Co-evolution between Stages of Institutionalization and Agency: The Case of the Music Industry’s Business Model

Moyon, Emilien; Lecocq, Xavier

doi:https://doi.org/10.7202/044658ar

In the last decade, the advent of disruptive innovations such as digital technologies and the Internet has unleashed a revolution in the music industry transforming consumer behavior, business models and the competition structure (Alderman, 2001; Hensmans, 2003; Michel, 2005; Michel, 2006, Tschmuck, 2006). While new entrants like Napster or Apple considered these changes a great opportunity, incumbents, and mainly the major companies[1], were more pessimistic about the future.

Neo-institutional theory is often being used to explain change and the emergence of new institutions within a field. An institution is a set of routines or practices that are being reproduced over time and that acts as a cognitive framework structuring the organizational field. In this frame, institutional change has traditionally been described as a social mechanism operating at the field-level and often neglecting the influence of actors. However, recently authors have re-introduced the role of actors promoting change, for instance through the concept of institutional entrepreneurship (DiMaggio, 1988; Fligstein and McAdam, 1995; Holm, 1995; Fligstein, 1997; Dorado, 2005). By considering the institutional entrepreneur as a heroic figure, other forces of the field are often neglected or considered to passively follow the entrepreneur’s dynamic (Demil and Lecocq, 2006).

Without denying the influence of the institutional entrepreneur working in favor of change, we consider that focusing mainly on this category of actors offers a limited understanding of the global process of institutionalization and does not fully take into consideration the dynamic aspect of the field with its conflicts, negotiations and trade-offs. To better understand institutional change, both the influence of heterogeneous actors at the field level and the field’s influence on the actors’ behavior need to be considered. This constitutes a more fine-grained approach encompassing the intertwining of the two levels of analysis (agency/institutions) in a dynamic perspective. In this research, we study actors who want to preserve traditional institutions and their strategic behavior to hinder and then benefit from institutional change, following the work of Hensmans, (2003) and Lawrence and Suddaby (2006). However, we move to a more dynamic approach of the relationship between agency and institutions by looking at the co-evolution between incumbents’ strategies and stages of institutionalization of the field. To do so, we combine Oliver’s (1991) typology of strategic responses to institutional pressures and Greenwood and al.’s (2002) framework of institutionalization process to shed light on the complexity of the actors’ strategic behavior (actor level) and their evolution during the different stages of institutional change (field level).

The music field is particularly conducive to illustrate institutional work to maintain institutions (Lawrence and Suddaby, 2006). We focused on the dominant business model of the traditional music industry which has been challenged over the last ten years. The outcome of this research gives a better understanding of the interactions between the agency at the actor level and the coercive pressures at the field level. We identified three ways incumbents impact institutional change process within the field. First, the incumbents develop (very heterogeneous) strategic responses that contribute to the discrimination between alternative practices[2]. Second, they actively participate in the building of institutions by modifying alternative practices introduced by new entrants. Third, active strategic responses from incumbents affect the duration of the overall process of institutional change. Also looking at the coercive pressures of the field, we identified three ways the evolution of institutional change process affects the incumbents’ action. First, they tend to adapt their strategy depending on the outcomes of the alternative practices; we observed more and more passive strategic responses while alternative practices were gaining legitimacy within the field. Second, the evolution of the institutional process leads to a modification of traditional practices that no longer suit the institutional context. Third, incumbents used their legitimacy to benefit from successful new practices and especially to capture revenues from emerging business models.

Agency and Stages of Institutional Change

Traditional approaches to neo-institutional theory primarily focus on the influence that institutions have on the behavior and structure of organizations (Meyer and Rowan, 1977; DiMaggio and Powell, 1983; DiMaggio, 1988; Scott, 1995). The fundamental concepts of legitimacy (Suddaby and Greenwood, 2005), “taken-for-grantedness” (Tolbert and Zucker, 1983) and isomorphism (DiMaggio and Powell, 1983) demonstrate how actors in a given field follow the same patterns and how institutions have the tendency to remain inert. The reciprocal relation and the influence of actors in the change process have been commonly overlooked until recently when researchers began to pay particular attention to the role of actors and the impact that they can have on institutions (Oliver, 1991; Hensmans, 2003; Greenwood and Suddaby, 2006; Lawrence and Suddaby, 2006).

