Organic Food in Spain: Market Segmentation Essay

Abstract In recent years, consumers concerns on environmental and health issues related to food products have increased and, as a result, the demand for organically grown production. Higher costs of production and retailer margins generate a gap between real prices and those consumers are willing to pay for organic food. In this paper, consumer willingness to pay for organic food in two Spanish regions is analyzed. Markets in both regions are segmented considering consumers lifestyles. Results indicate that consumers concerned about healthy diet and environmental degradation are more likely to buy organic food and are willing to pay a higher premium. Organic attributes are easily identified in perishable products as the premium consumers would pay for organic meat, fruits and vegetables is higher. 1 PUBLICADO EN International Food and Agribusiness Management Review, (2001), 3, 207-226. 1 Market segmentation and willingness to pay for organic products in Spain 1. Introduction Food consumption in most developed countries has attained a saturation point in quantity terms, and consumer food choices are broader than in the past. The result is a more diversified consumption. In this saturated market environment, distribution channels, marketing activities, diversification strategies and food quality are increasingly important. In addition, consumers have become more concerned about nutrition, health and the quality of food they eat. The increasing importance of health, and the impact food production has on the environment, on consumer food choice is well documented in the literature (Jolly et al. , 1989; Jordan and Elnagheeb, 1991; Oude Ophius, 1991; Baker and Crosbie, 1993; Grunert and Juhl, 1995; Kleijn et al., 1996; Viaene and Gellynck, 1996; Chupitaz and Keslemont, 1997). As a consequence, organic products production and consumption have grown in recent years. The number of papers which have been devoted to the study of organic food markets has increased (Lampkin, 1989; Beharrel and MacFie, 1991; Landell Mills, 1992; Tregear et al. , 1994; Lin et al. , 1996; Vetter and Christensen, 1996; Thompson and Kindwell, 1998; among others). Organic farming refers to a farming system which uses organic manure, and avoids or largely excludes the use of synthetic fertilizers, pesticides and chemicals. A recent study carried out by FAO (1998) has shown that an adequate management of organic farming generates a positive impact on the environment (e. g. , reduction of water â€Å"contamination†, increased soil fertility due to crop rotation). On the demand side, consumers have positive attitudes towards organic products as they perceive them as healthier than conventional alternatives (Beharrel and MacFie, 1991). In the case of Spain, although the production of organic food products has considerably increased during the past decade, demand is still very low as only 0. 5% of food expenditure is allocated to such products. The main obstacle with organic production seems to be the difficulty in selling organic products in retail food markets. Although consumers search for more diverse, higher quality and healthier food products, organic products face problems related to consumer product acceptability (new product, high price and deficiencies in distribution channels (Roddy 2 et al. , 1994)). On the production side, high costs, especially labor costs, and the difficulty of shifting from conventional to organic farming are also limiting factors (Vetter and Christensen, 1996; Hamiti et al. , 1996). Furthermore, food availability and seasonality influence marketing activities and make it difficult to establish appropriate retailing outlets. Higher costs of production and retailer margins jointly may result in higher prices than consumers are willing to pay for organic food attributes. The objective of this study is to estimate the willingness of different consumer segments in Spain to pay for organic products, in order to assess alternative price strategies carried out by producers. From other studies, it was expected that consumers would be willing to pay a premium for organic products. This expectation is based on the idea that these products are healthier and may diminish negative environmental effects associated with conventional agricultural production. In order to expand the scope of our results, in this study we have considered a wide range of food products: 1) vegetables; 2) potatoes; 3) cereals; 4) fruits; 5) eggs; 6) chicken; and 7) red meat. Among the different methodological alternatives to assess consumers willingness to pay, the contingent valuation (CV) approach was chosen (Hanemann, 1984, 1987). Although CV is a method primarily used for monetary evaluation of consumer preferences for non-market goods (e. g. , unpriced natural resources), it is also useful in this context because the organic market is still too â€Å"thin†, and organic products are not available in all retail outlets. The paper is organized as follows. In the next section, some descriptive statistics on the evolution of organic farming in the European Union (EU) and, particularly, in Spain are shown. A brief description of the survey instrument used is provided in section 3. Next, consumer market segments based on consumers’ lifestyles are defined and characterized taking into consideration both socioeconomic characteristics and attitudes towards organic food products and environmental concerns. In section 5 the willingness to pay of each segment and all products considered are calculated. A brief outline of the theoretical foundation of the CV method is also included. Finally, some concluding remarks are outlined. 