What Does Developing a Ranking of Leading Contemporary Art Galleries Unveil about the Importance of the National Factor? An Analysis of Art Basel Art Fair

: Since the French sociologist Raymonde Moulin developed her pioneering research in the 1960s, the art market has been continuously studied by social scientists. For several years now, art market studies have rapidly proliferated. Collectors and collections, though, have tended to draw more attention than professionals and dealers, which remain less analyzed. In this article, we intend to study the impact of the national factor on the gallery sector by using a ranking of the leading contemporary art galleries in the world that we ourselves developed. Having analyzed the construction of the most signiﬁcant rankings in the art world that all focus entirely or partly on artists, we decided to create one for contemporary art galleries. The methodology that we used will be presented in the ﬁrst part of the article. In its second part, we will analyze the territorial / national dimension that can be identiﬁed in the most important art fair in the world, Art Basel. We will comment on what is revealed by our ranking in terms of countries of origin associated with their share and diverse positions in the ranking. Finally, we will address the contribution that our ranking brings to the knowledge of so-called globalization.


Introduction
When Raymonde Moulin developed her seminal studies on the art market in the 1960s under the title Le marché de la peinture en France (Moulin [1967] 1987), 1 she limited her analysis to the French context. At the time, France still played a leading role on the international scene and on the market. Although it may sound hard to believe today given the later development of the international art market, Christie's and Sotheby's were minor players compared to French "commissaires-priseurs", the national equivalent of auctioneers. As the art market, in general, was highly concentrated in France, Raymonde Moulin focused her analysis on the French scene and market. Although the international context had evolved radically when she published her second very important book L'artiste, l'institution et le marché-which can be considered her masterpiece but unfortunately remains unpublished in English-in the 1990s (Moulin 1992), she mostly analyzed the French case once again. In this second book, she notably analyzed the steady development of the support of the French state to contemporary art in the 1980s, a direct consequence of the election of a left-wing president for the very first time in 1 Two decades later, an abridged English version of the book was published under the title The French Art Market. A Sociological View (Moulin 1987).
France after the second world war. Many State supported institutions were created and flourished during the 1980s and 1990s. In her analysis, Raymonde Moulin was especially interested in the process of the social creation of art value. She underscored how art value is created at the junction of institutions and the market. At the time, institutions, especially public ones, played a major role in the creation of art value. The mutual part of institutions and the market seemed rather balanced or institutions may even have played a more important role. Since that time, the weight of the market has increased significantly. In this article, we intend to study the gallery sector and the impact of the national factor on it by using a ranking of the leading contemporary art galleries in the world that we ourselves developed (Quemin 2020a). The methodology that we used will be presented in the first part of the article. In its second part, we will analyze the territorial/national dimension that can be identified in the most important art fair in the world, Art Basel. We will comment on what is revealed by our ranking in terms of countries of origin associated with their share and diverse positions in the ranking. Finally, we will address the contribution that our ranking brings to the knowledge of so-called globalization.

The Impossibility of Estimating the Volume of the Private Art Market and That of Transactions by Galleries to Elaborate a Ranking of These Structures
With the phenomenal development of the contemporary art market during the last three decades, the creation of art value has not been left unaffected. Today, the influence of the market is always stronger (Quemin 2020b) as the regular increase of auctions clearly shows. In the meantime, as commercial galleries organize private sales, their activity remains much more discrete, not to say secretive. Social scientists, especially economists, should never forget this point when they consider or even produce data that pretend to cover the entire art market and not only auctions, whose prices are public and whose transactions can be known with a very satisfactory level of precision (although there already exists some limited cases in which the prices can be manipulated). When it comes to private sales organized by commercial galleries, though, researchers should be wise enough to accept that it is in fact impossible to produce satisfactory data concerning this segment of the art market. As we will argue, the very fact that such data is published on a regular basis does not give it any accuracy or even minimal credibility. One should accept the limitations of the tools that are offered by the social sciences and economics to even estimate with a decent margin of error what will remain obscure as long as private transactions remain as such. A perfect illustration of this is offered by the two competing estimations of the global art market that, for one single year, were produced by two leading art fairs in the domain, Art Basel and TEFAF, which each commissioned an economist to determine its value.
