The research leading to these results has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement n° 311778

To be quoted as: Virginia Cravero, Emilia Schmitt, Laurette Gratteau, Ulysse Le Goff, Anaëlle Tanquerey-Cado, Dominique Barjolle (2015). Sustainability assessment of a local wine value chain in Switzerland. GLAMUR project. FiBL, Frick, Switzerland.

1-2

Sustainability assessment of a local wine value chain in Switzerland. Emilia Schmitt, Virginia Cravero, Laurette Gratteau, Ulysse Le Goff, Anaëlle Tanquerey-Cado, Dominique Barjolle – FiBL

www.glamur.eu 3

Table of contents 1 Introduction ...... 7 1.1 Structure of the report ...... 7 1.2 Introduction to the Swiss wine sector ...... 7 2 Context of the case study ...... 10 2.1 Selection of the case studies ...... 10 2.2 Global-local issues in the value-chain ...... 13 2.2.1 Marketing issues ...... 13 2.2.2 Environmental issues ...... 13 2.2.3 Normative framework and specific EU and national polices ...... 14 2.3 System boundaries, characteristics and mapping ...... 15 2.3.1 Local wine value chain ...... 15 2.3.2 Global wine chain ...... 17 2.3.3 Scope of the value chains under study ...... 18 2.3.4 Production and processing ...... 18 2.4 Critical and relevant issues ...... 20 2.4.1 The flow of information ...... 20 2.4.2 Workers’ conditions ...... 21 2.4.3 The economic profile of each value chain’s actors ...... 21 3 Research Design ...... 22 3.1 Research Questions ...... 22 3.2 Attributes ...... 23 3.3 Performance Indicators...... 23 4 Methods ...... 25 4.1 Indicators: methods used for their definition and calculation...... 25 4.2 Benchmarks: methods used for their definition and calculation ...... 25 4.3 Methods used for data collection ...... 29 4.4 Methods of participatory checklist used ...... 30 5 Results ...... 30 5.1 Scores of performance ...... 30 5.1.1 Attribute Biodiversity ...... 33 5.1.2 Attribute Creation and distribution of added value ...... 33 5.1.3 Attribute Food safety ...... 34 5.1.4 Attribute Governance ...... 34 5.1.5 Attribute Information and communication ...... 34 5.1.6 Attribute Pollution ...... 35 5.1.7 Attribute Resource use ...... 35 5.1.8 Attribute Territoriality ...... 35 5.2 Relevant descriptors ...... 36 5.3 Data quality check ...... 37 6 Discussion ...... 38 6.1 The methodology and its effectiveness and limits ...... 38 6.2 Addressing the attributes, indicators and research questions ...... 39 6.2.1 Biodiversity (environmental dimension) ...... 39 6.2.2 Creation and distribution of added value (economic dimension) ...... 41 6.2.3 Food safety ...... 43 6.2.4 Governance ...... 44 6.2.5 Information and communication ...... 46 6.2.6 Pollution ...... 47 6.2.7 Resource use ...... 49 www.glamur.eu 4

6.2.8 Territoriality ...... 50 6.3 Discussion of the local performance ...... 53 7 Conclusion ...... 57 8 References ...... 58 9 Annexes ...... 61 9.1 Context of the red local wine value chain ...... 61 9.2 Additional References ...... 64 9.3 Data quality check ...... 65

Tables Table 1 Local value chain in Valais using the WFSC framework (World Food System Center, 2013, adapted)...... 12 Table 2 Synoptic table and link to RQ ...... 24 Table 3 Indicators’ assessment, benchmarks’ definition and calculation...... 26 Table 4 Indicators’ data and performance scores for the local wine value chain ...... 30 Table 5 Levels of CO2 generated per tonne and km for various modes of transport (van der Zanden, 2009)...... 47 Table 6 Additional references for indicators...... 64 Table 7 Data quality check (Lewandowska, 2004)...... 65

Figures Figure 1 Valais vineyard distribution in the canton (IVV, 2010) ...... 15 Figure 2 Flowchart of the Valais wine value chain (Etat du Valais, 2006)...... 16 Figure 3 Flowchart of the Swiss local wine value chain...... 17 Figure 4 Flowchart of the French global wine supply chain ...... 18 Figure 5 Wine supply chain’s scope of the study...... 18 Figure 6 Wine making process. Adapted from (Gallien, 2011)...... 19 Figure 7 Localisation of stakeholder in the Valais canton...... 29 Figure 8 Indicators’ performance...... 31 Figure 9 Performance of grape-winemakers according to attribute...... 32 Figure 10 Local value chain performance according to dimension...... 33 Figure 11 Gross added value estimation according to the product and the distribution channel in CHF...... 42 Figure 12 Environmental impacts of red wine in Switzerland, Chili and Spain (SOFIES, 2009)... 48 Figure 13 Overall assessment of the Valais wine (Lehmann, 2000)...... 51 Figure 14 Pyramidal distribution of Valais’s wines (IVV, 2010)...... 52 Figure 15 Positive, negative and neutral influences between indicators and descriptors created with NodeXL...... 54 Figure 16 Swiss map where production sites are marked...... 61 Figure 17 Swiss wine production for red, white varieties (OFAG, 2013)...... 61 Figure 18 Swiss wine production for red and white varieties (OFAG, 2014)...... 62 Figure 19 Imports and exports data in 2012 concerning quantity (kg) and value (CHF) (AGIR, 2013) ...... 62 Figure 20 Swiss wine consumption of autochthon and imported white and red wine (OFAG, 2013)...... 63 Figure 21 Structure of the wine value chain in Valais (Catry, 2009)...... 63

www.glamur.eu 5

Summary This case study analyses the red wine local value chain of the Valais canton in Switzerland. The Valais is the first producer of wine of the country being characterized by more than 50 different varieties cultivated. The wine sector in Valais is composed by three main stakeholder groups: grape producers, winemakers and cellar men. Cellar men commercialize less than 15% of the volumes produced whereas the cooperative and traders buy 79% of grape produced and commercialize 85.5% of the volumes produced. As this study focuses only on a local chain, stakeholders interviewed concern especially grape growers/winemakers who commercialize their own wine directly at the cellar or to restaurants in very short supply chains. Semi-structured interviews have been completed with 26 stakeholders in order to collect primary data regarding 19 indicators extrapolated from 8 attributes. 12 indicators of 19 obtained a performance score higher than 50%. Results suggest that at the environmental level, more efforts are required to reduce GHG emissions linked to packaging and wine transportation and to better use the resources available such as water. At the social level, the employees’ conditions could be improved decreasing the number of black seasonal contracts. However, the high wages generated show high value retention at the production stage. At the health level, more international standards could be applied helping the entrance of these local wines into the international markets. The comparison with the French global value chain will be done in the comparative report.

www.glamur.eu 6

1 Introduction The “Global and local food assessment: a multidimensional performance-based approach” (GLAMUR) project has for goal to compare local and global food value chains in different sectors and European countries. This report refers to the task 3.6 “Case studies reports” of the project, consisting in the analysis of the local Swiss red wine value chain. The ultimate aim is to compare this wine value chain with the global value chain in France using the GLAMUR methods. The team involved in this part of the study (local wine) is the FiBL (Swiss team) collaborating with INRA Montpellier (French team), for this further comparison.

1.1 Structure of the report The report is structured in seven main parts: Chapter one introduces the Swiss wine sector as well as both local and global value chains. Chapter two provides practical information on the case study declined into: (i) the explanation of how the chains were selected, (ii) the global-local main issues emerging in the value chains, (iii) the system boundaries, characteristics and mapping including the scope of the analysis, and the (iv) the critical issues refined from the previous reports (quickscan). Chapter three presents the design of the research questions, the attributes selection procedure and the indicators construction process together with the methods of selection applied and a synoptic table linking attribute, indicators, research questions and methods of data collection. Chapter four goes on by presenting the final list of indicators, their calculation and the methods used to collect the data. Our results are presented in chapter five and they are discussed and analysed in detail in chapter six. We conclude this report in chapter seven and more information is available in the attached annexes. 1.2 Introduction to the Swiss wine sector The first traces of viticulture in Switzerland date back to the Roman era. Afterwards, viticulture was practiced in the middle Ages and modern times as horticulture. Changes took place between the mid-eighteenth century and the mid-nineteenth (Agricultural Revolution) causing a rise in yields followed by a drastic reduction in early 1880 caused by the Phylloxera crisis and the decrease of surface area dedicated to wine making (Lüdi , 2013). Until the early 1980’s, the Swiss public authorities supported high levels of production for the wine, which resulted in relatively low-quality and cheap wines. However, changing consumption trends and global oversupply of wine drove the price down too low. This difficult situation forced the sector to change for lower www.glamur.eu 7

quantities and higher quality, by reducing the yields. Rising costs of subsidies to keep this level of production were not seen as strategic and worth it anymore (they dated back to the Second World War). Agreements start to appear at the canton level, to allow areas to be associated with quality wines, by restrictions on the yields in Genève in 1988, followed by Valais in 1990, before all wine- producing cantons. The first wine AOC are in the continuity of these rules and start appearing after 1992 and a federal resolution. From then on, the Swiss wine sector leaves low-quality and starts to produce some of the highest quality in Europe (Boisseaux & Barjolle, 2004). In 2012, the viticulture surface in Switzerland overlaid 14’920 hectares (ha). The surface destined for white variety was around 6’320 ha (42%) while red varieties occupied 8’600 ha (58%). The distribution of wine production in Switzerland is unequal, being especially present in the south-west part of the country (see Figure 16, Annex 9.1). Indeed, the largest vineyard area in Switzerland is represented by the Canton of Valais with 4976 ha of grape cultivation followed by the Canton Vaud with 3784 ha, the Canton of Geneva with 1435 ha and Canton of Ticino with 1076 ha. In the Germanic part of the country, this sector does not represent extended surfaces compared to the French-speaking regions: the Canton of Zurich has 611 ha, the Canton of Schaffhausen 482 ha and the Canton of Grisons 452 ha (OFAG, 2014). The most representative varieties cultivated to produce white wines are: the Amigne, the Arvine, the Chardonnay, the Chasselas, the Pinot Blanc, the Pinot Gris, the Sauvignon Blanc, and the Sylvaner. Concerning red wines, the most representative varieties are: the Cabernet Franc, the Cabernet Sauvignon, the Pinot Noir, the Merlot, the Gamay and the Syrah. Regarding the cultivated vineyard surfaces, the most produced varieties are the Pinot Noir followed by the Chasselas and the Gamay (Spring & Dupraz, 2011). The country presents both white and red autochthon varieties. In the white category, these varieties are: the Ermitage, the Humagne Blanche, the Petite Arvine, the Savagnin Blanc (or Paien). Regarding red wines, autochthon varieties are represented by the Cornalin, the Diolinoir and the Humagne Rouge (Spring & Dupraz, 2011). In contrast with the high number of varieties cultivated, wine production in Switzerland is decreasing since 1985 (see Figure 17, Annex 9.1). In 2013, the total Swiss production was 839.6 thousands of hectolitres (OFAG, 2014) (see Figure 18, Annex 9.1) against the 1352 thousands of hectolitres produced 28 years before.

www.glamur.eu 8

Moreover, exportations in 2010 represented only 1.9% of the whole production (OFAG, 2011). At the same time, the indigene production does not cover the consumption demand, thus importations have always been essential in covering the 62% of the demand in 2010 (OFAG, 2011). Thus, Switzerland imports wines coming from important producing countries in term of quantity and value such as Italy and France. Swiss wine exports, which are one hundred times lower than its imports, interest mainly the German and French markets (AGIR, 2013) (see Figure 19, Annex 9.1). Therefore, Swiss wine production is essentially focalised on indigene consumption. This is specifically the case of white wines, which are mostly consumed by Swiss customers (see Figure 20, Annex 9.1) (OFAG, 2013). It is said that the Swiss wine market is “little but mature and sophisticated” (Catry, 2009). The Swiss wine value chain under study is limited to the grape production coming from the Valais Canton in Switzerland. This canton is the biggest Swiss producer in terms of hectolitres (377'047 hectolitres produced in 2012) (OFAG, 2013). This zone is characterised as producing mainly red grape varieties together with the willingness to continue producing original varieties such as Arvine, Amigne, Heida, Cornalin and Humagne. Therefore, the canton is characterised by a significant grape variety: more than 40 varieties are authorized to entry in the AOC/PDO scheme jointly with 10 other varieties coming from the south of Europe that can be used to produce second- class wines (Laesslé, 2012). Concerning the number and type of actors of this value chain, it is important to highlight the impressive proportion of very small parcels maintained by 22’000 owners. 13’000 are operators composing two main groups: the non-professional operators (94%) cultivating 30% of the cantonal vineyards with an average surface of 0.17 ha (Catry, 2009) and the professional operators (6%) cultivating the 70% of the cantonal vineyards with an average surface of 3.4 ha. In the latter group, 17 of the most important operators cultivate 10% of the cantonal vineyard within an average surface of 30 ha. Within the professional operators group, 29% are grape growers and 23% are poly-farmers. Both deliver the raw material (grapes) to other key actors: the winemakers and cellar men (representing the 33% of the professionals operators) working together with retailers such as wine shops and restaurants (representing 16% of the professionals operators) to valorise the final product (see Figure 21, Annex 9.1) (Catry, 2009).

www.glamur.eu 9

2 Context of the case study This chapter will present first the criteria of distinction between the local and the global chains in the wine sector and then the two chains selected in detail. 2.1 Selection of the case studies

The “local” and “global” wine value chains have been selected according to the 4 distinction criteria quoted in the GLAMUR theoretical framework: 1. The physical / geographical distance between production and consumption 2. The type of governance and organization of the value chain (degree of control of “local actors” and “global actors”) 3. The kind of resources, knowledge and technologies employed 4. The way value chain actors shape product identity with regard to the reference to the territory of production for food plays a relevant role or not.

