GLOBALIZATION OF MARKETS FOR CONTEMPORARY ART: Why local ties remain dominant in Amsterdam and Berlin

Since the 1990s¹ , a global institutional framework of art fairs, art biennials and commercial institutions such as galleries and auction houses has emerged outside of the traditional centers of the art market, Europe and the United States. This has enabled global commerce in art in which the importance of physical distance between consumers, distributors and producers of art seems to be weakening. Cross-border traffic of art, artists and art institutions, not only within the traditional centers, but also between those centers and peripheral regions such as Asia and Latin America, is indeed on the rise (see e.g., Artprice 2008; McAndrew 2008). Moreover, in international art market rankings, regions such as China and India, which used to be peripheral until the 1990s, now figure prominently.² A growing, if still scarce, literature on the globalization of ‘high culture’ has shed some light on some aspects of such cross-border traffic of art, focusing on institutions and actors such as museums (Bevers 1995; Quemin 2006), biennials (Buchholz and Wuggenig 2006), curators (Bydler 2004), or arts journalism (Janssen et al. 2008). Moreover, sociological and anthropological studies of the contemporary art market (see e.g., Moulin 1967 [1987]; Plattner 1996; Velthuis 2005), have focused on issues such as valuation procedures (Beckert and Rössel 2004), gatekeeping roles (Moulin 1967 [1987]; Bystryn 1978; Crane 1987), pricing (Velthuis 2003), market structure (Bystryn 1978), or status dynamics (Yogev 2010). However, despite the importance of such research areas, systematic studies bringing together analyses of globalization and its impact on art galleries are so far lacking, or have been limited to the participation of galleries in international art fairs (see e.g., Quemin 2008).

In spite of this lack of research, bold claims have been made on the globalization of high arts and of culture. Some have argued that the high arts have been globalized since the late nineteenth century (Waters 2001), that is much earlier than popular culture, and that ‘in almost no other sphere of culture is the shrinking of North and South, of East and West so intense as in the fine arts’ (Kramer 2001: 178). By contrast, others, including the late French sociologist Pierre Bourdieu, have declared cultural globalization a ‘myth’ (cf. Buchholz and Wuggenig 2006). They point at the continuing hegemony within the global art world of artists born in the United States and a small number of European countries, while artists from Africa, Latin America and Asia are hardly represented.

This article seeks to redress a gap in the literature on globalization and the primary market of art galleries, as well as shed some light on the broad range of arguments made on behalf of the globalization of culture and the arts. It offers an exploration of the cross-border traffic of artists within the primary art market; the segment of the market where new works of art are sold for the first time. Drawing upon both qualitative and quantitative data it investigates the nationality of contemporary artists represented by art galleries in Amsterdam and Berlin asking the following questions: To what extent are artists born in outside of Europe and the United States visible in these galleries? How should the regional focus within these commercial institutions be explained? And how should differences across emerging and established galleries in Amsterdam and Berlin be accounted for?

In the first section of this paper, the data and methodology are presented that will be used to test these bold claims.³ In section 2, quantitative data indicate that the global orientation of Amsterdam and Berlin galleries is partial at best. In section 3, two mainstream explanations as well as their limits are discussed for the continuing dominance of Western artists: a cultural affinities model and Wallerstein's (1974) core-periphery model. In the final section, I develop a complementary explanation in terms of organizational practices, business repertoires and role models within market settings.

Quantitative data for this study have been collected through artfacts.net, a commercial data provider to the primary art market, established in 2001. This website contains information about 234,190 artists, 223,293 exhibitions in 19,672 exhibition venues such as museums and art galleries spread over 169 countries.⁴ The data are provided by art institutions, art galleries, and individual artists and are further collected and verified by the website staff.