In 2000, TEFAF started publishing an annual report on the global art market with an estimated figure of its year after year volume. From 2008, the work was done by the economist Clare McAndrew. In 2016, though, she accepted an offer from another art fair, Art Basel, and its partner, the Swiss bank UBS, that had recruited her for her talents. Clare McAndrew's move compelled TEFAF to turn annual-report duties over to another author: Professor Rachel Pownall. When two competing reports were published and both pretended to estimate the global art market on the same year, they delivered substantially different results. For one single year, 2016, the difference in "estimations" made by two competing economists was huge. Rachel Pownall's estimation (Pownall 2017 Pownall!). Besides this enormous differential, there was no way to determine which of the two was the most serious (even if one of them was considered more or less pertinent). As given the preoccupying lack of rigorousness that was thus bitterly demonstrated by the comparison of the two reports and totally diverging figures, TEFAF immediately decided to give up publishing an annual report with an estimation of the global art market. Thus, Art Basel and its partner UBS, could pursue without any contradicting competitor. Although it was demonstrated that figures produced to "estimate" the global art market that included private transactions totally lacked rigor, the problem was partly solved by the immediate stop of the production of the TEFAF figure. The Dutch fair announced that it would instead pivot to producing "very in-depth and focused reports that would focus on a variety of subjects" (sic). Still, Art Basel, UBS, and Clare McAndrew did not seem to be puzzled by the huge gap in estimations from their former competitor: They went on publishing their annual report with renewed "estimations" of the global art market, year after year. Still, the least that can be said is that the gap between Rachel Pownall's and Clare McAndrew's results raised pertinent questions about how to measure an opaque market. It is very regrettable that the (small) debate that was raised by the publication of two very different "estimations" has, in effect, stopped before it would take on too disturbing proportions.
We found it necessary to mention and develop the previous point in order to underscore that, even though authors are drawing on similar data, in the case of aggregate data on the art market and the integration of auctions sales, which do not vary according to authors as their prices are public, the differences in findings as reported by various publications are just enormous. Still, it is necessary to mention that in addition to huge differences in estimated totals, the details for each gallery could be even stronger and the various mistakes that are made can partially compensate each other when the amounts are summed up.
With this development, we wanted to show that properly estimating the global art market by including private transactions to determine the economic importance of each gallery is a mirage rather than a possibility as private sales can never be known with sufficient accuracy.
It is often said in the art world that two of the leading galleries in the world, Gagosian and Pace, have been selling more than a billion dollars of works of art annually for years. Still, once again, it is impossible to check the information. The very fact that many leading galleries in the contemporary art world have branches in different countries make it even more difficult to know their activity, including where the works are really sold, which is not always reflected in the figures they produce when they pay their taxes. Everyone who is sufficiently integrated in the art market and who is a proper insider also knows that not all sales are declared. As this gallerist who owns branches both in Europe and in the United States admitted during an informal conversation with us (which makes admitting illegal practices easier than formal interview situations): "Our business is developing at a fast pace, but we still remain far from some of our competitors with whom we share some artists. Last year, we sold for 8 million euros of works of art, X sold for 30 it is said.
Q: Well, this is the official figure, it seems, he probably sold much more than that, don't you think? -Yes, you are absolutely right, and he probably declares that some works that were sold in France were sold in (another European country where the gallery also has a branch), although the sales were made in France, but this makes it possible for him to underestimate the amount of his sales here. But there are also all the sales that can be "forgotten" (he smiles)".
While we developed the project of creating a ranking of contemporary art galleries and considered their respective economic weights as the best way of proceeding, both our knowledge of existing figures and of the social world of galleries convinced us that it was, in fact, impossible to do so.

