 |
Revue française de sociologie | 157-182 Distribution électronique Cairn pour les éditions Ophrys. © Ophrys. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit. |
Price Formation : the Case of the Burgundy Wine Market
Yuna Chiffoleau
Unité Mixte de Recherche – Innovation Institut National de la Recherche Agronomique (INRA) 2, place Viala – 34060 Montpellier cedex I – France chiffole@ensam.inra.fr
Catherine Laporte
Unité Mixte de Recherche – CESAER Établissement National d’Enseignement Supérieur Agronomique de Dijon (ENESAD) 26, boulevard Dr Petitjean – 21079 Dijon cedex – France
1Price formation processes are the object of renewed research today in both economics and sociology. The question is obviously not new; it was at the center of fundamental works in political economy, such as Adam Smith’s explanation of mechanisms for adjusting the “market price” of a commodity around its “natural price”, an explanation involving observation of merchant behavior (Smith [1776] 2000). Classic sociologists such as Weber and Simmel also examined the question, but from a more historical and philosophical viewpoint, favoring an integrated vision of “markets” (Swedberg, 1994). Analysis of price formation was later left to economists, most of whom adopted a simplified, formal, idealized representation deriving from Walras’s model of the competitive market. In this framework, prices result from the “intersection” of supply and demand curves while the real mechanism that makes that intersection possible is omitted from the analysis, or rather evoked in very general terms by analogy with an auction supervised by a hypothetical auctioneer (Guerrien, 1992). Developments in the two disciplines during the last twenty years have led to reconsidering this simplified vision, and price formation has once again become a full-fledged research question. The extension, and criticism, of the neo-classical model have led economists to take into account the plurality of exchange situations as well as the specific role of information (Walliser, 2000). Meanwhile “new economic sociology” has made it clear that the social construction of markets and prices is one of sociology’s important themes (Granovetter and Swedberg, 1992).
2In addition to developments particular to each discipline, this increased interest in price formation analysis is linked to the development of “economics of quality” studies. According to Lucien Karpik’s definition (1989), the quality economy refers to situations involving exchange of differentiated products of uncertain quality; quality may then be assessed by means of several dimensions, and is the basis of competition among suppliers. Taking diversity of and uncertainty about the qualities of exchanged goods into account has led sociologists and economists to adopt a more complex representation of price formation. It has come to be understood as the result of a process, a series of decisions and interactions among abstract or concrete agents. This aspect, broken down to varying degrees, is present in studies of incomplete contracts that examine different negotiation sequences (Hart and Moore, 1988), as well as in specific analyses of social interactions around a physical market (Jorion, 1990). Furthermore, the process is understood to take place within an institutional, cultural, and/or relational structural framework that may be inherited or constructed through the very interactions of the process. This dimension is highlighted in new economic sociology by means of the notion of “embeddedness” and close attention to social networks (Granovetter, 1985). It is also being explored in economics, with increased reference to relational structures (Kirman, 2001) in addition to a willingness to take into account the different institutional arrangements and contexts affecting transactions. However, there are still very few studies that actually show how prices are determined in situations of uncertainty about quality. A recent overview of research studies about “quality” (Musselin et al., 2002) calls for examining this question directly, namely in order to specify the nature and role of networks in the price formation process.
3Our article, which falls into the framework of studies of differentiated commodity price formation as a contextualized process embedded in social networks, moves in this direction. We have sought to analyze situations in which seller posts prices that are the result of a decision-making procedure, prices that can then be modified during a phase of negotiation with buyer. The point is to show how, in a context of uncertainty about commodity quality, the nature of these procedures, together with social networks and statuses, can intervene in price formation, revealing the complexity of the social organization of a quality goods market. The example studied is the sale of Burgundy-produced wines, where producers post what is called a tarif; i.e., a list of prices for their range of wines. We thus refer to a particular “market” arrangement where “sellers” are in fact a great number of producers, quality is variable and uncertain for buyers, and transaction prices depend on offering prices displayed by sellers.
4In the first part we consider how economists who were dissatisfied with the model of perfect competition handled the question of differentiated commodity price formation, and present the approach developed in new economic sociology. In the second part we show how Burgundy winegrowing and producing is an exemplary case for analyzing price formation as the result of a decision-making procedure that originates with the seller while being socially embedded, and we explain our method. The third part presents our main results, which we discuss in the fourth and last part with reference to the hypotheses and research program of new economic sociology.
Quality commodity price formation : from economic models to economic sociology propositions
Taking quality into account in economic models
5In the canonical economic model of the perfect market, the commodity exchanged is homogeneous, its characteristics are known to all, and price is determined by supply and demand; that is, economic forces that escape agents’ control. But confronted with the fundamental question of commodity differentiation and assessment of “quality” in real markets, economists have proposed different models that break with the model of the perfect market to varying degrees.