For the Swiss wine sector, the first criterion is valid due to the fact that wines issued from local wine value chains are produced and consumed only at regional or national levels. In contrast, global wine value chains are characterized by productions issued in determinate sites but consumption reaching other countries and continents. Thus, high distances between these two steps are found and supported by high volumes of exports, meaning that the consumption step can be established abroad. The second criterion is also useful because the degree of control possessed by local actors is higher than the degree showed by global actors. Hence, the local wine value chain in Switzerland is characterized by actors who in many cases control the whole value chain by producing, bottling and commercializing the final products themselves. Actors working in the global wine value chain do not possess this flexibility in taking decisions due to the fact that regulations and decisions are taken by organizations and/or cooperatives and not by single producers. The third criterion is not entirely helpful to distinguish local wine value chains from global ones because nowadays for this sector resources, knowledge and technologies used by both value chains are quite similar: indeed the inputs as well as the varieties and the machineries used do not differ significantly. However, some varieties are autochthon and thus can be associated with local value chains as it occurs in Switzerland.

www.glamur.eu 10

Those varieties possess a strong linkage with their respective territories allowing achieving final products with a particular identity (criterion 4). It is important to highlight the efforts made by concerned actors in order to maintain and promote these identities. Their overarching goal is indeed to promote high quality products rather than using notoriety of varieties purely to increase prices. In global value chains, the product identity seems to be used mainly as a marketing tool. In summary, from the 4 interrelated axes cited above, the physical/ geographical distances between steps and actors and the kind of resources used such as the grape variety (local or generic) have been taken into account when selecting the case studies. Nevertheless, other specific characteristics have been identified to distinguish both local and global value chains for the wine sector. These are: 5. The mode of handling (bottle or bulk) 6. The mode of selling (direct sales or supermarkets) 7. The number of intermediaries in the chains 8. The volumes of production

Concerning the mode of handling of the final product, bulks are mainly used by global value chains producing high volumes. Indeed the function of bulks is to sell more volumes of wine at lower prices saving the cost of transportation of other

containers such as bottles. This has for effect to mitigate CO2 emissions as well (Observatoire Viticole, 2010). On the contrary, bottles are mainly used to give value to wines associated to a particular area, meaning that their function is also to shape the identity of the product. In the case of Swiss and French wine, the mode of selling can also reflect the localness or globalness of value chains: direct sales are used by local value chains as a tool to build loyalty and create personal contacts with customers, whereas in global value chains there is no such commitment. The number of intermediaries also determines a higher degree of globalness or localness of value chains: more than 100’000 people are directly and indirectly part of the French global value chain located in the Languedoc-Roussillon region (Ministère de l'agriculture, 2013) whereas in Switzerland, the local wine value chain counts a smaller number and type of actors (13.300 operators) including grape producers who are also wine makers and cellar men at the same time. The volumes of production also allow differentiating both global and local value chains: more than 300 million litres are produced in the Languedoc region in France (INTER OC, 2011) whereas in the Valais canton only 4.1 million litres are produced (Office de la viticulture, 2013).

www.glamur.eu 11

Table 1 below shows the World Food Supply Chain (WFSC) framework in which it is possible to localise physically each step of the local value chain further analysed in the next chapters.

Table 1 Local value chain in Valais using the WFSC framework (World Food System Center, 2013, adapted).

Spatial Scale Production inputs Agricultural Production Primary processing Secondary processing Distribution Retailing Consumption wine shops Water 5 km restaurants Land cooperative Local grape cultivation 0 km wine C 0 km cellars o

Machineries Agricultural Input n Equipment companies s Regional u Supermakets, m supermar restaurants Bottles transports ket e Machineries Labels fuel platforms Electricity r National s industrial electricity Input Machineries companies Corks Workers Bottles Continental Inputs (foots vine, fertilizers, additives) Fuel Global

www.glamur.eu 12

2.2 Global-local issues in the value-chain This sub-section discusses the main issues related to local and global value chains. These issues were identified starting from WP2 and the scoping and framing of the Swiss wine sector. A few preliminary interviews were conducted and extended media review completed the analysis. The critical issues concern the marketing management, the availability of information for customers and the environmental impacts connected.

2.2.1 Marketing issues Concerning the marketing, both value chains have opposite strategies: the local value chain targets high quality products whereas the global value chain applies basically a strategy based on competitive prices and high volumes of production. Due to the saturation of the Swiss internal market, the local value chain wants to position its products around 15 euros/bottle, being helped by the notoriety and a favourable image of its products. This strategy is also feasible thanks to the type of consumers purchasing these products: these are mainly Swiss clients receptive to a product differentiated based on quality (Catry, 2009). Catry (2009) concluded that in order to justify such strategy, an amelioration of the book of specification for PDO Valais wine is suitable. Nowadays, this debate is oriented to the creation of a new label (Label Marque Valais) for wines produced from authentic grape varieties characterized by limited production and tolerance of mixing with other wines to only 5%. Other wines known as "creative" and cited as "basic AOC" could use oenological internationally recognized instruments, such as mixing or using the MCR (rectified concentrated must), to meet consumers’ preferences and to place Valais wines at the same level that their competitors (Thomas, 2014a). The adherence of the whole supply chain to this marketing strategy is conceivable at long term. However, nowadays ‘priority for wine makers and cellar men is to obtain customer’s loyalty, in order to sell bottles produced under non- optimal climatic conditions (Laesslé, 2012).

2.2.2 Environmental issues The upstream part of the local supply chain generally practices environmentally- friendly approaches to preserve the environment; nevertheless, at the bottling level, it relies principally on glass bottle production as a vehicle to sell the final product. This strategy has negative consequences for the environment because of the pollutants produced during the process. Moreover, the recycling is highly www.glamur.eu 13

energy-demanding. Some producers started to use washed and/or recycled and/or light bottles. However a complete shift of thought is necessary to join the strategy based on Bag in Box (BIB) packaging acquired by the global value chain. Moreover the transport of individual bottles is much less efficient and pollutes more by unit and kilometre compared to bulk transportation.

2.2.3 Normative framework and specific EU and national polices Concerning the local value chain, one significant difference is the introduction of quotas of grape production per m2 as a tool to regulate the quantity and quality of grapes produced. Each year, the Inter-profession (IVV) establishes it, reducing the threshold when they consider it appropriate. The decrease of the threshold is not well accepted by producers because a decrease in the quotas means a net loss of income per square meter (Laesslé, 2012). In addition, the final product in the local value chain possesses a Protected Designation of Origin (PDO or AOC) whereas the global product shows a Protected Geographical Indication (PGI), illustrating different normative frameworks. The former is based on a tight link between the raw material, the final product, and the territory it is produced, with a limited geographical zone; while the latter is based on the production or notoriety or raw material used, associated to a determinate area. In general terms, Swiss legislation varies from the EU: for example, the latter transferred to the wine makers groups the task of being organized under a PDO or IGP syndicates whereas in Switzerland, all decisions are taken by the cantons. Therefore, in France exists the consumption service that consumers can contact in case of recrimination (Thomas, 2014b). Syndicated must control the quality of the raw material analysing also exported bottles. Afterwards, they compare these samples with those obtained from the original cellar verifying if the same wine can be found in national supermarkets. In Switzerland no legislations concerns such control because of the small volumes exported (Thomas, 2014b).

www.glamur.eu 14

2.3 System boundaries, characteristics and mapping 2.3.1 Local wine value chain According to the following figure, Valais’ vineyards are spread along several municipalities (Figure 1, in green), where vineyards are occupying hillsides from 450 to 800 m of altitude.

Figure 1 Valais vineyard distribution in the canton (IVV, 2010) The climate of this area is favourable to the grapes grown. The surrounding mountains impede the rains from coming and provoke the “foehn” effect, which allows drying the grapes, increasing their maturation and concentrating their sugar. Indeed, the Valais canton is the driest Swiss canton with 2100 hours of sunlight per year. The other important climatic phenomenon is the thermic contrast between warm days and fresh nights, improving at the end of the grapes’ maturation cycle the complexity and fineness of aromas (IVV, 2010). Concerning the red wine value chain, it is composed of three main steps: the production of grapes by grape growers, their transformation into wine by wine makers, cellar men and cooperatives and their sale through several businesses (Figure 2, below).

www.glamur.eu 15

Figure 2 Flowchart of the Valais wine value chain (Etat du Valais, 2006).

In detail, 21% of the grape cultivated is sold to wine makers and cellar men producing only bottles and commercializing about 14.5% of the whole wine production. The majority of grape growers (59%) sell the grapes to enterprises whereas the remaining 20% sell them directly to the cooperative. The enterprises and the cooperative commercialize together about 85.5% of the whole wine production selling both bottles and bulks. The commercialization then reaches the market by big distributors such as supermarkets (33%), hotels-restaurants- cafes (HORECA) (33%), direct sales (22%) and other shops (12%) (Etat du Valais, 2006). Other more "peripheral" actors do exist, having an important role in the construction of the resource dynamics: These actors are represented by cooperatives, research institutions, administrative-political actors and inter- professions (Laesslé, 2012). The local wine value chain studied for the case study includes grape growers who are also wine makers and cellar men at the same time, selling the final product directly from their cellar or to wine shops and restaurants. Regarding the upstream part of the chain, inputs come from the region (water, land) but also Europe (workers, machines, fertilizers) and the world (fuel) whereas the downstream part is essentially composed of local and Swiss consumers (Figure 3).

www.glamur.eu 16

Figure 3 Flowchart of the Swiss local wine value chain.

2.3.2 Global wine chain This chain is represented by a red wine from Languedoc region exported to Switzerland (Figure 4). This decision was taken due to the high number of actors, relationships but also high volumes of production and sales that characterize this supply chain. Indeed, the Languedoc-Roussillon vineyard is the biggest French exporter abroad selling to Switzerland almost 100’500 hectoliters (hl) per year. Moreover, Switzerland is among the first ten importers of still wines (PDO and PGI) coming from Languedoc-Roussillon. We have identified operators playing major roles in the global chain in France and Switzerland. The French team identified in the global chain two subcategories:  Wine sold in bulk by cooperatives and unions of cooperatives, and imported by wholesalers and distributors from Switzerland. These wines are produced by grape grower members of the cooperative. The wine is sold in bulks to French and Swiss traders, wholesalers and bottlers (e.g. GCF group, Bataillard, Scherer & Bühler or Haecky Gruppe). These importers provide other wholesalers not specialized in wine, like retailers and supermarkets. The wines sold by the latter present in the label a geographical indication (DOP or IGP) or the label “Vin de France”.  The export market specialized in bottled wine mobilizes other operators within the global chain. Bottled wine is thus sold on the global market by big wine estates and small cooperatives. Production strategy enhances the quality image and the link with the “terroir”, identified by DOP wines, vine varieties PGI and mentioning the name of the producer. Swiss operators provide specialized stores and restaurants.

This global wine chain is reported by the French INRA team and will be compared to the local Swiss chain during WP4.

www.glamur.eu 17

Figure 4 Flowchart of the French global wine supply chain

2.3.3 Scope of the value chains under study The scope of the study is delimited from agricultural inputs to consumption. The main steps studied are agricultural production (grape production), primary and secondary processing (wine production). Bottling, retailing and consumption or exportation will also be taken into account. Figure 5 below shows a simplified wine chain in Switzerland and the steps that we integrated in this study.