A list was compiled of all 136 Amsterdam private art galleries on artfacts.net. Of the 575 private art galleries listed for Berlin, 115 were selected randomly. Subsequently galleries that did not represent any artists, or predominantly sold work by deceased artists, were excluded from the dataset. Seventy-nine Amsterdam and 73 Berlin galleries remained, which represented a total of 3072 artists. For these artists, the country of birth was collected. Only for a small group of artists (495 out of 3072), their current city and/or country of residence was known. These data will be used to lend further support to findings.

For each gallery, data regarding participation in international art fairs were collected in order to determine their reputation within the art market. A gallery was considered established if it participated, at least once, in either Art Basel or Art Basel Miami Beach between 2000 and 2008, which was the case for 17 galleries (11%). These two fairs are widely regarded, both by my respondents and by observers of the art market, as the most prestigious fairs (see e.g., Thornton 2008). Their highly selective application process gives participating galleries a strong reputation signal (e.g., Quemin 2006, 2008). Unlike local reputation signals, such as membership of a national gallery association, this reputation signal is widely acknowledged both in Europe and the United States.

No longitudinal data regarding the group of artists represented by a gallery can be retrieved from the website, so the data necessarily present a snap-shot of the current situation rather than (unfolding) trends. Selection biases may exist because not all galleries are listed on the website and because some may be more keen than others to ensure their data are complete and up to date. In order to identify such potential selection biases, for Amsterdam galleries data on the nationality of the artists have also been collected from gallery websites and email contact with the galleries. This ‘shadow-dataset’ generates results that are approximately equal to those presented in section 3 (for results, see Visser 2009), indicating that the selection bias in the data of artfacts.net is marginal.

The production of qualitative data for this article started with a dataset of 20 Dutch and 39 German art dealers who were randomly sent an interview request in the spring, summer and fall of 2009. A total of 24 art dealers responded, and 21 of these were interviewed (14 in Berlin, seven in Amsterdam).⁵ Semi-structured interviews were conducted which lasted between 0.5 and 1.5 hours, always in the galleries’ premises.

Far from being a global world, with artists from all regions being represented by Western dealers, the quantitative data suggest that local ties predominate in both cities. Firstly, on average, Amsterdam galleries represent 21.3 artists, of whom at least 9.5 (45.2%) are foreign; Berlin galleries represent slightly fewer artists (18.5) of whom at least 9.1 (49.8%) are foreign.⁶ The nationality of these foreign artists shows the presence of a strong home bias (Table 1). Of the 1234 artists represented by Amsterdam dealers of whom the nationality is known, 39.1% is born in The Netherlands. In Germany, the home bias is even stronger (43.6% of 1157 artists). The current residence status of artists indicates that of the 250 artists represented by Amsterdam galleries, 36% lives in The Netherlands. For Berlin galleries, the percentage of artists who live in Germany is 46.5% (n=245).

Secondly, the dominance of American artists is striking, ranking third and second, respectively, in Amsterdam and Berlin. Thirdly, it is noteworthy that China and Japan are the only non-Western countries to appear in the top 10. In Berlin, all countries included in the list are members of the Organisation for Economic Co-operation and Development (OECD). Artists from peripheral regions such as Latin America and Africa are hardly represented by Amsterdam and Berlin galleries, accounting for around 1% and just above 2% of the total, respectively. The only exception is Asia, which in Amsterdam galleries is ranking second after Europe. Galleries do in other words operate internationally, but it would be an exaggeration to claim that their business is truly global.

These findings are in line with the few empirical studies on the globalization of contemporary art. French sociologist Alain Quemin, for instance, studied the composition of artists whose works are represented in the Fonds National d'Art Contemporain: France's main public collection of contemporary art which contains about 70,000 works of art. Five countries – the United States, Germany, the United Kingdom, Italy and Switzerland – account for 60% of all acquisitions for the museum collections between 1991 and 2004 (Quemin 2006). Artists from emerging countries such as Russia, China, India or Brazil are represented with a handful of acquisitions each. Similar patterns are found at exhibitions in major museums such as the Tate Gallery in London, the Centre Georges Pompidou in Paris, and the Museum of Modern Art in New York.

