Green or roasted coffee? How a collective of organic producers challenges the quality construction by overseeing quality attributes, relational approaches and knowledge

ABSTRACT This paper studies the conditions under which quality conventions in coffee can be revised. It shows how value is captured by a collective composed by Colombian producers overseeing coffee quality attributes, engaging in relational approaches and building knowledge with international buyers. Interviews were conducted with selected organic producers in Santa Marta, Colombia and buyers (e.g. in Germany, the Netherlands). Due to the sound relations between the staff of the producer organization and a German buyer, the manager of the collective proposed to also export roasted coffee in addition to green coffee. Although roasting at origin is not common in coffee exports, in the case analyzed existing hesitations were abandoned. Eye-level relationships, trust based on a relational approach and knowledge were key to achieving a collective quality construction and subsequent value capture by producers. Over time, these producers have improved their material quality techniques and become more familiar with the places where their coffee circulates and is valued, allowing them to influence the commercial spheres of the coffee business. The discussion is not about changing quality conventions, but on how quality conventions are institutionalized and the narratives behind them.


Introduction
The analysis of governance and diverse power struggles in global value chains (Grabs & Ponte, 2019) is often connected to value capture foreseen in economic expectations based on profit maximization by lead firms (e.g.Gereffi et al., 2005;Kaplinsky, 1998;Sturgeon, 2009;Talbot, 1997).Indeed, global value chains are more than just means for the global exchange of economic goods and capital flows, as they also connect 'people across culture, class, age, gender, race, ethnicity, citizenship and religion and create interrelated webs of social relationships' (Reinecke et al., 2018, p. 460).This paper reveals how value capture in the coffee sector is related to the management and controlling of coffee quality attributes (e.g.material, symbolic), highlighting the relevance of the relational and knowledge dimensions of actors in Colombia, Europe and the USA.Specifically, the paper seeks to understand how producers are able to capture value to contest the coffee paradox, e.g. the extent to which producers are able to create and control the value set in the material, symbolic and in-person service quality attributes (Daviron & Ponte, 2005) of coffee as a crop and end product.
To drink a cup of coffee, coffee cherries must undergo several processes and movements leading to diverse outcomes and value-added (Figure 1).After harvesting the cherries, various processing methods (e.g.washed, natural) are followed and the outer layers of skin and pulp from the cherries are removed (ITC, 2021).Green beans are sorted according to certain criteria, e.g.size, number of defects, etc (ITC, 2021).At this point, the coffee is ready to be roasted and packed, e.g.ground or as whole bean for sale to the end consumer.The extended and long-existing coffee trade practices rely on shipping green coffee to importing countries, where it is roasted, packaged, marketed and sold primarily through coffee shops, supermarkets or e-commerce (ITC, 2021).
Thus, division of labor and knowledge have become established in coffee production, processing, trade and consumption (Figure 1).Usually, the value capture by producers is minimal and depends on the value-added they are able to incorporate to the harvested coffee cherries.About 10% of the income generated by the coffee industry remains in origin countries, compared to 30% approximately 30 years ago (Giuli, 2021).Information on the degree of concentration in the coffee trade sector is scarce.The leading companies are privately owned and therefore not obliged to share data and figures publicly (Panhuysen & Pierrot, 2018).The Specialty Coffee Transaction Guide (Emory University et al., 2023) provides relevant price benchmarks for differentiated coffees, but does not provide the value share of downstream actors (e.g.roasters).However, it creates an anonymized and aggregated dataset that provides context for coffee price negotiations based on quality scores, volumes purchased and country of origin (www.transactionguide.coffee).
International buyers and processors are the ones who ultimately capture the value of the work done by producers at origin when leading actors (e.g.roasters, retailers, brand owners, shop owners) sell roasted coffee or (sophisticated) brewed beverages to consumers.All these imbalances in value capture (e.g.producers mainly exporting a raw material, roasters selling end products, coffee shop owners selling beverages) denote the coffee paradox coined by Daviron and Ponte (2005).Value is realized through the achievement of coffee quality attributes.The quality attributes (material, symbolic, in-person service), already defined and embedded in the coffee quality conventions, are achieved on the basis of the knowledge of the (chain and non-chain) actors involved, as well as access to financial and technological resources.The added value then results from the exchange of the different coffee outputs, based on the quality attributes, from one actor (e.g.producer) to another (e.g.roaster).In this exchange some (invisible) work is required to create the added value, e.g. the unpaid work of female workers at the agricultural nodes of the value chain.The well-informed segment of aspirational consumers pays not only for the material quality of coffee, but also for its symbolic quality attributes (e.g.ethical, environment-friendly practices) usually communicated via labels and for in-person service quality embedded in coffee cups sold at shops (Daviron & Ponte, 2005).
Institutional settings guiding the coffee sector.In 1989, the coffee market liberalization was orchestrated by the dismantling of the quota system and price bands agreed upon among origin and importing countries, embodied in the International Coffee Agreement (Acheson-Brown, 2003;Daviron & Ponte, 2005;Gilbert, 1996).Since the late 1980s, producers have faced constant pressure to conform to several voluntary standards despite the diverse national institutional settings supporting (or not) the compliance to voluntary standards, e.g.Organic, Fairtrade, Rainforest Alliance & UTZ (Dietz et al., 2018;Glasbergen, 2018;Grabs, 2020;Naylor, 2014;Ramirez Gomez et al., 2022;Raynolds, 2009).
More recently, initiatives have been established to calculate the 'True Price' and pay it to producers (True Price, 2017), to stop the use of the C-price for Arabica and Robusta coffee (set on the New York and London Stock exchanges, respectively) to cope with the speculation behind it (Boydell, 2018) or to establish a Coffee Equity Fund (Panhuysen & Pierrot, 2020).As a countermeasure, origin countries have begun to develop and administer their own voluntary standards (Hospes, 2014;Tepox-Vivar & González-Cabañas, 2021).Emerging alternative markets are guided by principles, e.g.collaboration, partnership, or relational coffee models between upstream, downstream actors and related stakeholders (Edelmann et al., 2022;Fischer, 2022;Guimarães et al., 2020;Quiñones-Ruiz, 2021;Vargas-Prieto & Castañeda-Guzmán, 2019).Likewise, there are initiatives to calculate the Return to Origin (RTO, defined as the percentage of coffee retail sales that goes back to the coffee supply chain (www.transparenttradecoffee.org)), or to revise existing quality conventions and business models for trade in de-commodified coffee (Muñoz-Rodríguez et al., 2019).Precisely this paper aims to examine the conditions by which quality conventions in coffee can be revised.
The next section presents the analytic landscape guiding this study, namely the coffee quality attributes, highlighting the relevance of the relational approach and knowledge building, followed by the methods analyzing the value chain of a collective of producers in Colombia and some selected international buyers.Results are presented followed by the discussion and conclusion.