These non-isomorphic perspectives on institutional change emphasized the importance of the political dimension and of strategic agency within the field (DiMaggio, 1988; DiMaggio and Powell, 1991; Holm, 1995; Clemens and Cook, 1999; Fligstein, 2001; Seo and Creed, 2002). In a first stage, research that reintroduced actors’ influence within institutional processes primarily focused on the concept of the institutional entrepreneur (DiMaggio, 1988; Fligstein and McAdam, 1995; Fligstein, 1997; Beckert, 1999; Dorado, 2005). The institutional entrepreneur may recognize change as an opportunity and therefore chooses to adopt an active behavior towards a given change. Its range of action widely depends on the opportunities that are presented in the field. Institutional opportunities are defined as “the likelihood that an organizational field will permit actors to identify and introduce a novel institutional combination and facilitate the mobilization of resources required to make it enduring” (Dorado, 2005: 391). These opportunities depend on the degree of institutionalization of the field which determines the influence institutions have on actors. Dorado (2005) identifies three different types of fields depending on the combination between the levels of multiplicity of referents and institutionalization. The field is defined as opportunity opaque when there are little or no opportunities due to few institutional referents and a high level of institutionalization. On the other hand when there are many referents and a substantial level of institutionalization, actors will have many opportunities in a field described as opportunity transparent. Finally, the third type of organizational field, opportunity hazy, is considered complex and unpredictable and therefore is not an appropriate context for institutional agency.

The institutional entrepreneur has been the focus of many researchers trying to explain institutional agency. However, focusing on an entrepreneur changing a field may lead to conceiving him as a hero (Demil and Lecocq, 2006) and to neglecting the essence of neo-institutionalism, i.e. the weight of institutions and the holistic approach (DiMaggio and Powell, 1983). Thus, in order to better understand the change process, researchers have started to consider the role of other kinds of actors, who want eventually to preserve institutions (Oliver, 1991; Hensmans, 2003; Lawrence and Suddaby, 2006; Scott, 2007). Indeed, the inertness of institutions cannot be simply explained by social mechanisms like isomorphism (DiMaggio and Powell, 1983) which do not take into account the strategic behavior of actors who want to maintain a status quo.

Oliver (1991) questioned the passive role of organizations by presenting a typology of different strategic responses to institutional pressure. These different responses are classified according to their virulence from conformity to manipulation (Table 1). The author suggests that the type of behavior depends on the nature and origin of institutional pressure. For example, when the pressure comes from a dominant actor in the field other organizations will be more likely to respond more passively. Moreover, contrary to the traditional institutional entrepreneurship approach, Oliver (1991) also evokes that actors may unintentionally influence the process of institutional change.

Table 1

**Strategic Responses to Institutional Processes (Oliver, 1991)**

More recently, authors reconsidered agency with concepts like “institutional construction” (Scott, 2007) or “institutional work” (Lawrence and Suddaby, 2006). The later distinguishes various types of influence that actors can have on institutions. Some actors may choose strategic moves to break down the present institutions or to create new ones (typical behavior of the institutional entrepreneur), while other actors who have vested interest try to protect the present institutions. Different ways of preserving institutions have been identified by Lawrence and Suddaby (2006): coercive pressure towards institutions (enabling, policing, deterring) or reproducing norms and beliefs to maintain institutions (valorizing/demonizing, mythologizing, embedding/routinizing).

Thus, research on agency within neo-institutional theory has moved to a better integration of actors’ strategic behavior, regardless of their position towards institutions. However, beyond the story of actors’ confrontation with institutions, we still need to know if there is a systematic relationship between strategic behavior and the evolution of the field. For instance, one may ask if the position of actors progressively evolves throughout the different stages of institutional change according to the level of legitimacy of traditional and emerging institutions. A dynamic approach reintroducing the permanent interaction between the two levels of analysis (actors and stages of institutionalization) would help to better understand the complexity of the change process.