Similarly, German sociologists Ulf Wuggenig and Larissa Buchholz point at the continuing hegemony of a small number of Western countries. For instance, in the Kunstkompass, an annual top 100 of artists based on their reputation within the art world, published by the German business magazine Kapital, artists from the United States and Europe have invariably dominated this ranking from 1970 to 2005. Although their share has slightly decreased since the mid 1990s, it has never fallen under 80% over the 35-year time span.

Although never applied to geographical patterns within the art gallery market, two models of cultural globalization explain the persistent dominance of a small number of Western countries. The first or ‘cultural affinities’ model, focuses on the diffusion processes of cultural goods, and argues that it depends on the cultural affinities and similarities between producers and consumers of cultural goods. This model has been predominantly used to explain why American television series and Hollywood movies are shown and appreciated to a far lesser extent in Europe and non-Western regions than is often presupposed: ‘audiences generally prefer local programs, because they find it easier to identify with the style, values, attitudes and behaviours expressed’ (Hannerz 1992; Sinclair et al. 1996; Bielby and Harrington 2005; Crane 2002: 10).

According to this model, contemporary art made in non-Western regions would fail to be exhibited and sold in the West as a result of the cultural disparities between these regions. Thus, Japan, in spite of its economic wealth, holds a peripheral position in the art world because it does not have the adequate occidental cosmology (Buchholz and Wuggenig 2006). When applied to the present case, this model would explain how the interpretational repertoires of the Western audience, which Amsterdam and Berlin galleries are directed at, do not allow for an easy appreciation let alone acquisition of art created outside of the West. Aware of this cultural disconnect between audience and artist, art dealers may be hesitant to exhibit non-Western art.

This ‘cultural affinities’ model finds some support in my quantitative data.⁷ It explains the strong presence of Belgian artists in Amsterdam galleries, since in Flanders, in the northern part of Belgium, the same language is spoken as in The Netherlands and the two regions have a partially shared history and identity. It also accounts for the strong presence of Polish, Czech, Lithuanian and Russian artists in Berlin's galleries, as these countries have historic cultural and political ties with Germany and continue to host German-speaking minorities. Similarly, the significantly larger share of Swiss and Austrian artists in Berlin galleries makes sense given the cultural and linguistic affinities between the two countries. However, not all the differences can be understood by means of the cultural affinities model, such as the strong position of France or of Italy, in the case of Berlin. Moreover, particularly striking about the data are the similarities between the patterns found in both countries (cf. Meyer 1997), namely, the dominance of artists born in the United States and in the gallery's home country.

If ‘culture’ has limited explanatory power empirically, it also fails to make sense theoretically. It may help explain how language may be a barrier for cross-border traffic, and how references to local societal or political circumstances may prevent one national audience from appreciating popular culture produced abroad. However, contemporary art does not present such linguistic and cultural barriers, but it is part of a global visual language (Zijlmans and Damme 2008; Belting and Buddensieg 2009).

A second, rival explanation for the patterns I find is the core-periphery model, which sees the dominant position of American and European artists in the Amsterdam and Berlin gallery scenes as a reflection of the persistent political and economic power of these regions vis-à-vis other regions of the world (Wallerstein 1974; see e.g., Quemin 2006; Barriendos 2009). Cultural globalization is thus equaled to westernization or Americanization, leading to a further concentration of market power with a small group of Western actors and institutions.

