The Coevolution of New Organizational Forms in the Fashion Industry: A Historical and Comparative Study of France, Italy, and the United States Marie-Laure Djelic; Antti Ainamo Organization Science, Vol. 10, No. 5, Focused Issue: Coevolution of Strategy and New Organizational Forms. (Sep. - Oct., 1999), pp. 622-637. Stable URL: http://links.jstor.org/sici?sici=1047-7039%28199909%2F10%2910%3A5%3C622%3ATCONOF%3E2.0.CO%3B2-%23 Organization Science is currently published by INFORMS. Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/informs.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org. http://www.jstor.org Tue Feb 12 07:30:03 2008 The Coevolution of New Organizational Forms in the Fashion Industry: A Historical and Comparative Study of France, Italy, and the United States Marie-Laure Djelic Antti Ainamo ESSEC, HR Management Department, Av. B. Hirsch, BP10.5, 95021 Cergy-Pontoise, Cedex, France, djelic @essec.fr Helsinki School of Economics and Business Administration, Runeberginkatu 14-16, FIN-00100 Helsinki, Finland, ainamo @ hkkk.$ Abstract In many industries, the contemporary context of acute environmental dislocation shows the limits of traditional organizational recipes. In direct response to environmental challenges, companies are experimenting with new organizational solutions. While flexibility, or the capacity to redefine organizational form to follow changing purposes, is undeniably a common trend, these experiments otherwise differ greatly. Diversity is such, in fact, that it is difficult to clearly identify and define a unique organizational paradigm for the future. To explore the connection between environmental dislocation and organizational transformations, we adopt a historical and comparative perspective. Our empirical base of evidence is the luxury fashion industry in three countries, France, Italy, and the United States. For many years, this industry was defined by stable environmental conditions, and a craft model of organization remained dominant. We show that, over a more recent period, increasing environmental turbulence has brought about a redefinition of the rules of the game. A common response has been for organizations to move towards greater flexibility or modularity and to experiment with network forms. However, we also show that the paths or trajectories leading to organizational flexibility have varied significantly across countries, reflecting historical legacies and institutional constraints. W e identify in fact three different network forms in that industry, which represent national ideal types-the "umbrella holding" company in France, the "flexible embedded network" in Italy, and the "virtual organization" in the United States. We argue that the process of change in the luxury fashion industry has been one of coevolution, where environmental transformation and organizational change have fed upon each other through time. Pioneer firms in the luxury fashion industry originally devised organizational solutions within the bounds set by nationally defined constraints and opportunities. Becoming institutionalized, these early solutions in turn shaped the ORGANIZATION SCIENCE,O 1999 INFORMS Vol. 10, No. 5, September-October 1999, pp. 622-637 environment for individual organizations and organizational populations, creating new sets of opportunities and constraints. In a path-dependent manner, different models of organization and national competitiveness thus emerged. In conclusion, we are brought to question the likelihood of full and stable convergence towards a unique organizational form or paradigm. There appears to be, in each national context, a process of construction of new organizational solutions that starts from local foundations. Embedded as they are in powerful historical and institutional legacies, organizational differences are there to stay, we believe, beyond the period of transition and acute environmental dislocation. (Fashion Industry; New Organization Forms; Coevolution; Historical Perspective) 1. Introduction Globalization, acute competition, the information technology revolution, and increasing customer sophistication are radically redefining environmental conditions. In many industries, these environmental challenges are showing the limits of traditional organizational recipes. Insisting on a necessary fit between organizational solutions and environments, organization theorists argue that periods of severe environmental dislocation call for flexible organizational solutions that can adapt to changing purposes (March 1991, Lewin and Stephens 1993, Brown and Eisenhardt 1998, Volberda 1998). At the same time, a look at organizational practice shows that there are multiple paths to organizational flexibility. Such diversity 1047-7039/99/1005/0622/$05.00 1526-5455 electronic ISSN MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industry makes it difficult to clearly identify and define a single organizational paradigm for the future. To explore the connection between environmental challenges and organizational transformations, we focus in this paper on the luxury fashion industry. Adopting a historical and comparative perspective, we find that, for many years, luxury fashion companies operated within a relatively stable environment, and the "craft" model of organization, characteristic of French pioneer firms, remained dominant. More recently, environmental conditions have started to change, becoming increasingly turbulent. In the face of global environmental challenges, luxury fashion companies have tended to move towards greater organizational flexibility. However, a comparison of the industry in France, Italy, and the United States shows that the trajectories leading to such organizational flexibility have differed significantly across national boundaries. We identify, in fact, three national patterns or "ideal types," each one corresponding to one of our national cases. We thus find strong support for a coevolution perspective where environmental transformation and organizational change interplay through time, feeding upon each other (Koza and Lewin 1998; Lewin et al., this issue; Koza and Lewin, this issue). Through our historical and comparative approach, we contribute to discussions about the mechanisms of coevolution. In each of the three countries examined, we point to a path-dependent and historically constructed process of interaction between-or coevolution of--organization forms, global environmental trends, and national institutional legacies. Using our findings to speculate further, we also take up the issue of the possible consequences of the coevolution process. We thus touch upon a key dimension of the debate on new organization forms-that of the ultimate congruence or convergence of forms. We believe that, in the luxury fashion industry, the current period of acute environmental turbulence is in fact a period of transition, characterized by search, exploration, and multiple but temporary solutions (Lewin et al., this issue). A matter open for discussion is whether, after this period of transition, the organizational landscape will end up converging upon a unique, widely legitimated, and institutionalized organizational paradigm (Lewin et al., this issue). Building upon our empirical results, we propose at the end of the paper our own view on the question. 2. Organization Forms and Environments In relatively simple and stable preindustrial times, small and craftlike organization forms were dominant. The first ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 and second industrial revolutions, at the beginning and end of the nineteenth century, created the conditions for mass production and economies of scale. The large size of many new companies then turned coordination and control into key organizational challenges. The solution that emerged was the standardization of organizational routines combined with a hierarchical and rigidly centralized form of control and reporting. This tightly coupled organizational solution, labeled here for short the bureaucratic paradigm, was gradually established throughout the first part of the twentieth century as a "one best way" that could be equally suited to all companies and situations (Taylor 1911). 2.1. Bringing the Environment Back In In time, though, this illusion of a possible "one best way" ran against a stubbornly complex and multifarious empirical reality. "Prebureaucratic" craft organizations survived, particularly in those industries where cost competition was not significant and customers valued quality craft production. There was increasing evidence, in fact, that the bureaucratic paradigm was not efficient in all situations. This led to the idea of a contingent fit between organizations and their environments. The survival and effectiveness of organizations appeared to hinge upon the right match between organizational capabilities and environmental peculiarities (Burns and Stalker 1961). While this idea has shaped organization theory to this day, the nature and direction of the fit and the mechanisms for change still very much remain a matter for debate. On one side, some have argued that environmental characteristics essentially determine and shape organization forms. Contingency theorists (Woodward 1965, Lawrence and Lorsch 1967), population ecologists (Hannan and Freeman 1984), and more recently, organizational neoinstitutionalists (Powell and DiMaggio 1991, Scott et al. 1994) all propose variants of this argument. Others have put forward an entirely different claim. Strategic choice (Child 1972) and resource dependency theories (Pfeffer and Salancik 1978), the cognitive (Weick 1969), and the more recent postmodern argument (Clegg 1990) all have, in one way or another, defended the idea that organizations choose and shape, at least in part, their own environments. A third framework is currently emerging that appears to bridge the above controversy. Adopting a longitudinal perspective, coevolution theorists argue that environmental transformation and organizational change interplay and feed upon each other through time (Koza and Lewin 1998; Lewin et al., this issue; and Koza and Lewin, this issue). In periods of relative environmental stability, existing and dominant organization forms define organizational populations and shape in part environmental landscapes. In turn, environmental transformations tend to 623 MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industry affect organizational populations and forms. In periods of relative stability, change takes place but only in an incremental way, in a manner analogous to species variation. To use the seminal and by now classic formulation of March (1991), organizations and populations "exploit" existing resources, dominant solutions, and institutionalized search routines. In times of significant environmental dislocation, on the other hand, individual organizations and organizational populations appear to be threatened. This triggeqs "exploration," that is a radical search for entirely new kinds of solutions. 