Figure 5 Wine supply chain’s scope of the study. 2.3.4 Production and processing Valais’ wines are classified in three categories according to the quality of the grapes and the quantity harvested. Category I includes grapes allowing the elaboration of PDO wines; the category II includes the grapes producing “vins de pays” (VDP) and the category III includes the grapes producing “vins de table” (VDT) (Le Conseil d'Etat du canton du Valais, 2004). To distinguish the first category, the canton applies “quantitative limits of production” (QLP): hence, to produce PDO wines the maximal quantity harvested per square meter can be 1,2Kg/m2, whereas for the VDP the maximal quantity harvested per square meter can be 1,6 Kg/ m2 (Le Conseil d'Etat du canton du Valais, 2004). The Inter-profession can reduce these limits of production even more according to annual cantonal production. The lower QLPs for category I wines should guarantee a higher quality of the wine sold under the cantonal PDO and protect the cantonal strategy to focus on quality in marketing. www.glamur.eu 18

It is forbidden to mix the PDO, VDP and VDT wine with different grapes, musts or wines with a different origin or provenance, or foreign wines. The addition of a limited percentage of grapes, musts or wine of the canton coming from other grape varieties is allowed to a maximum of 15%. For PDO wines, mixing must be done only with other PDO grapes, musts or Valais’ wines (Le Conseil d'Etat du canton du Valais, 2004). Concerning the labelling, several rules have to be applied. For PDO wines it is compulsory to indicate the protected denomination of origin. Regarding the procedure of wine-making, it may differ a little between winemakers but these few steps are systematic for the red wine from Valais (see Figure 6 below). After grapes (1) being harvested (2), the grapes are pressed and the juice remains in contact with the skin (maceration) which gives the product its colour and some important aromatic components (3). This process is usually accelerated by stirring. The resulting must is then fermented by yeast (4). The production of CO2 resulting from this is allowed to get out of the tank and the general temperature of the process is controlled. Afterwards, the malolactic fermentation (6), which is done by bacteria, reduces the acidity of the wine. The wine is (7) cleared and stabilized through physical and chemical processes (including addition of sulphites within the limits of the PDO) before aging in oak barrels (8). The blending (9) of different wines is limited by the PDO and must consist of wines produced in the Valais region. The wine is finally filtered, bottled (10) and cellared (11) before being eventually sold (12).

Figure 6 Wine making process. Adapted from (Gallien, 2011).

www.glamur.eu 19

2.4 Critical and relevant issues This paragraph summarizes the critical issues in term of sustainability that were identified in the case study. They were identified first via extended literature and media review and through preliminary interviews with key-respondents in the concerned sector (see chapter 5. Methods). The main critical issues regarding the local value chain’s sustainability performance concern: (i) the flow of information between central and peripheral actors linked with the fraud management, (ii) the workers ‘conditions and (iii) the economic profile of each actor of the value chain. Other secondary issues were as well identified, such as the not-optimal transportation of small quantities of single bottles, the remaining risks of environmental pollution because of chemicals spraying, sometimes from planes, the risk of health contamination from chemicals in the fields or from the wine when additives are concerned, etc. However it was identified that the first three issues are especially relevant and have reach a concerning level by their occurrence in the news and actors debates. The other issues are maybe less of a concerning level compared to other wine producing regions.

2.4.1 The flow of information The poor flow of information linked to the fraud management that affects the Valais wine system refers to (i) the low level of communication within the value chain between several organisms of control and wine makers, (ii) the inexistence of a database to store all the controls results, (iii) the multitude of different systems within each municipality, (iv) the lack of impartiality by certain control bodies and (v) the absence of cellar accounting at some wine makers (RTS, 2015) Therefore, when an infraction is committed it is noticed by the Swiss wine control (CSCV) and, when judged a serious offense, reported to the cantonal chemist. The latter has to denounce it to the criminal justice which will take the proper decision. Nevertheless, a scandalous event concerning a big winemaking company brought to light the lack of communication between the cantonal chemist and the criminal justice, and consequently the lack of information between the cantonal chemist, the CSCV and the Federal Office for Agriculture (FOAG). Moreover, the link between the grape harvesting controls and cellar controls does not exist, worsening the quality of information and communication. Similarly, concerning the labelling of final products, the fact that some labels present wrong cellar names to sell non-authorized wines has been denounced (RTS, 2015).

www.glamur.eu 20

2.4.2 Workers’ conditions Concerning the social and ethical dimensions, another critical issue is related to the condition of workers within the upstream part of the value chain. The majority of workers employed during harvesting time are coming from the EU but also from other parts of the world. For the EU workers, their legal declaration is easy thanks to the free movement of persons. However, in the Valais canton they are often underpaid compared to other sectors and paid in black. Therefore, the lack of salary and employees’ declarations has been denounced by several managers (Le Nouvelliste, 2014).

2.4.3 The economic profile of each value chain’s actors Within this sector, the economic profile of each stage differs significantly and the chain is overall not very balanced. The vineyard situation is the more delicate: although the gross value added of the vineyard activity (148'000'000 CHF, approximately) contributes more than 50% of the gross value added of the entire sector (about 280'000'000 CHF) controlling more than half of the added value of the sector and contributing more than the other actors in the economy of the canton, producers within it have to deal with high costs of winemaking. This reality is explained by the distinctiveness of Valais: vineyards are located on sloping terraces, small areas with high operating cost (hours / ha) and they are unable to improve their income by better negotiating with cellar men and enterprises. To cope with this situation, the ideal solution is to increase the added value of the whole sector by increasing the prices paid by consumers. If that is not accepted by the market, the industry will have to consider a different distribution of value in the sector between its actors (Catry, 2009).

www.glamur.eu 21

3 Research Design This section includes four main parts: (i) the list of research questions, the selection procedure for (ii) attributes and (iii) indicators and (iv) the synoptic table for the final indicators’ set. 3.1 Research Questions There are three overall research questions in the GLAMUR project:

1. What are the key food chain performance issues with regards to a global- local comparison? 2. What is the methodological strength and weakness of overall applied pairwise comparative analysis? 3. What are the specific interactions of the food chains under study and the policy settings?

This report will only answer partially to the first question as it can gives answers on the performance of only one local food chain. The comparison with the global wine chain will be conducted in a second step. The methodological strength and weaknesses will still be discussed in the discussion.

For the wine case study, two research objectives have been defined: 1. To analyse the main structures, interrelationships and complementarities between the global and local supply chains taking into account nature of upstream and downstream relations. 2. To examine along the whole supply chain the main performance issues related to diversified attributes and topics such as climate change, resource use, biodiversity, territoriality, food quality and traceability.

These research objectives will be fully answered during the comparative WP4 report with the French team. For this report, specific research questions emerged:

1. To what extent do the territoriality and the biodiversity benefit stakeholders concerning the communication around the product? 2. How is the creation and distribution of added value affecting performance? 3. How is the governance’s structure impacting on workers’ conditions and fraud management in the chain? 4. How are issues related to pollution, resource use and food safety impacting the product evolution? www.glamur.eu 22

3.2 Attributes In the GLAMUR project, 24 attributes of performance have been listed in common between the participating countries in order to compare local and global food chains. An attribute is “a quality or feature regarded as a characteristic or inherent part of something” (Kirwan et al, 2014). For each comparison, a selection within these 24 attributes is made. For the wine sector, attributes have been chosen in a participatory way with the INRA team from France and after several discussions with field’s experts. Therefore, attributes have been judged by actors consulted during preliminary semi-directed interviews. In addition, literature has been analysed highlighting the critical issues related to the value chains (chapter 2.4). However, the final list of attributes only concerns those allowing the cross- countries comparison. Thus, from the list provided by the GLAMUR project, eight attributes have been retained (Table 2). The remaining sixteen attributes have not been chosen for several reasons. For example, “food security” has not been retained due to the fact that wine is not considered as an essential compound of nutritious diets ensuring populations’ food security (although the presence of beneficial nutrients in it) and excessive alcohol consumption is actually a societal problem. In the case of “technological innovation”, literature revealed no differences between local-global chains; concerning the nature of the product, “animal welfare” attribute has been rejected as well because the wine value chains does not take advantage from any kind of animal to obtain the final product.

3.3 Performance Indicators The construction process of performance indicators started with a deeper literature-analysis. This process allowed the distinction between performance indicators (retained as essential for the performance assessment) and descriptors (retained essential for a proper contextualization within value chains). This phase contemplates also experts’ feedback in particular when evaluating indicators’ feasibility (easy access to indicators’ data through interviews and secondary data). Performance indicators have been identified to cover the main aspects related to the research questions illustrated previously.

www.glamur.eu 23

Table 2 Synoptic table and link to RQ

Dimension Attribute Brief attribute description from Used indicators Research Data collection methods WP2 questions Environmental Biodiversity “...ability of food supply chains to 1  Species Conservation preserve the stock of natural  Interview with grape practices resources…” (p.109) growers/winemakers/cell  Cultivars diversity ar men Economic Creation and “… concerned with looking at both how  Gross Income 2  Interview with grape distribution of value is created, but also how it is  Distribution of added growers/winemakers/cell added value distributed within the FSC.” (p.30) value across the chain ar men  Contribution to employment Health Food safety “…attempts to guarantee the safety of  Food safety standards 4  Interview with grape food products …” (p.82) and practices applied growers/winemakers/cell  Artificial additive ar men Economic/ ethical Governance “…First, there is something about 3  Decision making  Interview with grape regulation and governance structure…” mechanisms growers/winemakers/cell “…Second, there is something about ar men power and democracy…” (p.39)  Fraud management within the chain  Market Management Social Information and “…within the Italian report refers to the 1  Interview with grape communication amount and quality of information that is growers/winemakers/cell communicated to consumers together with  Availability of information ar men the product being sold, so that they are Product Labelling able to make a more informed  purchasing decision.” (p.76)

Environmental Pollution “Pollution encompasses any input into the  GHG emission for 4  Interview with grape natural environment which causes adverse transportation growers/winemakers/cell changes to ecosystems...”(p-74)  GHG emissions for ar men production  Environmental pollution (water, soil, air) Environmental Resource use “…the use and management of the flows 4  Interview with grape of available resources through global growers/winemakers/cell and local food chains. It has two main ar men elements. The first element concerns  Water Use Practices resource consumption. The second related  Material Use practices element concerns the tools (techniques) used to measure the resource use performance of food chains.” (p.100) Ethical Territoriality “…encompasses the capability of a 1  Interview with grape supply chain to represent and promote  Association of product growers/winemakers/cell the localness of a product and its link with with the territory ar men a specific terroir or place of  Social cohesion and production.”(p.91) Conviviality

www.glamur.eu 24

4 Methods This section includes the methodology used to set indicators and benchmarks, collect data and process them into scores of performance, including methods of participatory checklist. 4.1 Indicators: methods used for their definition and calculation As previously mentioned, indicators have been selected, taking into account critical issues related to them, as well as the availability and quality of data collected. For each attribute, the teams chose an average of 3-4 indicators. This process has been participatory between both Switzerland and France and between each team with their supply chain’s stakeholders, including wine experts. Both performance-based indicators and practice-based indicators have been used (FAO, 2013) as well as qualitative and quantitative ones. As this report does not study the global value chain, only the local chain’s scores of performance are shown. To assess some indicators, categories have been used. Each category, if fulfilled, is equivalent to one point. Thus, each producer gets a number of points between 0 and maximum benchmarks. The score is thereafter the average of all producers’ points. Categories used are shown in the table 3 below. Additional references about indicators and benchmarks’ construction are available in the annex 9.2.

4.2 Benchmarks: methods used for their definition and calculation In SAFA, benchmarks are “values or qualitative descriptions of activities, used as the basis by which the performance of an enterprise is evaluated within an indicator domain to facilitate a rating of sustainability performance” (FAO, 2013). Benchmarks are thus necessary in order to understand the nature of the scores of performance (good, medium or bad). Most of them have been conceived after literature review whereas others have been formulated using primary data collected during interviews. In both case, some of them such as those under the governance attribute include categories. In the latter case, each category scores 1 if it is fulfilled. Moreover, all data used to construct benchmarks comes from the areas under study. Table 3 presents the processes used for the assessment of each indicator, including the relevant stages of the chain to which it applies, and presents the

www.glamur.eu 25

benchmarks adopted for each indicator, including low-high performance values used for calculation purposes.

