Commodity chains, quality conventions and the transformation of agro-ecosystems: olive groves and olive oil production in two Andalusian case studies

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ “This is an Accepted Manuscript of an article published by Sage in European Urban And Regional Studies on 2012, available at: https://doi.org/10.1177/0969776411428560” 1 Commodity chains, quality conventions and the transformation of agro-ecosystems: olive groves and olive oil production in two Andalusian case studies Daniel Coq-Huelva, Manuel David García Brenes and Assumpta Sabuco-i-Cantó European Urban and Regional Studies 2012 19 (1) 77-91. DOI: 10.1177/0969776411428560 Abstract The olive oil commodity chain (OOCC) is very complex and has several kinds of agents with different positions and roles. Four main phases -agricultural production, olive milling, oil refining and commercial distributioncan be identified. The positions and drivenness levels associated with them are very heterogeneous. Given the way that the OOCC is currently organised, refinery companies and large retailers are the agents with the highest drivenness levels. At the same time, there are persistent disputes among the different agents in the various phases, between not only refineries and large retailers but also milling cooperatives. In this context, refineries and cooperatives have made several attempts at industrial upgrading. Quality conventions are basic elements in the daily working of the OOCC. Different quality conventions are promoted by various kinds of agents. This article analyses not only the different phases of the OOCC, but also its associated conventions. The first two phases can only be analysed in specific case studies. For this reason, localised olive oil production systems, based on different quality conventions, will be analysed in two case studies: Beas (Huelva) and Arjona - Porcuna (Jaen). Each of these towns has very different agro-ecosystems and uses different agricultural handling systems and olive milling-techniques. Keywords: Commodity chains, convention theory, olive oil, quality, southern Spain 2 Introduction Olives are the main agricultural and agro-industrial focus of the Andalusian economy. In recent years, Spain has produced about 30 percent of the world’s olives, of which over 80 percent are produced in Andalusia. Currently, Andalusia represents about 25 percent of world olive production. This article has three main objectives: firstly, to describe how the olive oil commodity chain (OOCC) operates in Spain, outlining the positions of the different participating agents as well as highlighting existing conflicts and how governance has evolved recently; secondly, to analyse the different industrial upgrading initiatives, emphasizing the strategies carried out by the cooperative millings, as well as the role of conventions; thirdly, to analyse existing relationships among these industrial upgrading initiatives and other perceptible transformations at local levels, such as expansion or contraction of crop area, changes in agrarian handling systems, modifications in farm design and evolution of agro-ecosystems (olive varieties, planting frameworks). After examining the OOCC’s overall evolution, two case-studies will be analysed: Beas (Huelva) in the West of the region and Arjona -Porcuna (Jaen) in the centre. There are vast differences between these towns. The aim of this text is to analyse these differences and to identify different territorial patterns associated with the way that production is organised. Moreover, these local differences cannot be understood without considering how the OOCC functions overall. 3 The article is organized as follows. First, we describe the theoretical basis of the research conducted, and then we briefly present the research methodology employed. In the third section we analyse the OOCC. Subsequently, my analysis will focus on how the first two stages (farming and olive milling) are carried out in the case studies. After discussing the way in agricultural production is organised, we compare the different olive “first manufacturing” processes. Theoretical base It is necessary to focus on the vertical coordination mechanisms existing between agriculture, the agro-food manufacturing industry and distribution in order to understand the dynamics linked to olive oil production and distribution. For this purpose, the concept of the global commodity chain (GCC) will be used. There is a vast academic literature about GCC 1 , which was first developed in the 1970s. Over the years, the literature has analysed different topics over different periods (Bair, 2005, 2008). GCC analysis has several theoretical influences: transaction cost economics, network theories, old institutional economics, and French regulation school, among others (Raikes et al, 2000; Gereffi et al, 2005). 