Rankings in the Contemporary Art World
While undertaking a previous research on star contemporary artists (Quemin 2013a(Quemin , 2013c(Quemin , 2018, we had been struck by the development of rankings in the contemporary art world. Today, these numerous rankings are proper players (Espeland and Sauder 2007) in this social world (Becker 1982). Although they are often criticized, their persistence year after year and their proliferation are clear signs of the social demand that they generate. Although there had been some experimentations in France as early as the 1950s, the most ancient actual ranking is the German Kunstkompass ("the compass of art" in German) that was created as early as 1970 by the economic journalist Willy Bongard and that, for decades, was published in the German economic journal Capital (Quemin 2002;Buckermann 2016). The Kunstkompass takes the form of a list-ranked from 1 to 100-of the most visible or famous artists in the world for a single year. In spite of what is often thought, the ranking is not based on economic success and has nothing to do with the market. The methodology of the Kunstkompass, which was formalized and made public under a detailed form on a few occasions only, remains rather basic. Every year, artists get points of visibility through three different types of occasions: Solo shows, group shows and reviews in leading art magazines (Quemin 2002). The more prestigious or important the institution, the higher the number of points it receives. At the end of the year, all points are summed up and the ranking of the "best" 100 artists in the world is published. For decades, the publication of the Kunstkompass was scrutinized by players of the art world. Still, it lost most of its influence with the creation of a competing ranking of artists, that of Artfacts at the beginning of the 1990s (Quemin 2013a). The German firm uses an algorithm that encompasses a much wider number of occasions of visibility than the Kunstkompass: Solo and group shows in museums and art centers, publications in magazines, but also even exhibitions in art hotels and non-profits. Moreover, a huge difference between the two methodologies that are used to build each indicator relies not only on the scope of information that is gathered-the Artfacts ranking constitutes a salient example of big data-but also on the fact that the Artfacts ranking, unlike the Kunstkompass, also integrates data that concerns the market. The number of points obtained by each artist in the ranking thus depends on the two segments that contribute to the creation of art value to make a reference to Raymonde Moulin's perspective and formulation (Moulin [1967(Moulin [ ] 1987(Moulin [ , 1992. Today, the database lists more than 550,000 artists worldwide, out of which 100,000 are ranked.
Another ranking that is quite popular in the art world and that can be briefly mentioned here is the Power 100 that has been published every year by the British journal Art Review since 2002. It aims at identifying and ranking the top 100 most influential players in the visual arts each year and covers members that go far beyond artists. They also include collectors, curators, and gallerists among other positions in the visual art world.
As we finished our research on star artists, we became fully conscious that rankings constitute very fruitful instruments in order to better understand the art worlds, how they are organized and hierarchized. They offer a quantified source of information that the social scientists can analyze and a key to the functioning of art worlds and their values. The very fact that a social scientist not only uses an existing ranking but creates one may seem unusual. Still, it may be necessary to underline that although this is not common in the sociology of art domain that a sociologist creates a quantified tool to better analyze the population on which he or she is working, this is very common in other domains of sociology, for instance, when sociologists create social mobility tables or homogamy tables (Quemin 2020a).
Although we found around a dozen competing rankings of visual artists (or sometimes even of contemporary works of art!), which show the strong social demand for such indicators, we were struck when discovering that no real rankings of gallerists exists. A partial exception should be noted. The site Artfacts not only references and ranks artists as we mentioned before, it also references other kinds of actors of the art world among which are galleries. For these players too, ranks are mentioned. Still, Artfacts never took the initiative to publish a proper ranking of them, even though it would be very easy to list them and compile their ranks. The reason may be that the gallery that is supposed to be the most prominent one in the world today would inevitably discredit the ranking to all players of the art world. According to the algorithm, the number one gallery in the world today is . . . Leo Castelli! As a matter of fact, the gallery was the most influential one in the world from the 1960s until the 1980s, but, although it still exists, it has become of minor importance for years. This has clearly to do with a hysteresis effect (Bourdieu [1981] 1993) of the algorithm. This very example clearly shows that although big data can be very useful in order to better understand the functioning of social worlds, they have to be used with a critical perspective. Sometimes, the obvious biases that are associated with them have to be controlled and corrected.