Analytic landscape
Global value chains are suitable examples to reveal coordination mechanisms to be adopted by chain actors and observe how value is created and captured based on the quality attributes pursued by chain actors From a sociological notion of convention, we refer to a normative principle that prescribes a concrete action in a given situation of social interaction usually accompanied by a regularity of behaviour (Miller Moya, 2008).The actions can be tacitly and jointly accepted by a collective and the source of its obligatory nature lies in the existence of mechanisms of approval or disapproval of behaviours (Miller Moya, 2008).Specifically, Boltanski andThévenot (1991, 2006), Diaz-Bone (2018) guide the discussions on the 'quality turn' in agro-food studies in the convention theory from a sociological notion.
Under the lens of convention theory quality is regarded as a device to be able to conform, as a result of communication on and negotiations of product quality criteria.These criteria are synthesized in specific practices and procedures created as objective references what prevents the notion of quality to be reduced to an abstract or subjective idea.Product quality 'is not an ontologically given property but rather a social construction' (Diaz-Bone, 2018, p. 80).Specifically, to understand quality in coffee, Daviron and Ponte (2005) categorize three quality attributes: (1) physical/material, (2) symbolic and (3) in-person service attributes to support coordination among coffee chain actors.Material attributes are embedded in the product, they are intrinsic and can usually be measured by human senses such as vision, taste, smell, hearing and touch or by using precise devices and protocols (Daviron & Ponte, 2005, p. 34).
The Specialty Coffee Association (SCA) has set specific lexicon, protocols and evaluation forms and best practices concerning green coffee grading, cupping or brewing (SCA s.a.).These are common tools to evaluate the material quality of coffees (M1 + M2, Figure 2).According to the SCA, coffees reaching 80 through 100 points during cupping evaluations and attaining certain attributes (e.g.acidity, body, sweetness, etc.) are considered as specialty coffees (SCA s.a.).It is worth-noting that the material attributes of the cup are not objective per se, they are objectified based on the standardization of the characteristics of the beans mainly made by downstream actors (e.g.buyers/roasters) and related stakeholders in core economies.Symbolic quality refers to attributes often embedded in trademarks, geographical indications and voluntary standards generally institutionalized through certification schemes and communicated via labels e.g.Organic or Fairtrade (Daviron & Ponte, 2005).While 'trademarks enable the 'consumption of an enterprise', geographical indications promote the "consumption of a place"' (Daviron & Ponte, 2005, p. 37).These values are embedded in symbolic quality attributes (S1, Figure 2).Certifications allow additional economic and symbolic values to be attached to agricultural products at the retail stage, without necessarily improving the income of producers (Dong & Liang, 2023).
Likewise, we attempt to study the 'ignored' efforts and conditions needed to generate value usually not codified in the quality definition, e.g.dealing with efforts, local conditions (e.g.difficult access to infrastructure), and narratives (e.g.gender relations) silently faced by producers: symbolic quality attributes (S2).This shall present what it is invisible for industrial processors and consumers to achieve purpose and fairness in production, processing, trade and consumption.Service quality attributes refer to the in-person attributes portraying the relation between café workers and consumers at coffee shops (Daviron & Ponte, 2005).We also add the online service attributes when roasted coffee is sold online portraying the feedback ((dis)satisfaction) of consumers based on the service provided during the transaction experience, e.g.user-friendliness, information provided during online purchasing of (specialty) coffees.The value generation and capture can be identified in the attainment of the quality attributes along the coffee chain.
The proposed framework seeks to also uncover relational processes dealing with the building of close relationships and trust (Edelmann et al., 2022) as well as the way knowledge is created, shared and controlled by actors (e.g.producers, processors) to generate and capture (e.g.economic) value.The relational processes and (fragmented) knowledge observed across global value chains permit to analyze them as social entities.The social approach to assess global value chains (Reinecke et al., 2018) provides an alternative lens to contest the still historic narratives of how global value chains shall function, be organized and analyzed.In this paper the authors consider the quality attributes to analyze value generation and capture and do not engage on the upgrading of primary products as an approach for global value chain analysis (e.g.Gereffi et al., 2005;Gibbon, 2001).