6One family of models is based on the idea that commodities exchanged are comparatively assessed in terms of attributes that define their quality. These commodities are thus in competition on an overall market defined by a general product category. If information is perfect, as in Lancaster’s basic model (1966), the price will tend to be aligned with production cost for each quality level. This condition is rarely met, however, because buyers generally cannot assess quality before purchase. Quality can either be checked after purchase (experience commodities) or be assumed (credence commodities) (Nelson, 1970). Taking quality into account leads to postulating the existence of information asymmetry between seller and buyer : seller usually has certain knowledge of quality, buyer doesn’t. But these conditions create permanent instability between quality assessment and price definition (Akerlof, 1970), instability which in turn makes transactions impossible. Trading quality commodities presupposes that sellers and buyers find a solution beforehand that will eliminate the asymmetry. Such solutions include reputation, use of extra-market information vectors, etc., and because they determine the way actors reason, they are understood to influence quality-price relations and the process that ultimately leads to identifying market price. Quality therefore determines price, but the way it does so depends on the information system structure; that is, the vectors that actors choose to clear up uncertainty about quality.
7Another family of models makes the assumption that quality differentiation is reflected in the individualization of particular products attached to a given company. This approach, developed by Chamberlin (1953) in the monopolistic competition model, resolves the question of uncertainty at the outset. The business reasons like a monopoly in response to demand for its product; it meets buyers that it knows and who “know” its product. Price is no longer the result of reasoning with reference to an extended market made up of comparable products but rather the constraint that a business exercises on substitutability of singular products and the power associated with the business’s individualized relation to its customers.
8These two families of economic models thus refer to two different types of social construction for determining the prices that businesses charge for differentiated commodities : 1) the system of information on quality, which allows for resolving the question of asymmetry on condition that that system is founded on shared information vectors; 2) the special relations that exist between a business and its customers, which give it the means to radically distinguish its products from those of its competitors. But these constructions are postulated before any modeling is done and must be accepted as such by the analyst.
9Other models or analyses acknowledge and examine construction of contracts, conventions, institutional arrangements or “networks” that allow for assessing quality and/or stabilizing relations between the business and its customers. Studies whose primary reference is incomplete contract theory or transaction cost economics attribute primary significance to contract commitments that reduce information asymmetry on commodity quality (Ménard, 2000). The specificity of investments made by each party to reduce uncertainty justifies construction of “hybrid forms” that stabilize both qualities and transactions. The economics of convention program clarifies the conditions leading to the emergence and coexistence of a plurality of “quality conventions” making possible agreement among agents on an extended market (Eymard-Duvernay, 1989). Other studies that engage both sociology and economics have also worked to shed light on “intermediation” arrangements between producers and customers that work to singularize a product and create buyer attachment to it (Callon et al., 2000). Most of these studies focus on the nature of contracts or economic justifications for product quality-related institutions, however, without specifically analyzing the mechanisms involved in price formation.
10These analyses provide only limited means of understanding how prices are determined. First, they generally only look into one component of the process, either adjustments leading to balanced prices in a given market situation (extended or individualized), or the definition of quality-related institutions.
The view afforded by economic sociology
11Economic sociology has been constructed in large part in opposition to the standard economic theory model, using a postulate that complexifies analysis of economic action and seeks to reintegrate its social dimension. This approach has taken on new force since the early 1990s, when it was formalized through the “new economic sociology” program (Granovetter and Swedberg, 1992), which brings together sociologists and “heterodox” economists
12Other studies have taken on this question more directly, and from the business’s point of view, seeking to make a connection with business embeddedness in social networks (relational embeddedness). In the case of businesses that serve businesses (advertising, legal defense, etc.), Uzzi and Lancaster (2001) have shown how the different types of relations in which these firms are embedded allow for diffusion of non-public information and development of informal governance, which in turn affect offering price. Non-market ties with customers develop trust and reduce transaction costs while inducing suppliers to share their gains with customers by reducing prices; ties with third parties (competitors, experts, etc.) provide suppliers with information that enables them to differentiate their products vis-à-vis their competitors and in so doing raise offering price; lastly, relations with customers of other firms charging high prices increase the status of the business, thereby giving it the means to raise its prices.
13Anglo-Saxon sociologists not only underline the importance of relational embeddedness but also recognize “status” as a key variable in market functioning, in particular price formation, though the definition and role of that variable are matters for debate. Podolny applies the principle that a firm’s status is linked not only to the relations it maintains with customers and third parties, as mentioned by Uzzi, but also its “reputation” as defined by economists (Shapiro, 1983), namely user assessment of qualities that the firm has already made available to customers (Podolny, 1993). In the example of California wineries,
14These studies in Anglo-Saxon economic sociology thus confirm, specify, or debate elements already put forward by such economists as Akerlof or Chamberlin. Moreover, they further a closer interlinking of economics and sociology by showing that the impact of relational embeddedness on price in both the case of service firms and wineries is not independent of economic variables particular to the given business (size, turnover, etc.) or its market environment (price of raw materials, regularity/exclusivity of seller-customer exchange, role played by intermediaries, etc.). In line with White’s market model, they also suggest that business interactions 1) are “decoupled” in institutions, which in turn work to frame their economic practices and 2) are linked to supplier “conventions” (White, 2002b). Podolny shows how wineries’ market relations work to construct and order the appellation system for California wines, a system which thus becomes autonomous as a market institution and comes to provide the basis for the different companies’ status. Moreover, without going so far as to define conventions as collective cognitive touchstones (Eymard-Duvernay, 1989), Podolny also distinguishes different winegrower motivations, identifying two types : love and money. Motivation has an influence on how a winery is integrated into the market system and ultimately on pricing approach (Scott Morton and Podolny, 2002). While relational embeddedness and status are operative in the way a given firm chooses its offering prices, the point is also to show how economic characteristics on the one hand; institutions, moral values and conventions on the other, condition their nature and impact.