2 In the case of agri-food commodity chains (AFCC), there is also a strong connection with political economy of agriculture (Friendland, 1984; Bonnano et al., 1994). Four elements are essential for the operation of the GCC: input-output structure, territoriality, governance structure and institutional context (Gereffi, 1999): The element with the highest level of theoretical development has been governance structure (Humphrey and Schmitz, 2001). Gereffi (1994) initially distinguished two main types: producer-driven chains (PDCs) and buyer-driven Chains (BDCs). 3 4 “Producer-driven commodity chains refer to those industries in which transnational corporations (TNC) or other large integrated industrial enterprises play the central role in controlling the production system (including its backward and forward linkages). This is most characteristic of capital- and technology-intensive industries…Buyerdriven commodity chains refer to those industries in which large retailers, brand-named merchandisers and trading companies play the pivotal role in setting up decentralized production networks in a variety of exporting countries” (Gereffi, 1994: 97). The difference between PDCs and BDCs has been widely contested; nevertheless, it is useful to explain some important governance traits of the GCC (Raikes et al, 2000; Gellert; 2003). Some of the elements highlighted include the possible existence of several driver agents, the relevance of agency capacities, the changing nature of governance structures and the diversity of elements that can explain “drivenness” within the GCC (Raikes el al, 2000). On the other hand, there are many types of commodity chains. Some are the result of the evolution of “old” productive processes. 4 Others have been developed more recently, as a consequence of subcontracting, outsourcing and vertical disintegration processes that are associated with post-Fordism (Gereffi and Kornoviech, 1994) 5 . There are also manufacturing, agricultural and raw material commodity chains. Although GCC literature has been traditionally centred on analysing industrial commodity chains (Raikes et al, 2000), there are an increasing number of analyses of the AFCC (Dolan and Humphey, 2000, 2004; Gibbon 2001b), especially in the case of some agricultural products (cocoa, coffee) produced and exported by countries in post-colonial contexts (Fold, 2001; Gibbon, 2001a). One of the AFCC’s main characteristics is the rootedness 5 of the crops in specific ecological niches. This fact explains the relevance of environmental issues, which have been relatively neglected by the GCC literature (Talbot, 2008). Industrial upgrading in a GCC must be understood as a process of organisational learning, which allows new competences to be developed and, simultaneously, enables the upgrading firm to increase its income (Gereffi, 1999; Fleury and Fleury, 2001). There are different kinds of competences (technological, logistical, managerial, marketing, etc.) that can be developed in industrial upgrading processes (Fleury and Fleury, 2001). Competence-building success depends, among other elements, on the GCC’s own internal operation (Gereffi, 1999; Dolan and Tewary, 2001), but firms’ strategies are also essential (Dolan and Tewary, 2001; Fleury and Fleury, 2001). In any case, a large proportion of industrial upgrading cases stem from outsourcing and externalisation processes, which are especially frequent in BDCs (Gereffi, 1999; Gibbon, 2001a). Quality is a central argument in AFCC analysis and can have a strong influence on its overall governance and/or on generating and exploiting specific niche markets (Morris and Young, 2004). Quality can have different meanings and can be supported by a wide (and even competing) range of criteria (Watts and Goodman, 1997). Therefore, it must be understood as a social construct, which is influenced by economic and socio-cultural elements (Marsden and Arce, 1995; Ilbery et al, 1998). Convention theory permits a better understanding of the interaction between the prevailing cultural norms and the behaviour of commodity chains’ agents (Murdoch and 6 Miele, 2004). GCC organisation is partially based on conventions understood as “tacit agreements”, which reduce the uncertainty associated with the organisation of productive processes (Storper and Salais, 1997). Convention theory permits the conceptualization of the relationships among different GCC stages, both as collaborative and as conflicting processes, in which conventions are continuously accepted, contested and traded (Wilkinson, 1997). Specifically, the role of quality conventions will be emphasised. Eymard-Duvernay’s work (1989) framed quality conventions within a typology based on Boltanski and Thévenot’s “èconomies de la grandeur (1991)”. • Market quality conventions. The common principle can be defined as product competitiveness. Product definition is characterised by a low informational complexity, price being the only standard reflecting its quality. • Industrial quality conventions. The common principle can be defined as efficiency measured with physical or technical standards. Economic transactions would be subordinated to maintaining efficiency. There is a relatively high informational complexity. However, it can be standardised through the use of scientific and technical criteria. • Domestic quality conventions. The common principle can be defined as maintaining tradition. Evaluation is based on how people are connected to a common root or source. In these cases, the continuous relationship between people and tradition ensures product quality. 