Methods
A case study (Eisenhardt & Graebner, 2007;Poteete et al., 2010) portrays a collective of producers (Red Figure 2. Linking quality attributes for coffee with a relational approach and knowledge building.Source: Adapted from Quiñones-Ruiz (2020).
Ecolsierra [La Red] based in Sierra Nevada, in northern Colombia) and three relevant and accessible international buyers from Europe and the USA (Tables 1  and 2).
Research has been conducted with the collective of producers -La Redsince 2012 to understand compliance processes to voluntary standards such as geographical indications (Quiñones-Ruiz et al., 2015) and relational approaches between producers and international buyers and roasters (Quiñones-Ruiz, 2020; 2021).The exchanges and trust built between the lead researcher and La Red staff members were key to select producers compliant with voluntary standards, such as Organic, and achieving at least an 80 SCA score (SCA s.a.) in previous harvests (e.g.2017/2018) to get to know their material and symbolic quality.Questions posed to producers related to the way coffee is processed (e.g.washed, natural, honey), 1 the types of varieties, what producers consider important to achieve quality, the challenges they face in growing and processing, etc. La Red approached some international buyers based in Germany, the Netherlands and the USA to conduct interviews to allow the authors to plan interviews with them.The questions posed to the buyers also covered what buyers consider important to achieve quality, what it is important to them when buying coffee, how they think producers understand quality, whether feedback is given to producers on the quality of the coffee, etc. Field work included: (1) visits to selected producers (some producers at their farms in Minca, Siberia, San Javier), ( 2) selected staff at Ecolsierra Export SAS in Santa Marta; and (3) visits to selected international buyers suggested by La Red.Another related chain actor (e.g.roaster) was also contacted.Planned visits to international buyers and to the SCA fair in Seattle in 2020 were cancelled due to the COVID-19 pandemic.
Data collection was based on semi-structured interviews with producers (P1-30) and staff in Colombia from June to August 2019, interviews with international buyers (B1-3) from October to November 2019 in the Netherlands and Germany, respectively, and with a trader in the USA via Skype in May 2020.In Colombia, semi-structured interviews were conducted in Spanish (transcription in Spanish and excerpts in English); in Europe, in English or German (translation/transcription in English).They lasted between 30 and 186 min and were mostly recorded upon consent.Declaration of consent under data protection law, set by the research institution, was accepted and signed by the interviewees.Respondents signed confidentiality agreements.Likewise, data were collected from websites, formal and informal documents (e.g.Development Plan, SCA results) provided by association members.Relevant literature on quality and the analytic framework (e.g.Quiñones- Ruiz, 2020) provided the deductive codes.The interviews include producers of La Red (umbrella association and their respective grupos de base) and staff members of Ecolsierra Export SAS (La Red's export company) who might have diverse responsibilities or roles (Table 1).Atlas.ti software was used to code the semi-structured interviews based on the categories of the analytic framework, e.g.quality attributes (Figure 2) and identified inductive codes.Content analysis of the interviews was conducted following Mayring (2000).Furthermore, reflective loops were conducted with selected interviewees/key informants.In a nearly 3.5-hour Zoom meeting in July 22, 2021, the results were jointly reflected upon and refined with La Red (staff, producers).Further virtual exchange took place.Tables 2 and 3 provide an overview of interviewed chain actors.La Red was founded by a group of producers coming from other Colombian states.It started in 1997 with 22 members (socios fundadores).The founders arrived in the areas near the foothills of the Sierra Nevada de Santa Marta in the department of Magdalena, where they settled on various farms and eventually became organic producers.The approximately 340 associated families, mostly medium-sized farms, are gathered in 22 groups and can sell their parchment coffee across seven purchasing points (Table 1).Since 2014, La Red has also exported directly through its trading company Ecolsierra Export SAS to international buyers to ensure traceability of its Organic and Fairtrade certified coffee (redecolsierra.org),some of the international buyers are described in Table 2.