15French sociologists who have participated in their discipline’s increased interest in markets are moving forward along these lines. Karpik’s work on the market for lawyers in France (1989) constitutes a fundamental contribution. The author examines how social networks intervene in lawyers’practices to determine the “right price” (in this case fee level); that is, the one they can rightfully lay claim to given seniority, skill and area of specialization. Social interactions among members of the profession give rise to “producer-networks” that diffuse information not only on price but also the person charging it, which also informs the receiver about the status-price relation and leads to developing a generalized scaled price system which is then to be interpreted as an institution that has been “decoupled” from interactions within producer networks. Producer networks are themselves entertwined with “exchange networks” in which judgments on lawyers’offerings in terms of proposed services and prices circulate. From these networks emerges a legitimate quality ordering that is only temporarily stable, varying with clients’ repeated testing of offerings and maintenance of interpersonal trust. But Karpik also shows how professional relations between lawyers and their clients are run through with demands for morality and justice that help maintain trust and work to fix “reasonable” fee levels. This leads him to evoke the principle of an “economy of moderation” that presupposes forms of balanced exchange between the profession and the public, forms motivated by values of equity or friendship.
16The contribution of Anglo-Saxon and French economic sociology traditions to analysis of price formation has the advantage of bringing together the progress made and questions posed by extension of the neo-classical model and attention to the role of social structures, particularly relational embeddedness. The influence of different types of social networks, their possible “decoupling” in institutions, and their moral foundations must be examined in greater detail, following Karpik’s analysis in particular. The example of Burgundy wine lends itself to a comprehensive approach to the procedures suppliers use to determine their prices, an approach which will deepen our understanding in precisely the way proposed.
A comprehensive approach to price determination for Burgundy wines
An exemplary case
17Winegrowing Burgundy is a strong entity characterized by a shared winemaking process : wines are generally made with a single variety of grape, reds from Pinot Noir and whites from Chardonnay. With the exception of sparkling wines, they are all dry, still wines. Yet there is great variety : there are more than 20,000 bottled wines for sale at wineries alone, differentiated by general geographic origin (and bearing the corresponding appellation), vintage and domaine name.
18This product diversity goes together with strong consumer price heterogeneity that is not random but rather tightly linked to the French system of appellation d’origine contrôlée (AOC) set up in the lates on the basis of a hierarchical distinction among specific, limited geographical areas or terroirs. Differentiation in this case can be analyzed as a compromise between the natural capacity of the given soils to produce wines with complex flavors and the power relations that existed at the time among wealthy landowners and small producers (Jacquet and Laferté, 2005). The hundred or so official geographic origins for Burgundy wines fall into four hierarchically ordered levels : regional, “village wines”, premiers crus and grands crus. As Catherine Laporte has shown (2000), this system, which partially eliminates information asymmetry on quality between producer and consumer, ultimately functions as a basis on which to construct a price system : each appellation level and each denomination of origin corresponds to a price range.
19Though the appellation system as an institution “decoupled” from producer interactions seems after-the-fact to be what frames price formation in Burgundy, there is still significant price variability within individual appellations. A given appellation is not sufficient to define or characterize a type of quality for the set of products endowed with it. Quality depends on the technical choices each producer makes within the limits set by each appellation “decree”.
20Indeed, Burgundy domaines
21Though the move from bulk to bottle sales does not raise technical or financial difficulties, it nonetheless requires finding and “loyalizing” customers, as well as being able to fix “fair” prices. This last point leads wholesalers to complain that some producers underprice certain appellations.
22Each producer may nonetheless obtain a great deal of varied information for a given year, information of the sort that enables her to determine her offering price. Above and beyond the information specific to his own company (production costs, product characterization, stock, etc.), there is the professional press, which diffuses studies on consumer trends and competition by sales circuit and country, and specialized guidebooks which regularly publish evaluations of producer wines as well as “consumer prices”; on-site stock quantities are regularly made public, and customers themselves, particularly importers and wholesalers, provide information to producers, etc. The procedure by which offering price is determined involves producer’s taking into account certain elements or signals as well as the way he combines them to make his decision. Given producer diversity, especially in terms of “appellation portfolio” and sale system, there is no reason to think all producers proceed in the same way.