Table 3 Indicators’ assessment, benchmarks’ definition and calculation.

Stage of Attribute Indicator Unit Description Low High chain 1 point if participating to voluntaries schemes for species conservation + 1 integrated and organic management against parasites + 1 wildflower strips + 1 safeguarding of flora and fauna Species Grapes networks + 1 adapted Qualitative Practices Based Conservation production pulverisation to not injure bees + 1 0 13 Indicator (PBI) practices participation to “Projets Qualité du Paysage” (QP) + 1 grassing + 1 per type of ecological surfaces (

Biodiversity hedge / bush; isolated trees; dry stone wall; pond / ditch; meadow/ fallow)+ proportion of ecological surface n° in Valais: 20 red cultivars + 29 Cultivars Grapes white cultivars + absence of other 1 50 diversity production productions (0 point) or presence of other productions (1 point) CHF Benchmark min: 50% below the Grapes and average gross national agricultural 47’0 188’3 Gross Income wine income in 2013, benchmark max: 82 30 production 50% above Difference in the share (%) of

the final selling price between Benchmark min 1 means perfect Distribution of Core value inequality; benchmark max:0 1 0 value wine producers, and 1 added value chain means perfect equality in the value traders/retailers/intermediari added es. Contribution UTA/ha Grapes and N° of annual work unit (AWU)

Creation and distribution of added added of distribution and Creation to wine taking into account the Swiss 0 2 employment production AWU per hectare. Food safety Qualitative PBI Grapes and Only HACCP (1 point), for quality standards and wine ISO 9001 (1 point) , for food safety 0 3

practices production ISO 22000 (1 point) Sulphites (g/hl), sugar (kg/hl) For sulphites: benchmark min: 40 g/hl, benchmark max: 0 g/hl; for 40 0 Artificial Wine sugar: benchmark min: 4.25 kg/hl, Food safety Food additive production benchmark max: 0 Kg/hl. 4.25 0 Performance is average of both.

1 Core Value chain : (Neven, 2014), p. 10 : « Four core functions (links) are distinguished in the chain: production (e.g. farming or fishing), aggregation, processing and distribution (wholesale and retail) »

www.glamur.eu 26

1 point if: 1) using software helping with the decision making process 2) associations helping with the decision making process: Vitival, Vitiswiss, IVV, 3) Joint decisions Decision making for natural resource use making Core value (or the majority 4) joint decisions 0 5 mechanisms chain mechanisms for price

establishment are applied 5) joint decisions mechanisms concerning the management of the organoleptic qualities of wines are applied 1 point if 1)sufficient n° of controls done per year concerning the whole supply chain including financial situations 2) reliability of

controls along the whole chain Fraud Qualitative (impartiality of inspectors) 3) n° of management PBI Core value inspectors 4) proper exchange within the 0 7 chain and traceability of information chain Governance between the different inspectors 5)

creation of tools such as a centralised database to reinforce the traceability system 6) proper severity scale for offenses 7) proper empowerment of inspectors 1) Type of clients (if only 1 type= 0 point; more than 1 type= 1 point); 2) share of local and global clients --> 0 point if not diversified, 1 point if 3)contracts with engagement Market Core value and good relationships 4)price 0 5 Management chain negotiations and evolution 5)strategic sales (1 point if more than one (public testing sessions, exhibition, publicity, working image, distribution network, touristic activities)

1 point if 1) website available, 2) Availability of personal contact with clients, 3) Core value information organization of tasting sessions, 4) 0 5 chain newsletter, 5) point of sale Qualitative PBI information. 1 point if 1)minimum legal information, 2) presence of second Product Wine cellars, label with more information 0 2 Labelling cooperative concerning mode of production, how to consume the product, Information & communication & Information methods used to produce it

Type of transportation from best to GHG Core value worse: truck (1 point), bottles truck emission for Qualitative PBI 3 0 chain ( 2 point), car or little truck (3 transportation Pollution point)

www.glamur.eu 27

For the viticulture: 1) reduce use of tractors and machineries, optimisation of paths, spacing for planting 2) Reasoning cultural practices (temporary or permanent Wine grassing). production For the vinification: 3) Control of GHG and energy and water consumption emissions for 0 5 distribution For the packaging: 4) production (bottling) Commissioning / capping (choice of lightweight glass bottles), PET- BIB packaging 5) packaging/pallets (weight reduction of cartons, reduction of over packaging such as individual boxes, use of wood boxes) 1 point if 1)Limitation of 0 9 phytosanitary products (preventive methods included) 2)Soils’ Grapes, protection practices 3)grassing wine 4)organic matter amendment Environmental production 5)Amendment winemaking pollution and residues as compost 6) reuse of (water, soil, distribution vine shoots 7) preference for air) (bottling) recyclable material 8) consideration of weather condition when using phytosanitary products 9) alternative methods to stop diseases (e.g sexual confusion) 1 point if: 1) recovery of water 0 7 2)use of efficient irrigation technologies 3) dosing spray (drip Water Use Grapes and irrigation or flood irrigation) 4) Practices wine selection of adapted flora production 5)optimization of the pressure and tubing's diameter 6) reuse of water

for installation cleaning 7) no Qualitative PBI irrigation 1) type of bottles (0= new bottles, 0 8 1=if light bottles; 2=recycled

Resource use Resource bottles; 3= washed bottles; 4= only Wine BIB) 2) type of corks (0= if production Material Use capsule, 1= cork) 3) type of tanks and practices (0 if wood thanks, 1= inox tanks ) distribution 4) use of efficient machines 5) (bottling) share of machines 6) use of renewable 7) reuse of materials 8) buildings insolation 1) denomination (0 point= "vin 0 3 rouge" ou "vin de table" or "vins

avec indication de provenance" Association of (vin de pays), 1 point= AOC the product Qualitative PBI Core value Valais, 2 points=AOC region with the chain denomination, 3 points= AOC territory Territoriality municipality denomination (Grand Cru included); 2) practices linked to the territory (dry stones walls); www.glamur.eu 28

3) red autochthon varieties cultivated (Cornalin, Humagne rouge, Diolinoir)

1 point if participating to: 1) Vitival, 0 6 Social 2)Vitiswiss, 3) association with Core value cohesion and local producers 4) other producers' chain Conviviality associations 5) scientific studies 6) syndicates

4.3 Methods used for data collection Data has been collected through semi-structured interviews with grape producers-winemakers who are also cellar men selling wine directly from their cellars. To have a proper overview of the value chain, some data have been also collected from wine shops, restaurants, one big retailer, one input producing enterprise, one bottle producing industry and the only cooperative (see Figure 7).

Figure 7 Localisation of stakeholder in the Valais canton.

In order to test the quality and pertinence of the questions, four preliminary semi- structured interviews have been conducted with producers-winemakers: thus it was asked to the interviewees along the chain what they thought had been missed as part of the data collection process. Afterwards, twenty-two interviews took place with small wine producers who also cultivate grape and distribute it. Interviews’ questions have been modified www.glamur.eu 29

according to the stakeholder’s category interviewed (bottle industry, agricultural inputs, producers, wine shops, retailer, canton). Thus, six different semi- structured interviews have been prepared.

4.4 Methods of participatory checklist used Participatory methods have been used in a first time to assess the coherence of questions addresses. Indeed, preliminary semi-structured interviews with producers and experts helped to refine the questionnaire. Moreover, interviewees helped to illustrate the multiple connections between actors and issues, drawing a clear and precise context and improving the reliability of the results.

5 Results

Attribute Indicator Local data Local score (%) Species Conservation practices 6,81 52 Biodiversity Cultivars diversity 14.15 28.3 Gross Income 155’410 76, 7 Creation and distribution of added Distribution of added value across the chain 0 100 value Contribution to employment 1,14 56.4 Food safety standards and practices 1 33,3 Food safety Artificial additive 2,7 88,4 Decision making mechanisms 3 60 Governance Fraud management within the chain 2,5 35, 7 Market Management 3,02 60,4 Availability of information 3,92 78,3 Information and communication Product Labelling 1,14 56,8 GHG emission for transportation 1,54 48.7 Pollution GHG emissions for production 1.74 34,8 Environmental pollution (water, soil, air) 5,38 59.5 Water Use Practices 1,95 28 Resource use Material Use practices 3,50 43.7 Association of the product with the territory 2,21 74 Territoriality Social cohesion and Conviviality 3,32 55 Table 4 Indicators’ data and performance scores for the local wine value chain This section illustrates data and scores of performance obtained after collecting data (5.1). Therefore, relevant descriptors are presented complementing the indicators of performance (5.2) to finally show the quality of the data used (5.3).

5.1 Scores of performance The following table (Erreur ! Source du renvoi introuvable.) shows the benchmarked scores of performance for the local value chain. The details of the indicator construction and benchmarks are in table 3 (above in part 4.2). www.glamur.eu 30

Social cohesion and Conviviality

Product Labelling

Availability of information

Market Management

fraud management within the…

decision making mechanisms

Artificial additive

Food safety standards and… Territoriality Contribution to employment Resource use Distribution of added value… Pollution Gross Income

environmental pollution…

GHG emissions for production

GHG emission for transportation

Material Use practices

Water Use Practices

Cultivars diversity

Species Conservation practices

0 25 50 75 100

Figure 8 Indicators’ performance. At the graphical level, performances are show in Figure 8 above. www.glamur.eu 31

As it is possible to see in the chart above (Figure 8), the best performances are shown by the following indicators: 1. Artificial additive (food safety attribute) 2. Availability of information (information and communication attribute) 3. Gross income (Creation and distribution of added value)

Whereas the value chains shows to need some improvements for the following indicators: 1. Cultivar diversity (biodiversity) 2. Water use practices (resource use) 3. Food safety standards and practices applied (food safety attribute)

However, for cultivar diversity is it necessary to highlight the fact that the poor score is due to the high threshold presented by the benchmark maximum. In general, 12 indicators of 19 obtained more than 50% of score. More efforts are required at the environmental level to reduce GHG emissions and to better use the resources available such as water (see also Figure 9 and Figure 10, below). At the social level, the employees’ conditions could improve as well as at the health level, and more international standards could be applied.

100 Biodiversity

Resource Use 75 Pollution

Creation and distribution of added value 50 Food safety

Governance

25 Information and communication Territoriality

0

Figure 9 Performance of grape-winemakers according to attribute. www.glamur.eu 32

100

75

Environment Ecomomic 50 Health Social/ethical

25

0

Figure 10 Local value chain performances according to dimensions.

5.1.1 Attribute Biodiversity The local chain performs not very well for both indicators analysed. Indeed, for “species conservation practices” only the minority of stakeholders interviewed applied each practice considered important for this purpose. Hence, this issue reflects the poor score of performance obtained. Nevertheless, practices applied by the majority of the sample are numerous, allowing a significant preservation of flora and fauna species. Concerning the “cultivar diversity” indicator, its performance is also negative due to the high maximum benchmark used: indeed, in the whole canton, more than 50 grape varieties are cultivated whereas our sample harvests an average of 14 different varieties. 27% of the interviewees also affirmed to have other productions (mainly horticulture) within the vineyard, increasing the biodiversity of the analysed area.

5.1.2 Attribute Creation and distribution of added value This attribute performs well for each indicator. The sample analysed has gross incomes higher than the national average. They also present a satisfactorily distribution of added value with the other stakeholders such as the cooperative, wine shops and restaurant and cellar men. The distribution of added value has been calculated using the relative difference between the price paid to the producer and the price paid by the consumer, in agreement with the INRA team. Most of the wine is sold through direct selling in the Valais region, which explains

www.glamur.eu 33

the 100% obtained for this indicator. We could not assess the distribution of added value in the other sub-chains.

5.1.3 Attribute Food safety This attribute performs badly within the indicator “Food safety standards and practices applied” but it is very performant concerning the indicators “artificial additive”. In the first case, only basic norms of food safety such as HACCP are applied by grape growers/winemakers whereas in terms of artificial additives, Swiss regulations are severe allowing only the use of minimum quantities. HACCP constitutes an internationally recognized method to define risks and prevention measures, which is capable to demonstrate with records that products are not harmful for consumers. It is largely used to reduce food risks during winemaking and it could be implemented also to other viticulture operations (OIV, 2012). This method is essential to identify critical control points that could affect the final product’s quality. When fertilizing, for example, risks are associated with the presence of biogenic amines in the must, or fungi producing micotoxins or heavy metals. Preventive measures can be applied such as the OIV guide application for a sustainable viticulture. Other risks are associated with herbicides application, phyto-sanitary treatments, irrigation, mechanic or manual harvesting, leaf removal and transportation towards the cellar. However, only the first four are considered as critical control points to manage (OIV, 2012). Through the interviews, it was asked if other type of certifications such as IS0 9001 and 22000 were applied. The former is conceived to provide a constant quality helping reducing the loss, whereas the latter can be useful for an organization in the food chain to demonstrate its ability to control food safety hazards ensuring the quality of food at the time of human consumption (AMIPEQ, 2014; ISO, 2009).