Producers informing about coffee quality attributes and knowledge
Producers understand the importance of material quality when describing the factors that influence or affect coffee quality (M1 and to a less extent M2), they also acknowledge the additional struggles to cultivate and (primary) process coffee (S2) (Table 3).Producers stated the relevance of the particular location and agro-ecological conditions (e.g.soil, water), the main problem in harvesting the cherries (avoiding picking green ones, M1), additional monetary and transaction costs (e.g.time and physical efforts, incentives) borne by producers to motivate workers (e.g.pickers) to harvest ripe cherries for quality coffee.
Regarding organic fertilizers, the general manager explained that the project to create its own Biofábrica 2 arose due to the high costs of buying organic fertilizers and the need to meet the associates' requirements (based on agroecological analyses and not only soil analyses).28 producers (93%) stated cultivating the Castillo variety -15 use mainly Castillo and 13 specifically reported they combine it with other varieties (e.g.Arabico Rojo, Caturro, Tabi, Catimor).Two mentioned to grow Colombia variety rather than Castillo on their plots.They also explained how the Colombia variety affects cup quality (M2).Although producers govern most farm activities to achieve good material quality, they see a need to learn more about cupping (M2).Furthermore, 27 producers (90%) reported consuming their own coffee.Only one producer specified to get pasilla (lowquality coffee) for his own consumption.P25 stated: 'One is a coffee producer and to leave the worst for oneself, no! … no and I left that routine and I take the good coffee […]'.
The general manager also mentioned the challenges in drying coffee, as it usually rains during harvest time and the poor road network makes the situation worse.Unlike other Colombian regions, this zone has only one main harvest season from October to January.Although producers use marquesinas (solar dryers), they cannot dry all coffee in their facilities.Hence, they must hire expensive transportation to quickly bring the still-wet coffee to La Red's drying facilities to avoid rotting.Producers have adjusted their drying practices.They used to dry coffee only on the patios, but today many of them build solar dryers with the social and organic price premium from Fairtrade/Organic certifications.A lack of electricity or blackouts are also major problems in coffee primary processing.Another common challenge in coffee culture is access to labor during harvest.Nevertheless, Sierra Nevada producers seem to have social proximity and empathy with them and in some cases the categorization 'family' is used to denote labor relations.Some explained that they had previously worked as laborers (jornaleros (day workers), mayordomos) until they could save money to buy some land and become owners (P3, P10, P11, P25, P28).Understanding M1 + M2 is not enough, since the local conditions (e.g.environmental, social, infrastructure personal stories) categorized in S2 have to considered to be able to achieve the required material quality.14 producers would suggest international buyers to attain a sense P7 uttered that a good coffee 'is a coffee that has all the processes well done, that is, from the beginning, from the harvest to the drying, that is a good quality coffee, and when it comes to the [cup] tasting, that really depends on the client, because we have had surprises that there are some who like it one way, others another, so it can really be a very good cup of coffee, but that depends a lot on the buyer'.Source: Own elaboration.
of empathy for their work, so that the conditions under which the coffee is produced can be recognized and valued (S2), to be translated into higher prices and be passed onto consumers.Producers aims to voice their discomforts and efforts (categorized in S2) to downstream actors.Although La Red doesn't own a coffee shop (inperson service), it has been selling its brand Café Tima for domestic consumption since 2004/2005.Café Tima accounts for about 1% of coffee production.Future projects to open local coffee shops are being discussed.Moreover, producers have started to consider coffee as a beverage that should be well taken care of.Specifically, 8 producers (26%) perceived coffee as a beverage or food-product.The person in charge of institutional strengthening firmly emphasized the need to enhance a culture of coffee consumption.According to the general manager, Colombia is still known as a country where it is difficult to find a good coffee in hotels.Recently, the domestic quality consumption of coffee has developed and the specialty coffee scene is booming (ITC, 2021).
The role of knowledge (at farm level, primary processing, within the organization) to generate and capture value: As for the average SCA score, the quality manager, who is also an associated producer, stated that 'the average quality of our coffee is, between 81 and 83 [SCA]' (Table 3, right column).However, 84 and 86 SCA points have been achieved.With the support of an external person, she conducted experiments with selected producers called 'Guardianes de la Sierra' to promote the innovative capacity of producers.Nevertheless, the experiment with new processing techniques (honey processing) did not work properly because a general protocol was established without considering farm specifics (e.g.altitude, infrastructure).A farm located at 800 meters above sea level (m.a.s.l.) has different requirements than one at 1600 m.a.s.l.The quality manager stated that everything was poorly planned with P2 (well-known for high cup quality).He did not have the drying capacity for the honey processed coffee (new processing method for them) and the cupping revealed defects such as traces of fungus and mold.The quality manager clearly acknowledged the mistake of having only a general protocol.The fact of engaging in new production practices to achieve M1 + M2 is not easy, experiments and failures are needed to attain the desired outcome, e.g.absence of certain physical defects, obtaining specific flavors when cupping the coffee, to be able to capture more value for their coffees eventually valued with higher prices by buyers.
Likewise, the general manager highlighted the need for innovation, however, without putting pressure on producers (en camisa de once varas), as producers can improve M2 even with small changes (e.g.corrective farm practices) and 'without systematizing everything'.The general manager further stated the importance of having a 'segmentation within the organization' to understand the diverse qualities by producers' collective and to find suitable markets (e.g. by attending international fairs).La Red can pay the producer a better price for M1 delivered at the time of physical analysis (factor de rendimiento), but not for a 'wow [cup] profile' (cupping score, M2) if La Red cannot find a buyer for that coffee.La Red has developed strong social cohesion and trusting relationships with its members.All interviewed producers stated being satisfied with La Red and selling the required amount of coffee at the purchasing points.
P13 also stated the importance of foreign or virtual expositions such as participating in international fairs: 'We went to visit them [potential buyers/clients] at the stands, and we managed to make connections [with little English].They [the fairs] are expensive, because exchanging [Colombian] pesos for dollars or euros is expensive, but it has been very productive.I think that if we had not done that, we probably would not be in the position we are in now, because it is not easy to get customers, and one has to go out and look for them and tell them the story and try to convince them to come and they have come, in fact, this year [2019] several have come to visit us and they have liked our process, and above all, because it seems to me that the history of La Red is very beautiful, that is, from its beginnings, [the founders] meeting in the park or in the street, with paper and pencil, with no money, just ideas and nothing else, and from there little by little it began to grow, obviously with the support of many people and foundations, such as the Fundación ProSierra and the Comité de Cafeteros, … , and well, that's where we are going, the idea is to continue growing.