Developing a Ranking of Contemporary Art Galleries
As given the absence of any satisfactory ranking of galleries, our idea was to develop our own ranking in order to better understand how the social world that galleries constitute is organized. For this, we considered that we could use Artfacts data but in a more critical and controlled way than when the firm directly allocates a rank to galleries with its algorithm with no critical view of the process. When we worked on rankings of artists for several years (Quemin 2013a), we came to the conclusion that rankings are much more trustworthy for star artists than for less famous ones. By comparing the ranks of similar artists both in the Kunstkompass and in the Artfacts ranking for a given year, we could show that although the two rankings used very different methodologies, ranks were rather similar at the top of the two lists and even strikingly similar at their very top. This finding made it possible for us to formulate that at the very top of the social world that we observed (and this observation can probably be widened to other social worlds), fame tends to impose itself to its observers. People are so recognized, so famous, that whatever the methodology that is used to identify them as occupying the top positions in their social world may be, whether this methodology is rather basic or very elaborate, the obtained results will always be rather similar.
As given the previous finding, we decided to determine the rank in terms of artistic "quality" (Misdrahi Flores 2013) and recognition of the artists of all the galleries through the visibility of their "best" artists. When we made interviews with gallerists, they all constantly underscored that what makes the value of a gallery is fundamentally the quality of its list of represented artists, its roster. As this (salaried) gallery director who works for one of the most important ones in the world once told us in an informal interview: "Yes, I know that I have excellent contacts with some major collectors and, quite often, people ask why I do not decide to create my own gallery. Of course, I could find a backer and open a space, but so what? How could I convince the best artists to work with me? In this gallery, I mainly work with stars, they work with us because we also are a star gallery. But I am not naïve, collectors may like me a lot, what they buy are works by artists. Artists are the core of our business, if you cannot attract the best of them, your gallery is doomed to remain a minor one". To say it quite simply, players of the art world tend to consider that a good gallery is a gallery that represents good artists (the best gallery being the one that represents the best artists). Reversely, a good artist is generally considered to be an artist who is represented by a good gallery. From a sociological perspective, galleries and artists tend to qualify each other's quality simultaneously.
We could have considered the entire roster of galleries. However, we decided to limit our analysis to their top 10 artists, as we were often told in our qualitative interviews that the top of the list is absolutely vital to galleries. As this international gallerist states: "Apart from Gagosian, who precisely selects most of his artists because they can nearly all be sold at very high prices, all other galleries make most of their sales and profits with a very limited number of their represented artists. Sometimes it can be 5, some other times, it can be 10 or 15, but the full list of artists or even an important part of it is never what is at stake for us in the business".
The methodology that we adopted reflects how galleries generally make most of their profits with a very limited number of their artists or that a few of them play a major role for the gallery both in terms of sales and visibility and prestige. This is why we decided to consider the average or medium-Artfacts-rank of the top 10 artists of each gallery. Limiting ourselves to 10 artists and not to the entire roster of the gallery was also more pertinent as galleries can adopt strategies for the future development of their business that could otherwise interfere with the real "quality" (Misdrahi Flores 2013) of their list of represented artists. For instance, some galleries tend to integrate not only star artists who, today, are nearly inevitably old (in their 60s, 70s, 80s when they are not in their 90s) (Quemin 2013a), but also young promising artists whose ranks are inevitably much lower. Such a strategy aims at securing the future development and is all the more likely that galleries are strong enough to cover such costs or investments. Still, considering all artists would lower the rank of such galleries as it would not reflect their real positions in the art world and on the art market.
A second criteria should be noted: We also decided to include living artists only and not deceased ones, except when galleries mentioned explicitly on their websites that they represent their estates and hence have an undisputable link to the artist. Having been very familiar with the art world and with the art market for many years now, we were astonished when we discovered that some galleries, especially among the most prestigious ones, tend to cheat (or embellish the truth) when they present their roster on their website. They sometimes include names of deceased artists that will add some prestige to the gallery. Picasso is a perfect example of this strategy. Although his name is added to the list of represented artists of some heavyweights in the gallery sector, absolutely no gallery is a partner of the Picasso Administration. In order to prevent our ranking from being affected by such strategies of some leading galleries, we decided to limit our calculation to the medium range of the top 10 living artists of each gallery and include deceased artists only in the case of estate representations, as it is a kind of representation with which one cannot cheat. A next step in our methodology was to decide on the galleries that would be ranked. Throughout our previous research (Quemin 2013a(Quemin , 2020a, we have become consistently more interested in the players that occupy top positions in the art world. They are the most visible socially, they generally are the ones who master the rules of the social game that they play and share with other actors, but they are generally more difficult to meet than more ordinary players. For about 10 years now, our activity as university professors has been combined with a second occupation as art critics and art journalists. This has opened many doors for us. Thus, we could make intensive field work and, among other things, socialize with prominent players of the international contemporary art world, be they museum directors and curators, artists and . . . gallerists. Once again for our research on galleries, we decided to focus on the most important structures, notably because they are the ones that best master the rules of the social and economic game to which they participate. They can also tell us more about how the game functions, and hence, about the mechanisms that create art value (Moulin [1967(Moulin [ ] 1987(Moulin [ , 1992. Focusing on the top players of the art world also enables us to make the connection with celebrity studies (Adler 1985;Turner 2004;Kurzman et al. 2007;Ferris 2010;van de Rijt et al. 2013), a domain that is not often associated with sociology of art (Lang and Lang 1988;Bowness 1989;Quemin 2013a).