5.1.4 Attribute Governance Two out of three indicators perform well within this indicator. These are “decision making mechanisms” and “market management” and both perform more than 50%. At the contrary, the fraud management system presents some deficiencies that affect the performance of the indicator. Nevertheless, the overall attribute’s performance is higher than 50%, suggesting that improvements can still be done.

5.1.5 Attribute Information and communication This attribute has been analysed under two main indicators: “availability of information” and “product labelling”. The first indicator refers to stakeholders’ willingness to communicate with customers through direct contact or different platforms such as websites, newsletters and tasting sessions. It takes also into www.glamur.eu 34

account the flow of information within the different stakeholders. The face-to-face interaction proved to increase the quality and quantity of information, and thus customers’ satisfaction. Information concerning the whole chain is also delivered by the Federal Office of Agricultural (FOAG) jointly with local institutions through reports available online. However, for the large public, television reports, journals and magazines remain the best approach to inform the population. Concerning the second indicator, it has been analysed the type of information present in the labels. Hence, this indicator only take into account the information provided for customers when purchasing the final product.

5.1.6 Attribute Pollution This attribute has been declined into three indicators: “GHG emissions for transportation”, “GHG emissions for production” and “environmental pollution”. Due to the lack of precise primary quantitative data, categories have been conceived for all the indicators. Concerning the first two, “GHG emission for transportation” slightly performs better whereas in the case of the third indicator, part of the categories used are strictly related with the practices applied to preserve the biodiversity, hence, the local value chain performs better compared to the two first indicators.

5.1.7 Attribute Resource use Both resource use indicators “water use practices” and “material use practices” perform under 50%. Also in this case, seven and eight categories respectively have been used to determine the final scores. Globally, the attribute reflects a bad performance of the local value chain regarding the best practices adopted to save water and purchase and maintain material.

5.1.8 Attribute Territoriality This attribute wants to highlight the tight link between the local wines and their territory. For that, two indicators have been created: “Association of product with the territory” and “Social cohesion and Conviviality”. Within the first, the willingness to rediscover autochthon varieties, which in some cases implies several technical problems for their cultivation (e.g. Cornalin), denote a clear strategy towards the territory’s promotion. This strategy is also necessary from the market point of view: to maintain market share, the canton Valais has to offer unique products by the conservation of autochthon species (Boisseaux & Barjolle, 2004). In relation to traditional grape varieties, traditional practices still applied such as the dry stone walls shape the territory in a unique way. Dry stone walls have thus www.glamur.eu 35

socio-cultural and heritage functions expressing different facets of the Valais canton. They are also important form the touristic point of view as they attract visitors. Indeed, according to the agricultural service, dry stone walls provide a tourist development supported by the social animation entailed. They can also contribute to economic activity in peripheral zones (SEREC, 2001).

5.2 Relevant descriptors Relevant descriptors have been defined using the first list of indicators, judged essential to describe the value chain but not to assess its performance. Thus, three main descriptors have been identified:

 Communication along the chain  Traceability  Public health

The first one illustrates the quality of the communication between each member of the value chain. Indeed, it was asked to each interviewee (grapes producers, wine makers, cellar men, the cooperative and wine shops) if the communication with other stakeholders was mainly business to business (BtoB) or encompassing engagement and commitment. This descriptor is thus essential to understand the nature of connection between actors. The results obtained after collecting data show that engagement and commitment are common within the local value chain. Concerning traceability, this descriptor is still very important, especially since the scandal of improper wine mixing the value chain had to cope with recently. It is important to highlight that the number and nature of controls are suitable; however, the non-impartiality and non-communication between different organs of controls allowed some wine producers to act illegally. For that, regulations within the value chain are now into a process of improvement. Public health is related to the wine producers and consumers’ conditions when producing and consuming wine, respectively. Both are hotspots illustrating that the use of pesticides and chemical substances during grape production can damage producers’ and their family’s health. Concerning consumers’ health, the discussion is rather focused on the negative and positive effects that wine consumption carries: from the content of anti-oxidative polyphenols to the proper quantity of alcohol that can be ingested.

www.glamur.eu 36

5.3 Data quality check

The quality of the data has been assessed by using the Pedigree matrix described by Lewandowska (Lewandowska, 2004). Data obtained the maximum score thanks to the fact that primary data has been possible to collect and use (see annex 9.3).

Therefore, data is:  Reliable,  representative (from all sites relevant for the supply chain considered, over an adequate period),  with less than 3 years of difference to the time period of the dataset  from the area under study,  from enterprises, processes and materials under study (demonstrating the technological correlation).

www.glamur.eu 37

6 Discussion This section discusses three main parts: (i) the methodology, its effectiveness and limits; (ii) the research questions, context and critical issues shaping the value chain’s performance; (iii) the role of policy setting around the local value chain. 6.1 The methodology and its effectiveness and limits The methodology used to assess the performance of the local value chain has been effective for this purpose. The common list of attributes provided by WP2 allowed including the main critical issues within the assessment of the local wine value chain. Moreover, the participatory approach applied first with key actors and then with the French team, allowed to define the most important indicators which in the future will be the subject of the cross-comparison, which will be facilitated thanks to a common indicator definition. Qualitative research proved to be coherent with the study because “unlike quantitative researchers who seek causal determination, prediction, and generalization of findings, qualitative researchers seek instead illumination, understanding, and extrapolation to similar situations” (Hoepfl, 1997). Also, an accurate literature analysis has been done to define the proper attributes and indicators. However, for certain attributes only two indicators have been selected, excluding other aspects that could be significant for the performance assessment. The main reason has been the lack of data that guided the team to focalize on those indicators providing representative and pertinent data according to the pedigree matrix (Lewandowska, 2004). Nevertheless, those issues have been analysed as descriptors helping to correctly analyse the results obtained. Indicators have been mainly evaluated qualitatively due to the answers obtained with semi-structured interviews, which are often not precise enough to use in quantitative indicators. Moreover, some specific questions, concerning income for example, have been skipped by some interviewees. At the same time, answers regarding the water, electricity and fuel consumptions have not been possible to obtain due to stakeholders’ lack of knowledge or lack of willingness to share accounting data. For benchmarks, it has been possible to find secondary data regarding the specific value chain under study, increasing the reliability and precision of scores obtained.

www.glamur.eu 38

6.2 Addressing the attributes, indicators and research questions This section presents a precise analysis of each of the attributes and their related indicators taking into account the research questions previously listed (see section 3.1). For that, the context, the critical issues and policy regulating the value chain are included in the discussion.

6.2.1 Biodiversity (environmental dimension) Concerning the first indicator “Species Conservation practices”, local producers are aware of the importance to have biodiversity surrounding the grapes culture in order to improve the ecological quality of the vineyard. The majority of producers’ interviewed already implement important practices such as organic management of parasites, the preservation of fauna and flora networks, the maintenance of wildflower strips, grassing, the preservation of ecological infrastructures. These practices are included in the required ecological services (RES) which allow having direct payments (FOAG, 2015) . However, more efforts can be done as being part of projects supporting landscapes’ quality (such as the agro-environmental viticulture network) and increasing the percentage of ecological surfaces within the vineyard, beneficial for the promotion and conservation of the biodiversity. Several motivations can be contemplated by producers such as (Emery, 2013): 1. Actively participate in the maintenance of biological and landscape diversity of quality in the vineyard; 2. Prepare the ground for possible practices of ecological compensation network’ areas; 3. Obtain financial contributions (only for parcels meeting the quality criteria EQO) 4. Become familiar with maintenance practices saving soil from herbicides (for vineyards with predominant chemical weed control).

A critical issue influencing the biodiversity of the region is the presence of dry stone walls: Indeed, for many animals (including "auxiliary crop") the vineyard is an important medium in terms of food, but it often lacks shelter or nesting site for the majority of them. For that, the presence of scree or dry stone walls offers a very favourable habitat: the interstices serve both as cottages, shelter and breeding. They can also accommodate a diversified food. Finally, they are used to support special vegetation typical from clean-poor soils (e.g. sedum). Moreover, due to its dimensions and sometimes substantial linear form, the wall plays a role as very interesting corridor for many species: many species of lizard, www.glamur.eu 39

worms and snakes are common hosts. In the many cavities, there are also wild bees, beetles and ants. Countless species of insects live there all along the year; among them are "ancillary cultures"(see specific fact sheet) contributing to the environmental management of vineyards restricting the use of pesticides. Meanwhile, dry stone walls are a suitable environment for fifty species of lichens sheltering in turn very specific snails. Finally, insectivorous birds and carnivores find there a rich and varied food (LPO Alsace, n.d.). Concerning the second indicator “cultivar diversity”, a high number of cultivars entail positive effects in terms of agro-biodiversity. More than 50 grape varieties are cultivated in the whole canton (Office Cantonal de la Viticulture, 2008) though the average of varieties cultivated by our sample of producers is 14 per vineyard. This number is higher than other European countries probably due to a marketing strategy and tradition. The low performance in this indicator is thus explained by a high maximum benchmark that shows the high regional agro-biodiversity. Biodiversity is linked to other dimensions such as the social dimension with the existence of dry stone walls, which are part of the local heritage, the ethical due to an increasing number of consumers willing to purchase wines respectful of the environment and the economic thanks to the tourists seeking for natural attractions. Consequently, also marketing and political spheres can influence positively or negatively the biodiversity performance.

www.glamur.eu 40

6.2.2 Creation and distribution of added value (economic dimension) This attribute has been declined in three main indicators. Concerning the gross income, interviewees present higher values compared to national average calculated by the institute Agroscope (2013). This can be due to the fact that producers have their fixed clients and they produce high quality wines giving major profits. Moreover, in many cases they don’t have to support the financial charge of annual employees and rather turn to seasonal low-paid workforce. The wine sector also benefits from a high-value product that pays the producers well compared to other agricultural sectors, such as milk, and this is why the performance is high in comparison with the national average used as benchmark. This result is in contrast with the findings of the Viti2015 written by the professor Bernard Catry from the University of Lausanne. Indeed, he argues that “the winemaker (including grape grower) step is not profitable. This ultimately threatens the sustainability of the sector. For that, there is a need to increase incomes by incrementing the wine quality and therefore the price of the wines. Otherwise, it will be necessary to revisit the distribution of profit margins within the profession” (Catry, 2009). Therefore, grape growers-winemakers significantly contribute to the canton economy though they do not create an economic surplus unlike their closest partners’ in the market: cellar men and distributors. This aspect leads to another critical issue: the distribution of added value within the whole value chain. According to Catry (2009), there will be no future for the sector without a strong vineyard dimension. Nevertheless, improving the winemaker step could cause the destabilisation of other supply chain’s actors, thus is necessary to increase uniformly the added value of the whole supply chain by raising the product’s quality and consequently the prices of Valais wines (Catry, 2009). He proposed to raise the price of the wine of about 1.10 of Swiss francs (CHF) per litre to increase the sector’s turnover by 10%. This would be profitable for each stakeholder including grape growers-winemakers who will earn 60 cents more per kilo of raisins (Catry, 2009). Thus, the challenge is to continue to improve the quality of wines from the Valais to differentiate them from other wines, while correcting the low profitability of the wine stage. This can be done by including economic incentives for grape growers-winemakers to invest sufficiently in their exploitations, which is not currently the case; decreasing the vineyards’ value. The most critical threats are the concentration of big retailers and new wines entrance in the sector (Catry, 2009). www.glamur.eu 41

It is also important to highlight that the added value changes according to the type of product sold and the type of distribution used to do it (see Figure 11).

Figure 11 Gross added value estimation according to the product and the distribution channel in CHF.