A determining procedure based on “integration variables”
23We wanted to understand how wine producers in Burgundy in all their diversity proceed when it comes to fixing bottle prices posted at the domaine. We were especially interested in testing and specifying the influence of domaines’relational embeddedness on its price-formation reasoning and price levels, while also wishing to consider economic factors in a bi-disciplinary approach. This required :
24Given our aims, these two variables were of interest because they are both economic and sociological; they are also valid indicators of a given business’s relational embeddedness. Given that wines are rarely downgraded, appellation portfolio goes a long way to explaining a domaine’s turnover. But it also represents the business’s “inscription” in a hierarchically ordered market institution, the appellations system, and creates a status dimension. As for sale system, which is characterized by how wine production is divided up between bulk and bottle sales on the one hand, French market sales and exports on the other, it may be considered first as an economic dimension : according to experts, bottling and exporting constitute the most profitable sale system. But this system also has a sociological dimension in that, as a system of market relations, it creates a set of interlocutors and a body of information with which the producer then remains in contact. Appellation portfolio and sale system are thus the “integration variables” that we have used in the following analysis.
Regularities and variants in price formation procedures
Principles common to all producers : adjusting base price and selecting decision-making criteria
25All producers construct their tarif or rate yearly in accordance with a general rule : prices must be relatively stable in order to “keep the clientele’s trust”. Nineteen of the producers we interviewed mentioned this explicitly : “We reason in comparison with previous years to remain stable on prices”
26Producers generally determine a tarif for each distribution circuit (direct sales, export, cafés-hotels-restaurants, etc.). This applies to all appellation levels and possible level subdivisions. It involves establishing a single reference rate :
27Producers’considerations in determining this base rate are extremely heterogeneous. Some correspond to figures internal to the business (production and packaging costs, quantity) or external, readily and less readily accessible figures (inflation, exchange rates, other producers’ prices). Others represent context assessment (state of the market) or a given characteristic (vintage quality) deemed indispensable to proper reasoning, though it is not always possible to identify how the assessment was made. “State of the market” in particular can be associated with overall wine supply, one or more appellation categories, or defined with regard to a given country, etc.
TABLE I. - Citation frequency for criteria used in determining base rate (total of 31 rates; criteria categorized on the basis of respondents’ spontaneous statements)
28We identified :
29When bulk sale rate intervenes in price construction it is always a decisive criterion. Inversely, state of the market, quality, and other producers’ prices may be put forward as 1) decisive criteria, 2) criteria used to fine-tune prices only, or even 3) cited as not included in decision-making. This reference diversity may be explained by assessment difficulty. Some producers also think these criteria are simply not relevant for constructing their base rate : “Vintage quality is not important for price. In any case, younger years are cheaper than older ones”; “Price is not linked to quality. We raised our prices in 2000 despite the year, just because we hadn’t raised them in three years”; “The wine market situation and our prices in comparison to others’ are not important”. The diversity of criteria is explained not solely by market opacity, difficulty accessing information, or calculation complexity. It also seems to reflect divergence in viewpoints about what should be taken into account when establishing prices.
Price determination procedures as distinguished by decisive criteria
Procedures focused on income
30For producers whose decisive criteria are “bulk rates+packaging costs” and “production costs” only (Table I), it is possible to define a precise formal procedure for price calculation : “We calculate our prices in relation to average price per pièce
31In fact, in the first type of procedure producer’s calculation is aimed to attain minimum income. The “bulk price” criterion only makes sense if coupled with an assessment of packaging cost. The alternative of selling by the bottle has to ensure income at least equivalent to what bulk sales will bring in : “We calculate in relation to an average bulk price, plus 10 francs per bottle for packaging costs. This gives us bottle price, which we then multiply by 280 bottles per pièce. That way we know whether we’ll make more profit by the bottle or in bulk.”
32Likewise, when the decisive criterion is production cost, the rate determined has to cover at least production factors : “Sales price is cost price minus rent cost plus selling cost plus margin. We try to promote the value of our work without counting rent cost.”
Procedures attentive to transaction context
33Six producers establish their sale prices with reference to “the others’ prices”, which may be quoted as those of “friends”, “colleagues”, “neighbors”, or “competitors”. In any case, “the others” are not just any others but, as White has suggested, individuals considered “equivalent” in terms of quality and customers. Five producers explicitly mentioned this, one of them explaining as follows : “The rate is a function of quality and vintage renown, then the current vintage on the market and the market in general. But the base is determined with reference to friends who make the same quality wine and are targeting the same clientele.” Though product characteristics are considered explanatory price factors, certain peers are in fact a privileged source of information for eliminating uncertainty about the product.
34Three other producers said they determined their prices first and foremost with reference to the “state of the market”. They referred to different competitive spaces and/or promotion of more or less defined products : “The rate is 100% the result of the market situation, of what the press says about the competition”; “We look at the market situation, product scarcity and demand”; “The rate is determined by the export market; that is, demand, exchange rates, etc.”.
Procedures focused on vintage quality
35Only one producer claimed to establish his base rate exclusively on vintage quality. But his estimate is in reference to the preceding year, and the impact is slight because the producer is also concerned not to vary prices too much : “Every year the price is set by looking at quality for that year compared to quality for year n-1… Still, we try to have a stable rate.” Another producer also cited quality as the main criterion, but combined with stock level.
36Our first level of analysis thus shows diversity of procedures for setting prices within a single “market” –a fact that the standard economic approach cannot acknowledge. The point is now to see how these procedures explain prices charged by the different domaines.