7 Each of these conventions is based on a “repertoire of justification” understood as instituted and historically constructed ways of evaluating actions taken in the name of a common or collective good. Three main “repertoires of justification” will be highlighted (Murdoch and Miele, 2004): • market performance based on profitability or economic value in a competitive market. • industrial efficiency based on long-term planning, physical production growth and investment. • domestic worth based on local embeddedness. The simultaneous coexistence of different quality conventions (and therefore, different “repertoires of justification”) is frequent. This fact is justified by the complexity of business relationships. In addition, conventions often change because of transformations in the GCC structure, in agents’ behaviour or because they are agreements that are frequently contested (Storper and Salais, 1997; Ponte and Gibbon, 2005). Therefore, quality conventions are influenced not only by a distant and alien power, but also by several micro-powers (Allen, 1994; Foucault, 2004). Agro-ecosystems’ characteristics may change as a result of the prevalence of certain types of conventions, which, in turn, are in constant interaction with the commodity chain dynamics. Methodology The analysis is two parts. The first contains a synthetic exposition of the OOCC’s main characteristics. Qualitative information and statistical data are used to reconstruct the olive oil value chain and to synthesize the main relationships between its different 8 phases. Various statistical sources have been used. The Food and Agriculture Organization´s statistics production and trade (FAOSTAT) have been one of the most important sources. Some specific Spanish statistics have also played a very relevant role 6 . Private statistical sources, which provide information about firms and other private agents, were also used 7 . The second part is an analysis of two local production systems. A qualitative research methodology, based on semi-structured interviews, was used. A total of 60 interviews (related to the topics discussed here) were conducted among men and women (in the latter case, 25 interviews). Although increasingly displaced from the workforce and decision-making, women have traditionally had a very important role in the harvest, mainly as temporary workers (“jornaleras”). Women from different age groups have been interviewed because age is a fundamental aspect of the women’s position in the labour organization of agriculture. Large, medium and small farm owners were also interviewed (29 interviews) 8 . They were exclusively male. 9 Small farm owners are part-time farmers (with other businesses in building or in the service industry) or old retired farmers who have a small olive grove as a supplementary source of income. Almost all farmers are members of local cooperatives. Additionally, five cooperative“managers” were interviewed (they were also farmers). The interviewers asked about the interviewees’ relevant experience with olive groves. They had a “life history” structure, but without the depth and range of a true “life history”. The interviewees each answered, in their own words, some specific questions 15 important element of ownership and management structure. At the same time, however, they have strong connections with transnational capital. The first foreign company with a significant market share was Lesieur Group (France), in the 1980s. During the 1990s, international groups (especially Ferruzzi and Unilever) had considerable participation (about 50 percent of market share) in the refinery industry. However, various groups sold their businesses at the beginning of the 2000s (Rosell and Viladomiu, 2006). At present, some of the main Spanish refinery companies (see Table 2) are undergoing an accelerated internationalisation process. The most relevant case is the SOS Cuétara group, which has developed a policy of purchasing olive oil companies in recent years. This process has been focused especially on Italy, where SOS Cuétara has bought three leading firms (Bertolli, Carapelli and Minerva) with an important market share in some of the most important consumer countries (not only Italy, but also other countries, like the United States). Table 2 The relationships between cooperatives and refineries are market relations based on market conventions. They, therefore, have a market performance repertoire of justification. The easy codification of olive oil characteristics allows a continual turnover of suppliers, in order to find the best price and highest profitability in the short term. The industrial specifications required for homogenisation are minimal. This facilitates strengthening market conventions. 