Empirically, this model might explain why American artists rank third and second, respectively, in Amsterdam and Berlin galleries, why artists from non-Western peripheral regions with little economic and political power are represented; as well as the rising economic superpower China as the only non-OECD country to make it to the top 10 list. But, like the cultural affinities model, it has several limitations. Theoretically, it fails to explain how economic and political power on a macro level, translates into decision-making processes by art dealers on a micro level; why, in other words, art dealers would prefer to select artists born in these core countries. Also, it does not explain why Japan, the world's second largest economy, only ranks 8th and 9th, respectively, in the top 10 list of artists represented by galleries in Amsterdam and Berlin (see Hannerz 1992 for a related critique of the core-periphery model). Moreover, the relatively strong presence of artists born in small countries such as Belgium, Switzerland, or Austria (which would be even stronger once the numbers of artists would be corrected for population size) does not make sense from this perspective. The strong home bias of art galleries is difficult to understand by means of a core-periphery model. In short, whilst the cultural affinities and the core-periphery models explain part of the patterns found within Amsterdam and Berlin gallery programs, a complementary explanation is needed.

My alternative explanation for the home bias of Berlin and Amsterdam galleries is not based upon cultural or power differences, but on organizational practices, institutional specificities and role models of actors within markets. The argument here is that to understand transnational flows of cultural goods, the specific models, institutions and practices that characterize each cultural industry – from movies to architecture, from popular music to visual arts – need to be taken into account (cf. Kuipers 2011). Both the cultural affinities and the core-periphery models, with their implicit universalistic assumption that one theory can account for differences across all cultural industries, fail to do so. In particular, I argue that widely diffused organizational practices and institutional characteristics of the contemporary art market account for the limited extent to which non-Western artists are represented. In other words, the predominance of local ties is the result of institutional isomorphism which pose at least six distinct barriers to globalization (cf. DiMaggio and Powell 1983; Meyer et al. 1997).

The first barrier is related to search processes. Unlike in many other cultural industries, no centralized or quasi-centralized venues, transfer markets or institutions exist where producers and distributors of art meet and business is done. Art fairs do not fulfill this role, since they function predominantly as meeting places for galleries and their clientele. They are avoided by artists since, as the American artist Paul McCarthy stated about Art Basel, visiting the fair as an artist is like ‘watching your parents having sex’.⁸ This means, especially in the case of upcoming dealers, that they look for artists to represent in institutional isolation. In doing so, they have strong local informational advantages. Since they frequent local art schools, studio's, museums etc. more often than those abroad, they are much better informed about the local supply of artists who fit their own program or preferences in both artistic and commercial terms. This informational argument holds especially for young, starting galleries, which need to ‘handpick’, as one interviewee put it, all their artists. Art dealers recounted how they had encountered ‘their’ artists in local situations such as at a traffic light in town, through side jobs, or through mutual friends. Three young dealers I interviewed explained that their galleries grew out of a group of befriended artists who were already showing together at a non-profit exhibition space. In other words, the gallery emerged out of local ties.

After an artist has been selected, a dealer may not be willing to engage in a long-term business relationship with him/her right away. As art dealers frequently emphasize, apart from an artistic fit, a social ‘click’ needs to exist as well (Plattner 1996; cf. Velthuis 2005). In order to find out if it does, i.e., if their expectations match and the temperaments of the artist and the dealer work out together, repeated face to face encounters may be warranted. This courting process, dealers emphasized in my interviews, is facilitated if the physical distance between dealer and artist he is eying is limited. Since artists are usually added to the program on the basis of recommendations by other artists who the gallery already represents (see Plattner 1996; Institut für Kunst und Medienmanagement 2004), this initial local orientation may subsequently persist because of path dependent effects: local artists recommend artists from their own network who are likely to be locals as well. As a result of these search processes, networks effects, and local informational advantages, artists from non-Western regions, even if they ‘speak’ a global language of art and appeal to Western audiences, may remain unknown to art dealers and thus fail to be represented in Western art capitals.

Once galleries are more established, this local search procedure becomes less important: they are more likely to abstain from ‘picking up a young artist from art school’, as one dealer put it. Instead, they focus on working with artists who enjoy public visibility; for instance, because their work is part of museum shows and prestigious curated exhibitions, such as one of the many Biennials for contemporary art, or because they have been profiled by the international art press. Subsequently, they try to lure these artists away from the galleries, which represent them.