TABLE II. - Classifying price determination procedures by decisive criteria
From analysis of procedures to the social organization of quality markets
Relations among price determination procedure, price, and structures
Procedure and “private customer” rate
37“Private customer” rates
TABLE III. - Price determination procedure and “private customer” rate
Procedure, price and business
38The graph allows us to visualize 1) primary reasoning criterion, 2) price level (from A, high, to C, low), 3) economic value of appellation portfolio,
39Procedures as related to appellation portfolio and sale system bring to the fore three relatively homogeneous business blocks :
40In addition to being correlated at the price level, price determination procedures therefore seem closely linked to domaine’s appellation portfolio and sale system. There are exceptions, however : at equal portfolio levels, for example, price determination procedures may diverge. This is the case for two domaines with average appellation portfolios; one business reasons in terms of production costs and charges high prices; the other decides with regard to other producers’ prices and charges low prices. The explanation perhaps lies in the fact that they use different sale systems : the first bottles 100% of its wine and exports a high percentage; the figures for the second are average. The point now is to understand in greater detail the influence on price formation processes of the two “integration variables”, appellation portfolio and sale system, which are both economic factors and indicators of a business’s degree of relational embeddedness.
From economic variables to the social foundations of quality commodity markets
Behind the sale system, the role of professional relations and routine
41A business’s sale system reflects its relational embeddedness in the market system. Sale system involves the set of interlocutors with which the business has regular relations : importers, if it exports its wine; intermediaries and consumers if it sells its wine by the bottle in France; wholesalers if it sells it in bulk. While sale system has some effect on the information that can be mobilized to determine the base rate, its precise impact can only be discerned by looking at the content and meaning that producers give to their various market ties.
42For producers who say they simply accept the inherited rate or who reason in terms of income, the most frequent exchanges are with customers who have come to the domaine. These exchanges are a “chore” for some, a “pleasure” for others. They do not discuss price with these customers. Exchanges with other types of customers are much less frequent and centered above all on those whom they “have been working with for a long time”. They look to sell their bottles to customers who are of no interest to wine wholesalers; namely employee councils they have met through their grape harvesters. They are not looking for a specialized clientele and therefore do not attend international trade fairs such as the Vinexpo, reserved for well-known professionals, preferring instead to promoting their wine at gatherings such as the Grands Jours de Bourgogne, attended by either the public at large or primarily local inhabitants. They may even favor agricultural fairs, such as the Mâcon fair, over gatherings with other wine producers. Most importantly, the point of these gatherings for some of these producers is “to have a good time with friends”, also “to maintain good relations with fellow producers”, rather than any strategy for finding customers, even less an occasion for discussing price. Moreover, for most of them, bulk wine sales still represent a non-negligeable proportion of their turnover.
43Producers who fix their rates in direct reference to bulk price or production costs later reproduce their reasoning for bulk at the level of bottle pricing. For them, domaine-specific quality is not taken into account, since appellation suffices to define price. The result is that for all these producers, general price level is low, firmly situated at the bottom of the price range for the given appellation. Information about this range is obtained through wholesalers or the village trade union, which they participate in fairly often “to discuss technique with the pals, not business”. For them, bottle price has to guarantee what they consider acceptable income. Specifically, it is the fact that these producers both sell wine in bulk and are dependent on wholesalers that explains their behavior in determining bottle price. But because they proceed on the bottled wine market in the same way as they do on the bulk wholesale market, these producers also suggest the weight of routine in market operations. Some have only started selling by the bottle recently, and now sell much less of their wine in bulk. The decision to sell some by bottle represented a real break in their production habits; it has allowed them to handle the sporadic crises that occur in the bulk market. Still, a change in sale system does not necessarily mean a radical change in the way offering price is reached. Habits must be changed, routine abandoned, and this may require adapting to a new social status, a new profession : “There is a difference between being a winemaker and a winegrape grower”; “We’re a small structure, and we are winegrowers before being salesmen.” The earlier relational embeddedness, linked to bulk sales, led to developing routines that have limited the impact of these producers’ entry into the system of market relations associated with bottled wine sale.
44Two of the domaines that determine offering price with reference to production cost do not follow this logic, and they charge high prices. These domaines belong to major wholesale firms. Their price level may be explained by the fact that 1) the cost taken into account includes remuneration for all production factors–not always the case for independent houses– and overhead is likely to be extremely high; 2) margin is calculated by the parent company, whose primary occupation is sales, or by a specialized salesperson permanently employed by the domaine. The network mobilized to construct price is thus more complex, and leads to sharper adjustment of base rate with reference to production cost : “The rate is set on the basis of cost price indicated by the group’s accounting department […] The group demands minimal margins.”