2.2. Organization Forms for the Future There is little doubt that the end of the twentieth century is a period of significant environmental dislocation, at least as much as the end of the nineteenth century had been in its time. Many industries and companies are having to face increasingly turbulent, ambiguous, and hypercompetitive environmental conditions (Volberda 1996, 1998; Brown and Eisenhardt 1998). In those conditions, the capacity to search for or "explore" appears necessary to ensure organizational survival. And such capacity requires flexible and organic organization forms (March 1991, Lewin and Stephens 1993). Although they should not give up some of the clear advantages that come together with standardization and exploitation, companies should be moving towards-or preserving-looser types of integrating mechanisms and more flexible organizational features (Volberda 1996, 1998). They should find ways to promote cultural and political variety and still avoid inefficiencies, fragmentation, and political strife (Webb and Pettigrew, this issue). Design theory suggests a solution for handling this apparent contradiction-a redefinition of the organization as a "nearly decomposable system" (Simon 1996). According to Simon, "the potential for rapid evolution exists in any complex system that consists of a set of stable subsystems, each operating nearly independently of the processes going on within other subsystems" (Simon 1996: p. 193). In such complex systems, each organizational part or module may be better adapted either for "exploitation" or for "exploration." Pioneering experiments by leading firms appear to point towards this kind of flexible combination of subsystems or modules, where a core competence corresponds to each module (Taylor 1991, Miles and Snow 1986). Near decomposability, or modularity as it is commonly labeled, thus seems to be key to managing complexity in tomorrow's organizations. And network forms of organization, allowing modularity, are in fact emerging and thriving in many industries (Nohria and Eccles 1992; Koza and Lewin, this issue). 624 However, while there is widespread agreement among organizational practitioners and theorists alike that modularity can indeed make it possible for organizations to reconcile flexibility with cost efficiency, the "network form" is, as of now, more of a ragbag than a clear paradigm. Many different organizational experiments do fit under the label. It seems, in fact, that the road to the future is not straightforward but rather leads to multiple "migration paths" or trajectories of change. A coevolution perspective with a historical and comparative dimension can help account for this-each trajectory only makes sense, in fact, within a particular institutional context and in connection with specific historical legacies. We illustrate this below, using the case of the luxury fashion industry. The question remains whether these multiple forms, solutions, and trajectories may nevertheless end up converging or whether differences are there to stay, beyond the period of transition and acute environmental dislocation. We give our own view on the matter in the discussion at the end of the paper. 3. The Luxury Fashion Industry Outlining the boundary of the luxury fashion industry is not an easy task. Although a number of apparently "objective" dimensions come to mind when trying to define luxury-such as, for example, price, intrinsic aesthetic value, quality, rarity-those dimensions are not, in themselves, fully satisfactory. They correspond to a traditional and somewhat outdated understanding of the luxury good, where value stemmed from the intrinsic features of a product (CERNA 1995, Jellinek 1997). 3.1. The Luxury Fashion Industry-Definition Our understanding of the luxury fashion industry has in fact significantly evolved, particularly over the last 20 years, at the same time that the boundaries of the industry were expanding. There is no such thing anymore as an objective and unanimous definition of aesthetic value. Rather, there are a potentially infinite number of lifestyles, discourses, and definitions revealing a multiplicity of aesthetic worlds (CERNA 1995, Djelic and Gutsatz 1998, Ecole de Paris 1998). A key defining characteristic of luxury fashion companies, in this context, is that they do not deliver only products but diverse sets of representations as well. Luxury fashion companies are also brand names and brand names are "spaces for dreams" (Ecole de Paris 1998) where customers can satisfy not only their material but also their symbolic needs. Luxury fashion products are generally intended for use, but they are also associated with some intangible dimensions that pertain to the realm ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industry of meanings or aesthetics and give these goods their marginal value (CERNA 1995, Gutsatz 1996). The definition proposed here reflects and reveals the fact that the borders between the luxury and the nonluxury segments of the industry may not today be as watertight as they were 30 or 40 years ago. For luxury brands, competition may in fact increasingly come from beyond the boundary of the luxury goods industry (Nueno and Quelch 1998). The worldwide luxury goods market now represents a total of $35 billion (U.S.) (Financial Times, 27 February 1998). On that market, French companies have contributed a little less than 50% of total turnover in the mid1990s (calculated from CERNA 1995). If one considers only the fashion segment of the luxury market, though, French companies emerge as much less significant than their notoriety would lead one to expect. The combined turnover of Yves Saint Laurent Couture, Dior, Givenchy, Kenzo, and Christian Lacroix reached a mere $0.5 billion (U.S.) in 1997. In the meantime, the Italian Armani or the American Ralph Lauren each had a turnover of around $1 billion (U.S.) (Les Echos, 30 April 1998). 3.2. The Luxury Fashion Industry-Methodology This paper takes the fashion segment of the luxury goods market as its empirical focus and builds upon the work both authors have been doing on the fashion industry (Ainamo 1996, Djelic and Gutsatz 1998). In particular, it draws selectively upon the first results of a large-scale, comparative project on the luxury goods industry. The object of that project has been to identify and describe evolutions through time and tendencies in the luxury goods industry with regard to environmental trends, organizational forms, and managerial competencies. Upon completion of the project, the empirical base will include around 40 companies in three different countriesFrance, Italy, and the United States-the distribution by country paralleling more or less closely the weight of each country in the industry. For the purposes of this paper, we considered only the fashion segment of the luxury goods industry, and we decided to focus more particularly on a few companies for each country. These particular companies were selected for their ideal typical nature andlor for their overall weight in the national or worldwide markets. The results presented in this paper reflect information obtained on this small sample of companies through interviews, press reviews, and the analysis of available written documents. The data collection and the chosen methodology have had both a comparative and an historical dimension, allowing us to identify similarities but also differences across time and space (Chandler 1990, GuillCn 1994, Djelic 1998). We were able in this way to identify national patterns and ORGANIZATION SCIENCE NO^. 10, NO. 5 , September-October 1999 to point to path-dependent and historically embedded trajectories in each country. At the same time, though, our methodological choices have constrained our ambitions. And the most we can claim is that these national patterns are ideal types, to which exceptions may easily be found in each country. Table 1 gives a summary description of the sample of selected companies. In France, the Louis Vuitton Moet Hennessy group (LVMH) brings together under the same umbrella company some of the world most famous fashion names (Givenchy, Christian Lacroix, Kenzo, CCline, and Christian Dior). Together, these houses or brand names represent a significant share of the French luxury fashion industry, and the peculiar structure of LVMH makes it a unique entity worth looking at in greater detail. We also chose to focus on Hermits as possibly the most vocal advocate of a French tradition of luxury and, as such, an interesting ideal type for the purposes of our study. For contrast, we finally looked at Pierre Cardin, the French company that may stand the furthest away from this French model and tradition. In Italy, we selected Armani on the grounds, firstly, of its overall weight in the Italian, but also in the worldwide fashion industry. The recent and astounding success of Gucci and Prada, which have become important players on the fashion market in only a few years, explains why they were also included in our sample. In the United States, the sheer size of Ralph Lauren and Calvin Klein and the apparent success of their relatively similar strategy have made them interesting ideal types. For good measure, we also included Donna Karan, which, with a similar strategy, is turning out to be much less successful. Our historical and comparative data collection allows us to document major transformations, particularly over the past twenty years, in the environment of luxury fashion companies. In 4, we point to a number of global trends affecting all of these companies across national boundaries. We then present, in 3 5, the different responses which these companies have come up with to adapt to environmental transformations. Beyond a common tendency towards increasing flexibility, these organizational responses are very much embedded in national institutional contexts and peculiar historical legacies. 