2. Relative importance of organic agricultural production in the European Union and Spain 3 Land under organic farming in EU countries has dramatically increased from 425 thousand ha in 1992 to 2.9 million ha in 1999 (Table 1). Although the area has grown seven fold, it still represents only 2. 2% of total cultivated agricultural land. The proportion of land devoted to organic production on total cultivated land varies from country to country. The highest values are found in Austria (8. 4%), followed by Finland (6. 3%), Denmark (5. 5%), Sweden (5. 5%) and Italy (5. 3%). The lowest are found in the rest of the Mediterranean countries (Portugal, Greece and Spain) in spite of the important increase of land devoted to organic products in these countries. Among EU countries the most spectacular increase in land devoted to organic production has taken place in Italy, although a high percentage corresponds to pastures. (Insert Table 1) Organic farming in Spain has been developed only recently but has increased rapidly during the past few years. Land devoted to organic production has increased from 7,900 ha, in 1992, to 352,000 ha, in 1999, when it accounted for 1. 4% of the total utilized agricultural land, still under the European average. Three-fourths of Spanish organic farming is concentrated in three regions: Extremadura (47%), Andalucia (17%) and Castilla-Leon (12%) (Table 2). In relative terms, regions with higher percentages of organic farming on total utilized agricultural area are Canarias (7. 7%), Extremadura (6%) and Comunidad Valenciana (2. 6%). Normally, organic production in each region is highly related to the crops which are traditionally produced there. At the national level, cereals and olives are the main organic crops followed by nuts and fruits. However, it is noticeable that a high percentage of land devoted to pastures is under organic production, mainly concentrated in the central and southern part of Spain. This explains the relative importance of Extremadura and Andalucia in organic production. Navarra, on the other hand, is the region in which more diversification exists, producing a wide variety of organic agricultural products (fruits, vegetables, cereals and meat). (Insert Table 2) Available information on consumption is less precise. According to the European Commission, EU expenditures on organic products accounted for 1. 5 million Euros in 1991, representing 0. 3% of total food expenditures. The proportion varies from country to country. Expenditure on organic products in Denmark lies between 2% to 3% of total food expenditures; in Germany, it represents 1. 2%; in the United Kingdom, Ireland, France and the Netherlands it is 4 less than 1%. In the case of Spain, the relative importance of organic food products is lower, accounting for only 0. 5% of total food expenditures. Comparing the Spanish figures on production and consumption of organic food products, it is clear that production has increased faster than consumption. The demand for organic food in central and northern European countries is growing more rapidly than in Spain. The main consequence is that, today, more than 50% of organically produced fruits and vegetables are sold in foreign markets leaving, to a certain extent, consumers unattended in the domestic market. Producers get a premium when selling in foreign markets and they expect and want to earn the same premium in the domestic market. However, as in Spain most of the conventional food products are less expensive than in other European countries (mainly, pasta, fruits and vegetables), the gap between conventional and organic products prices is higher in Spain, limiting the expansion of organic food consumption. The study tries to provide a better understanding of Spanish consumers in relation to organic food. Assuming that higher prices is the main limiting factor for increasing organic consumption, special attention will be paid to the maximum premium consumers are willing to pay for such products. As a previous step, consumers will be segmented according to their lifestyles and, then, market segments characterized taking into account consumers socioeconomic characteristics and attitudes towards organic food products and environmental concerns. 3. Data The data used here come from a survey conducted in July-August 1997 in two Spanish regions: Navarra and Madrid. Navarra was selected not only because it is one of the most important producing regions in Spain, but also because it produces a wide variety of organic food products. Madrid is one the most important regions, together with Cataluna, in terms of organic food products consumption. Samples in both regions were selected using a stratified random sample of food buyers on the basis of age and district of residence2 . Four hundred respondents were randomly selected and personally interviewed at home in each region. Respondents were the main purchasers of food products within the household. The first question in the questionnaire was respondents’ 2 Census data were available in both regions. 