16 Large retailing. The fourth actorlarge retailing (LR)- has high negotiation power within the commodity chain. Large corporate retailers accounted for 87 percent of olive oil sales in the Spanish domestic market between 2000 and 2005 (Ministerio de Agricultura, Pesca y Alimentación, 2006). These companies have a high negotiation capacity and the possibility to influence how the rest of the agents behave. Spanish consumers are very sensitive to olive oil prices, because it is a basic element of their diet. Usually, large retailers use low olive oil prices to attract consumers to their shops, for example by promoting their own cheap brands. Almost half of the olive oil sold in Spain in 2005 was sold under a distributor brand (Alimarket, 2006). This strategy creates strong pressure for price reduction. Large retailers have stable relationships with large refineries, mainly for logistic reasons. However, the pressure is sometimes so high that large retailers have also developed relationships with cooperatives. Despite their logistic deficiencies, these cooperatives can be an important ally in negotiations with large refinery companies. Half of large retailers are controlled by Spanish capital and the other half by international capital (see Table 3). In this sense, the retail OOCC structure shows a strong global capital presence. Table 3 There are no major differences between the conventions applied by industry and retailers. Market conventions with some industrial specifications (in extra virgin olive oil) drive exchanges. 17 The LR’s increasingly important role has meant a change in the OOCC’s governance structure. Prior to the develoment of LR, the OOCC’s governance structure fitted PDC model, where refinery companies have higher drivenness and negotiation power. LR has pushed for a progressive transition to a mixed model, where refinery companies and large retailers both have a high and shared drivenness. Although some elements, such as the relevance of distributor brands or the increasing LR relations with some specific cooperatives, are characteristic of BDCs, the OOCC still does not have a BDC governance structure. In this sense, the OOCC has a bipolar governance structure (Fold, 2002), which is very common in AFCC (Talbot, 2008). There has even been a sharp increase in packaged exports in recent years, but exports of packaged olive oil accounted for only about 30 percent of total exports in 2006 19 . This kind of high-value exports is led by refinery companies, although, paradoxically, an important proportion of the exported oil is not refined olive oil but virgin olive oil. Refineries have developed strong competences in marketing and commercialisation. This process is, on the one hand, a sort of industrial upgrading for refinery companies and, on the other hand, an attempt to compensate for the increasing negotiation power of large retailers in Spain. The conflicts within the commodity chain are not only between refinery companies and large retailers. Mills (or, at least, some mills) are also trying to undertake specific industrial upgrading processes, following two different approaches: 1. by developing their own export channels, 18 2. by developing short marketing channels addressed to specific niches in the Spanish domestic market. These two strategies are present in the case studies analysed below. Cooperative mill’s industrial upgrading also implicates farmers and, in some sense, the territories in which farmers develop their activities. Nevertheless, quality strategies, used to promote specific market niches in olive oil are not simple. Olive oil is perceived as a quality product in comparison to other vegetable oils (sunflower, corn, palm etc). From a marketing point of view, insistence on the concept of quality means differentiating a product that has already been differentiated. In addition, quality can be socially constructed following different approaches. Changes in agricultural organisation and the direct social implications One of the main characteristics of farming activities is the significant local adaptations they undergo. These local particularities can affect some relevant productive elements. For this reason, significant social construction processes are frequently developed in agricultural activities (Arce and Marsden, 1993). The two case studies have been chosen to demonstrate some of the agro-ecological and economic complexities of the crop in Southern Spain 20 . From an agro-ecological perspective, certain