4. Environmental Trends and Challenges in the Luxury Fashion Industry The luxury fashion industry originated in France in the middle of the nineteenth century when a British tailor who had settled in Paris-Charles Frederick Worth-decided to sign his products, thus claiming the originality 625 MARIE-LAURE DJELIC AND ANTTI AINAMO Table 1 Luxury Fashion Companies-A FRANCE LVMH (sales 1997: US$450m) Hermes (sales 1997:US$120m) Pierre Cardin (ma.) Coevolution in the Fashion Industry Sample ITALY UNITED STATES Armani (sales 1996:US$750m) Gucci (sales 1997:US$lOOm) Prada (sales 1997:US$190m) Ralph Lauren (sales 1997: US$9OOm) Calvin Klein (n.a.) Donna Karan (sales 1997:US$540m) Note. Sales figures are for apparel only. They do not include accessories, leather, cosmetics, or perfumes. of his creations and the specificity of his image. Until the 1960s, the industry amounted in fact to a handful of Parisian haute couture houses serving a privileged clientele and characterized by labor intensive, craftlike processes. This small but dominant group of French players all but managed to close off the boundaries of the industry. 4.1. The Institutionalization of Stability before 1970 Organized though a professional association-the Chambre Syndicaie de la Haute Couture-French traditional players were for a long period of time able to control entry into the luxury fashion industry by institutionalizing a set of strict rules (Grumbach 1993). Under those rules-for the most part still in force today-the label haute couture could be delivered only to those companies that fit the following requirements: (a) employing at least 20 persons in the production of clothes in the company's studios, (b) presenting for each season-spring and fall-a collection of at least 75 designs, (c) presenting these collections with the help of at least three live models, and (d) doing so in the house itself, in special areas designed for this purpose (Crane 1997). The stringent nature of these rules in effect made it nearly impossible for foreign competitors to obtain the label haute couture and thus, at least until the 1960s, to gain the legitimacy that was required to impose themselves within the industry. Thanks in part to their first movers' advantage, French traditional players in the luxury fashion industry thus managed to define and control their own environment. For many years, they were able to institutionalize and enjoy a stable and predictable environment, creating around themselves a comfortable buffer zone. Behind these protective boundaries, French haute couture houses did set creative and fashion trends that had a worldwide impact. They also very much dictated technological and organizational solutions, institutionalizing them as key defining dimensions of the industry's identity (Grumbach 1993, Dumas 1998). However, over the past 30 years, the predominance of French haute couture houses has come to be contested. 626 It has become increasingly difficult for these traditional fashion companies to ignore and fend off the pressure stemming from emerging environmental trends. These environmental trends, playing out at a global level, have created many new and significant challenges, to which traditional players in the luxury fashion industry have finally had to react. 4.2. Global Pressure from the Late 1960s Many of the global trends that have come to redefine the environment for luxury fashion companies are quite similar in fact to those shaping other industries. An evolution of the customer base and of its lifestyle has combined with technological transformations and a globalization of markets. These global trends have been further reinforced by an increasing competition and by the emergence, in particular, of companies challenging the predominance of traditional French haute couture houses. These outsiders have welcomed, seized upon, and pushed along a redefinition of the rules of the game in the luxury fashion industry that could only be to their benefit. Market demand for luxury fashion has changed significantly over the last 30 years or so. Traditional marketsWestern Europe and the United States-have become mature. The original elite target groups for French haute couture houses, that had worn only or mostly made-toorder clothes, have gradually declined (Crane 1997). This could in part be explained by increasing costs and prices but also by changes in lifestyles. While this small elite was running dry, a growing mass of middle- and uppermiddle class customers, with rising buying power, was looking for signs of "distinction" (Bourdieu 1979). Through their consumption of luxury goods, these new groups of customers have been searching for symbolic satisfaction in ways that have often differed from those of traditional elite group members (Gutsatz 1996, Ecole de Paris 1998). When traditional elite groups valued distinction, hierarchy, order, or nobility, highly educated middle-class customers seem to care for entirely new sets of values-solidarity, meaning, genuineness, and ethics. Together with the education level of Western customers, expectations also have grown. Customers have become increasingly able to manage their own choices; they ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industry are autonomous and demanding. They have their own particular lifestyle and representation of the world, with which they expect the products they consume to resonate. Traditional haute couture houses, which used to rely on the intrinsic features of their products to trigger sales, have found that the product, nowadays, is far from selfsufficient. Customers look for perfect quality and service but also for price and immediate satisfaction as well as for a fit between the space for dreams characteristic of a brand and their own symbolic needs and representations. While markets have radically changed in nature, they have also expanded in geographic scope. The oil shock and the consequent influx of revenues in the Middle East, the economic boom in Asia, and the fall of the Berlin Wall, have pushed back the frontier for luxury goods companies. These new markets, characterized by an emerging elite hungry for symbols of status, have become highly attractive for fashion houses. But they have also contributed to making demand less homogeneous, reinforcing a trend already characteristic of the industry's traditional markets. Thus, like many other industries, the luxury fashion industry is facing a global challenge. The internationalization of markets is such that luxury fashion products and brands can potentially become universal. At the same time, though, luxury fashion companies will increasingly have to be sensitive to the multiplicity of needs, and they will have to adapt to the local peculiarities of demand. For those fashion houses embedded in a tradition of craftsmanship-the French players in particular-the expansion and diversification of markets have brought along another significant challenge. To handle change and diversity in their markets, luxury fashion companies have had to consider the integration of new technological tools and the adoption of modern management practices. At first sight, these appear to threaten the craftsmanship tradition. Managing the integration and striking a balance between tradition and innovation, mass production and customization, have thus become key challenges in the luxury fashion industry, in particular for those players deeply embedded in the industry traditions (Ecole de Paris 1998, Dumas 1998). 4.3. Increasing Foreign Competition from the Late 1970s The redefinition of markets and the strength of global pressure have created a window of opportunity in the luxury fashion industry, through which competition has rushed in. By playing upon and using these environmental transformations to their own advantage, new competitors or challengers have in fact come to reinforce them. Competition to traditional haute couture houses originally emerged from within France as early as in the ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 1960s. A number of small companies were then set up that did not fit the requirements for getting the haute couture label. These crkateur companies thus remained at the margins of the industry. They made the most of their marginal position and as a matter of fact invented a new industry, the luxury ready-to-wear. They seized upon the unmistakable decline of the haute couture business, betting upon instead and tapping into a widening group of middle- and upper-middle class customers that were benefiting from a period of significant economic growth in France (Grumbach 1993). The organized haute couture community still managed to keep control over this French competition. French crkateur companies remained relatively small in size and never became a significant threat to the haute couture community (Crane 1997). The main and real challenge came, in fact, a little later from beyond French national borders. Starting in the late 1970s and early 1980s, new players originating mostly from Italy and from the United States imposed themselves at the luxury end of the fashion market. These challengers perceived evolving market and market demand in the luxury fashion business more as opportunities than as threats. By anticipating and building upon the global trends identified above, Italian and American entrants have made significant forays into the luxury fashion industry. Bypassing the stringent requirements for entry into the haute couture industry, they have bet instead upon luxury ready-to-wear. These outsiders have welcomed and pushed further along a redefinition of the rule of the game that could only be to their benefit. In the process, they have reinforced existing global trends and significantly contributed to reshaping the luxury fashion industry (Gutsatz 1996, Ecole de Paris 1998, Martin 1998). Defining customers as coconstructors or cocreators and not merely passive consumers, Italian and American challengers have brought about a redefinition of the relationship between fashion companies and their markets. These challengers take into account customers, their various profiles, and