The division between upcoming and established galleries, which has frequently been discussed in academic literature (see e.g., Bystryn 1978; Santagata 1995; Velthuis 2011) is, in other words, homologous with a division between local and global search processes. This homology finds some support in the quantitative data: 34.8% of the artists represented by established art dealers in Amsterdam and Berlin were born outside of Europe. For upcoming art dealers, this percentage was significantly lower: 28.2%. Also, the home bias is slightly weaker for established galleries: 63.6% of the artists they represent was not born in the country where the gallery is located. For upcoming art dealers, this percentage was 58.2.

Secondly, the support system of art galleries poses barriers for representing foreign artists. Whereas established galleries, which represent artists with an international reputation and four or five-digit prices for their works, can afford to have several or, in rare cases, dozens of employees, easy access to financing in order to buy works for inventory, and a cash-flow that is stable and large enough to pay for various professional services, the financial situation of starting galleries is frequently so fragile, that artists need to provide practical services to the gallery in order to safeguard its survival. During one of the interviews I conducted, for instance, which took place in a new space the gallery was about to relocate to, two artists were painting the walls. My respondent, like other dealers, emphasized that his gallery was a ‘communal enterprise’ in which ‘every helping hand’ is welcome. Dealers also explained that they needed the artists close by so that they could at least install their own shows in the gallery space. Artists need to live within close proximity to their galleries in order to be able to provide them with such practical type of support. For starting galleries, mundane factors such as transportation costs further restrict their willingness to represent foreign artist.

Thirdly, role-definitions of how dealers interact with ‘their’ artists and collectors, which have emerged historically (see e.g., Velthuis 2005) and are only slow to change, restrict the global orientation of galleries. Most crucially, the dealer sees it as his/her task to support artists materially, morally and artistically. My respondents said they wanted to ensure that artists could work comfortably, without being disturbed by practical concerns. They were ready to engage in discussions with their artists about artistic issues and were eager to know how their work was progressing aesthetically. This both caring and curatorial role, they admitted, was easier to enact for artists ‘who live around the corner, where you call and pass by right away’, as one of them put it, than for those living in other continents:

The existence of these barriers, which are related to the necessity of dense interactions and mutually supportive relationships between artists and their dealers, is supported by a rich literature in geography on the clustering of cultural industries (see e.g., Scott 1999, 2006; Cooke and Lazzeretti 2007). This proximity provides organizational flexibility and makes it easier for actors within cultural industries to adapt to permanently changing market circumstances. In a study of the Young British Artists (YBA) in London in the 1990s, While (2003) emphasizes this density of social networks and associations between local artists, gallery spaces and collectors as a prerequisite for their success. My data indicate that especially Berlin is another cultural cluster, which serves as a magnet for artists and other creative actors (see also Institut für Kunst und Medienmanagement 2004). Of the 250 artists represented by Berlin galleries whose current city of residence is known, 33.9% lives in Berlin. This percentage is a smaller albeit still substantial in Amsterdam, with 19.6%. Recurrent interactions and clustering of dense artist-gallery networks are also warranted to make often subtle changes to the artist's output, whether in terms of style, material, subject matter or number of works produced. Given the high value placed within the art world on the artist's autonomy and the sensitive nature of outside influences on his/her output (see e.g., Becker 1984), such changes can be communicated more easily through face to face interaction.