In a previous research that we conducted on art fairs and gallery participations for the year 2008, we could produce an estimation of around 2200 contemporary art galleries in the world (Quemin 2013b). As the market strongly developed in the following decade, the number of proper contemporary art galleries can be estimated to be somewhere between 2500 and 2800 today. Still, as we intended to analyze the top of that social and economic world, we had to apply a filter that would designate the galleries that can be considered as the "best" ones in the world before ranking them. For this, we decided to limit our analysis and ranking to galleries that participated in Art Basel fair in June 2019. For many years now, Art Basel (in Basel, Switzerland, the original art fair created two outposts: The first one, Art Basel Miami Beach, in the US in 2001-2002, and the second one, Art Basel Hong Kong, in Asia in 2013) has been the unrivaled number one art fair in the world. The prestige associated with being selected and admitted to the art fair and the high volume of business that is guaranteed by participating in the event make that no gallery that can get admitted would turn it down. Each year, competition to get selected is ferocious. Around 800 galleries apply, but only less than 300 get admitted, around 250 in the main section ("gallery"). It should be mentioned that many galleries do not even apply as they know that they have absolutely no chance of getting selected. Being admitted to Art Basel is widely considered as a sign of consecration for galleries and being accepted as a clear sign that one belongs to the elite of the profession (Quemin 2013c). (See Appendix A for the top 50 galleries of our ranking).

Art Basel: An International Art Fair, Not a "Global" Event in Terms of Participating Galleries
Even before taking into account the positions of the galleries in the ranking, it can be noted that, although Art Basel is generally considered a "global" event, one that is supposed to address galleries Arts 2020, 9, 105 7 of 16 and collectors from the entire world, the representation of the various countries is extremely uneven when considering its exhibitors (Table 1). Although the entire world is vast and comprises around 190 nations, galleries that are selected to participate in Art Basel come from 28 different countries only, with very strong inequalities in terms of nationality, with the USA and a very limited number of European countries taking the lion's share (Quemin 2002(Quemin , 2006(Quemin , 2013b. Out of the 247 galleries that participated in the gallery section at Art Basel in 2019, no less than one-quarter were American (60 galleries or 24.3%). Germany came second with 52 galleries (21.1%), ahead of the United Kingdom (12.1%), France (8.9%), Switzerland (7.3%), and Italy (5.7%).
Although Art Basel is generally considered a "global" event, given its strong international dimension, only six countries accounted for 79.4% of exhibitors. North American and European countries represented as much as . . . 91.9% of the exhibitors! The share of the rest of the world was strikingly limited. Even Asia, in spite of the success of Chinese artists on the international art market in recent years, accounted for only 4.0% of art galleries participating at Art Basel. South and Central America, in spite of the long tradition of Brazilian contemporary art galleries, represented only 2.8% of participants (Brazil took the lion's share with six of the seven galleries for all of South America and Columbia was the only other represented country with one single gallery only)! All of Africa was represented by only two South African galleries (less than 1% of participating galleries, 0.8%). The Middle East was present only through Lebanon (0.4%) and the whole continent of Oceania was totally absent.