symbolic needs or expectations. Quite similar in that to fashion companies at the lower end of the industry, these new entrants to the luxury fashion industry do not only design, manufacture, and sell their products. They also interact with their customers and systematically segment their markets and channels of distribution (Nueno and Quelch 1998). The challenge here thus becomes to accept and integrate, at least to some extent, different symbolic needs and expectations without losing control of the identity of the brand. At the same time, the multiplicity and diversity of symbolic needs has very important consequences. A "streetwise" creator, having a good sense of the symbolic needs 627 MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industry of a particular target group and able to secure significant financial backing, can create a new brand or renovate an old one and successfully develop and exploit it. The astounding revival in the 1990s of brands like Gucci and Prada or the success of the Italian Etro, which in less than ten years has been transformed from an obscure fabric designer to a cult fashion brand, bear witness to that (La Tribune, 7 April 1998). Quick success, though, goes together with quick failure, as when the space for dreams associated with a particular brand no longer resonates with the symbolic needs of a large enough group of customers. The recent bankruptcy of the American Isaac Misrahi, which experienced rapid growth and success in the early 1990s, or the difficulties of the British company Laura Ashley are clear illustrations of that (Wall Street Journal, 20 April 1998). 5. Organizational Responses: The Institutional Embeddedness of Trajectories Facing radical market transformation, increasing technological complexity, and heightened competition, luxury fashion companies have had to react and adapt. In the context of significant environmental dislocation, many have felt the need for greater organizational flexibility. The meaning of organizational flexibility, though, and the paths or trajectories leading to it have varied significantly, particularly across national borders. In the luxury fashion industry, one can in fact easily identify national paths or trajectories to flexibility. Differences between these national ideal types can largely be traced to historical legacies and institutional constraints, which, at least thus far, have been defined, constructed, and reconstructed at the national level. Exceptions to these national ideal types can naturally be found in each country, but they remain on the whole marginal and isolated. 5.1. French Craft-Based Organizations-The Weight of Tradition Until World War 11, French haute couture was a highly stable industry where made-to-order clothes for women were the principal source of revenue. The business of designing, malung, and selling fashionable custom-made clothing was a craft and labor-intensive process that did not require significant investments. The clientele was small and homogeneous, mostly drawn from within the ranks of the French aristocracy and upper-middle class (Grumbach 1993, Crane 1997). This privileged elite was furthermore fairly loyal, and clients tended to dress exclusively with one single house. In the tradition pioneered by Charles Frederick Worth, 628 a couturier or artistic designer typically owned each house. The image and success of each house was closely dependent upon the unique creative power of its designer, who was also often in charge of running the business. This designer has traditionally been a "cult" figure, around whom everything seemed to revolve and upon whom everything seemed to depend. As artist and craftsman, he or she ensured that the process as a whole, from creation to manufacturing and distribution, would be fully integrated (Dumas 1998). This step-by-step monitoring, which amounted in fact to a quality control integrated upstream, was essentially achieved through a reliance on traditional savoir faires-or traditional craft techniques-inherited from the past. Thus, while the success and predominance of French haute couture houses originally built upon their explorative capacities in design and creation, the dominant logic when it came to technology and organization was an exploitation of old and seasoned recipes. To this day, the very identity of French haute couture houses has appeared to lie in their reliance upon traditional practices at all stages of the integrated production process. As already underscored, the predominance of French haute couture houses was further strengthened for a while by the existence of a protective set of institutions and rules that all but sealed off the boundaries of the industry. Each house had in itself little slack, but the institutionalized organization of the French fashion industry created somewhat of a buffer zone around members of that industry. This system emphasized stability and conformity, preventing radical breaks and rejuvenation. As a consequence, the creative spirit within French haute couture houses has gradually tended to dry up. Losing somewhat their creative edge, French haute couture houses thus entered the 1980s in an already weakened position. By that time, furthermore, the global challenges described above were becoming too strong to be ignored. Facing competition and increasing levels of environmental turbulence and ambiguity, French haute couture houses had to think of possible ways to respond. They had, in particular, to consider how they might integrate a degree of organizational flexibility. They came up with essentially two kinds of answers-outsourcing and what we call here the "umbrella holding." Outsourcing and the Risk of Losing One's Identity. Breaking up the integrated, in-house process and outsourcing or licensing parts of it represented an obvious way to secure advantages of scale and scope while still integrating a degree of organizational flexibility. This solution has tempted several French haute couture houses. Companies like Pierre Cardin or Christian Dior were among the first to license their lines or products on a large ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industq scale. But, starting in the 1970s, most haute couture houses were engaging in at least partial licensing or outsourcing for their perfumes, accessories, and ready-towear lines. Although easy and tempting, such a strategy undeniably entailed a number of risks. In the craft-based and tradition-grounded French luxury fashion industry, stable organizational solutions were more than a virtue. They were part and parcel of the product and of the brand image. Because it was bound to disrupt the integrated, inhouse process, any attempt to develop productive capacities, to exploit particular products or the brand name, could threaten the specificity of the products and the identity of the company. This has been the argument put forward to this day by a company like Hermirs in refusing any tampering with its image through licensing or outsourcing. A source of great pride for Hermks-and an important dimension of its self-definition as a "luxury" company-is undeniably the fact that never in its history did it give in to the temptation of outsourcing or licensing (Dumas 1998; Les Echos, 26 March 1998). In striking contrast, Pierre Cardin did bet on systematic licensing. This company explored new solutions to the detriment of traditional craft techniques and organizational routines, which had for a long time made up the core of its identity as a haute couture house. It privileged, in the process, the systematic exploitation of existing products and of its brand name over the explorative core competence, which had originally built and maintained its identity as an innovative fashion company. This eventually led the Chambre Syndicale de la Haute Couture to exclude the Pierre Cardin house. The company Pierre Cardin still exists and it is in fact quite prosperous-a total turnover of $6.5 billion (U.S.) in 1997, profits undisclosed. The Pierre Cardin brand name, though, has undeniably lost some of its luster and its products are not so readily identified as luxury goods (Les Echos, July 11, 1998). Constrained by peculiar historical legacies and institutional conditions, French haute couture houses have thus had limited room for maneuver. They have traditionally been defined by the very integration of their activity and by the full control which the main designer has kept over it. In this context, the classical and easy path to flexibility-outsourcing parts of the process and some of the activities of the organization-has appeared to threaten the very identity of these companies. This does not mean, though, that under conditions of severe environmental turbulence the French luxury fashion industry is necessarily doomed. In the face of global challenges, some French luxury fashion houses appear to follow their own, quite unique path to organizational flexibility. This ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 path is leading them directly from a craft-based model of organization to a form of flexible or modular organizational arrangement, which we label here the "umbrella holding." The "Umbrella Holding. " The strategy of diversification devised before the First World War by the French couturier, Poiret, anticipated what we call the "umbrella holding" trajectory. Poiret had early on been aware that the haute couture industry faced an uncertain future. His idea was to launch a diversification of the product range and in particular to add subsidiary lines of products to clothing (Grumbach 1993). This type of strategy seduced other couturiers. It became increasingly successful and widespread after the Second World War, when the market for made-to-order clothes gave clear signs of rapid decline. The main subsidiary product was perfume, but many haute couture houses also diversified into cosmetics, accessories, and most importantly into different types of ready-to-wear lines of clothing for women, but also for men, and in some cases for children. When it gained significant scope, such a strategy required integrating a degree of flexibility and modularity within the boundaries of the organization. Each product line was different, and each unit producing and marketing it became relatively autonomous, talung a life of its own under the umbrella of the brand name. Some