45For producers stating that they base price determination on state of the market or vintage quality, relational embeddeding within the market system is very different. These producers have a clientele of “loyal connoisseurs” that they maintain by means of regular invitations to private evening gatherings, regular information diffusion on domaine activity, etc. They receive passing customers by appointment only, if at all. They attend the most renowned wine fairs, such as Vinexpo and The London Wine Fair, and often delegate attendance at less prestigious ones to sales personnel. Their main source of income is exports, especially to distant countries, and some of them highlight the role of relations with importers in price determination : “I have real discussions with the importers. They each come here at least once a year.” With importers they exchange information about the market in general but also on “the prices the others charge”, those who are “in the same market slot”. They thus confirm our results with producers who spontaneously explain their prices in relation to others’. Importers are “nodes” within what Karpik defines as “producer networks”. Moreover, producers seek to have both business and friendship relations with them, sometimes going so far as to implicate their families : “They come here, or I go there with my family”; “We call each other often.” In addition to attracting and keeping the buyer, the point here is to strengthen the tie by means of a “multiplex” relation (Degenne and Forsé, 1994) in order to be sure to obtain “the right information”, as Uzzi suggests in his analysis of service businesses. These producers’membership in renowned “cru” trade unions also allows them to meet their direct competitors. Formed by cooptation from among the most prestigious domaines, the trade unions are not really open to bulk producers. Their regional responsibilities in such bodies as the Bureau Interprofessionnel des Vins also give them privileged access to information on colleagues.
46Among producers reasoning in terms of quality, some also refer to “the way the specialized press have spoken of them” to clarify their reasoning, or they mention even more directly their special relations with guide writers, referring not to the Guide Hachette, which nearly all winegrowers have potential access to, but foreign specialists such as Robert Parker. They thus bring to the fore another type of “market professional” who contributes to framing their economic action (Cochoy and Dubuisson-Quellier, 2000): prescribers (Hatchuel, 1995).
47Producers who say they reason in terms of “others’prices” either belong to the preceding group or are in an intermediary situation. Some, looking to develop bottle sales and thereby become independent from wholesalers, observe the prices charged by domaines they want to resemble; this allows them to set their own prices slightly below those domaines’in order to be sure to sell and thus minimize risk. Furthermore, producers who are today ranged with those who reason in terms of the market or quality state that they also began by positioning themselves in relation to others’ prices.
48Producers who reason with reference to others, to the market, or in terms of quality are associated then with a type of relational embeddedness that allows them to set their prices within a specific competitive environment that for some is defined from the outset by their peers, for others is “filtered” by intermediaries or prescribers. Direct or indirect reference to this kind of embedding systematically goes together with fairly high prices–with the exception noted above, a producer saying he is forced to set a low initial rate because of his father’s long sales practice with wholesalers. In contrast to other producers who charge low prices, this producer is working to gradually raise his prices on the basis of major efforts to improve quality, efforts of the sort he first observed being made by his “reputed” neighbors. This last case moves us to enrich the analysis by taking into account the second integration variable, appellation portfolio, to show more precisely how price determination procedure, price level, and company “status” interact.
Price setting procedure and price level as effects of status
49As explained, our study is constructed on the basis of two “integration variables”, indicators of a domaine’s relational embeddedness. Specifically, the point is to show how these variables constitute the basis of domaine status and affect determination of their prices. Expert opinion holds that a high percentage of bottled wine and the practice of exporting it are signs of prestige in quality markets, while selling in bulk is non-prestigious. Domaine appellation portfolio, meanwhile, greatly conditions a business’s potential clientele and the clientele’s relation to the products. Prestigious appellations and therefore prestigious portfolios are what attract well-known restaurants and intermediaries, who are also in contact with renowned wine sellers; renowned appellations are what stimulate the interest of wine guide writers and are ultimately what make final consumers willing to purchase expensive wines. Appellation portfolio also affects assessments by private customers buying at the domaine: attention to quality and product expectations are different at a “small” producer’s who has only regional appellation wine from what they are at a master winemaker’s renowned for his grands crus (Laporte, 2000). Moreover, despite the fact that private customers taste before buying, they seldom have a level of expertise high enough to allow them to distinguish quality by means of organoleptic criteria (Combris et al., 1997). To found their judgment, they place their trust in appellations. Since they are at least partially familiar with the Burgundy appellation hierarchy, portfolio is crucial in defining a domaine’s “reputation”.
50Several producers with a limited portfolio actually say that their clientele does not attach much importance to the particular quality of their products and therefore not at all to vintage. There is therefore no reason for these businesses to take quality or vintage into account as a priority in setting prices. Prices are therefore low, and the first requirement is that price meet producers’ income expectations. It is determined so as to satisfy this objective, as limited by production costs and with possible attention to the alternative of selling in bulk. On the other end of the spectrum, as soon as appellation portfolio allows a producer to stress the quality of his products, production cost no longer appears a decisive decision-making factor and prices are high. Given the potential clientele, production cost is not a constraint; nor does selling in bulk appear a possible solution for ensuring equivalent income. Price determination is then linked to conditions for assessing the quality-price relation on the basis of what peers do; these producers seek to position themselves in relation to peers or assessments by market professionals.
51The two variables –domaine’s appellation portfolio and its sale system– thus work in three ways to define its status and ultimately its price level : 1) they affect the business’s ability to by-pass relations with wholesalers and develop ties with wealthy buyers or reputed prescribers, thus enabling it (or not) to charge high prices; 2) they amount to “signals” for buyers and intermediaries seeking to assess quality, moving them to accept high prices (or not); 3) they condition producer’s readiness (or reluctance) to promote the quality of his products by means of price.