The polycentric or monocentric structure-in terms of number of represented cities-of the countries that comprised the highest numbers of the "best" galleries in the world is also worthy of mention. Although Los Angeles may be considered a real international capital city for contemporary 2 We specify in what cities the galleries are located when not all of them are located in the capital city of the country. art galleries, even in comparison to other countries (it constitutes a significant number of galleries participating to Art Basel (7)), at the national level its importance is very secondary when compared to the influence of New York City that comprises no less than 50 of the "best" contemporary art galleries in the world, making it the undisputable contemporary art capital of the world.
Germany comes second and so does the city of Berlin with 35 galleries out of the 52 for the whole country, but their distribution across Germany is much more balanced than in the US case. The situation is rather similar in Switzerland and in Italy, whereas the United Kingdom, France, and Spain share the American model of concentration with even more marked traits, as the capital city of each of these three countries gathers all but one gallery that participate in Art Basel. This analysis tends to show that territory plays a double role: At the international level, as the different countries hold very unequal positions, but also at the national level, as the configuration of the art market can vary. As the previous figures show, the "best" art galleries in the world are concentrated in a very small number of countries. In some of these countries, nearly all galleries, or at least a very high proportion of them, can be found in one city only. This result shows the very strong weight of the territorial factor, be it national borders or metropolises in a given country, even at a time when the so-called globalization is supposed to have diminished its impact.

Questioning "Globalization" More Precisely Through the Lens of Our Ranking of Galleries
In the next step of the research, we decided to focus only on the "best" galleries that participate in Art Basel in terms of their roster. Hence, we decided to consider the "cream of the crop", blue chip" galleries that are all heavyweights in the gallery business. Is their national belonging (in terms of main location in the world) significantly different? As given the very strong concentration that we previously noticed, it would seem that this tendency could not be more pronounced.
If we consider the top 50 galleries in the ranking that we mentioned (in terms of ranks of their top ten artists), their nationalities are the following (Table 2). The USA still reinforces its share among this group of major galleries, its figure rising from one-fourth of the galleries to one-third. The shares of Germany, the United Kingdom, France, Italy, and Switzerland are barely affected. What is more noticeable is that the share of North America and Europe that was already very high at 92% rose even further to 96%. Only one South Korean gallery and one South African gallery managed to make their way to the top 50. As we can see, the general context appears extremely hostile to non-Western galleries and those of the Southern hemisphere. This explains why they were so rare in the first list and why they are even rarer in this second one.
However, being exposed to less competition with local, national, or even "regional" 3 galleries to attract the best artists than what can be found in the most important metropolises for contemporary art (among which New York City occupies the first place) might play as an advantage, even if it is only a secondary one.
In a next step of the research, we decided to consider the main location of the top ten players of your ranking of galleries (Table 3). Although numbers are very small and must be interpreted cautiously, the United States tends to further reinforce its disproportionate share in this elite group of contemporary art galleries (from one-fourth when considering all galleries participating in Art Basel to one-third when taking into account the top 50 of the best participants, to 4 out of 10 when focusing on the top 10 players only). The respective shares of Germany, the UK, and France are similar, with two galleries each among the top 10 players in the world. The top 50 galleries can be found in 14 different countries only and the top 10 galleries in 4 countries only, which, from the start of our analysis, all appeared to constitute a small nucleus of the international network of contemporary galleries. Also, the more prestigious and restricted the group of galleries becomes, the higher the share of the USA is found. Even among participants in the most exclusive art fair in the world, it seems that all galleries are not equal.

Reconsidering the Theories of Globalization
Some 20 years ago, we started working on the international dimension of contemporary art (Quemin 2002). At the time, the ideology of globalization was extremely popular, both in the contemporary art world and in the social sciences. Analyses of globalization had developed and come to prominence in the 1990s and had even become central to sociology in the second half of this decade (Bartelson 2000;Bellavance 2000;Therborn 2000b). Still, very few authors cared to apply their theoretical developments to facts and the empirical data supporting some of the leading theoretical positions was often limited or even insubstantial (Ohmae 1990;Sassen 1992Sassen , 1999Sassen , 2000Lash and Urry 1994;Appadurai 1996Appadurai , 2000. In the June 2000 issue of International Sociology, devoted to the topic of globalization (Therborn 2000a), and in a book specifically focusing on cultural globalization (King 2000), there was an almost total absence of empirical data. The authors resorted to abstract considerations and very rarely supported their analyses with statistical evidence. Thus, they undermined the way that data can, in fact, provide at least a partial answer to some of the key questions raised by theoretical explorations of globalization.