companies-Christian Dior or Yves Saint Laurent for example, but not Hermks-subcontracted parts of the production for subsidiary lines. In this context, the all-important role of the main designer or couturier was to maintain tight control over the subsidiary lines and the subcontractors, so that quality problems would not impact upon the brand name. The diversification strategy in the French luxury industry has thus traditionally amounted to creating, under a single brand name, several relatively autonomous organizational entities, each being fully in charge of a given product line and only controlled from the core. By hindsight, at least, it is thus not surprising that a peculiar conglomerate or holding company such as LVMH (Louis Vuitton Moet Hennessy) would emerge in that industry. The strategy of Bernard Arnault-founder and present Chairman of LVMH-has been to bring together, under the same umbrella holding company, a number of formerly independent luxury houses, and to exploit their brand reputation while keeping each house largely autonomous. The idea is that each brand is valuable on its own and that more harm than synergy could result from too close an integration. The portfolio of LVMH currently includes luxury fashion houses such as Givenchy, Kenzo, Christian Lacroix, CCline, and Christian Dior but also a number of luxury companies producing accessories 629 MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion lndzrstly (Louis Vuitton, Loewe), perfumes, wines, or spirits. Recently, LVMH has diversified further into selective retailing and distribution, an activity that directly complements the production of luxury goods. The most famous example of such vertical diversification has been the acquisition in 1996 of Duty Free Shoppers (DFS), a worldwide chain of duty free outlets (Choez 1998; Le Figaro, 16 March 1998). The umbrella holding or the peculiar French path to flexibility is sketched out in Figure 1. As of now, LVMH is still an aggregate in search of identity. A number of recent developments, though, seem to indicate that this aggregate could in time develop into a truly flexible internal network. Bernard Arnault and his top management are working to create and foster interfaces between the various modules or entities. The objective is to optimize the overall performance of the aggregate or internal network by rationalizing a number of potentially common activities-marketing, distribution, or even production (Martin 1998). At the very same time, the top management of LVMH is very conscious of the need to preserve the traditional craft-based nature of each module. This, after all, is where the value of French luxury fashion brands lies. While the umbrella holding trajectory is still very much in construction, we believe that it may be the only chance of survival in the long term for individual French fashion-and in particular haute couture-houses. 5.2. Italian and American Fashion Companies-The Challengers Most French haute couture houses have thus been finding it hard, outside or within LVMH, to strike a balance between their traditional identity and the changes required by evolving markets and market demand. In the meantime, new entrants from Italy and the United States have seized upon these trends, constructing them as opportunities rather than threats. These challengers chose not to compete head-on with traditional French players. Rather, they welcomed and fostered a radical redefinition, for the industry, of the rules of the competitive game that could only be to their benefit. They did so by betting on and building upon what had merely been a side strategy for French couturiers. They systematically turned subsidiary lines-luxury ready-towear, perfume, cosmetics, or accessories-into the core of their business. In the process, they developed what were, for the luxury goods industry, original and innovative organizational solutions. To make the most of the expansion of markets, Italian and American challengers increased their productive capacities through mechanization and standardization as well as partial outsourcing. To handle the greater sophistication of customers, the 630 multiplicity of their profiles, and the changing nature of their needs, they designed flexible and speedily reactive organizations, directly tuned into the "world of the street." In the recent past, two variants of the flexible network organization-one Italian, the other American-have thus emerged in the luxury fashion industry. These variants differ essentially along two dimensions. In the Italian case, the networks of partners and subcontractors tend to have a smaller and much more stable membership than in the American case. The extent of outsourcing is much more significant in the American case than in the Italian case, with only a minimal range of activities being kept in-house. While both the Italian and American variants have emerged in the context of significant and global environmental dislocation, each exemplifies a particular trajectory towards flexibility and modularity. In turn, these two trajectories are quite different from the one identified in the French case. These trajectories, we show below, have been shaped in part by the early choices of national pioneer companies and by the national institutional environments in which these companies were embedded. In fact, the difference between both variants can be traced to the historical and institutional legacies peculiar to each national environment. The Italian Model: The Flexible Embedded Network. A peculiarity of the Italian economy-particularly in the region that has come to be known as the Third Italy-has been the survival and dynamism of traditional industrial districts. The prosperity, since 1945, of industrial districts in this region has in fact quite strikingly contrasted with their near disappearance in some other Western European countries, particularly in France (Djelic 1998). These industrial districts are made up of large numbers of small entities, whether craft workshops or small industrial firms (Brusco 1982, Goodman and Bamford 1989, Lazerson 1995). Engaging in multilateral forms of cooperation, these entities have traditionally created dense networks backed by adequate institutional and political support, at the regional and national levels (Weiss 1988). Embedded as they often were in these traditional industrial districts, Italian fashion houses have been able to come up with innovative organizational solutions. When, in the French case, the path to luxury fashion has been design and creation, manufacturing has been the core competence Italians have built upon (The Economist, 11 April 1998; La Tribune, 7 April 1998). The Italian fashion industry has emerged as a set of flexible and relatively stable networks, tightly embedded in local or regional industrial communities. These networks have enabled ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 MARIE-LAURE DJELIC AND ANTTI AINAMO Figure 1 The French Trajectory-The Coevolution in the Fashion Industry Umbrella Holding Finance, Personnel... C O M M O N SERVICES (or potentially so) Logistics, Marketing, Advertising, Distribution, Manufacturing, Subcontracting CORE PERIPHERY Individual House Individual House Individual House Manufacturing, Sales, Brand Management Manufacturing, Sales, Brand Management Manufacturing, Sales, Brand Management Some subcontracting : second lines or accessories Italian fashion houses to reconcile flexibility with almost limitless production capacity. Italian fashion companies have subcontracted a large share of their manufacturing activity, keeping in-house only a few key product lines. Their networks of industrial partners have typically been small and stable, allowing them to keep tight but still informal control. By privileging networks over organic growth, Italian fashion companies have been able to combine the best of two worlds. On the one hand, they have been able to easily and flexibly manage variation in demand. On the other hand, they have not had to sacrifice craft production or the quality of products. By setting up networks, Italian fashion companies have kept the advantages of small size and flexibility-creativity, adaptability, and speed of reaction to market changes. By limiting, and sometimes altogether avoiding, the licensing strategy, they have been able to keep stricter control over products and brands (Women's Wear Daily, 10 December 1997). The tight nature of the network linking Italian luxury fashion companies to their subcontractors and the regularity of partnerships have made direct control and monitoring relatively easy. Italian fashion companies have naturally handled problems of interface by setting up and systematically using quality-control tools. The foundation of trust, though, stemming from a common institutional and cultural background, has by far been the most powerful coupling mechanism (Brusco 1982, Granovetter 1985, Powell 1990, Lazerson 1995). This particular path to flexibility or modularity has been common to Italian fashion companies from the low ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 NO subcontracting Some subcontracting : second lines or accessories to the high end of the industry. It is characteristic of a company like Benetton, but it also defines the organization of luxury fashion companies such as Armani, Gucci, or Prada (The Economist, 11 April 1998; Le Monde, 18 January 1998). By following this particular path towards the network form, Italian fashion companies have decoupled somewhat their brand name from the product and the production process, in striking contrast to French haute couture houses. Italian fashion companies thus pay close attention to brand management in itself, distinct from the monitoring of product, creation, or distribution. As shown in Figure 2, core activities have increasingly been handled as separate modules-and these core activities have included design and creation, brand management (marketing, advertising, promotion), or quality control. Gucci's Tom Ford explains the rapid success of his company in the 1990s by the creation of a coherent brand and by its close monitoring. In the case of Gucci, reaching Figure 2 Core and Peripheral Modules in the Italian Trajectory MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industry coherence has meant revamping all products and product