P5 stated:
The coffee business is there, in the finished coffee, because one is a farmer, but one is no longer in the jungle or in the 'uga uga', so one can now go to the internet, one can now look at the coffee process.
There one learns a lot, and then one already knows that the best income from coffee is the finished coffee; for example even if I sell my [roasted] coffee cheap I see a profit, I sell it at 12,000 pesos [approximately 2.60 euros] per pound, with its packaging and its label, and I only see a profit of 40% when I sell it in beans [parchment coffee] […], there is still a long way to go for that, this is something that one has to search little by little.

Quality attributes acknowledged by downstream actors
While producers mentioned the subjectivity of quality (depending on what buyers look for as these actors objectify quality through standardization processes based e.g. on the compliance to the SCA guidelines), international buyers/roasters provided precise statements on the importance of quality (M1) to primary producers (Table 4).Producers are primarily seen as suppliers of a raw material, of the material quality (M1) of green coffee in order to make a living from it.In doing so, producers shall take care of the environment (S1), which is a primary concern of downstream actors.The interviewed downstream actors (Table 4) consider certain symbolic values (e.g.S1), especially B3.1, B3.2 and R. All of them are organic certified.Particularly, B3.1 and La Red attempt to define the True Price of green coffee which also includes intangible components (S2) (B3.1, general manager).After B3.1's (and a photographer's) last visit to La Red and some producers in Sierra Nevada, B3.1 organized a two-week popup exhibition at Amsterdam Central Station, displaying images and other materials (e.g.coffee bags and baskets, a small de-pulping machine).Visitors had the opportunity to make virtual calls to producers, assisted by La Red (international business expert), who was responsible for translating questions from English to Spanish and viceversa.Frequently asked questions were: How many children do you have?What do you like most about being a coffee producer?What is the most important thing about coffee to you?What Table 4.The diverse perceptions of good quality coffee by international buyers, trader and roaster.

Buyers, trader and roaster
How buyers, trader and roaster think producers conceive quality B1 A good coffee is the one that in the end tastes good in the cup (M2) A good [physical] quality where the producers invest efforts (M1) and can obtain a sustainable living (S2).

B2
I like rather fruity coffees, for example, very much.They rather so really have these floral aromas (M2) … from the production conditions, a qualitative bean and where the bean can be separated as a reasonable product at the end, yes.So, a bean without defects, […] with a uniform sorting, without diseases (M1).And also, one for which the producers get a reasonable price.B3.1 Uh, a good coffee for me is a very sustainably harvested coffee (S1) … which for me, it's with acid [acidity], because I like it […] (M2).I am not very knowledgeable about coffee.But I do know that I appreciate specialty coffees.And I also find the way of preparing is very important (in-person quality).
It's a coffee he or she understands himself or herself.So, a coffee that is produced … taking care of the environment (S1) but also knowing what a producer is doing and the effect his or her work has or the soil or the climate has on the coffee (S1).And also, one that is paid according to the quality of the coffee (M1).B3.2I would start with the taste to talk about a good coffee.It should be balanced.It should be a little bit bitter and a little bit sour and have some body in it.You have to be taken through a wave of flavors in your mouth and have a good aftertaste (M2).But of course, […] the background of the coffee is very important too.It cannot be that the farmer is not paid well for his coffee.And it cannot be that the planet is exhausted for growing coffee (S1 and S2).
A coffee that he knows his customers value and also that he knows what his coffee is worth.I think that would be the main thing (S2) T A good coffee for me is one that has a good balance between its aroma and flavor, that has a body with positive qualities and has lot of cleanliness (M2) Lack of defects, clean, sweet (M1 + M2).