As we underscored in some of our former publications (Quemin 2002(Quemin , 2006(Quemin , 2013b a recurring line of questioning in the debate over the specific dynamics of cultural globalization is the extent to which this phenomenon maintains or generates forms of inequality, and if so, of what kind? (De Swaan 1995;Cowen 1998;Therborn 2000b). One line of argument is put forward by Saskia Sassen and Zygmunt Bauman who suggest that under conditions of globalization, national borders are being increasingly erased and substituted with growing fluxes or international exchanges (Sassen 1996;Bauman 1998). A second line of argument explains how the global is structured more around patterns of flows than 3 In the contemporary art world, the world "regional" does not generally refer to a territory inside a country, but to a group of neighboring countries that can be as vast as a continent. around pre-constituted entities. Manuel Castells, for example, refers to flows rather than organizations (Castells 1991), while Jan Aart Scholte emphasizes deterritorialization as characteristic of globalization, the increase of flows making national entities less significant (Scholte 1996).
A line of analysis of particular relevance here is that represented by the works of Immanuel Wallerstein (Wallerstein 1991(Wallerstein , 2000 and Pierre Bourdieu and Loïc Wacquant (Bourdieu and Wacquant 1999), which point to an imbalance in international cultural exchanges and to the existence of domination effects. Immanuel Wallerstein argues that transnational cultural exchanges merely reflect the contradictions and imbalances in the world economy with a few Western economies dominating the rest of the world. However, for Pierre Bourdieu and Johan Heilbron, and also for Gisèle Sapiro, the cultural sphere exhibits a certain form of autonomy (Bourdieu 1990) from the economic sphere; this can be seen for example, in the forms of transnational cultural exchanges affecting the fields of literature and of translations (Heilbron 1999(Heilbron , 2001Sapiro 2008). This (bourdieusian) school of thought uses quantitative data to highlight both the fact that 'Imbalances (. . .) characterize the very structure of international exchange' (Heilbron 2001, p. 146), and that 'instead of an equilibrium between import and export, the reality of transnational exchange is a process of uneven exchange' (Heilbron 1999, p. 439). The conclusion here is that cultural exchanges operate in relation to a clearly discernable geographical center, which is distinct from its periphery and are characterized by domination effects. Today, this approach constitutes the most convincing use of empirical data to examine the workings of cultural globalization with special reference to the field of literacy (Heilbron 1999(Heilbron , 2001Janssen et al. 2008;Sapiro 2008). In the field of visual arts (Zarobell 2017), a similar conclusion seems to be at work, as our figures show. The most recognized artists on the international contemporary art scene are from a very small number of Western countries; a finding that exemplifies the specific workings of cultural hegemony within this field (Quemin 2002(Quemin , 2006(Quemin , 2013a. The data produced and analyzed in this article show the very hierarchized nature of the contemporary art world, in this case, that of galleries. Territory and nations play a major role in this hierarchy and, although internationalization is extremely strong at the top of the gallery sector, globalization remains a myth.

Conclusions
In this article, we had a double aim. Reflecting on rankings as a contribution to a domain that remains rather unexamined in the social sciences and that can be designated very generally as celebrity studies. From a methodological perspective, we elaborated an original ranking of contemporary art galleries trying to identify the "best" or the most renowned ones. This made it possible for us to go further, when analyzing galleries participating in Art Basel Art fair, the most important of these events in the world: We could analyze the impact of the country of origin more precisely than by only considering participation to the fair. Not only does the nationality of galleries affect participation, but the higher the degree of prestige among galleries is, the stronger the share of the United States becomes. Besides, as we decided to question the concept of globalization with empirical material, we could show that the internationalization of the contemporary art world has little to do with the so-called globalization and that territory, and especially national borders, still play a prominent role. Even today, as the very recent empirical material that we produced in this article shows, among all players of the contemporary art world, countries or nations should not be underestimated. Besides, the highest segment of the art market is still controlled by an extremely limited number of countries among which the USA plays a role that is absolutely central.