5 degree of knowledge of what an organic product was. If the respondent did not know anything about them, he/she was not interviewed. So, only respondents with at least some knowledge of organic products were considered 3 . Only 10% of respondents in Navarra and 5% in Madrid claimed to buy organic products regularly, while another 55% in Navarra and 42%, in Madrid, were occasional buyers. These percentages are similar to those obtained in other studies ( ullen and Wholegenant, 1991; M Hansen and Sorensen, 1992, 1993). 4. Market segmentation for organic food As mentioned above, the first step in this study was to group consumers into homogeneous clusters. Consumers were segmented according to their lifestyles as previous research has shown that the willingness to pay for an organic product might be influenced by individual ways of living rather than by the usual socioeconomic variables (Hartman and New Hope, 1997). Once the market segments were obtained, they were characterized taking into account both consumers socioeconomic characteristics and their attitudes towards organic food products and environmental issues. Among socioeconomic characteristics, age, gender, education level, family size and income were judged to be the most relevant. Lifestyles as well as attitudes towards environmental issues, on one hand, and towards organic food products, on the other, were measured by three tested scales (the complete scales are shown in the Appendix). Respondents were asked to evaluate, assigning a value from 1 to 7, their agreement with different sentences. These three scales have been reduced by a principal components analysis to extract main dimensions on each aspect. 4 The main results from these analyses follow. i) Lifestyles. Table 3 shows the correlation between the original variables (sentences) and factors obtained from principal components analysis in Navarra and Madrid. In Navarra, lifestyles was summarized in three factors which account for 53% of the total variance (Table 3, top). The first one, ‘Natural food consumption’, explains 30% of the total variance and is linked to 3 Only consumers with at least a certain knowledge on what organic products are were interviewed. In a previous qualitative research, 90% respondents said they had heard or knew something about organic products. More or less the same percentage was found in the survey. Thus, the sample seems reasonably representative. 4 The non-linear principal components analysis (De Leeuw and Rijckevorsel, 1980) was also used as an alternative to condense the information. Results were almost identical to those presented. 6 consumption of fresh fruits and vegetables and the concern for reducing the consumption of red meat, processed food or food with additives. The second factor, ‘Life Equilibrium’, explains 14% of the total variance, and refers to the existence of a real interest on keeping a balance between work and private life, living in a methodical and ordered way and trying to reduce stress. Finally, the third factor contributes to 9% of the total variance. It is called, ‘Health care’, and shows the interest of consumers for keeping themselves healthy, through practicing sports, following a natural diet, controlling salt ingestion and regularly checking their health. In Madrid, four factors were selected which account for 62% of the total variance (Table 3, botton). The first three factors are similar to those obtained in Navarra, but in a different order, indicating that their relative importance differs. The first factor here is, ‘ Health care’, which explains 28% of the total variance and is linked to consumers awareness to regularly check their health. The second factor, ‘Natural food consumption’, contributes to explain 13% of the total lifestyles variance and, as in the case of Navarra, is linked to variables indicating consumers consumption of healthier food products. The third factor, ‘ ife Equilibrium’, has a similar L interpretation to the case of Navarra. Finally, the fourth factor, ‘Mediterranean diet ’, is related to a high consumption of fruits and a moderate consumption of meat. (Insert Table 3) ii) Attitudes towards environmental issues. In relation to attitudes about environmental issues, two factors were selected in both regions which explained 63% and 70% of the total variance in Navarra and Madrid, respectively (Table 4). The first factor, ‘Environmental conservation’, is related to variables indicating that consumers have an active interest in reducing the effect of environmental degradation by recycling products and using recycled products, etc. The second factor, ‘Environmental concerns’, is linked to those variables showing consumers awareness about the negative effect of development on the environment. (Insert Table 4) iii) Attitudes towards organic food products Table 5 shows the results of the principal components analysis carried out on variables related to consumers’ attitudes towards organic food products. The original information was summarized in two and three factors in the cases of Navarra and Madrid, respectively (Table 5). In Navarra the two factors account for 45% of the total variance. The first factor, ‘ Positive aspects’, emphasizes the quality, taste, healthiness, attractiveness and absence of harmful effects 7 of these products. The second factor, ‘Negative aspects’, is related to the perception that organic products are only a new fashion and more expensive than the conventional ones. In Madrid the three