R
… which is nice in taste (M2) and which I enjoy to drink (in-person quality).And … well I think taste is at the top (M2).And then of course you would like to drink your coffee in a responsible way (S1).But, to be honest, priority is taste (M2).
A good coffee for them would be a coffee where they earn a good living with (S2).
would you like consumers to know about your coffee?How is your coffee produced or when is it harvested?What are the processes involved with coffee?B3.1 also shot videos and shared them with the general manager, who then presented them to producers.All buyers (B) and the roaster mentioned the importance of making a living by selling good (quality) coffee [while the buyers represented a business or company, the collective of producers also runs a business!].M1 was recognized by international buyers/roasters actors as necessary to obtain good coffees.Nobody specifically stated the importance of M2 for producers to acknowledge good coffee.Only one employee of a trading company, based in the USA but coming from the origin, a daughter of coffee producers (previously responsible for quality assurance in a local association), mentioned that producers should also consider attributes as sweetness (M2) when assessing the quality of their coffee.
B2 particularly highlighted the importance of organizational structures of the associations at origin (e.g.decision-making processes, transparency in the money producers directly receive, pre-financing (60%-100%)).B2 stated that, unlike Fairtrade, they do not use auditors, but rely on 'direct exchange at eyelevel' and ask questions, but not 'with such a strange neo-colonial attitude'.Likewise, B2 uttered the fact of being organized as a collective in Germany: 'That means that we not only try to pay fairer wages […]'.At the collective, 'everyone earns the same.It doesn't matter whether we're packing […] or doing controlling […].It's all one wage level.And we also make all decisions together at the plenum.In other words, according to a consensus principle'.
The roaster stated the need to change the traders with whom they source their green coffee, as he feels 'a little black box because we are doing business with the trader and what is happening behind the trader and everything there, it is more or less unknown to us'.He further added: 'as a purchaser I am responsible to get the best out of the market.And the best does not always mean the lowest price.The best means […] a good price and all the other elements which are involved […] risk, quality, etc. […] and price is one of them'.
During the interview, he specifically thanked for exchanging with him to understand more 'what happens at the other side of the chain' (S2).

A collective of producers reshaping the coffee chain by controlling quality attributes and relating to international buyers
Generally, downstream actors (e.g.large companies -Nespresso, Starbucksand related players) in importing countries set voluntary standards institutionalized in certification schemes or protocols (e.g.Organic, C.A.F.E.Practices, SCA guidelines) to be first followed by communities of producers in origin countries, as coordination devices, to be able to access differentiated markets.Producers are the first in institutionalizing specific practices at the moment of cultivating and reaching the primary processing of the coffee cherries to mainly obtain and sell/export green coffee (still a raw material).Usually, these practices deal with the attainment of material quality (M1 + M2) but also symbolic attributes, embedded in wellknown certifications such as Organic and Fairtrade (S1) as well as the additional efforts borne by producers due to the specific context in which they are located (S2).
La Red is able to generate and capture value by managing and controlling M1 + M2 and to certain extent S1 + S2.The quality manager is able to create several lots to comply with the demands of international buyers, normally set according to SCA scores and specified organoleptic characteristics.Thus, producers have begun to understand the coffee business through La Red.
The trader and quality manager uttered the importance of providing feedback (for further coordination) on M2 to producers via international buyers/roasters to achieve better material quality.Both highlighted the relevance of the SCA as an authority for specialty coffees and suggested adding information to the evaluation protocols about the context in which the coffee was produced (e.g.type of organizations/producers, main difficulties, in other words: S2).Likewise, the trader critically stated that roasters and those responsible for post-purchase coffee evaluation shall provide mandatory feedback to producers not only on the safe arrival of coffee, but also on material quality (M1 + M2), so that producers can adapt their practices if needed.Likewise, at origin, the quality manager emphasized the importance of providing all information about coffee processing (e.g.postharvest) by producers.This allows the producer organization to better advise them and provide feedback on M1 (e.g.defects) and M2 (e.g.cup quality).
Knowledge and communication are important, both between producers and their organizations, and between organizations and international buyers/ roasters.
Producers have been able to manage symbolic attributes (S1) institutionalized in certifications/protocols (e.g.Organic, SCA) and build long-term relationships, e.g.direct trade with international buyers and roasters.The trust and positive experiences between La Red and B1, allowed the general manager to ask B1 considering buying roasted coffee in addition to green coffee.La Red proposed B1 to cup their roasted coffee, as they had the facilities, machinery, knowledge and expertise to process roasted coffee.Although, B1 considered buying roasted coffee from La Red, B1 first acknowledged that it was not the usual trade practice in the coffee branch, as international buyers and roasters typically buy a raw material.Although coffee is mainly traded as green, B2 has already imported roasted coffee twice since they started this project in 2019: two containers with a large amount of green and small amount of roasted coffee (Figure 3).B1 requested some samples of roasted coffee to learn about the quality of the roasting by La Red and the taste (M2).Though adjustments were needed to meet the German taste, B1 was satisfied with the roasting profile.B1 traveled to Colombia in May 2019 to conduct joint coffee tasting/cuppings, visit selected producers, plan and negotiate logistics and learn to select the best packaging material for the export of roasted coffee, schedule production and delivery cycles and set prices.It required substantial efforts, communication (e.g. more than 300 emails) and a collective knowledge building.New knowledge was needed to achieve such endeavor.Deciding on the packaging and delivery cycles of the roasted coffee was key.La Red's production and delivery cycle (exports of roasted coffee) had to be carefully redefined.B1 stressed the trust in La Red which allowed to develop a project encompassing collective knowledge building and teamwork capacity.
In May 2020, B1 launched two types of La Red roasted coffee, one following the German flavor profile for their zone and another with the Colombian profile, to also show German consumers how Colombians drink their coffee.In a follow-up exchange with B1 in May 2022, B1 confirmed to still import green and roasted coffee from Colombia, which was also documented on the company's website.Knowledge building was based on bidirectional knowledge sharing, which is a critical stage of knowledge transfer (Tangaraja et al., 2016).This included the provision, collection and joint application/testing of knowledge among staff belonging to diverse organizations/continents.The relational and behavioral attributes generated a formative concept (Tangaraja et al., 2016) that went beyond the individual codification of practices required for coordination.B1 considered La Red as a downstream actor (rather than just supplier of raw material) able to roast for them the way German-based roasters do, even if B1 might save some roasting costs due to lower labor costs at origin.The share value accrued by the collective is shown in Figure 4. B1 has also been approached by Ethiopian producers to consider/implement similar processes/experiences.
Adding value to green coffee at origin, e.g.roasted coffee is considered difficult due to factors such as quality preservation (shelf life, freshness), marketdriven blends or just-in-time delivery (ITC, 2021).E.g. coffee shall be consumed quickly to achieve optimal M1 + M2; blends can be very difficult to handle in origin countries due to seasonality, supply and demand fluctuations and geographical distances.Supermarkets and retail centers do not maintain large inventories as they face rapidly changing demand and must quickly adapt to roaster deliveries (ITC, 2021).However, the case of La Red and the German buyer shows that all posed challenges can be handled by carefully managing quality attributes, having access to specific knowledge but above all when chain actors have built trust and a relational approach (Figure 2).The social proximity and cognitive abilities (Edelmann et al., 2022) shared among actors are key to develop such endeavor despite the geographical distances of actors.Direct and proximate relationships alone do not replace traditional production and trade conventions.The discussion is not about changing existing quality conventions, but about how quality and trade conventions are institutionalized and the narratives behind them.The collective captures value not necessarily through the recognition of the unpaid work at farm level (S2), but through the extra work and effort to capture value, namely through roasting, access to knowledge and direct contact with downstream actors.

