAN UNIVERSITY OF LIFE SCIENCES Stable Preferences in Neoclassical Theory § What does “stability of preferences” mean? ? ? § “Constancy of tastes is a simplifying assumption, not an article of faith” (Silberberg and Suen 2001: 6) § The assumption of stable preferences for specific goods has been challenged in (for example) applied demand studies – Parametric as well as nonparametric methods (Varian) have been used – Stable preferences have been rejected (e. g. , Moschini and Meilke 1989; Jin and Koo 2003) and not rejected (e. g. , Chalfant and Alston 1988) § Alternative interpretations of stability of preferences are provided by: – Becker (1976: 5) who discuss stability of “underlying preferences” such as health, prestige, and envy 35 – Lusk and Briggeman (2009) who discuss stability of “food values” such as safety, nutrition, and taste www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Stability in Preferences over Specific Goods § We look at specific goods § In an empirical study of the Marseille fish market, Härdle and Kirman (1985) found: – Downward sloping aggregate but not individual demand curves – “Sophisticated and complicated individual behavior may lead to simple aggregate properties” – Similar to what we find in an experimental setting 36 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Demand Curves § Individual demand curves are defined as: § The fish was sold in packages of about 300 grams – May be insufficient for a family meal – Participants could choose how many packages they wanted to buy for the market price before they started to bid – The quantity (x) counted as a 1 regardless of # of packages § The market demand curve is constructed by ranking the individual bids from the highest to the lowest, i. e. : 37 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES Demand Curve SA May Are Individual Bids Stable? No 116 participants who participated in May and December 2008 38 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES Demand Curve WC May Are Individual Bids Stable? No 116 participants who participated in May and December 2008 39 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES The Bids May be Affected by Changes in External Variables! Are Individual Bid Differences Stable? No 116 participants who participated in May and December 2008 40 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES Inverse Survival Function SA May Are Individual Hedonic Scores Stable? No 116 participants who participated in May and December 2008 41 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES Inverse Survival Function WC May Are Individual Hedonic Scores Stable? No 116 participants who participated in May and December 2008 42 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES The Hedonics May be Affected by Changes in External Variables! Are Individual Bid Differences Stable? No 116 participants who participated in May and December 2008 43 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Stability of Individual Bids, 116 Participants SA WC FC MO PA Constant bid 21 17 17 22 79 Change > 1 SD 20 39 36 30 18 Constant ranking 53 47 43 61 69 Total 11 § Constant bids were typically zero bids § The SD are in the range of € 3. 74 to € 6. 91 for different fish types § Only 11 participants had a constant ranking of the bids § Only 53 participants ranked salmon identically § All correlation coefficients are statistically different from zero and one (95% CI 0. 31 -0. 75) 44 § Same conclusions for hedonic scores www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Stability of Market Demand § Next, we constructed the market demand curves for December and for May – Rank the participants bids from the highest to the lowest independently according to: 45 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES Demand Curve SA May and December Are the Market Demand Curves Stable? Yes 116 participants who participated in May and December 2008 46 www. umb. no

SCHOOL OF ECONOMICS AND BUSINESS NORWEGIAN UNIVERSITY OF LIFE SCIENCES Demand Curve WC May and December Are the Market Demand Curves Stable? Yes 116 participants who participated in May and December 2008 47 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Stability of Market Demand § Wilcoxon signed rank test did not reject stable median bids with the exception of FC – The median bid for FC increased significantly – May be explained by the significant increase in hedonic score § A Brown-Forsythe test did not identical variances of the bids § A Kolmogorov-Smirnov test did not reject identical PDFs for the bids 48 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Conclusions Paper § Unstable individual bids § Aggregation creates stable market demand § Random day-to-day variations in individual preferences have minor effects on the stability of the market demand § Good news for producers and retailers who are mainly interested in predicting the market demand 49 www. umb. no

NORWEGIAN UNIVERSITY OF LIFE SCIENCES School of Economics and Business Can We Explain the Changes in Individual Preferences? 50 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Why Unstable Individual Demand Preferences? § According to economics, changes in demand can be caused by: – Changes in relative prices and income – Changes in quality (of heterogeneous goods) – Storage – Preferences – Socioeconomics § The marketing literature focus on changes in preferences from purchase-to-purchase occasion – Changes in external factors such as new information, available products, presence of other people, seasonality, etc – Changes in internal factors; consumers seek variety 51 www. umb. no