As we can appreciate in the chart above, according to different grape varieties (Fendant, Johannisberg, Pinot Noir, Petite Arvine) and mode of distribution (VD= direct sell and GD= retailers), the added value is distributed differently between the grape growing (VITI, in blue), the wine making process (VINI, in light blue) and the packaging/distribution sectors (CONDICOMM, in turquoise). Thus, the direct sales produces more added value compared to retailers. Indeed, the formers allow selling the wine at the proper price without applying discounts as is done by retailers (Catry, 2009). In general, the higher added value can be found at the grape production stage (2.74), followed by the packaging/distribution step (1.37) and the wine making process stage (1.02) (Catry, 2009) . The differences in profitability between the various actors of the sector are often due to the power of negotiation with their partners (clients or suppliers). In addition, the fragmentation of vineyards makes the increase of grape growing/winemakers’ incomes by negotiating the prices of sale with cellar men or traders difficult. It was identified in the critical issues that revenues of grape growers and wine growers would be critical. However, the results found show that the sample of producers studied do not present very critical situations in terms of revenues, at least for the exploitation managers interviewed, who manage to generate a comfortable income. Regarding the contribution to employment, the viticulture www.glamur.eu 42

remains the largest agricultural employer in Valais (Catry, 2009). Winemakers significantly contribute to the canton’s economy, especially in terms of jobs. This is due to the low level of vineyards’ mechanisation, not facilitated by the topography and by the relatively high average age of grape growers. Nevertheless, this value chain still relies on seasonal employees not always declared to the cantonal services. Thus, they are paid in black and they don’t have any work permit. Though their gross salary of 15 CHF per hour is higher compared to other sectors such as the vegetable, the wage is still too low to live properly in the country in addition to the difficult condition of work. The Valais canton suggests that the federal law has to require the announcement of any employee engaged by the employer from the first day of work, as is already done France (Berreau, 2014).

6.2.3 Food safety Due to high level of competition in the wine market, food safety management such as HACCP help to maintain competitive advantages by creating consumers’ faith in the safety of the products (van Schoor & Groenewald , 2002). However, at a local and small scale, standards such as ISO 9001 or 22000 are not contemplated mainly due to the high investments required between internal and external costs (The 9000 store, 2015). These evidences explained why this indicator presents a bad performance. The implementation of ISO standards has been just confirmed by the cooperative and other “big” stakeholders who can afford the entire procedure. Concerning the indicator “artificial additive”, its performance is very satisfactory (88.4%) thanks to the high quality of products that do not need artificial additives to be improved. In both sulphites and sugar cases, quantities used are significantly below the maximum allowed especially concerning added sugar. The PDO regulations prohibit the use of added sugar (Le Conseil d'Etat du canton du Valais, 2004). Concerning sulphites, these are used as an antimicrobial agent inhibiting yeast and bacteria and as an antioxidant to protect flavours and defend against browning. Hence, they increase the shelf-life of the final product (Leake, 2014). Nowadays, the consumers’ concern about the presence of sulphites in the final product is rising and it has supported the decision to indicate in the label their presence when the concentration used is higher than 10 mg/L (Le Conseil d'Etat du canton du Valais, 2004). This is due to the fact that sulphites can provoke an inflammatory reaction caused by poor tolerance of the chemical compounds contained in sulphites (Hordé, 2014). However, the possibilities to produce wine exempt of sulphites are quite low, with

www.glamur.eu 43

only one biodynamic producer mentioning a trial to produce a sulphite-free wine, while still producing most wines with sulphites.

6.2.4 Governance This attribute has been analysed under three main indicators: “Decision making mechanisms”, “fraud management within the chain” and “market management”. Regarding the first indicator, the local value chain needs to improve the following issues: (i) In order to facilitate the decision making process, software such as “Aide à la decision” (OAD) could be used. Indeed, it is applied to illustrate phyto-sanitary situations, development and forecasting of pests and it is accompanied by decision rules helping the user in his intervention’s choices. Towards sustainability, reasoning support tools could be analysed, in particularly those that integrate not only valuation models and risk prediction, but also agronomic variables and economic approach (e.g. cereals’ values). These tools generally allow analysis at a scale from the farm to the plot (Agriculture.gouv.fr, 2012). The “suzukii-fly” crisis that occurred in autumn 2014 revealed that such tools are not in use and grape producers found themselves vulnerable, turning to private companies for help, instead of receiving valuable coordination and advice from the cantonal or inter-professional services. (ii) Price decision making is strictly linked to the power of negotiation. The latter depends on the degree of concentration of each type of stakeholders: The more stakeholders are concentrated, the more negotiation of good conditions with upstream or downstream stakeholders is possible. This is the case of cellar men who are able to negotiate optimal conditions with growers/winemakers and customers, including retailers (Catry, 2009).

On the contrary, joint decisions are taken concerning natural resource use and the management of organoleptic qualities of wines under the coordination of the inter-profession (IVV), official interlocutor of the canton of Valais. The inter- profession is an organization composed by equal number of grape growers, grape growers-winemakers, the cooperative and cellar men (traders) defending the general interests of the sector (economy, quality, promotion) (Les vins du Valais, 2013). Other associations such as Vitival facilitate the decision making process concerning the development of sustainable and responsible viticulture (Vitival, 2015). www.glamur.eu 44

Clements et al. (n.d.) suggest that decision making in a supply chain must correspond to (i) high level of integration between the same group of stakeholders and stakeholders of other groups, and (ii) direct communication without presence of intermediaries. Regarding the indicator “fraud management within the chain”, the case study highlighted controversial aspects: on one hand, the number and the nature of controls are satisfactory and reliable; on the other hand, the high number of different type of inspectors and the absence of exchange of information between them jointly with their small power and the inefficient severity of offenses’ scale, reflect the necessity to improve the local value chain’s fraud management system (RTS, 2015). To avoid creating excessive administrative burdens, it should be possible to create plots of lots. This method is expected to maintain traceability to the plot without increasing the administrative work. (IVV, 2014). The three main benefits expected are:  Computerized data allow faster statistics.  Greater accuracy to better target controls, according to the principle of criticality or risk.  Ensure traceability to the vineyard plot offers more customer transparency on the source.

Nowadays, data related to the surface (yield limit) and production (harvest report) are managed separately by two different bodies: the Department of Agriculture and cantonal chemist, respectively. To improve that, the Valais’ canton is creating a centralised database that will help to store and centralise all data to increase traceability and facilitate the access to information (IVV, 2014). The third indicator “market management” has been constructed to analyse the local chain’s resilience at the market level. Stakeholders principally rely on local customers (including customers from other cantons) whom loyalty is extremely strong in many cases. For that, strategic sales (public tasting sessions, exhibition, publicity, working image, distribution network, touristic activities) together with price negotiations are constantly put in practice. Concerning global customers, these are not a reliable financial source for stakeholders because of their little knowledge about Valais’ wine and their willingness to mainly experience the natural attractions such as the Alps. The industry believes that the real competitors of Valais wines are foreign wines. In this sense, collaboration with colleagues from other Swiss wine regions and Swiss Wine Promotion (SWP) are preferred. The IVV has also no export strategy: Choosing a target market from a geographical point of view is impossible, www.glamur.eu 45

because there is no Swiss dominance on international wine markets. To promote exports, IVV should be focused on creating an image of Valais wines by its participation in various events, competitions, and tastings and media actions. Consequently, Swiss’ recognition will increase too. For that, the IVV wants to set up a reflection on this topic with the SWP (Swiss Wine Promotion) (IVV, 2014).In summary, thanks to different types of local customers, sales strategies and the good reputation of wines, the local value chain’s resilience is assured. In addition, the high differentiation of product allows increasing the market share within the country. The Inter-profession IVV will improve the marketing strategy towards the German part of Switzerland, finding co-branding with other branches (tourism – other agricultural products – Fashion – etc.) and launching a campaign “From the vine to the bottle” during the harvest season (IVV, 2014).

6.2.5 Information and communication This indicator has been created to verify the presence of an integrated system between the local value chain’s upstream and downstream parts, taking also into consideration the type of communication existing along the chain: relations can be Business to Business (B to B), Business to Consumers (B to C) and/or based mainly on engagement and commitment (Svensson, et al., 2004). Thus, essential information for grape growers, wine makers and cellar men are elaborated by the Inter-profession who takes position on all legislative, regulatory and administrative issues regarding the viticulture in Valais (IVV, 2014). The flow of information between these two groups is bidirectional, characterized by written contracts and focused towards the same objectives. Contracts are also used by the cooperative and big retailers whereas between producers and little shops or restaurants it is easier to find simple oral stipulations. Grape growers/winemakers mainly rely on direct contacts to provide information to customers: once per week it is thus possible to taste wines directly from the cellar before purchasing it. This service is not proposed by big retailers which present difficulties to build loyal relationships with customers. Small producers make efforts to avoid this situation by creating also personalised web pages where essential information about the vineyard can be found. As the availability of information for consumers is high, local wine bottles do not present much information at the labels level. Interviewees affirmed using labels only due to the laws that impose them. Thus, no information concerning the mode of production or how to consume the product is present and rear labels are rarely used. This fact could be the cause of such little volumes exported.

www.glamur.eu 46

6.2.6 Pollution This attribute has been defined as critical in order to improve the sustainability of the local value chain within the environmental dimension. Transport and

packaging cause high quantities of CO2 emitted into the atmosphere. Transport generates more that 50% of all CO2 emitted when producing, packing and distributing wine. The critical aspect is related to the mode of transport used and not the distances travelled (van der Zanden, 2009) (see table 5, below): Table 5 Levels of CO2 generated per tonne and km for various modes of transport (van der Zanden, 2009).

Interviews highlighted that local producers mainly used car or little truck to reach customers. This is mainly due to their mutual proximity and the low volumes sold. Vice versa, it has not been possible to take into account the transport mode and distances covered by customers. Nevertheless, it is easy to imagine them using

private cars to reach the cellar points and thus increasing the CO2 emissions. Improvements are noticed in the retailing sector where wine is transported in bulk

and bottled close to destination, saving 35-50% in transport CO2 (van der Zanden, 2009). However, the total distance in the Valais core value chain (from grape production to the final product ready to be sold) is very short. In most wineries interviewed, this distance is close to zero km as the grapes are grown next to the vinification installation and the shop. Nevertheless, distances are travelled for the provision of inputs such as bottles and the delivery of bottles to the customers. It has been revealed by the bottle factory that most material comes from Italy. The delivery of bottles however remains mostly within the Valais canton or reaches neighbouring cantons. The local chain thus still relies on road transport; for that, in order to decrease GHG emissions, local producers could use eco-friendly packaging. By reducing the weight and increasing the percentage of recycled material used for the www.glamur.eu 47

packaging, CO2 footprint will be reduced significantly. Thus, light packaging with high recycled material content such as recycled PET (rPET) could represent a good solution to be taken into account (van der Zanden, 2009). However, interviews highlighted that little of these innovations are applied. Glass bottles remain the main used packaging emitting 30-40 % of the total emissions needed to elaborate a wine bottle of 0.75 L. Moreover, a full bottle worsens the

CO2 footprint about 5 to 20% during transportation (Ducruet, et al., 2011). These results have been confirmed by an LCA study on wine produced in the Geneva canton (SOFIES, 2009) (Figure 12, below). The chart represents the red wine’s environmental impacts in milipoint (mPt) when bottling (in green), vinificating (in red) and producing (in blu) in the Geneva canton (first to columns from the left).

Figure 12 Environmental impacts of red wine in Switzerland, Chili and Spain (SOFIES, 2009).

Thus, the major environmental impacts come from the bottling including the production of bottles. For that, nowadays the Bag-In-Box (BIB) packaging starts to be taken into consideration, mainly by enterprises selling in bulk. BIB

packaging decreases significantly the CO2 emissions due to its low weight (200- 250 g). Moreover, CO2 emissions produced for the fabrication of 5L BIB is eight time inferior than the production of 6 bottles of wine (550 g of CO2 for the first against 4000g of CO2 for the second). The negative point of this technology is the fact that BIB is permeable to gases, decreasing the wine’s quality and entailing microbiological risks. However, if transported, distributed and consumed rapidly, it could represent a valid option (Ducruet, et al., 2011). The impact on the image

www.glamur.eu 48

of the product would however be an important question and the acceptability of such packages is not guaranteed. During wine production, from grape harvesting until bottling, other practices can

be applied in order to decrease CO2 emissions. For example, according to the French Institute of Vine and Wine (IFV), the reduction of tractors’ use, paths’ optimisation, row spacing planting, permanent or temporary grassing, cartons’ weight reduction, the decrease of over-packaging and the use of wood boxes

have strong impacts on CO2 emissions decreasing (IFV, n.d.). The case study highlighted that the quantities of herbicides applied in the vineyards decline slowly. This is clearly due to operating constraints intensified by the particular conditions in Valais (dry climate, light soils fragmentation of the vineyard ...) (Clavien, n.d.). Organic domains are doing better in terms of biodiversity and no-pesticide used but their number remains low. However, preventive methods, soils’ protection, alternative methods to control diseases (e.g. sexual confusion) are applied by the majority of the stakeholders interviewed. Moreover, what supports the local chain is the fact that only little mechanization can be applied due to the challenging topography of the area. Thus, indirectly less pollution is produced.