Discussion and conclusion
The analytic landscape (Figure 2) served to illustrate how producers navigate in the understanding of quality attributes to generate and capture value, individually and through collectives (La Red).The understanding and controlling of material quality (M1 + M2) depend on the appropriate environmental practices and processing stages along the chain that producers manage and dominate.Through robust collectives, coffee producers (and workers) have expanded their knowledge of physical quality (M1) and are beginning to grasp taste and other quality attributes (M2).
However, a deeper and common understanding among contrasting chain actors can only occur when upstream and downstream actors (e.g. from producer groups, industrial processors, consumers) finally value the actual conditions under which coffee is produced (S1 + S2).Much of the missing infrastructure that government is supposed to provide as public goods comes at the expense of producers to achieve a desire quality.Some buyers/roasters have begun to learn about the growing conditions by visiting farms and observing what matters most to them, first the environment, and only recently developing close and more eye-to-eye relationships with producers.However, producers also need to familiarize themselves with target markets to understand what buyers/roasters and consumers are looking for and visit their business and infrastructures.Anthropological studies trace coffee producers to international buying partners to define coffee not only by taste, but also by a complex system of values worked out among producers, roasters and consumers (Fischer (2022) in Guatemala; West (2012) in Papua New Guinea or Besky (2020) for tea in India).After reflective loops with La Red staff, it was believed that many international buyers still evoke pity (lástima) for what is happening at origin (a form of selling social marketing), but in many cases don't pay higher prices.The term 'symbolic attribute' was also challenged and it was suggested to call it an immaterial attribute to make it visible, it is not abstract and needs to be objectified and paid for.
Apart from identifying the practices to coordinate quality and trade among coffee chain actors, the analytic landscape serves to uncover the relational approach to adjust practices despite the geographical distance among actors.Value capture is achieved by controlling the quality attributes, achieving a relational approach embedded in social proximity and joint knowledge building.By engaging on a collective project, La Red was able to deliver high-quality roasted coffee to a trustworthy buyer, promote reciprocal understanding and transfer symbolic values directly to final consumers.The initial knowledge and expertise of La Red was recognized by B1 in launching a trial of roasting at origin despite the challenges to roast and export coffee from origin countries (ITC, 2021).By daring to propose something different, producers were able to govern their work, create value and decide how remuneration and resources can also be distributed.It showed producers' self-determination and how democratic governance of firms might result in a network connectivity at different levels (locally and internationally) achieved beyond the global south/global north divide.In this juncture, chain actors are seen as partners.Nevertheless, lengthy relational processes based on robust internal governance, strong personalities, relation of producers to their natural and human resources as well as knowledge building (e.g.beyond primary production and compliance with certification schemes) are key prerequisites for producers to capture value.La Red was able to start a network that can eventually be institutionalized in the long-run, going beyond specific projects or one-time activities.The exchanges in the presented networks generate an intersection of actors' knowledge bases.Likewise, trust is a result of the social proximity already practiced among the actors.
Building trust, relational approach and knowledge sharing among chain actors confer opportunities to institutionalize quality and trade conventions to contest the coffee paradox.Indeed, a collective (producers from Peru, Nicaragua, Guatemala, Mexico, Ethiopia) is able to govern coffee shops (in-person service quality) located in Sacramento (USA), directly reaching final consumers.Personal exchanges with small roasters show that roasting at origin is rejected because it might jeopardize their business, as it is not yet conceived as an additional product for conscious consumers looking for more options.Adding value, e.g.roasting at origin is not new and it is not the panacea, studying its impact on generating a workable income for producers, similar to the logic of downstream peers for making profits, and on knowledge building are however not truly considered as alternatives to address the perils in the industry.Nonetheless, few buyers see it as an additional product line or business alternative (e.g.Kaffee-Kooperative.de(joint-venture between Rwanda and Germany, but also including other origins), Moyee Coffee (joint venture between Ethiopia and the Netherlands)), all specialized in roasted coffee at origin.Recently, there are also small coffee buyers selling roasted coffee (e.g.Café Tres Cruces based in Paris, nulldiebohne in Vienna), also including children of producers living in core economies.
A female-roaster from El Salvador living in Vienna stated in an informal conversation the need to revitalize the coffee industry by roasting coffee at origin and delivering it to supermarkets in Europe, as many coffees on the shelves have 'no name-no origin' and are of average or bad M1 to which consumers have already become accustomed.This vision challenges the ITC (2021) report and related initiatives from international cooperation on value-added at origin.Will future industry studies and approaches (e.g. from the International Coffee Organization, SCA, International Trade Center) include valueadded alternatives based on e.g. the building of network connectivity at different levels and knowledge required to tackle the old perils and traditional sustainability discourses and start seeing producers as business partners rather than poor and marginalized actors who need to be helped?This standpoint provides another perspective to revise the existing governance and followed moral legitimacy (Reinecke et al., 2017).Upstream actors shall be involved, have a voice to be able to capture value and be able to challenge the well-accepted and historic handling, e.g. trade conventions mainly exporting green coffee.
In 1997, the founders of La Red would probably never have imagined running an export company and challenging the trade conventions for coffee trade.These developments, transcending the quality discourses, do no decouple their role as peasants (Narotzky, 2016) and their connection to the land as guardians.The discussion is not about changing the established quality conventions, but about revealing how quality and trade conventions are institutionalized and the narratives behind them, allowing community of producers to capture economic and symbolic values.Over time, these producers have improved their material quality techniques and become more familiar with the places where their coffee circulates and is valued, allowing them to influence the commercial spheres of the coffee business.
Although a case study, this work aims to contest the quality discourses and value chain narratives that artfully manipulate the moral legitimacy set by influential actors (and followed in the branch), who, intentionally or not, impede the achievement of a greater good.Though this might be seen as an atypical case of a collective generating and capturing value by controlling quality attributes and having its roasting company, knowledge and expertise, national policies also enhancing domestic quality consumption and development cooperation shall consider alternative imaginaries (e.g.knowledge building beyond physical material quality) to foster this type of accomplishments in rural communities in addition to the usual supportive price incentives for raw material.The experiences of La Red allow us to think about other products, equally immersed in global chains that were originally part of colonial settings, e.g.tea, cocoa/chocolate.Nonetheless, each of these products shall be examined on the basis of their specificities and the relationships that connect producers, industrial processors and consumers along local and global chains.