lines. A combination of coherent brand management with high-quality manufacturing, subcontracted to a tight network of regular partners, has also been the recipe behind the rapid and astounding success of a company like Giorgio Armani (Le Monde, 18 January 1998). The organizational model described above is naturally an ideal type and a number of exceptions may be found. Its key features, however, are important to explain the success of Jtalian luxury fashion companies in the 1980s and early 1990s. Success has bred confidence and, through confidence, stability. The fundamentals of the model have thus been institutionalized, and they are not being questioned. New challenges, though, are emerging. For companies like Gucci, Prada, or Armani, the main questions today seem to be how to sustain growth, ensure survival, and create the conditions for continuing performance independent of the current designer or management team. Rather than going public-a strategy more readily considered in the United States-Italian companies seem to take an interest in the model of the umbrella holding company as exemplified by LVMH. Patrizio Bertelli, Prada's Chairman, praises LVMH's strategy and seriously considers, apparently, following it himself: The idea is to acquire other luxury goods brands that could not only coexist, but exploit synergies wherever possible. For example, in the area of service and financial management-these areas of business can be in common. (Patrizio Bertelli in Women's Wear Daily, 10 December 1997). The American Model: Towards the Virtual Orga~zization. A specific feature of the American economy is the sheer size and buying power of its national market. That the United States invented mass production and mass marketing is therefore not surprising (Djelic 1998). With such a market to serve, and to reach economies of scale, American companies have for some time viewed specialization and subcontracting as possible and efficient strategies. The sheer size of the country, though, and the history of its settlement explain that subcontracting was not as likely as in Italy to be contained within strict regional boundaries. Even in those locally or regionally constituted communities that share a common basis of trust, interfirm networks have always included, in the United States, a majority of arm's-length ties regulated by spot contracts (Uzzi 1997, p. 42). In contrast, the Italian flexible and embedded networks have generally favored tight links and long-term partnerships. The particular path or trajectory followed by American fashion companies mirrors the peculiarities of the institutional context in which they have evolved. It also reflects the process through which these companies have 632 entered the luxury segment of the fashion industry. Ralph Lauren, Calvin Klein, and Donna Karan have managed to establish themselves in the luxury fashion industry without submitting to the stringent rules and requirements set up by the French community of couturiers. Even more than their Italian competitors, American fashion companies are new entrants at the luxury end of the fashion industry. They have managed to impose themselves without building upon a product legitimacy that had traditionally been symbolized, in the industry, by an haute couture activity. American fashion companies originally started from mass production at a lower end of the industry. To this day, Ralph Lauren, Calvin Klein, or Donna Karan do not have any haute couture activity. Over time, though, they have managed to scale up their products and their image by creating around their brand name a coherent space for dreams that echoed the symbolic needs of a particular customer group. This particular strategy has made it possible for American companies to price beautiful but rather standard ready- to-wear products at a level comparable to that of their French haute couture competitors-Christian Dior or Chanel for example. These particular historical and institutional legacies explain why it has been relatively easy for American players to fully decouple brand from product and to outsource most of the production process. Brand management has emerged, in this context, as the core or strategic competence of the organization. When design and creation are still at the heart of the French luxury fashion industry and manufacturing is key to Italian fashion, one could easily argue that, for American players, the source of competitive advantage has been brand management. There are, naturally, a number of small- or medium-sized American fashion houses, which play upon creative design and innovation rather than brand management. The names Anna Sui, Michael Kors, or Bob Mackie come to mind. These companies are closer in type to the French cre'ateur houses. Altogether, though, they weigh very little on the American and worldwide market and cannot compare to the likes of Calvin Klein, Ralph Lauren, or even Donna Karan (Women's Wear Daily, 7 November 1997). In the case of those leading American companies, the radical decoupling of the brand name from product and production has made possible and even triggered organizational modularity. In fact, the path to modularity followed by these companies has been characterized by speed and by its systematic nature. The emerging network form of organization is quite close, in the end, to what the literature has labeled the "virtual organization." American fashion companies have kept in-house only a minimal range of activities-those defined as strategic ORGANIZATION SCIENCE NO^. 10, NO. 5 , September-October 1999 MARIE-LAdRE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Indzlstry and core competences and amounting in effect to a monitoring of the overall coherence of the brand. For companies like Ralph Lauren, Calvin Klein, or Donna Karan, the value of the brand name is naturally related to the quality of products and processes. At the same time, though, and maybe more importantly, the legitimacy and the value of products and product lines stem from the sets of symbolic representations that are associated with the brand name. American fashion companies build upon concepts-those intangible goods that define brand names. From these concepts, they deduce product lines and products but also marketing strategies, the design of stores, and other communication tools (Nueno and Quelch 1998). For fashion companies like Ralph Lauren, Calvin Klein, or Donna Karan, managing the brand has thus implied much more than a simple attention to the quality and consistency of products and services. It has meant ensuring the overall coherence of the script, articulating a series of interdependent sequences of events, selecting some, dropping others, arranging and rearranging them according to changing conditions and needs. These companies have not been afraid of complex organizational arrangements consisting of large numbers of relatively autonomous modules. They can be described as "hubs," where the core competence is the coordination and management of interfaces between modules. Like the Italians, but unlike the French, leading American fashion companies have entirely licensed off or subcontracted manufacturing. In contrast to both the Italians and the French, the Americans have also outsourced, at least in part, design, creation, or brand management to freelance designers, communication and advertising agencies, or consultants. As Figure 3 shows, leading American fashion companies have retained in-house little more than strategic decision making. They have kept full control over important choices regarding design and Figure 3 product lines, the selection of materials, brand management (marketing, advertising, promotion, merchandising), the handling of quality-control tools, the monitoring of subcontracting partners or licensees, and, finally, the elaboration of a distribution strategy. American fashion companies like Ralph Lauren, Calvin Klein, or Donna Karan have thus evolved towards a network form of organization. Unable to build upon, though, a locally embedded community of potential partners and subcontractors as in the Italian case, they have created networks their own way. They have mostly relied on arm's-length ties and spot contracts and they have built in flexible membership so as to follow organizational needs and market transformations. The challenge with this type of network arrangement stems essentially both from the shifting geometry of the network and from the great number of interfaces between the various modules. To reach a satisfactory level of control, American fashion companies have had to foster and bring about a degree of standardization of these interfaces. They have done so by setting clear and detailed sets of instructions to map out the work of subcontractors and partners, but also by defining strict quality standards, deadlines, and control procedures. This monitoring of interfaces has generally been formalized through explicit legal contracts. It has been made possible by the integration of sophisticated information systems. In this particular context, the role of what we call here an "organizational pilot" appears to be quite significant. In companies like Ralph Lauren, Calvin Klein, or Donna Karan, the responsibility with respect to the overall coherence of the brand does indeed lie with such a pilot. He or she is at the same time the founder, the designer, and, maybe most importantly, the top manager. Naturally, the role of this organizational pilot is far from being a simple one. The recent difficulties Donna Karan has faced in her attempt to articulate the core of her company with its Core and Peripheral Modules in the American Trajectory I CORE (OUTSOURCED) PERIPHERY STRATEGIC DECISION MAKING 1"V.I- 1 1 71 Design, Creation designers) (subcontractors) ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 subcontractors and licensees (multiple licensees and outlet stores) MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industry peripheral modules clearly bear witness to that (Les Echos, 24 March 1998). 