parameters are especially relevant: • the age of the olive groves and planting density; • the existence, or not, of autochthonous varieties; • the type of landscape (mountainous, plains, etc.) 21 19 Some economics parameters are included in the analysis: • yields; • the amount of machinery used; • the possibilities of crop transformation; • the industrial upgrading strategies implemented by cooperative mills. It is also to mention the relevance of environmental issues, and their analysis, with regard to AFCC (Talbot, 2008). The olive groves of Jaen’s campiñas (Arjona and Porcuna) Arjona and Porcuna are two municipal districts with a similar population (in 2008 Arjona had 5826 inhabitants and Porcuna 6939). Agriculture is the most important economic activity in both municipal districts, it generated around 35 percent of employment in 2001 in both cases. These two municipal districts also have an important service sector, owing to state funding of public services (health, education) and the development of an emerging tourist sector. In Arjona, there is also a relevant furniture industry. The rate of unemployment is relatively high in both cases - 16.5 percent in Arjona and 22.4 percent in Porcuna in 2001- especially among women. Today, olive groves are a monoculture in these territories. Since the 1950s, they have gradually increased their share in the agrarian structure. Olive groves represented 81.0 porcent of Arjona’s overall surface area in 1995 and 89.5 percent in 2008. This growth 20 was even more intense in Porcuna, increasing from 68.5 percent to 87.5 percent. Before the 1950s, olive groves were interspersed with other crops, generally cereals and legumes, but their share has been progressively reduced. The level of genetic diversity is low. Only one type of olive variety is used in this area, Picual, because of its productivity, its fat yield and the temporal stability of production. This olive groves are generally non-irrigated. Irrigated olive groves as a proportion of the total olive grove area was only 6.2 percent in 2008. However, non-irrigated olive groves have relatively high levels of land productivity (3,500 - 5,000 kg per hectare). Land productivity has traditionally been one of the bases of the crop’s profitability. The other base is control over costs. Since the mid-1980s, more mechanization has been introduced, which has implied a strong reduction in labour costs (although the most important costs are still harvesting salaries). However, even when machinery is used, there are different styles of farming and these have different costs associated with them. For example, olives can be ground and tree harvested, or only tree harvested. Ground-harvesting uses blowers and vacuum cleaners. Treeharvesting consists of a system of shakers and nets, which are placed under the three. In the first case, the harvesting costs are lower and the fat yields are higher, but the olive oil that is obtained has higher levels of acidity and worse organoleptic properties. The prevalence of one specific style of farming basically depends on the ability to recover higher costs using higher prices that, in turn, are linked to the “industrial upgrading” strategies developed by the cooperative mills and the quality conventions with which they are associated. As will be discussed below, cooperatives from this area have failed in their attempt at industrial upgrading. They cannot valorise 21 improvements to olive oil quality, which are not demanded either by refinery companies, or by large retailers. Therefore, ground-harvested and tree-harvested olives are valued in the same way, although they do not possess the same industrial quality. As a consequence of this, both kinds of olives are stacked in undifferentiated containers. Tree-harvested olives have a higher quality, but, in reality, it does not make a difference because they cannot be valorised. " What happens is that … they don’t pay for quality. The money does not reach farmers, we have to work more, and I agree that the very moment that an olive falls from the tree, it loses quality ... but, in any case, we don’t see any of the profits in the end, the money goes straight to traders. … the money disappears ... " (Interview with the owner of a large farm, 45 years old). This conflict can also be observed in terms of conventions. Farmers and cooperative mills are interested in developing industrial quality conventions based on industrial efficiency repertoires of justification, which institute clear and exigent benchmarks for olive oil quality in terms of objective indicators, and even in terms of professional tasting. These conventions are tied to farmer’s dominant “habitus” (Bourdieu, 1998) based on productivism. However, the success of these industrial conventions depends on the success of the industrial upgrading initiatives carried out by the cooperative mills. Because these attempts have failed, stricter