Notes
1.In the washed processing method, coffee cherries are picked and delivered to a facility (e.g.wet mill, washing station).Coffee fruits float in water to remove defective and underripe cherries.Machinery (automated or manual) is then typically used for depulping, e.g.removing some or all of the skin and fruit mucilage from around the bean.In the natural process, harvested ripe coffee cherries are sun-dried on drying beds, tables or cement floors.In the honey process, fresh coffee cherries are depulped, but left to dry without washing.2. It is a physical space devoted to the manufacturing and storage of fertilizers and inputs based on bacteria, fungi and plants as bioinsecticides, compounds with rock flour and mineral broths.Many of these products reduce dependence on synthetic inputs and improve soil fertility (Anacafé, 2022).

Figure 1 .
Figure 1.The illustrative example of the fragmented knowledge-based global value chain for coffee.Source: Adapted from Tröster (2015).

Figure 3 .
Figure3.The reshaped value chain between La Red and buyer B1.Note: The value added, namely, roasting, takes place in the origin country.Source: Own elaboration.

Figure 4 .
Figure 4. Comparison of value shares of roasted coffee in Germany and Colombia.Source: Eidt (2020), own translation.

Table 1 .
Selected organic producers from La Red.
Note: N/A: No data available.a P24 is divided in P24.1 and P24.2 as the producer's wife was also present during the interview.Staff of the Ecolsierra Export SAS: Quality and roaster manager (thereafter quality manager), general manager, international business/logistics, certification, institutional strengthening.

Table 2 .
Selected companies buying from La Red.
Note: Buyer (B), Trader (T), Roaster (R), not applicable (N/A), Free on Board (FOB), Specialty Coffee Association (SCA) a Two persons were present during the interview B3.1 and B3.2.La Red recently negotiated a CIF (incoterm; Cost, Insurance and Freight) export with a newly acquired Austrian buyer.b R3 roasts for B3 but does not buy from La Red.

Table 3 .
Quality attributes recognized by producers during the interviews.