52It should be noted, however, that a valuable appellation portfolio is only associated with prestigious status if the producer actively promotes it, and this involves production practices, business strategy, and being actively integrated into “exchange networks”, as demonstrated by the following case :
5354This case shows how a commitment to quality, enacted and made visible through technical and organizational innovation, allows for constructing a reputation and changing a business’s status, which in turn creates resources for modifying the sale system and breaking with the inherited base rate. This “distinction” strategy (Bourdieu, 1979) leads to product “rarity”, further strengthening domaine status by favoring positive assessment and buyer and intermediary loyalty. Through this example, we confirm one of Musselin’s conclusions in her study of what would seem an extremely different market, candidates for academic hiring (1996). She shows how determining an academic’s “price” at the moment he or she is hired involves recognizing candidate’s current value, namely in terms of the “rarity” of his or her past scientific or scholarly achievements.
55Another example attests to the way a distinction strategy based on innovation can allow for eliminating the constraint of a fundamentally limited portfolio and ultimately increasing status, if the efforts are substantial enough :
56
Categorically refusing to sell to wholesalers, he based his sale strategy first on selling to private customers either at the domaine or by correspondence. He considers his customer network the fruit of fifteen years of persistent public relations efforts, which included receiving passing customers, then “connaisseurs” whom he selected, maintained, and bolstered by the organizing of conferences and tastings. He recently acquired the village château and runs a bread-and-breakfast for wealthy clientele, above all foreigners, to whom he sells by export.
57This case also illustrates Karpik’s understanding of the foundations of quality market functioning, namely the necessity of constant networking, presentation of products and self-promotion if one wants to become known and recognized in exchange networks (Karpik, 1989).
58We can now confirm the positive tie between high status and high prices demonstrated by Podolny and Uzzi, among others. But we still need to account for the importance of routine and of moral values. Moral values may operate as a strong constraint when it comes to promoting wine quality or getting beyond relational embeddedness, above all changing status; they also have the effect of distinguishing between different types of company projects : not all Burgundy wine producers seek to resemble “the elites”.
59The last case reminds us that a producer can in fact choose not to activate his status in the wine world, a decision that can be justified by the desire to affirm certain moral values :
6061This case overtly expresses what certain bulk producers already suggested in presenting themselves unabashedly as “small producers” much more interested in production activity than business operations. It also confirms the principle of a wine market marked by an economy of moderation, a principle that was already visible in the procedure common to all producers of only slightly adjusting base rate.
62Clearly, relational embeddedness is complex. Ties with various types of market professionals and within local networks are more or less decisive, while status appears a variable of choice that is as closely linked to individuals’ practices and projects and their moral values as to their resources. Trying to understand the “meaning” of these variables for suppliers enables us to reach a few conclusions that both confirm and supplement the results of other studies, Anglo-Saxon and French, while opening avenues for further research.
63How do businesses reach an offering price for their goods in the context of a quality market, where there is uncertainty about both product quality and the quality-price relation ? Fifteen years after Karpik’s pathbreaking article (1989), this question has not been studied much as a process or by means of an approach in both economics and sociology terms. As mentioned, this is what led a group of French researchers to call for research into this theme (Musselin et al., 2002). This article contributes to the debate by presenting the results of a comprehensive approach to price formation for Burgundy wines. In line with economic sociology postulates, we have apprehended the process as the product of a decision-making procedure on the part of a seller, a procedure influenced by sellers embeddedness in the social relations system.
64Our survey of a diversified sample of producers leads us first to confirm the principle of an “economy of moderation” put forward by Lucien Karpik, in the sense that all producers only slightly adjust last year’s price when determining their offering price, motivated by a concern to keep clientele trust since clientele is in part composed of friends. But this does not preclude diversity in price determination procedures within a single market, which in turn calls into question the standard economics hypothesis that all businesses choose to proceed in the same way (Guerrien, 1992). Producers’ price determination procedures and price levels appear closely linked to two differentiating variables : appellations portfolio and sale system. This study makes special use of these variables as potential indicators of producers’ relational embeddedness. Seeking to understand the variables’ “meaning” for suppliers allows us to show that these variables not only reflect and frame their market relations but also work to define their status, which in turn influences their price decisions, namely because it affects their ability and disposition to promote the quality of their products. However, we have also identified routines that condition the nature of relational embedding, status level, and the influence of the two factors, as do distinction strategies linked to an affirmation of moral values. In illustrating how these strategies may either provide the means to enhance status or be associated, on the contrary, with a claim to modest status, our study accounts for dynamic markets that are neither systematically reproduced or passively accepted, where positions are both mobile and the result to some extent of seller-actor’s will. In this sense it allows for enriching Anglo-Saxon sociologists’ relatively structuralist approaches, such as that of Podolny and Uzzi, which call for paying closer attention to intramarket status mobility due to innovation (Podolny, 1993).