market conventions with lower informative complexity and no long-term investment commitment are imposed. Quality and price dynamics have a direct effect on agro-ecosystem evolution. The productivist logic presses towards a farm restructuring based on a new round of 22 production intensification, which means reducing planting frameworks, increasing water demand, using more fertilisers, and so on. Modernised olive grove: Beas Beas is a municipal district with 4320 inhabitants located near Huelva, and it is well connected to this city. Despite being relatively close to an urban area, Beas is still strongly associated with agriculture, with 25 percent of its employment being generated by this sector. However, there are other important economic sectors, such as construction (21 percent of employment). The agricultural structure of Beas has dramatically changed in recent years. Olive groves have always existed in Beas, but they have expanded greatly, from 28.9 percent of the agricultural surface area in 1995 to 46.6 percent in 2008. Nevertheless, Beas is not a monoculture area. Other crops are also important, such as sunflowers (20.0 percent of the agricultural land) or wheat (19.5 percent). Most of the olive groves have been planted in recent years. These new plantations have been designed to ensure high yields. There has been an important expansion of irrigation, which requires elevated investments that can be provided only by wealthy landowners. Yields are 5,000-6,000 kg per hectare on irrigated land. Farm re-design affects not only irrigation but also other elements, specifically the replacement of olive varieties. The local olive variety -Verdial de Huevarhas often (but not always) been replaced by other varieties, such as Picual (typical of Jaen, but with little presence in this area until the 1990s) and Arbequina (which comes from Catalonia) 22 . 23 As I shall discuss below, the industrial upgrading strategy of Beas’s cooperative has been successful. This has implied significant changes in the dominant quality conventions. At present, the prevailing conventions are based on the reduction of acidity levels and impurities and on the improvement of other organoleptic characteristics. Therefore, an industrial quality convention is clearly being imposed. Paradoxically, however, olive oil is especially valued by customers because it is from Beas. Therefore, in some sense, a domestic quality convention is also operating. Nevertheless, there is no attempt to openly exploit a domestic convention of quality through the creation of a Protected Designation of Origin for example. Agricultural handling systems have been profoundly modified. Nowadays, almost all the olives are tree harvested. This implies the need to anticipate the harvest in order to reduce the number of olives that fall to the ground. It also implies the use of machinery (shakers and nets). The practice of collecting olives from the ground (using blowers) has fallen dramatically over the last few years, because there is a significant price difference between tree- and ground-harvested olives, owing to the local cooperative’s pricing policy. The role of farming cooperatives in the different areas This section analyses the different industrial upgrading strategies used by the cooperatives located in our case-study areas. 24 Cooperatives in Jaen’s campiñas This area has high production that cannot be absorbed by developing short market channels. This explains the strategy of local cooperatives based on the development of commercial structures that can sell large quantities of olive oil in foreign markets. “There are 350 mills in the province of Jaen, but it is impossible to put 350 brands on the market, with 350 dealers”. (Interview with a large farmer and agronomist, 45 years old) Two cooperatives in Porcuna and one in Arjona joined in a second-grade cooperative (Fedeoliva) dedicated to marketing olive oil that was launched in 1989 and brought together 15 first-degree cooperatives. Fedeoliva conducted an aggressive marketing campaign in the United States, aiming to position its brand (named “Tosca”) in the US market. There was no domestic element in its quality approach. In fact, its brand has a certain “Italian” sound that sought to link olive oil with the Italian cultural connection that prevails among US consumers. In 2003, Fedeoliva was the fifth company in terms of sales in the United States and the first in the states of New York and New Jersey. However, management irregularities and narrow trading margins provoked the company’s suspension of payments. Fedeoliva had a strategy based on marketing high-quality olive oil. Quality was understood in an industrial sense, based on the control of various technical parameters (acidity, level of impurities, etc.). These industrial conventions were transferred to 31 Gellert, P. (2003) Renegotiating a timber commodity chain: Lessons from Indonesia on the political construction of commodity chains. Sociological Forum 18 (1), 53-84. Gereffi, G. (1994) The organisation of buyer-driven commodity chains: How U.S. retailers shape overseas production networks. In Gereffi, G and Korzeniewicz, M.