factors explain the 59% of the total variance. In this case, the organic food positive aspects were divided in two factors, ‘Quality and healthy aspects’ and ‘External appearance’ (related to organic food products’ attractiveness and taste). The â€Å"negative aspects† factor is related to the same attitudes as in the case of Navarra. (Insert Table 5) Market segmentation The K-means cluster analysis technique (Malhotra, 1993) was used to identify market segments in relation to organic food products. Lifestyles factors (Table 3) were used as segmentation variables. Three segments in Navarra and four segments in Madrid were identified. Each segment was characterized taking into account: consumers’ socioeconomic characteristics (age, gender, education level, family size and income); factors related to attitudes towards environmental issues (Table 4); factors related to attitudes towards organic food products (Table 5); and the consumption level of organic food products 5 . Results from cluster analysis and market segments characterization are shown in Tables 6 and 7, for Navarra and Madrid, respectively. (Insert Table 6) In Navarra, the first segment accounts for 25% of the sample (Table 6). It includes those people who show a leaning towards natural food consumption and a balanced life, but with no excessive care for their health. This is a potential consumer group of organic products as most of respondents occasionally consume them and a high proportion is willing to taste them in the near future. For this reason, the segment is labeled †Likely consumers†. Consumers in this segment are mainly women, middle-aged, with an educational level of high school or less and they positively assess the beneficial aspects of organic food products. The second segment includes 52% of consumers. It is formed by respondents evenly split between men and women. They are also not well educated (high school or less) and 75% are of medium income. They are worried about health and the balance between private life and work, but pay less attention to follow a natural diet. The percentage of regular and occasional 5. Four categories were defined: 1) regular consumption, 2) occasional consumption; 3) no consumption but probably yes in the near future; 4) no consumption at all. 8 consumers is the highest among all segments; therefore this segment is called â€Å"Organic food consumers†. Nevertheless, this big proportion could reflect certain confusion among consumers between real organic products and those grown in home gardens, for self-consumption. This is very frequent in this area. Navarra is a very important producer region of fruits and vegetables and many people living in town keep strong links with rural areas. This fact could also explain the negative value assigned to the â€Å"natural food consumption† factor. Furthermore, no need for a special consideration of healthy diet is shown as it is intrinsic to consumption habits. The third segment accounts for 23% of respondents. This is clearly the least involved group on natural food consumption, life equilibrium and health care and includes the highest percentage of no consumers. The relatively low potential consumption together with the lack of awareness about health and more natural diets suggest that organic consumption in this group will not be stimulated. Therefore, this segment is labeled as â€Å"Unlikely consumers†. Half of consumers in this segment have less than 35 years old. Finally, it is important to note that among the â€Å"unlikely consumers† half of them have more than a high school. Results from segmentation in Madrid are different because of the heterogeneity of an over 5 million inhabitants town (Table 7). Four segments were identified. Half of the consumers within the first segment (23% of the population) are over 60 years old and there are more households with 2 o fewer inhabitants than in the other groups. Consumers in this group are better educated than in other segments. However, almost 90% of consumers are of medium or modest income. They are not very aware of environmental problems but are worried about health issues and try to follow a â€Å"balanced† life. Mediterranean diet is valued positively but consumption of organic products is not of much interest for them as they do not appreciate the positive organic food products attributes. Therefore, this segment is called â€Å"Unlikely mature consumers†. (Insert Table 7) The second segment is similar to the previous one in terms of the consumption level although the percentage of regular and occasional consumers is even lower. In relation to sociodemographic characteristics consumers in this segments are younger, than those of the first segment, the percentage of male consumers is higher and, finally, family size is larger. They are not very involved in environmental issues, have a negative image of organic products and are not worried about diet and health but try to maintain a certain equilibrium between working and 9 private life. This segment is labeled â€Å"Unlikely young consumers† and accounts for 20% of the population. The third segment (22% of the population) is also mainly formed by households with 3 or 4 members (usually a couple with one or two children). The percentage of consumers educated beyond high school is very low in this group (17%). Consumers within this segment are looking for a more natural diet in which Mediterranean products