6. Discussion and Concluding Remarks The environments in which organizations are embedded set constraints and define opportunities. They very much shape, as a consequence, the paths or trajectories which these organizations are going to follow and their structural evolution through time. Focusing on the luxury fashion industry as our empirical base of evidence, we found that it has had to face increasingly turbulent and ambiguous environmental conditions, particularly over the past 30 years. Before this period of significant environmental dislocation, French haute couture houses had been able, due to their early lead in the luxury fashion industry, to set the rules of the game and to control the boundaries of the industry. In the process, they institutionalized technological and organizational stability. In recent years, though, French fashion houses have found it more and more difficult to ignore the pressures stemming from global challenges. Market expansion and diversification have combined with radical technological evolution and increasing competition to alter the rules of the game. The main challengers have been Italian and American companies. Without an haute couture tradition, these companies have successfully managed to scale up their product lines and to reposition themselves at the high or luxury end of the fashion industry. These outsiders have seized upon global environmental trends to foster a redefinition of the rules of the game in the luxury fashion industry that could only be to their benefit. They have contributed, in the process, to increasing the level of environmental dislocation and turbulence in that industry. In the face of such environmental turbulence, most companies in the luxury fashion industry, including French haute couture houses, have felt the need to react. Organizational flexibility and modularity-or the network form-have emerged as a common answer. However, we have found that organizational modularity or the network form did not always mean the same thing and that there were in fact different paths or trajectories leading to organizational flexibility. Historical legacies, we have shown, and the peculiarities of national institutional contexts have to a significant extent shaped these trajectories. Opportunities and constraints have differed from one national environment to the next. Nationally defined opportunities and constraints had an impact upon the French pioneers, who created and institutionalized early organizational solutions in that industry. In turn, these early solutions shaped environmental conditions for organizational players, setting the boundaries of the industry, the rules of the game, and even the possible ways to go around them. 634 Altogether, we have thus been able to document in this paper that global trends or challenges on the one hand, historical legacies and the peculiarities of national institutional constraints on the other, have shaped and continue to shape paths to organizational flexibility as well as the network forms emerging in the luxury fashion industry. In France, the weight of organizational legacies and institutionalized practices turned out to create significant constraints for haute couture houses, narrowing and limiting the range of possible reactions in the face of global challenges. In particular, licensing and outsourcing, as strategies to integrate a degree of organizational flexibility, have appeared to be fraught with dangers. Such strategies were bound to disrupt the integrated, in-house production process and, as a consequence, they were likely to destroy the value of the products and to threaten the image and the very identity of French haute couture houses. This does not have to mean, though, that under conditions of significant environmental disruption, the French luxury fashion industry is necessarily doomed. As a matter of fact, a peculiar trajectory to organizational modularity has been emerging in the French case, which we have labeled the "umbrella holding." The umbrella holding brings together existing organizational entities and brand names. It becomes in the process an internal network that can rationalize a number of shared competences while respecting the autonomy and integrity of each component module. Considering the historical and institutional constraints characteristic of the French case, the umbrella holding appears to be a locally legitimate way to adapt to global change. At the same time, we argue that it might be the only chance of survival, in the long term, for individual French fashion houses. While French haute couture houses have generally been quite reluctant to change, Italian and American challengers early on seized upon environmental trends to make significant headway in the industry. Identifying global challenges as opportunities, they have managed to redefine in part the rules of the game. In order to be at the same time globally efficient and locally in tune with their customer base, Italian and American players have created flexible organizational solutions. Instead of bringing together several craft-based entities and brand names under a single umbrella holding, they have defined themselves and organized as networks of competences. A number of these competences-those considered particularly strategic-were integrated at the core, while others, more peripheral, were outsourced or licensed off. Beyond the similarities, though, there were enough differences between the Italian and the American cases that ORGANIZATION SCIENCE Nol. 10, NO. 5, September-October 1999 MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industry we could in fact identify two variants of the flexible network organization-one Italian, the other American. In the context of significant and global environmental dislocation, each variant exemplifies, we argue, a particular trajectory towards organizational flexibility. We have shown in this paper that differences between those two trajectories can be traced to the historical and institutional legacies peculiar to each national environment. Italian fashion companies have made the most of the industrial districts in which they were locally embedded to construct flexible but relatively stable networks with regular partners. In the Italian case, as a consequence, networks of partners and subcontractors have tended to be tight and strictly embedded within the local or regional community. The foundation of trust stemming from a common institutional and cultural background has made the management of interfaces fairly easy and not too costly. Evidence from the American case, on the other hand, points to a more open network form. In this case, the extent of outsourcing has been much more significant and only a minimal range of activities has been kept in-house. This would seem to indicate an evolution in the direction of what has been labeled in the literature the "virtual organization," where more and more activities are being outsourced. The open character of such an organizational solution means that an increasingly smaller core has been responsible for the articulation of a series of relatively independent tasks. Even in situations of high ambiguity and uncertainty, this form of organization could remain most flexible and adaptable. Modules could be rearranged, newly created, or altogether dropped to fit changing conditions and purposes. The American or "virtual" model gives a lot of weight to an organizational pilot or manager, in contrast to the key role of the product designer in the French tradition of haute couture or to the collective responsibility characteristic of the Italian flexible and embedded network. The manager becomes responsible for steering the organization, identifying opportunities, making choices, and mapping the migration path of the organization. Of course, this particular characteristic of the American solution is also its main weakness, and the identification of the skills and competencies necessary for such organizational pilots is certainly an interesting path for future research. To sum up, we have documented in this paper that, in the luxury fashion industry, organizational flexibility or modularity has emerged as a common answer to global challenges and increasing environmental dislocation. Probing further, however, we have been able to identify ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 three different paths or trajectories leading to such organizational flexibility or modularity-the "umbrella holding," the "flexible embedded network," and the "virtual organization." In each case, we have traced the peculiarities of these paths or trajectories-constructed in the paper as national ideal types-to peculiar historical legacies and institutional constraints. We have thus found strong support, in the case of the luxury fashion industry, for a coevolution perspective where environmental transformation and organizational change interplay through time, feeding upon each other (Lewin et al., in this issue). By pointing to path-dependent and historically constructed processes of interaction between organization forms, global environmental trends, and national legacies, this paper contributes to discussions about the mechanisms of coevolution. We have to conclude, though, that, in our minds, the existence of different trajectories makes convergence or congruence highly unlikely. The evidence we have built from does not allow us to claim that the different variants of flexible organizations might someday converge towards a global and uniform standard. In the debate on new organization forms, we thus question the likelihood of a future stabilization around a unique organizational solution that would become widely legitimated and institutionalized. Rather, we argue that differences are there to stay beyond the period of transition and acute environmental dislocation, embedded as they are in powerful historical and institutional legacies. On the basis of this study, we believe that more research should be undertaken in at least two directions. There is a need, firstly, to test the applicability of the three network forms or ideal types identified here in other contexts and in other industries. Another important issue seems to be the role of organizational pilots or managers. While the coevolution perspective provides an interesting theory of organizational and institutional constraints, there is a need to cross-fertilize it with a theory of agency. More work is undeniably needed, both empirical and theoretical, in that direction. Acknowledgments For critical comments on earlier drafts of this paper, the authors would like to thank Arie Lewin, Henk Volberda, the participants at the EGOS 1998 Meetings-Subtheme 3 on New Organizational Forms, and Jane Salk, as well as three anonymous reviewers for Organization Science. References Ainamo, A. 1996. Industrial design and business perfonnance: A case study of a Finnish fashion finn. Ph.D. Dissertation, Helsinki School of Economics and Business Administration, Helsinki, Finland. Bartlett, C., S. Ghoshal. 