(ed) Commodity Chains and Global Capitalism. London: Praeger. Gereffi, G. (1999) International trade and industrial upgrading in the apparel of commodity chains Journal of International Economics 48(1): 37-70. Gereffi, G and Korzeniewicz, M.(ed) (1994) Commodity Chains and Global Capitalism. London: Prager. Gereffi, G., Humphrey, J. And Sturgeon T. (2005) The governance of global value chains Review of International Political Economy 12 (1): 78-104. Gibbon, P. (2001a) Upgrading primary products: A global value chain approach World Development, 29 (2): 345-363. Gibbon P. (2001b) Agro-commodity chains: An introduction IDS Bulletin 32(3): 60-68. Guzmán Álvarez, J. R. (2004): Geografía de los Paisajes del Olivar Andaluz. Seville. Consejería de Agricultura y Pesca. Junta de Andalucía. Humphrey J. and Schmitz H. (2001) Governance in Global Value Chains IDS Bulletin 32(3) 19-29. Ilbery, B and Kneafesey, M. (1998) Product and place: Promoting quality products and services in the lagging regions of the EU. European Urban and Regional Studies, 5:. 329-341. Leon, L., De la Rosa, R., Rallo L., Guerrero N. and Barranco D. (2007) Influence of spacing on the initial production of hedgerow Arbequina olive orchards Spanish Journal of Agricultural Research, 5 (4): 554-558. 32 Marsden, T and Arce A. (1995) Constructing quality: Emerging food networks in regional transition. Environmental and Planning A, 27: 1261-1279 Ministerio de Agricultura, Pesca y alimentación (2006): La Alimentación en España. Madrid. Ministerio de Agricultura, Pesca y Alimentación (2007) Anuario de Estadísticas Agroalimentarias 2006, Madrid, URL (accessed 25 October 2011): http:/ www.marm.es/estadística/pags/anuario/2006/AE_2006_Completo.pdf. Morris, C. and Young, C. (2004) New geographies of agro-food chains. An analysis of UK quality assurance schemes In Hughes, A. And Reimer, S. (edts) Geographies of commodity chains. London: Routledge Murdoch, J. and Miele, M. (2004) Culinary networks and cultural connections. A conventions perspective In Hughes, A. And Reimer, S. (edts) Geographies of commodity chains. London: Routledge Pastor J. (2009): La ausencia de relevo generacional eleva la edad media de los olivareros a 55 años. Ideal. 13 September. Ponte, S. and Gibbon (2005) Quality standards, conventions and the governance of global value chains Economy and Society, 34 (1): 1-31 Raikes, P: Friis Jensen M. and Ponte S. (2000) Global commodidy chains analysis and the French filière approach: comparison and critique” Economy and Society, 29(3): 390- 417. Rosell, J. and Viladomiu. L. (2006) Aceite de oliva. Avance y repliegue del capital extranjero In: Etxezarreta, M. (Coord) La Agricultura Española en la Era de la Globalización. Madrid: Ministerio de Agricultura, Pesca y Alimentación. Sánchez Martínez JD, Gallego Simón VJ, and Araque Jiménez E (2008). El monocultivo olivarero jiennense: ¿del productivismo a la sostenibilidad? Boletín de la Asociación de Geógrafos Españoles 47:245-270. 33 Storper, M. and Salais R. (1997) Worlds of production: The Action Frameworks of the Economy. Cambridge MA and London: Harvard University Press. Talbot, J.M (2008) The comparative advantages of tropical commodity chain analysis. In: Bair J. (edt) Frontiers of Commodity Chains Research. Stanford CA: Stanford University Press Watts, M. and Goodman, D. (1997) Global appetite, local metabolism: Nature, culture and industry in fin-de-siècle agro-food systems In: Goodman D. and Watts M. (edts) Globalising Food: Agrarian questions and global restructuring. London: Routledge. Wilkinson (1997) A new paradigm for economic analysis? Economy and Society, 26: 305-339. 34 Figure 1. Olive oil commodity chain Table 1. Market share of different kinds of oils in Spain Market share (percent) Virgin and extra virgin olive oil 23.1 Refined olive oil 62.9 Pomace olive oil 1.3 Sunflower oil 11.7 Seed oil 0.8 Corn oil 0.1 Soy oil 0.1 Source: Ministerio of Agricultura, Pesca y Alimentación (2007). Farming: Olives Groves First transformation : Mills Retailers Controlled by farmers’ - coperative and individual capitalists. Territorial dispersion Edible vi rgin olive oil Wastes: Pomace (“Orujo” and “alpechin) Non edible virgi n olive oil Packaging: D i fferent oils Refin in g Olive oilil l Extracting - pomace Milling crude pomace oil Mixing virgin and refined olive oil Mixing virgin and r efined pomace olive oil SUBCHAIN 3. POMACE OLIVE OIL SUBCHAIN 1. REFINED OLIVE OIL wastes: “ Orujillo ” and pits. SUBCHAIN 2. EXTRA VIRGIN OLIVE OIL 35 Table 2. Main milling and packaging companies in the olive oil sector in Spain Market share Capital origin SOS-Cuétara 29.3 Spanish Sovena, S.A. 13.9 Nutrivente (Portuguese) Acesur 9.1 Spanish Ybarra-Migasa 8.2 Spanish Hojiblanca 5.6 Spanish Oleo Martos, S.A. 4.0 Spanish Urzante, S.L 3.5 Spanish Aceites Toledo, S.A 3.2 Spanish Borges Food, S.L 2.8 Spanish Ideal, S.L. 2.5 Spanish Source: Alimarket, 2006 Table 3. Main food distribution companies in Spain. 