Department of Economics and Resource Management NORWEGIAN UNIVERSITY OF LIFE SCIENCES Why Unstable Individual Demand Preferences? § Behavioral economics focuses on effects of arbitrary anchors and the framing (of experiments) on preferences: – For example coherent arbitrariness (Ariely, Loewenstein, and Prelec 2003) • Are you willing to pay a dollar figure equal to the last two digits of your social security number for a “cordless keyboard” ? • State your WTP for “cordless keyboard” • Peoples absolute valuations affected by arbitrary anchors (social security number) • Peoples relative valuations coherent (as supported by a stable demand curve) 52 § If your social security number affects your WTP, then many other arbitrary variables will affect them! www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Preliminary Model § Try to explain hedonic scores and bids § We use the bids from all the experiments – We have 2, 425 hedonics scores including 53 zero scores – We have 2, 845 bids including 811 zero bids – Data are censored at zero § Tobit models to take account of these zero bids 53 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Preliminary Model - Limitations § Some limitations in preliminary model: – A double hurdle model would be preferred, but we do not know how to estimate them given the panel nature of the data – Hedonic scores and bids both reflect preferences. Have not taken into account the endogeneity of hedonic scores in bid equations – Do not take censoring by market prices of field substitutes into account – Have not estimated relative bids and hedonic scores • Relative bids more consistent with theory 54 • Relative bids should wash out many of the systematic effects like CA and other random anchors – Have not looked at cross-price effects www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Preliminary Model - Variables § Hedonic scores for fish equation – Socioeconomic: Age and gender – Quality: Type of fish – Coherent arbitrariness: Test product orange juice or champagne? – Information: Country-of-origin and “wild or farmed” before tasting – Ordering: The first product typically receives the highest score – Time of the day: Lunch or dinner – Day of week effect: Monday-to-Friday or weekend fish § Additional variables in bid for fish equation – Prices of product: Bids are affected by the price of field substitutes – Quality: Hedonic score of fish 55 – Income – Information: Country-of-origin and “wild or farmed” before bidding www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Effects of Fish Type on Hedonic Scores Variable Parameter Constant 7. 49 WC -1. 36 § Reference product is salmon FC -1. 46 MO -0. 69 § Salmon is preferred to all other PA -4. 22 § All variables significant at the 1% level of significance fish types § Pangasius received an hedonic score that was 4. 22 lower than salmon 56 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Other Significant Variables, Hedonic Scores Variable Parameter Age 0. 01 CA (juice=1) -0. 34 § Taste scores increase by 0. 01 Order 2 -0. 43 Order 3 -0. 44 per year the participant's age increases Order 4 -0. 58 § We find CA also in hedonic Order 5 -0. 39 § All variables significant at the 5% level scores. If OJ was used as test product, hedonic score is reduced by 0. 34 § Ordering effects as expected § Gender, information, time-of- day, and day-of-week insignificant 57 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Effects of Quality on Bids Variable Parameter Constant -9. 30 Hedonic score 2. 75 § The variables WC, FC, and MO are insignificant, i. e. , the quality effects goes through hedonics and not fish type (except for pangasius) (Hedonic score)2 -0. 08 WC 0. 11 FC -0. 40 MO -0. 24 PA -1. 93 § The other variables are significant at the 1% level § Hedonics are very important § An increase in score from 5 to 6, increases bid by € 1. 87 § An increase in score from 8 to 9, increases bid by € 1. 39 58 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Other Effects on Bids Variable Parameter Price 0. 43 Price 2 0. 00 Gender (female=1) Infohed (info=1) § Significant at the 2% level § If the (expected) outside price increases with € 1, then bid increases with € 0. 43 (but at decreasing rate) -1. 89 0. 76 CA (juice=1) -1. 05 Days-to-weekend -0. 39 § Females bid € 1. 89 less § If information before tasting, bids increase with € 0. 76 § OJ as test product, decreases bids with € 1. 05 § Bids increase by € 0. 39 per day you get closer to weekend 59 § No significant effects of income, age, order of tasting, or time-ofday www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Conclusions Survey Market Potential for Farmed Cod § A positive view on fish from North Atlantic as compared with other areas – Slightly more positive towards wild than farmed fish § Wild fish scores best on health and safety § Farmed fish scores better on environmental issues § Cod is a Monday-to-Friday fish – Can it be upgraded? They try! – Sustainable labeling? 60 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Conclusions Hedonic Scores Market Potential for Farmed Cod § WC did on average better than FC § FC did better than WC in uninformed tasting § Segments are important – In December 2008, 60% preferred WC and 40% FC – Young participants preferred FC 61 www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Conclusions BDM Market Potential and WTP for Farmed Cod § WC received higher bids than FC on average – FC received higher bids than WC in uniformed bids • Can be explained by the higher hedonic score § Segments are important. In December 2008 – 36% bid more for FC than WC – 14% bid the same amount – 50% bid less for FC than WC § Strategies for FC – Difficult to take any higher price for FC than WC • Production costs critical 62 – Segments are important for FC – young people – To create value added product developments such as Ström may be a successful strategy www. umb. no

School of Economics and Business NORWEGIAN UNIVERSITY OF LIFE SCIENCES Some Other Conclusions § The market is characterized by – Unstable individual bids but stable market demand § Why unstable individual hedonics? – Fish type, CA, and ordering effects important § Why unstable bids? – Quality: Increasing hedonic scores increases bids – Outside prices: High outside prices increase the bids – CA: Expensive test product increases the bids – Females bid less 63 – Information about country-of-origin and “wild or farmed” increases bids – The bids increase towards the weekend www. umb. no

NORWEGIAN UNIVERSITY OF LIFE SCIENCES School of Economics and Business Thank You! 64 www. umb. no