play an important role, although they show a negative attitude towards the external appearance of organic food products. They are occasional buyers of organic food and are likely to increase their consumption if the natural attribute of such products is reinforced. Thus, this segment is called â€Å"Likely consumers†. Finally, the forth segment accounts for 35% of the population and is labeled â€Å"organic food consumers†. The percentage of regular and occasional consumers is the highest in relation to other groups. It is quite similar to the second segment found in Navarra although it is smaller. Consumers do not show special socioeconomic profiles. Only it is noticeable the high percentage of female consumers. People within this group participate more actively in environmental conservation tasks and are more concerned about food diet and health. As it can be observed from tables 6 and 7, market segments found in both regions present certain similarities although they are more heterogeneous in the case of Madrid. The â€Å"organic food consumers† segment is larger in the producing region (Navarra) as previously discussed. In any case, different market segments, with different socioeconomic and lifestyle characteristics, normally mean differences in preferences. In this paper, the main objective was to evaluate the consumer willingness to pay (WTP) for organic products. In the next section we will answer this question and will identify differences across segments in both regions. 5. Willingness to pay: the contingent valuation method Consumers willingness to pay (WTP) for organic food products is here measured using a direct valuation method: contingent valuation (CV). A mixed questioning procedure, normally called closed-ended with follow-up was used. This procedure consists of a dichotomous choice (DC) question and a maximum WTP question. In the DC question, consumers are asked whether or not they are willing to pay a premium, A, to buy an organic food product instead of a i conventional one. The amount Ai is a percentage over the price of the conventional product and 10 differs across consumers (5%, 10%, 15% and 20%)6 . Consumers responses are YES if they are willing to pay at least Ai for an organic product or NO, otherwise. Consumers are then asked for the exact premium they are willing to pay. The dichotomous individual response is linked to the maximum utility choice which allows us to calculate the WTP from appropriate welfare measures (Hanemann, 1984). Hanemann (1984) assumed that consumers know, with certainty, their utility function before being asked and after paying the amount Ai for buying organic food products. However, some components of these utilities are unknown or unobservable by researches who consider them as stochastic. This issue is the crucial assumption which explains the relation between statistical binary response models and the utility maximizing theory (Hanemann, 1984, 1987). Assuming a linear utility function and a logistic distribution function for the binary question, the WTP can be measured through the estimation of the following logistic function (Hanemann, 1984): Pi = (1 + e -(? + ? Ai) )-1 where, Pi : 1 if consumers are willing to pay the amount Ai and 0 otherwise Ai : the four premiums offered to consumers (5%, 10%, 15% and 20%) Therefore, the mean of WTP is calculated as follows: E(WTP) = ? ?0 (1 + e -(? + ? A) )-1 dA = -? /? (1) Survey data consisted of 400 questionnaires but only 360 were available in Navarra and 376 in Madrid for estimation purposes. The rest of questionnaires were dropped out because interviewers had not answered the dichotomous question. Model defined in (1) was transformed to a linear one: DCi = ? + ? Ai where, DCi : is the logarithm of the probability ratio (Pi /1- Pi ) (2). 6 Consumers are randomly offered a premium of 5, 10, 15 or 20% above the marketed price for a conventional product. Our 400 interviews resulted in 100 answers for each amount A i . 11 Tables 8 and 9 show the maximum willingness to pay for various organic products (vegetables, potatoes, cereals, fruits, eggs, chicken and red meat) in Navarra and Madrid 7 . As we were interested in knowing if WTP differences existed across market segments, we tested for differences in WTP using a covariance analysis method. The unrestricted model was equation (1) adding the appropriate dummy variables. Two types of restrictions were tested: i) no differences across all segments; and ii) differences between any pair of segments. Results from Likelihood Ratio tests are shown in Tables 8 and 9. (Insert Table 8) (Insert Table 9) Also, we tested for starting point bias, that is, if premium offered to consumers biased WTP results. A linear regression was used to estimate consumers WTP as a function of the offered premium. A significant relationship would mean that WTP results were biased. Only in the case of potatoes in Madrid, results were conditioned to the offered premium. In the case of Navarra, all estimated parameters in segments 1 and 2 are individually significant at 5% level (Table 8). As expected, ? coefficients are negative, meaning that the higher is the premium (Ai) offered to consumers, the lower is the probability of answering YES. The third column in each segment shows the willingness to pay for organic products. The WTP is significantly different among all segments at 5% significance level (first column in Table 