2006 Market share (percent) Capital origin Carrefour - DIA 16.3 International capital (French origin) Mercadona 16.1 Spanish) Eroski Group 7.7 Spanish Alcampo (Auchan group) 4.6 Internacional capital (French origin) Source: Distribución Actualidad, No.379,2007 p. 22 Table 4. Summary of the two local agro-food systems Jaen’ s Campiñas Beas Dominant conventions Market conventions (with some industrial elements) Industrial conventions (with some domestic elements) Conventions promoted by farmers Industrial conventions Industrial conventions (with some domestic elements) Type of value chain promoted by farmers Long value chain. Second degree cooperative Short value chain. Direct commercialisation Success of quality initiatives of farmers No Yes 36 Notes 1 This also includes literature on global value chains, which is a concept closely associated to global commodity chains (Bair, 2005) 2 It is impossible to provide a comprehensive review of the academic literature on the GCC and its theoretical influences given its vastness. Some reviews have been undertaken by Blair (2005) and Raikes et al (2000) 3 Gereffi et al (2005) have developed a wider classification of governance structures, differentiating five types of governance: markets, modular value chains, relational value chains, captive value chains and hierarchy. However, for the purposes of this article, the dichotomy between PDC and BDC works better, because many of the relationships between agents in the OOCC do not fit exactly into any of the categories proposed by Gereffi et al (2005) 4 Wheat, cotton or shipbuilding were analysed in Gereffi and Kornowietsky (1994) as commodity chains that have a very long history, sometimes coming from more than 200 years ago. 5 Textiles, electronics, informatics etc. are frequently referred to. 6 Ministerio de Agricultura, Pesca y Alimentación, Anuario de Estadisticas Agroalimentarias 2006. (Yearbook of Agro-Food Statistics). Instituto Nacional de Estadistica, Encuesta de Estructuras Agrarias (Survey of Agrarian Structures. http://www.ine,es/inebmenu/mnu_agricultura.htm; and Instituto Nacional de Estadistica, Censo Agrario 1999 y 2009 (Agrarian Census). http;/www.ine,es/inebmenu/mnu_agricultura.htm. 7 The yearbook Alimarket (Informe Annual, 1996 and 2006, AECOC) and the review Distribución Actualidad, No 379, 2007. 8 “Large” means owners with more than 50 hectares on one or more areas of land; “ nedium” means owners with 15-50 hectares on one or more areas of land; “ small” means owners with less than 15 hectares on one or more areas of land. 9 . There is a strong male dominance. Sometimes womwn own land, but usually their husbands organize the crop and women participate only in particular events, such as some specific cooperative meetings. There are also some women who manage their own land activities, but this is still very rare. 10 Ministerio de Agricultura, Pesca y Alimentación, Anuario de Estadística Agraria, Madrid: Secretaria General Técnica. Centro de Publicaciones, 1992; Ministerio de Agricultura, Pesca y Alimentación, Anuario de Estadísticas Agroalimentarias, Madrid, 2004. 11 Data obtained from FAOSTAT 12 Data obtained from FAOSTAT 13 Data obtained from FAOSTAT 14 Ministerio de Medio Ambiente y Medio Rural y Marino, Estudio de la cadena de valor y la formación de precios del aceite de oliva.2010, p.16. 15 Instituto Nacional de Estadística. Encuesta sobre estructura de las explotaciones agrarias, 2005. 16 This estimation has been made in Junta de Andalucia, “ El sector del aceite de oliva y la aceituna de mesa en Andalucia. Seville, 2008, pp. 24-9. 17 Andalusian olive production is obtained from Ministerio de Agricultura, Pesca y Alimentación (2007). Details of the subsidies associated with the CAP can be found on the website of Fondo Español de Garantia Agraria (La PAC y los Fondos Europeos Agricolas, http:// www.fega.es). 18 The most important is the “ Plan de Empleo Rural” ( Rural Employment Plan), which assures a minimum income to rural areas. 19 Database ESTACO, Instituto de Comercio Exterior, Ministerio de Industria, Turismo y Comercio. 20 The two case studies do not cover the entire economic and ecologic diversity of olive groves speciafically in southern Spain, but they are relevant examples of this diversity. 21 Guzmán Álvarez (2004) classifies the Andalusian olive groves into nive landscape categories. Three of the most important categoriesolive groves in campiñas (plains), olive groves on mountains (on a slope of 15-40 percent) and olive groves on smooth slopes (less than 15 percent-are considered in this article. 22 The variety Verdial de Huevar still has a significant role in this area.