6.2.7 Resource use The first indicator water use management has revealed to be not efficient within the wine production. This fact is mainly due to the availability of water from the areas surrounding the vineyards (IVV, 2010). The majority of producers don’t apply practices such as the recovery of water, efficient irrigation technologies or the reuse of water for installations cleaning. Therefore, the availability of water is not felt as a critical issue. However, other types of practices are applied such as the dosing spray (by drip irrigation or flood irrigation), the selection of adapted flora keeping the soil humid and the optimization of the pressure and tubing's diameter. Valais is still the drier canton of Switzerland and thus, producers should consider this issue and the canton could sensitize them more as future droughts could occur more frequently. Also concerning the second indicator concerning material use practices, performances are under the 50% mainly due to the kind of packaging used. However, efforts are made by using isolated buildings, the reuse of certain material and the share of machines between producers. Renewable energies are applied in only few cases. In the region, one company is charged of industrial washing bottles. An industrial washing bottle conserves resources of raw materials by direct reuse of packaging (Univerre, 2015). This technology uses less resources and energy than the glass production. At the economic level, it is www.glamur.eu 49

also less expensive than the fabrication of new bottles. At the sanitary level, no microbiological contamination is present. However, it needs important quantities of water for the bottles’ cleaning. Concerning stoppers, the material used is principally cork, bought by all stakeholders interviewed. Cork is a natural, renewable, biodegradable product obtained by non-sacrificed trees. It is also the result of a non-environmentally harmful transformation. However, this material has two main drawbacks: mechanical defect and the formation of trichloroanisole (TCA). The first problem is a technical issue which can be solves by modern technologies. The latter entails quality and organoleptic problems rising when the cork’s humidity is higher than 8% (Varela, 2004). Synthetic stoppers avoid the TCA formation, protecting the wine from oxidation and any reduction process. Thus, the levels of sulphurous anhydride and carbon dioxide do not change over time and there are no risks to lose aromas or wine freshness (Korked, 2015). In summary, combining environmentally friendly practices with high quality would be the best solution to apply. Nowadays more efforts need to be done to achieve this level.

6.2.8 Territoriality The link between the Valais region and his wine has been constructed rediscovering autochthon varieties, by using traditional practices such as dry stone walls and by using efficiently certifications such as Geographical Indications. Through interviews, it has possible to verify an increasing willingness to cultivate ancient varieties to propose products that can be identified with the territories and stakeholders behind them. Concerning the use of dry stone walls, as we can see in Figure 13, terraced vineyards are highly appreciated for their territorial (ET), social (ES) and environmental externalities (EE). Indeed, they allow maximum exposure to biological agents and thus a higher quality grape, despite some difficulty of access (Canton du Valais, 2015). Dry stone walls are also more attractive within the market (AM) occupying competitive positions (PC) (Lehmann, 2000).

www.glamur.eu 50

Figure 13 Overall assessment of the Valais wine (Lehmann, 2000).

Another element helping in the construction of territorial wines are the Geographical Indications such as PDO (AOC in Switzerland), aiming to protect and valorise of traditional products, are widely used since many years. What mainly differ Swiss PDOs from European PDOs is the fact that the latter makes reference to a determinate geographical zone which englobes different municipalities. Local winemakers in Valais, however, prefer to use the name of the municipality, also recognized as PDO areas. Consequently wine PDOs are highly fragmented and characterized by individual entrepreneurial approaches (Boisseaux & Barjolle, 2004). However, the Inter-profession IVV denounces that the current AOC legislation does not allow a true differentiation of Valais wines. For that, IVV suggests the creation of a distinct label named “Label Marque Valais” (see Figure 14) to use with those wines that fully meet the ethical values and authenticity. This new range of Valais wines meet high requirements in terms of quality, ethically, responsibility and authenticity (IVV, 2014). The book of specifications of the “Label Marque Valais” will include some criteria such as:

 Strict oenological rules (only 5% of tolerance when mixing with other wines)  Include the Vitiswiss Certification (Integrated Production and Sustainable Development) for environmental benefits  More traceability and retribution to producers  Increase the supply chain and canton added value  Increase traceability for consumers providing information by QR code. Vinification and bottling must be done in the canton

www.glamur.eu 51

Figure 14 Pyramidal distribution of Valais’s wines (IVV, 2010).

IVV wants also to revise the current PDO book of specifications to distinguish this wine category from wines labelled “Label Marque Valais”: The goal is to offer consumers a choice between authentic wines (under the new brand) and creative ones (under current brands). Thus, oenological internationally recognized instruments, such as mixing or using the RCM (rectified concentrated must) should be used for basic AOC to meet the needs of consumers and to balance Valais wines with their competitors (IVV, 2014). The necessity of a new brand is due to the willingness to offer better products to consumers and to distinguish them by a higher quality and a stronger link with the territory. This necessity was born to react to the fraudulent scandalous that recently damage the Valais’ reputation. The performance of this indicator would thus change, perhaps showing a higher performance score of the local value chain. The number of events and associations reflecting the link between wine and territory is high and englobe tourisms activities to promote the region and its products. Interviews highlighted that these activities are essential to create notoriety. Thus, the IVV's slogan "Valais, the favourite wine destination of Swiss” is supported by a majority of value chain’s stakeholders. Communication starts to be less focused on the grape varieties and more on the particularities of Valais wines. Wine tourism is also another important tool that needs to be developed to bring consumers closer to winemakers. For that, the Valais canton must highlight the unique characteristics of Valais wines: the www.glamur.eu 52

quality of reception in the cellars, the stunning natural scenery and great varieties diversity (IVV, 2014). 6.3 Discussion of the local performance This section illustrates a final overview of the scores of performance achieved by the study of the local red value chain in the Valais canton. It takes into account the scores’ coherency, the influence of descriptors over scores, and the existence or not of strong interactions between indicators. Moreover, the new research questions generated by the study and what types of output are expected from the French comparison are treated. Due to the fact that interviews have been conducted mainly with grape producers-wine makers, it has not been possible to draw general comments regarding the whole value chain. Moreover, for some indicators such as “Product labelling” and “GHG emissions for transportation”, more categories could be used to refine the final score of performance. In this case, secondary data could be used to assess the indicator’s performance or more exhaustive methods such as LCA could be used if time is available. Regarding descriptors, those have been essential to understand the scores obtained. In addition to those illustrated in the section 6.2, other descriptors such as topography, agricultural policy and subsidies, consumer behaviour and tourism have been included in the discussion (seeFigure 15, below).

www.glamur.eu 53

Figure 15 Positive, negative and neutral influences between indicators and descriptors created with NodeXL.

Black dots represent the descriptors, green arrows represent positive influences between descriptors and indicators, red arrows represent negative influences and black arrows represent a neutral or undefined influence (Figure 15). As it is illustrated, descriptors have mainly positive influences on the performance of attributes studied. However, the topography of the canton has a negative impact on the resources used due to the fact that it shapes the territory with steep slopes making the cultivation of grapes more difficult and implying the use of many resources to levelling the ground. The difficulties entailed by the topography, nevertheless, involved the creation of dry stone walls. As dry stone walls are recognized as unique structures shaping the identity of the territory, the topography has a positive influence on the territoriality of the Valais canton. The topography has also a neutral influence on pollution meaning that both positive and negative effects have been found. Indeed, thanks to the topography of the territory, only little mechanization is possible; however it incites the use of airplanes to spread substances on vineyards, producing higher risks of pollution. Tourism is an important tool to create added value and to reinforce the notoriety www.glamur.eu 54

of the canton. Nevertheless, no clear influence has been found between consumer behaviour and territoriality and creation and distribution of added value: Some consumers accept to pay high prices to buy high quality products directly from producers; however, lower quality wines sold in supermarkets continue to be purchased, increasing the retailers’ income and consequently the added value distribution’s inequality. Nevertheless, customers mainly purchase certifies bottles (PDOs ones) as quality guarantee, showing the willingness to buy products strongly linked with the territory. Agricultural policy and subsidies are positively related to pollution and biodiversity performances due to the willingness to decrease the use of pollutant substances used and the different voluntary and subsidy programs that encourage conservations areas for example. As subsidies interact with producers’ revenues, there is an influence, but it is very hard to define whether is positive or negative as Catry (2009) still denounces the incomes’ inequality between producers and retailers. Concerning interactions between indicators, some evidences have been found. For example, the fact that producers putting efforts on preserving the biodiversity implies economic benefits that increases their gross incomes. The biodiversity conservation is also significantly important from the territorial point of view: it shapes the territory and the grapes cultivated, increasing the link between the product and its territory. Another strong interaction has been found between “pollution” and “resource use” attributes. Indeed, the resources used to produce wine entail pollution at different levels of the value chain. The attribute “governance” proved to be essential to understand the functioning of the value chain because it entails all decisions taken at different value chain’s levels, for example, regarding food safety, the creation and distribution of added value and the territoriality. Within this attribute, the Inter-profession IVV plays an important role. The analysis conducted shaped also new research questions associated with other attributes contemplated in the WP2 comparative report (task 2.3) such as: labour relations, resilience and technological innovation (Kirwan, et al., 2014). Concerning the first attribute, it would be useful to better analyse the socio- economic welfare and recognition of workers associated with the health-related labour risks. Indeed, both issues are current hotspots that the value chain has to cope with. In the context of the actual agricultural policy, resilience would be associated to the economic dimension due to the high dependency to direct payments of many producers to properly run the activity. Resilience would also be associated to the market sphere due to the little capacity of the value chain to cope with external shocks within international markets. Concerning technological www.glamur.eu 55

innovations, it would be useful for the value chain to study the implementation of new packaging and way of transportation to reduce the pollution level. In general, wine supply chain is still polluting the surrounding environment by the fabrication and use of glass bottles and the use of transports such as trucks and cars.

Thus, the following research questions could be addressed:

1. To what extent the welfare and recognition of workers improve the social and health dimensions of the value chain? 2. Which mechanisms could be put in place to increase the economic resilience of the value chain? 3. To what extent could technological innovation decrease the value chain carbon footprint and pollution?

These attributes have not been included in the analysis because they are not being considered as essential to assess the performance of the value chain under study. This decision has been taken after the agreement between both Swiss and French teams to focalise the attention on those issues considered as crucial for the value chain analysis meanwhile reducing the number of attributes analysed. They are moreover especially difficult to benchmark and future progress in research on sustainability in these areas will help create more benchmarks and allow using more indicators. Then, they would refine the analysis by detailing more issues, thus giving more information useful to enrich the study. From the French comparison, it is expected to verify if the indicators chosen properly allow comparing the global-local performance.

www.glamur.eu 56

7 Conclusion The analysis of the local red value chain in the Valais canton highlighted that, at the grape grower-winemaker level, almost all the attributes analyzed present good performance with the exception of the attribute “resource use”. These indicators are under the 50% of performance. Indeed, water and material use practices must be improved through effective legislations jointly with producers’ willingness. In relation to material use, another critical issue to cope with is the production and use of glass bottle entailing high levels of pollution. Although a clear effort to preserve biodiversity and promotion of ancient grape varieties exist,

the CO2 emissions released during bottle fabrication and transportation are too high, worsening the environmental performance of the value chain. Thus, efforts are necessary to promote lower footprint packaging by sensitization of customers about their negative impact on ecosystems. At the same time, more investigations in the R&D area are needed to improve the safety and quality of products contained in more ecological packaging such as BIB. Concerning the social and ethical dimensions, major changes should be taken into account regarding the fraud management system in order to increase consumers’ trust and the workforce’s economic and health conditions. Nevertheless, the Valais wines’ notoriety is strong and well anchored to the territory as it has been reflected by the attribute “territoriality”. These issues have a positive impact in the economic sector, increasing the creation of added value. Nonetheless, more studies are necessary to analyze the distribution of the added value between the upstream and the downstream parts of the value chain to avoid inequality. The methodology used to assess the performance has been relevant, although for some indicators more categories could be created and the use of quantitative data could have been more appropriate, especially within the environmental dimension. No major difficulties have been experienced concerning the data collection. However biological adversities such as the arrival of the “Suzukii-fly” provoked several interviews’ delays that have limited the time to include more indicators in the discussion.