65It seems worthwhile to return to an approach based on “integration variables” that are both economic and sociological in character. Appellation portfolio and sale system are “intermediate objects” (Vinck, 1999) in this interdisciplinary contribution to analysis of economic action and, more broadly, to understanding of the social organization of quality markets. This encounter around objects may be continued in a way that more directly combines an approach to production markets of the sort proposed by White with analysis of institutions and conventions as constructed through interactions of the sort that also cause institutions and conventions to evolve. This will enable us to better apprehend the “intertwining” of different types of embeddedness–relational, cultural, and institutional– in economic action (Le Velly, 2002). Given the theoretical connections proposed by conventions economists (Favereau et al., 1994), the case of winegrowers seems to us good material for advancing in this direction (Garcia-Parpet, 2001; White, 2002b). The research can continue with a comparison of Burgundy and Languedoc, the latter associated with greater uncertainty about quality, weak institutional “decoupling” as a frame for supply diversity (Touzard, 2000), and what may be assumed to be different types of relational embeddedness.
66Translation : Amy Jacobs
Previously published : RFS, 2004,45,4
Annexe
APPENDIX
Domaine selection method and main characteristics of the sample
67The domaines studied were selected from a data base made up of Burgundy operators’ on-site “private customer” price lists in December 2001. All Burgundy domaines that sell by the bottle have these “private customer” lists because they sell at least part of their stock through direct on-site sales. The data base contains information on 20,367 wines from 1,245 Côte d’Or, Yonne, and Saône-et-Loire domaines. Two wines are considered different from each other when they figure as a distinct entry on a “private customer” price list. The data base provided the following information on individuals wines : price (tax included), geographic appellation, hierarchical level, color, vintage, domaine name.
68All domaines in the data based were characterized by their “private customer” lists :
TABLE IV. -
69The 28 domaines surveyed were chosen on the basis of expert advice in such a way as to reflect structural, appellation portfolio and sale system diversity.
70For sale systems, the sample encompasses the main points that producers have in common within the Burgundy region as well as the points on which they diverge :
Yuna Chiffoleau
Unité Mixte de Recherche – Innovation Institut National de la Recherche Agronomique (INRA) 2, place Viala – 34060 Montpellier cedex I – France chiffole@ensam.inra.fr
Catherine Laporte
Unité Mixte de Recherche – CESAER Établissement National d’Enseignement Supérieur Agronomique de Dijon (ENESAD) 26, boulevard Dr Petitjean – 21079 Dijon cedex – France
Notes
[ *] This article is a product of the deep friendship between its authors. Catherine Laporte left us suddenly on January 26,2005. I dedicate the English version of our text to her passion, generosity and talent (Y. Chiffoleau).
[ (1)] In the latter connection, however, we highly recommend Dubuisson-Quellier and Neuville (2003), a set of contrasting case studies whose authors, a group of sociology and management science researchers who favor a “pragmatic” use of the economics of convention and sociology of sciences, analyze arrangements for judgment framing that favor economic exchange in contexts of uncertainty about quality.
[ (2)] Different authors (see Steiner, 1999; Chantelat, 2002) understand the various “heterodox” economics approaches (economics of convention, neo-institutionalist economics, regulation theory, etc.) to be part of the field of new economic sociology, though opinion differs on which approaches fit and which do not.
[ (3)] Wineries are private businesses that produce and sell wine from a combination of estategrown and purchased grapes.
[ (4)] The decree covers only which varietals may be used, how to tend grapevines, maximum grape yield, and minimum alcohol content.
[ (5)] This term is reserved for companies that produce and sell vine from their own vineyard; it thus exclude cooperatives. There are approximately 2,800 domaines, which together represent 76% all Burgundy wine produced. 8% of the official surface area belongs to wine wholesalers; this represents 25% of grands crus areas.
[ (6)] This observation is consistent with marketing research hypotheses in which price determination is understood as a specific business capacity (Dutta et al., 2003).
[ (7)] Reference prices are defined for each Burgundy appellation and are published monthly by the Bureau Interprofessionel des Vins de Bourgogne.
[ (8)] All quotations are transcriptions of statements or comments from Burgundy wine producers.
[ (9)] Only three businesses set two different rates from the outset. We therefore have a total of 31 rates for the 28 domaines surveyed.
[ (10)] A pièce is 228 liters of wine in bulk, the equivalent of 300 bottles.
[ (11)] Our results show that “private customer” rate can more broadly be considered an indicator of producer prices, independently of year and sales circuit. This hypothesis was confirmed by several wine wholesalers.
[ (12)] Appellation portfolio value is calculated by annual average of bulk rates for appellations making up domaine land, weighted by surface area covered by each appellation.
[ (13)] One domaine was not taken into account because we lacked sufficient information on it.
[ (14)] Usually sale in bulk is considered less profitable but a surer financial guarantee than bottle sales : with bulk one is protected against cash flow problems.
[ (15)] Following the idea in Favereau et al. (1994) that “markets” as defined by White are associated with specific conventions, these two cases may lead to the identification of a specific market among businesses that determine price with reference to cost, a market linked to “merchant worth”. In our future studies we intend to look in greater detail into the question of profit margin, to specify how it is calculated and how it may define particular conventions and markets.