Fourthly, clustering may also be informed by the nature of the product involved: given the highly uncertain value of the contemporary art which upcoming galleries seek to market and the strong efforts galleries need to make to establish its artistic and commercial value (see e.g., Moulin 1994; Beckert and Rössel 2004), these art dealers need to have detailed knowledge of the aesthetic qualities of the works of art and of the artists’ motives to produce them. In order to acquire such knowledge from artists and to subsequently pass it on to collectors, face-to-face interactions may be crucial. By contrast, the uncertainty regarding the artistic and economic value of contemporary art marketed by established galleries has already been significantly reduced. These works have already been consecrated, to put it in Bourdieu's (1993) terms. They may have been exhibited in museums and written about by art critics. Moreover, works by the same artist may regularly be sold at auction, which function, as one art dealer put it, as a ‘barometer of value’. Others therefore referred to this class of art as ‘blue chip’. During the interviews they said that they frequently sold these works after having emailed a jpeg file to an anonymous buyer. Buyers could risk spending four- or five-figure price tags on pieces they had never seen in person since they knew well what the piece would be like and what its value is. As one of the directors of Berlin's most established art gallery put it:

Because the value of these pieces has already, to some degree at least, stabilized, face-to-face interactions between the artist, dealer and collectors are deemed less important.⁹

In more general terms, economic geographers like Michael Storper have noted that face to face interaction remains important, even in a globalized economy, because of the costliness of exchanging dense information related to among others deal-making, evaluation of products, motivating trading partners, and relationship adjustment (Storper and Venables 2004: 352). Fifthly, another characteristic of art dealers’ role model which produces local rather than global outcomes, is their aim to create a community of critics, collectors, fellow artists and other members of the art world around the artists they represent. In order to market art, fostering a tie between the collector and the artist is particularly important. Acquiring art from a gallery ‘is an act that not only confers the property rights on an object, but also grants access to a club distinguished by a specific aesthetic quality’ (Hutter et al. 2007: 249). One of the main instruments to create these communities or clubs, is to organize social events, such as visits to the artists’ studio, or dinners in which artists and collectors of the gallery join. One dealer said:

Although not impossible to achieve with the artists living around the globe, it is much easier, dealers emphasize, with the artists living close by.

A sixth, final organizational reason why local ties persist, is the importance of trust relationships within the art market and the face-to-face interactions they require, especially in the case of upcoming art dealers (see e.g., Moulin 1967 [1987]; Plattner 1996; Bonus and Ronte 1997). Although standard, legally binding contracts exist for consignment relationships, these are often conducted without them. Firstly, an artist cannot be contractually enforced to continue producing valuable works of art in the future. Secondly, if a contract is written up, it is difficult for one party to monitor the other and ensure that all terms of the contract are complied with. Thirdly, litigation is expensive in case of breach of contract and may damage the reputation of the artist and the gallery. Transaction costs involved in developing a contract, monitoring, and litigation, are, in other words, high in case of consignments. Finally, it is questionable whether a dealer can expect financial compensation from successful litigation. For these reasons, dealers often try to engage in long-term trust relationship with their artists, which serve as an alternative for imperfect contractual relationships (cf. Kollock 1994; Caves 2000).

Geographical distance does not always need to be a barrier to establishing such relationships. Dealers emphasize that they are frequently in touch by email or phone with the artists they represent, no matter where they live. Akin to the global micronetworks discussed by Knorr Cetina and Brügger (2002: 941–2) in financial markets, trust is further enhanced by reciprocal long-distance gift exchange between artists and dealers. A dealer may, for instance, assist the artist in practical matters such as acquiring canvases and other materials or shipping works of art, and, in exchange, an artist may provide the dealer with valuable information about upcoming artists who are looking for a (new) gallery to represent them. Also, especially at the higher end of the market, roles and routines, obligations and mutual expectations of artists and dealers are standardized and stabilized to such an extent that face-to-face interaction is no longer necessary to maintain these trust relationships. Here trust takes an institutionalized form (Zucker 1986; Dyer 2000). As one established German dealer, who represents artists from among others Iran, Peru and Poland, responded:

But especially for starting galleries, both artists and dealers are inexperienced and, do not interact on the basis of stabilized routines. Trust here takes the personalized form, which means that dense face-to-face interaction (for instance at openings of gallery shows and other local art events) is needed to establish trust relationships. Subsequently, if mutual expectations no longer seem to match, or one is upset about the other's behaviour, a face-to-face meeting (for instance over lunch or dinner) may be necessary to realign expectations and talk things out (cf. Storper and Venables 2004: 356).