www.glamur.eu 57

8 References AGIR, 2013. Valais du Vin. [Online] Available at: http://valaisduvin.com/2013/08/16/importations- exportations-les-chiffres/ [Accessed 16 01 2014]. Agriculture.gouv.fr, 2012. Agriculture.gouv.fr. [Online] Available at: http://agriculture.gouv.fr/Les-Outils- d-Aide-a-la-Decision [Accessed 01 12 2014]. Berreau, G., 2014. Le Nouvelliste. [Online] Available at: http://www.lenouvelliste.ch/fr/en-continu/vingt- travailleurs-au-noir-dans-une-cave-498-1363681 [Accessed 08 01 2015]. Boisseaux, S. & Barjolle, D., 2004. La bataille des A.O.C. en Suisse. Lausanne: Presses polytechniques et universitaires romandes. Canton du Valais, 2015. Site officiel du canton du Valais. [Online] Available at: http://www.vs.ch/Navig/navig.asp?MenuID=29152&Language=fr[Accessed 20 02 2015]. Catry, B., 2009. VITI 2015. STRATEGIE VITIVINICOLE VALAISANNE A L’HORIZON 2015, Sierre. Clavien, Y., n.d. Influence des pratiques d'entretien du sol et cartographie de vignobles. Clements , M. D., Wilson, M. M. & Bacanaru , E.-D., n.d. CILT. [Online] Available at: http://www.cilt.co.nz/Story?Action=View&Story_id=1089[Accessed 20 11 2014]. Ducruet, J., Deneulin , P. & Riedo, A., 2011. Conservation du vin en Bag-In-Box. Revue suisse Viticulture, Arboriculture, Horticulture, 43(5), pp. 290-295. Emery, S., 2013. Vignobles à biodiversité naturelle: des surfaces de compensation écologique à promouvoir. Revue Suisse Vitic.Arboric.Hortic., 45(1), pp. 67-68. Etat du Valais, laboratoire cantonal, 2006. Statistiques des vins 2006. Etat du Valais, 2000. Vers une agriculture valaisanne durable. FAO, 2013. Sustainability Assessment of Food and Agriculture systems guideline version 3.0, Rome. FOAG, 2011. L’année viticole 2010. FOAG, 2012. Rapport agricole 2012 (in French), Switzerland: Federal office of agriculture FOAG. FOAG, 2013. Das Weinjahr. L’année viticole 2012. FOAG, 2013. Rapport agricole. FOAG, 2014. L'année viticole 2013. FOAG, 2015. Office fédéral de l'agriculture. [Online] Available at: http://www.blw.admin.ch/themen/00006/00049/index.html?lang=fr[Accessed 24 02 2015]. Gallien, A., 2011. Intellego. [Online] Available at: http://www.intellego.fr/soutien-scolaire--/aide-scolaire-svt/la-production-de-vin-ou- vinification-dans-le-cas-du-vin-rouge-et-du-vin-blanc/53698[Accessed 01 02 2015]. Hoepfl, M. C., 1997. Choosing qualitative research: A primer for technology education researchers. Journal of Technology Education, 9(1), pp. 47-63. Hordé , P., 2014. Santé medicine. [Online] Available at: http://sante-medecine.commentcamarche.net/faq/2053-allergie-ou-intolerance-aux- sulfites [Accessed 12 02 2015]. www.glamur.eu 58

IFV, n.d. Comptabilisation des émissions de gaz à effet de serre: Application de la méthode Bilan Carbone à la filière viti-vinicole. INTER OC, 2011. Rapport d’Activité. IVV, 2010. Les essentiels des vins du Valais. IVV, 2014. Viti horizon 2020. Définition de la stratégie vitivinicole valaisanne à l'horizon 2020. Kirwan, J. et al., 2014. GLAMUR WP2 - Scoping / framing general comparative report on food chain performance (deliverable 2.3), UK: Countryside and Community Research Institute, University of Gloucestershire. Korked, 2015. Korked. [Online] Available at: http://korked.com/products/synthetic-stoppers[Accessed 18 02 2015]. Laesslé, M., 2012. Production locale de valeurs par la vigne et le vin : cadre conceptuel en termes de ressource. Le Conseil d'Etat du canton du Valais, 2004. Ordonnance sur la vigne et le vin. Le Nouvelliste, 2014. Le Nouvelliste. [Online] Available at: http://www.lenouvelliste.ch/fr/en- continu/vingt-travailleurs-au-noir-dans-une-cave-498-1363681[Accessed 02 02 2015]. Leake, L., 2014. Food Quality and Safety. [Online] Available at: http://www.foodquality.com/details/article/6523601/Wine_Quality_and_Safety_101.html?tzcheck= 1 [Accessed 12 02 2015]. Lehmann, B. e. a., 2000. Vers une agriculture valaisanne durable. Les vins du Valais, 2013. Les vins du Valais. [En ligne] Available at: http://www.lesvinsduvalais.ch/encaveur/sommes-nous.html[Accès le 17 01 2015]. Lewandowska, A., 2004. Comparative LCA of Industrial Objects. Part 1: LCA Data Quality Assurance - Sensitivity Analysis and Pedigree Matrix. The International Journal of Life Cycle Assessment, pp. 86-89. LPO Alsace, n.d. Biodivine. [Online] Available at: http://www.biodivine.eu/docs/murets-pierriers_alsace_biodiversite.pdf[Accessed 15 01 2015]. Lüdi , H., 2013. Viticulture. [Online] Available at: http://www.hls-dhs-dss.ch/textes/f/F13937.php?topdf=1[Accessed 17 10 2014]. Ministère de l'agriculture, 2013. Perspectives d’évolution de la filière vitivinicole dans la région Languedoc-Roussillon à l’horizon 2025. Neven, D., 2014. Developing sustainable food value chains. Guiding principles, Rome: FAO. Observatoire Viticole, 2010. COMPRENDRE LE MARCHE MONDIAL DU VIN EN VRAC. [Online] Available at: http://www.obs-viti-cg34.com/viti/download_file.jsp?id_doc=1618[Accessed 31 October 2014]. Office Cantonal de la viticulture, 2008. [Online]. Available at: http://www.vs.ch/NavigData/DS_68/M5929/fr/Ronde_des_c%C3%A9pages_avril_2008.pdf [Accessed 22 12 2014]. Office de la viticulture, 2013. Observatoire des vins en grande distribution en suisse. www.glamur.eu 59

OFS, 2015. Agriculture - Indicateurs. Production et revenus. [Online] Available at: http://www.bfs.admin.ch/bfs/portal/fr/index/themen/07/03/blank/ind24.indicator.240304.2403.html[ Accessed 12 12 2014]. OIV, 2012. Guide d'application du système HACCP (Hazard Analysis and Critical Control Points) à la viticulture de l'OIV, Switzerland. RTS, 2015. Affaire Giroud, du vin en eaux troubles. [Online] Available at: http://www.rts.ch/emissions/temps-present/justice-criminalite/6368255-affaire-giroud- du-vin-en-eaux-troubles.html[Accessed 01 2015]. SCA, 2009. Surfaces de compensation écologique (SCE). SEREC, 2001. Sauvegarde des murs en pierres sèches et du vignoble en terrasses valaisan. SOFIES, 2009. Performance environnementale des produits " vin rouge" et "tomate" à Genève. Spring, J.-L. & Dupraz, P., 2011. Caractéristiques des principaux cépages cultivés en Suisse. Revue Suisse de Viticulture, Arboriculture et Horticulture, pp. 75-77. Svensson, A., Nilsson, C. & Ohlsson, M., 2004. Key factors to long-term relationships between wine suppliers and restaurants. The 9000 store, 2015. The 9000 store. [En ligne] Available at: http://the9000store.com/ISO-9000-Tips-How-Much-Does-it-cost.aspx[Accès le 12 02 2015]. Thomas, P., 2014 a. thomasvino. [Online] Available at: http://www.thomasvino.ch/?p=12385[Accessed 30 01 2015]. Thomas, P., 2014 b. L'Hebdo. [Online] Available at: http://www.hebdo.ch/hebdo/actuels/detail/vins- suisses-bouteilles-etiquettes-affaire-giroud-fisc-aoc-viticulteurs-arisants-industriels[Accessed 12 10 2014]. Univerre, 2015. Univerre. [Online] Available at: http://www.univerre-prouva.ch/home [Accessed 12 11 2014]. van der Zanden, G.-J., 2009. The truth about CO2 emissions in the wine industry. van Schoor, L. & Groenewald , M., 2002. Wineland. [Online] Available at: http://www.wineland.co.za/technical/food-safety-in-the-wine-industry-advantages-of-the-haccp- system[Accessed 12 02 2015]. Varela, M. . C., 2004. Liège, une valeur ajoutée pour les vins ?. Revue des OEnologues, Issue 112. Vitival, 2015. Vitival. [Online] Available at: http://www.vitival.ch/fr/presentation/objectifs-37-0[Accessed 16 01 2015]. WFSC, 2014. ETH Zurich. [Online] Available at: http://www.worldfoodsystem.ethz.ch/content/dam/ethz/special- interest/dual/worldfoodsystemcenter- dam/documents/wfsc%20annual%20report%20web.pdf[Accessed 12 05 2014].

www.glamur.eu 60

9 Annexes 9.1 Context of the red local wine value chain

Figure 16 Swiss map where production sites are marked.

Figure 17 Swiss wine production for red, white varieties (OFAG, 2013).

www.glamur.eu 61

Figure 18 Swiss wine production for red and white varieties (OFAG, 2014).

Figure 19 Imports and exports data in 2012 concerning quantity (kg) and value (CHF) (AGIR, 2013)

www.glamur.eu 62

Figure 20 Swiss wine consumption of autochthon and imported white and red wine (OFAG, 2013).

Figure 21 Structure of the wine value chain in Valais (Catry, 2009).

www.glamur.eu 63

9.2 Additional References

Table 6 Additional references for indicators. Attributes Indicators or Source Documents/URL descriptors Biodiversity Species Canton Vaud association http://www.prometerre.ch/dossiers Conservation promoting earth’s professions /projets_qualite_du_paysage__qp practices (Prométerre) _ Cultivars diversity Official site of the Valais canton http://www.vs.ch/NavigData/DS_6 8/M5929/fr/Ronde_des_c%C3%A 9pages_avril_2008.pdf Creation and distribution Gross income Accounting results of Swiss of added value farms by Agroscope Contribution to Business’ results according to employment earned incomes: all regions mixed 2011/13 Food safety Food safety Will and Guenter (2007) Food www2.gtz.de/dokumente/bib/07- standards and safety standards and practices 0800.pdf practices applied applied Artificial additive Valais’ ordinance 916.142 (2004) and Laesslé(2014 ) Construction locale des valeurs du vin : une analyse en termes de ressource. Governance Fraud management IVV (2014), Viti horizon 2020. http://www.romanduvin.ch/wp- within the chain RTS (2015). Affaire Giroud, du content/uploads/2014/11/Rapporti vin en eaux troubles ntermediaireVitihorizon2020_2611 2014.pdf Information and Product Labelling COMMISSION REGULATION http://eur-lex.europa.eu/legal- communication (EC) No 607/2009 content/EN/TXT/PDF/?uri=CELEX :32009R0607&from=EN

Pollution GHG emission for Geert-Jan van der Zanden http://www.txb- transportation (2009), The truth about CO2 finewines.com/upload/The_truth_ emissions in the wine industry about_CO2_emissions_in_the_wi ne_industry_April_2009_223.pdf GHG emissions for IFV (n.d.),Emissions of http://www.vignevin.com/fileadmin production greenhouse gases accounting: /users/ifv/publications/A_telecharg Application of the carbon er/Itin24_BilanCarbone.pdf balance method to the wine industry Territoriality Association of Valais’ ordinance 916.142 http://www.agrivalais.ch/Upload/R product with the (2004) and apportfinalSEREC.pdf territory SEREC(2001),Improvement Of Agricultural Structures And Landscape In Wine Perimeters

www.glamur.eu 64

9.3 Data quality check Table 7 Data quality check (Lewandowska, 2004).

Attributes Indicators Quality class Biodiversity Species Conservation practices A Cultivars diversity A Creation and Gross income A distribution of added value Distribution of added value across the chain A Contribution to employment A Food safety Food safety standards and practices applied A Artificial additive A Governance Decision making mechanisms A Fraud management within the value chain A Market management A Information and Availability of information A communication Product labelling A Pollution GHG emissions for transportation A GHG emissions for production A Environmental pollution (water, soil, air) A

Resource use Water use practices A Material use practices A Territoriality Association of the product with the territory A

Social cohesion and Conviviality A

www.glamur.eu 65