1989. Managing Across Borders: The Transnational Solution. Harvard Business School Press, Boston, MA. 635 MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industly Bourdieu, P. 1979. La Distinction. Editions de Minuit, Paris, France. . 1983. The field of cultural production or the economic world reversed. Poetics: J. Empirical Res. Literature, the Media and the Arts. 12 311-356. Brown, S., K. Eisenhardt. 1998. Competing on the Edge: Strategy as Structured Chaos. Harvard Business School Press, Boston, MA. Brusco, S. 1982. The Emilian model: Productive decentralization and social integration. Cambridge J. Econom. 6. Bums, T., G. Stalker. 1961. The Marzagerilerlt of Itzrzovation. Tavistock, London. CERNA. 1995, L'industrie du luxe dans l'e'conomie fran~aise. Ministkre de L'Industrie, Paris. Chandler, A. 1990. Scale and Scope. Harvard University Press, Cambridge, MA. Child, J. 1972. Organizational structure, environment and performance: The role of strategic choice. Soc., 6 1-22. Choez (1998). Interview with M. Choez, General Manager of the Perfumes Division, LVMH. Clegg, S. 1990. Modern Organizations: Organization Studies in the Postmodern World. Sage, London. Crane, D. 1997. Globalization, organizational size, and innovation in the French luxury fashion industry: Production of culture theory revisited. Poetics: J. Empirical Res. Literature, the Media and the Arts. 24 3 9 3 4 1 4 . Dijksterhuis, M., F. Van den Bosch, H. Volverda. 1999. Where do new organizational forms come from? Management logics as a source of coevolution. Organ. Sci. 10 (5). This issue. Djelic, M. L. 1998. Exporting the American Model. Oxford University Press, Oxford. , M. Gutsatz. 1998. Organizational and managerial competences in the luxury goods industry-between tradition and postmodernism. Paper presented at the IVth International Conference on Competence-Based Management, Oslo, Norway (available on request). Dumas (1998), Interview with M. Dumas, Chairman, Hermks. Les Echos, Donna Karan International flouge lourdement dans le rouge. 24 March 1998. , Jean-Louis Dumas souhaite ajouter de nouvelles brancher B Herniks. 26 March 1998. ,Luxe: tlargir l'offre sans banaliser les marques. 30 April 1998. , Pierre Cardin n'exclut pas de s'introduire en bourse. 11 July 1998. Ecole de Paris. 1998. De'bat: Le luxe, domaine des rtves (March 23). The Economist, The Changing Fabric of Fashion. 11 April 1998. Le Figaro, LVMH: Courment reboudir. 16 March 1998. Financial Times, 27 February 1998. Goodman, E., J. Bamford, eds. 1989. Small Firms and Industrial Districts iiz Italy. Routledge, London. Granovetter, M. 1985. Economic action and social structure: The problem of embeddedness. Amer. J. Soc. 91 481-510. Grumbach, D. 1993. Histoires de la Mode. Seuil, Paris. Guillkn, M. 1997. Scientific management's lost aesthetic: Architecture, organization, and the Taylorized beauty of the mechanical. Admin. Sci. Quart. 42 682-715. Gutsatz, M. 1996. Le luxe: Reprtsentations et competences. Decisions Marketing 9. 636 Hannan, M., J. Freeman. 1984. Structural inertia and organizational change. Amer. Soc. Rev. 49 149-164. Jellinek, S. 1997. Le sage et la cynique: DBbat virtue1 sur des questions fondamentales de parfumerie. Dragoco Report 4 19-148. Koza, M., A. Lewin. 1998. The coevolution of strategic alliances. Organ. Sci. 9 (3) 255-265. -. 1999. The coevolution of network alliances: A longitudinal analysis of an international professional service network. Organ. Sci. 10 (5). This issue. Lawrence, P., J. Lorsch. 1967. Organizatiotz and Environment. Harvard Business School Press, Boston, MA. Lazerson, M. 1995. A new phoenix?: Modem putting-out in the Modena knitwear industry. Admin. Sci. Quart. (March) 34-59. Levinthal, D., J. March. 1993. The myopia of strategic learning. Strategic Management J. 14 95-1 12. Lewin, A,, C. Stephens. 1993. Designing postindustrial organizations: Combining theory and practice. G. Huber, W. Glick, eds. Organizational Change and Redesign. Oxford University Press, New York. , C. Long, T. Carroll. 1999. Coevolution of new organizational forms. Organ. Sci. 10 (5). This issue. Lewin, K. 195 1. Field Theoly in Social Science. Harper and Row, New York. March, J. 1991. Exploration and exploitation in organizational learning. Organ. Sci. 26 563-577. . 1995. The future, disposable organizations and the rigidities of imagination. Organ. 2 71-97. Martin (1998), Interview with M. Martin, Chainnan, Kenzo, LVMH. Miles, R., C. Snow. 1986. Organizations: New concepts for the new forms. California Management Rev. 28 62-73. Le Monde, Du gris, du chic, du confort. 18 January 1998. Nohria, N., R. Eccles, eds. 1992. Networks and Organizations. Harvard University Press, Cambridge, MA. Nueno, J., J. Quelch. 1998. The mass marketing of luxury. Bus. Horizons (November-December) 61-68. Pfeffer, J., G. Salancik. 1978. The External Control of Organizations: A Resource Dependence Perspective. Harper and Row, New York. Powell, W. 1990. Neither market nor hierarchy: Network forms of organization. Res. in Organ. Behavior 12 295-336. , P. DiMaggio, eds. 1991. The New Institutionalism in Organizational Analysis. The University of Chicago Press, Chicago. Scott, R., et al. 1994. Institutional Environments and Organizations. Sage, Newbury Park, CA. Simon, H. 1996. Sciences of the Artificial, 3rd ed. MIT Press, Cambridge, MA. Taylor, F. W. 1911. The Principles of Scientijc Management. Harper, New York. Taylor, W. 1991. The logic of global business: An interview with ABB's Percy Barnevik. Harvard Bus. Rev. (March-April). La Tribune, Etro reussit sa mutation vers le frtt-&-porter.7 April 1998. Uzzi, B. 1997. Social structure and competition in interfim networks: The paradox of embeddedness. Admin. Sci. Quart. 42 35-67. Volberda, H. 1996. Towards the flexible form: How to remain vital in hypercompetitive environments. Organ. Sci. 7 (4) 359-374. . 1998. Building the Flexible Firm: How to Remain Competitive. Oxford University Press, New York. ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 MARIE-LAURE DJELIC AND ANTTI AINAMO Coevolution in the Fashion Industq Wall Street Journal, Laura Asley finds aMalaysian Supporter. 20 April 1998. Webb, D., A. Pettigrew. 1999. The temporal development of strategy: Patterns in the UK Insurance Industry. Organ. Sci. 10 (5). This issue. Weick, K. 1969. The Social Psychology of Organizations. Addison Wesley, New York. Weiss, L. 1988. Creating Capitalism: The State and Small Business Since 1945. Basil Blackwell, New York. Women's Wear Daily, Fashion Waves. 7 November 1997. , Powering Prada. 10 December 1997. Woodward, J. 1965. Industrial Organization: T h e o q ~and Practice. Oxford University Press, London. Accepted by Arie Y. Lewin; received September 11, 1998. This paper has been with the authors for two revisions. ORGANIZATION SCIENCE NO^. 10, NO. 5, September-October 1999 http://www.jstor.org LINKED CITATIONS - Page 1 of 2 - You have printed the following article: The Coevolution of New Organizational Forms in the Fashion Industry: A Historical and Comparative Study of France, Italy, and the United States Marie-Laure Djelic; Antti Ainamo Organization Science, Vol. 10, No. 5, Focused Issue: Coevolution of Strategy and New Organizational Forms. (Sep. - Oct., 1999), pp. 622-637. Stable URL: http://links.jstor.org/sici?sici=1047-7039%28199909%2F10%2910%3A5%3C622%3ATCONOF%3E2.0.CO%3B2-%23 This article references the following linked citations. If you are trying to access articles from an off-campus location, you may be required to first logon via your library web site to access JSTOR. Please visit your library's website or contact a librarian to learn about options for remote access to JSTOR. References Where Do New Organizational Forms Come From? Management Logics as a Source of Coevolution Marjolijn S. Dijksterhuis; Frans A. J. Van den Bosch; Henk W. Volberda Organization Science, Vol. 10, No. 5, Focused Issue: Coevolution of Strategy and New Organizational Forms. (Sep. - Oct., 1999), pp. 569-582. Stable URL: http://links.jstor.org/sici?sici=1047-7039%28199909%2F10%2910%3A5%3C569%3AWDNOFC%3E2.0.CO%3B2-S Economic Action and Social Structure: The Problem of Embeddedness Mark Granovetter The American Journal of Sociology, Vol. 91, No. 3. (Nov., 1985), pp. 481-510. Stable URL: http://links.jstor.org/sici?sici=0002-9602%28198511%2991%3A3%3C481%3AEAASST%3E2.0.CO%3B2-R The Co-Evolution of Strategic Alliances Mitchell P. Koza; Arie Y. Lewin Organization Science, Vol. 9, No. 3, Special Issue: Managing Partnerships and Strategic Alliances. (May - Jun., 1998), pp. 255-264. Stable URL: http://links.jstor.org/sici?sici=1047-7039%28199805%2F06%299%3A3%3C255%3ATCOSA%3E2.0.CO%3B2-D http://www.jstor.org LINKED CITATIONS - Page 2 of 2 - The Coevolution of Network Alliances: A Longitudinal Analysis of an International Professional Service Network Mitchell P. Koza; Arie Y. Lewin Organization Science, Vol. 10, No. 5, Focused Issue: Coevolution of Strategy and New Organizational Forms. (Sep. - Oct., 1999), pp. 638-653. Stable URL: http://links.jstor.org/sici?sici=1047-7039%28199909%2F10%2910%3A5%3C638%3ATCONAA%3E2.0.CO%3B2-J The Myopia of Learning Daniel A. Levinthal; James G. March Strategic Management Journal, Vol. 14, Special Issue: Organizations, Decision Making and Strategy. (Winter, 1993), pp. 95-112. Stable URL: http://links.jstor.org/sici?sici=0143-2095%28199324%2914%3C95%3ATMOL%3E2.0.CO%3B2-1 Toward the Flexible Form: How to Remain Vital in Hypercompetitive Environments Henk W. Volberda Organization Science, Vol. 7, No. 4. (Jul. - Aug., 1996), pp. 359-374. Stable URL: http://links.jstor.org/sici?sici=1047-7039%28199607%2F08%297%3A4%3C359%3ATTFFHT%3E2.0.CO%3B2-9 The Temporal Development of Strategy: Patterns in the U. K. Insurance Industry David Webb; Andrew Pettigrew Organization Science, Vol. 10, No. 5, Focused Issue: Coevolution of Strategy and New Organizational Forms. (Sep. - Oct., 1999), pp. 601-621. Stable URL: http://links.jstor.org/sici?sici=1047-7039%28199909%2F10%2910%3A5%3C601%3ATTDOSP%3E2.0.CO%3B2-3