In short, a small group of established, ‘consecrated’ art dealers may have the financial resources, global networks of collectors and curators and access to the world's most important art fairs to market artists from all over the world. This reputation and the resources that come with them, enable them to select the artists they represent outside of the confines of their own local art worlds. However, these established galleries which actively participate in a global market are only slightly more than 10% of the entire gallery population (17 out of 152 in my dataset). The vast majority may be willing to represent foreign artists, and may even offer them incentives, because being able to offer an international program at commercially attractive art fairs gives the participating galleries a good reputation in the art world. In reality, however, the role models of dealers, the local support systems their galleries rely on, the way they select their artists, and the personalized trust relationships they engage in, constitute barriers to a global orientation.

The art market is another instance where globalization does not result in what has been called ‘deterritorialization’ (Appadurai 1996) or a waning importance of physical distance. Instead globalization here leads to an emphasis on local ties and a re-evaluation of physical distance (see e.g., Cox 1997; Sassen 2000; Storper and Venables 2004; Faulconbridge et al. 2007). This argument contrasts with recent media representations of art markets in the last decade: the topoi of multimillion dollar works of art created by Chinese and Indian artists, newly rich collectors from Russia snapping up pieces at auction, and established Western art institutions such as the Guggenheim Museum and the Louvre expanding into the Middle East. The findings presented in this article also contrast with the ideology of the global which western art dealers and other actors in the contemporary art world have long embraced. My respondents in Amsterdam and Berlin, for instance, invariably claim to disregard nationality and to focus on quality exclusively when selecting the artists they represent. From a Bourdieusian perspective, this shared ideology is hardly surprising: the global – equated here with selecting art on the basis of quality exclusively – can be understood as an opportunity for actors within the art world to deny the economy, to express their distance from branding art based on nationality, and by doing so, to accumulate symbolic capital (Bourdieu 1993).

This study does not want to deny that Western and non-Western regions are to a greater extent than before, becoming part of a global art world. But quantitative data obtained from artfacts.net suggest that this process is at best only in its initial stage when it comes to the promotion of non-Western artists and the marketing of their art in European art capitals. Amsterdam and Berlin art galleries choose instead to represent artists who are born in The Netherlands and Germany, respectively, and who live close by. Neither does it deny that globalization has strongly influenced how commerce in art is shaped. However, this influence has occurred in a different manner and for a much longer time period than has so far been suggested. What has been diffused is a modified version of a gallery model which originated in France in the nineteenth century (White and White 1965). This model, with its local search processes, trust relationships, and mutual support between artists and dealers they nowadays entail, poses boundaries to the art dealer's global aspirations and therefore produces local outcomes. Here a parallel may be drawn with the media industries, where the global diffusion of television formats such as the soap opera, has resulted in locally oriented cultural centers (see e.g., Sinclair et al. 1996). This is especially the case for upcoming galleries, which market art with a highly uncertain artistic and commercial value, and represent artists without a firmly established reputation; face-to-face interaction with ‘their’ artists is a prerequisite. Overall, further research should be conducted to evaluate the extent to which non-Western countries such as Russia, India and China that have recently integrated into a global art world, adopt these gallery models. If they do, paradoxically, further localization is to be expected.

Olav Velthuis is Associate Professor at the Department of Sociology and Anthropology of the University of Amsterdam. He is the author of Imaginary Economics (NAi Publishers, 2005) and Talking Prices. Symbolic Meanings of Prices on the Market for Contemporary Art (Princeton University Press, 2005), which received the Viviana Zelizer Distinguished Book Award of the American Sociological Association for the best book in economic sociology (2006). Velthuis is currently studying the emergence and development of art markets in the BRIC-countries (Brazil, Russia, India and China). Before, he worked for several years as a Staff Reporter Globalization for the Dutch daily de Volkskrant.