8). However, no differences were found between segment 1 and segment 2, indicating that potential and actual consumers are willing to pay a similar premium for all products. In both segments, the WTP ranges from 15% to 25% over the price of conventional products. Consumers included in the third segment, â€Å"Unlikely consumers†, are more reluctant to pay a premium for an organic product which is consistent with their attitudes and lifestyles. Most of the estimated parameters are not significant indicating that their WTP is zero. Only in the case of eggs is the WTP close to 10%. In general terms, consumers in Madrid were willing to pay a smaller premium for an organic product (Table 9). Furthermore, when compared to Navarra, results obtained in Madrid are substantially different as market segments differ. The most interesting result is that all segments are willing to pay a higher premium for fruits and vegetables. It seems that for 7 For comparison purposes the sample average WTP for different products in Navarra and Madrid are respectively: vegetables (13%, 12%); potatoes (9%, 9%); cereals (10%, 8%); fruits (13%, 13%); eggs (10%, 11%); chicken (13%, 9%) and; red meat (14%, 11%). 12 perishable products, like fruits and vegetables, the specific characteristics of organic production are more appreciated by consumers. Consumers’ WTP is significantly different across all segments when jointly considered. However, differences segment 1, 2 and 3 are quite small and not significant in most products. On the other hand, the â€Å"Organic food consumers† are willing to pay a higher and significantly different premium than the rest for an organic food product. Only for animal products (red meat, chicken and eggs), differences between real and potential consumers (market segments 4 and 3) are not significant. Among â€Å"unlikely consumers† (market segments 1 and 2), the older ones, worried about natural food consumption, are willing to pay a higher premium for organic meat products than the younger. 6. Conclusions Organic farming practices are becoming increasingly popular among producers although they still represent a marginal share of arable land. New possibilities to get subsidies within the Common Agricultural Policy has favored a rapid growth of organic production although a high percentage is devoted to pastures. On the demand side, the increase concerns about health, diet and environmental deterioration have, at least among some market segments, stimulated the demand for organic food. In Spain, organic food production and consumption have grown more slowly than in other â€Å"northern† European countries. One of the main obstacles for organic food expansion in Spain is the existing gap between conventional and organic food prices. Approximately 75% of organic production is exported to foreign countries where food prices are higher than in domestic markets. As the producers price strategy is oriented to fix more or less the same price level in both domestic and foreign markets (mainly in Germany and Denmark where the consumption of organic products has undertaken a noticeable increase in the last years), the result is that the premium Spanish consumers have to pay for organic food products is higher in comparison to other European countries. Two issues were explored which can affect the future development of organic production in Spain. The first was to detect market segments which could be potential consumers of organic food. The second issue was the identification of the maximum premium the various market segments were willing to pay for such products in order to help producers to take adequate pricing strategies in domestic markets. The study was carried out in two Spanish regions: one is 13 an active producing area ( avarra) while the other is the main food consumption market N (Madrid). Differences between regions were also analyzed. Finally, one of the main outcomes of this paper is that it covers a wide range of products in order to make comparisons both between products and market segments. In relation to the first issue, market segments were identified considering consumers lifestyles and, then, were characterized taking into account not only consumers socioeconomic characteristics but also consumption levels and attitudes towards organic food products and environmental issues. Similar segments were obtained in both regions although more heterogeneity was found in the case of Madrid. In general terms, three broad market segments were identified: consumers, likely consumers and unlikely consumers. The surprising result is that organic food consumers in both regions were larger than expected taking into account expenditure figures discussed in section 2. The answer to this surprising result is that in many areas consumers are supplying their own products. They consider them organic, simply because no fertilizers are used, when really they are not, as there is not any official certification. In spite of this result, some concluding remarks can be outlined. In general terms, consumers socioeconomic characteristics are not very relevant when explaining differences among market segments. Lifestyles and attitudes towards environmental issues are key factors explaining organic food consumption and have to be considered when designing appropriate promotion strategies by producers or marketers. In relation to the second issue, three main results were obtained.