With their empirical study of the accounting profession, Greenwood and al. (2002) outlined a model of institutional change in which six key stages are presented (Table 2). This change process is provoked by social, technological or regulatory environmental outbreaks (Munir, 2005) that may impact the organizational field and weaken prevailing institutions. Greenwood and al. (2002) describe the process by which some institutions disappear (deinstitutionalization), while emerging structures spread and gain legitimacy throughout a field (preinstitutionalization, theorization, diffusion) to become institutionalized (reinstitutionalization).

Table 2

**Stages of Institutional Change (Greenwood et al., 2002)**

In this research, we apply Greenwood and al.’s (2002) framework to the music industry in order to understand the way former institutions were being abandoned while new ones were gaining legitimacy. At the actor level, we are also interested in analyzing more specifically the strategic behaviour of actors. We have chosen to focus this research on the role of the majors that attempt to preserve their position within the field. To do so, we use Oliver’s (1991) typology of strategic responses to institutional process. The combination of these two conceptual frameworks (Oliver’s typology on strategic responses and Greenwood and al. framework on the institutional evolution) gives a new perspective on institutional change that emphasizes the dynamic between change in the organizational field and agency. The integration of the two lenses mentioned above is quite easy from a theoretical point of view as they deal with different levels of analysis and they allow consideration of both the strategic behaviour of the firms and the evolution of institutions within the field. However, combining these frameworks is more difficult from an empirical point of view as it requires data collection and analysis about general processes and individual actions.

The Music Industry

Research Design

Over the last decade institutions in the music field have been increasingly challenged and this has therefore weakened their legitimacy. The apparition of new actors and the design of rival business models have threatened the central and privileged position of the majors and some institutionalized practices that they protect. The opposition between the majors and new entrants results in severe turbulence in the music industry which makes it particularly interesting to observe institutional change (Hensmans, 2003) and particularly the role played by incumbents.

The research design includes two separate steps. Preliminary field research has been conducted to provide a thorough background in the traditional music industry, to understand the interaction between actors and the institutional context and to calibrate the different stages of institutional change (Greenwood and al., 2002). We reviewed several pieces of literature which focus on the music industry from authors like Anand and Peterson (2000), Alderman (2001), Dubosson-Torbay and al. (2004), Huygens and al. (2001), Knopper (2009), Lampel and al. (2008), Lopes (1992), Molteni and Ordanini (2003), Michel (2005), Michel (2006), Spitz and Hunter (2005) and Tschmuck (2006). Then seventeen semi-directive interviews were then conducted with experts[3] from the field who have been confronted with change in French and Canadian music market. We also participated in several conferences[4] on the music industry during which we were able to continue our research with more informal interviews. This preliminary work first helped us to immerse ourselves in the music industry and to identify one core institution and four institutionalized practices that were built over time, reproduced and “taken-for-granted” by the different actors from the field. Second, the semi-directive interviews were also necessary to capture the actors’ representations. Specialists’ perspectives on change helped us to identify key actors and to determine the different stages of institutional change. For instance, interviewees were asked to name events that induced the change process. Most of them considered the introduction of peer-to-peer[5] (from this point on referred to as P2P) networks and the spread of mp3 technology to be the main precipitating jolts that induced change. More surprisingly some interviewees also mentioned the CD burner as having a major impact in the consumption patterns of music. Because the spread of mp3 and the introduction of both P2P networks and CD burners occurred in 1997, we considered it as a turning point in our research and therefore we make reference to the time before 1997 as the traditional era and from that point on the change era. This delimitation is consistent with Lampel and al.’s (2008) perspective of exogenous jolts that are repeatedly causing important discontinuities since the birth of the music industry.

Concerning the second step of the research design, we collected secondary data in order to identify more specifically the strategic behavior of two major companies –Universal Music, UM from now on, and BMG- facing institutional change. Based on preliminary research, we chose to focus our data collection on these two majors that held a representative position of this category of actors within the field. When interviewees were asked to describe the role each major company was playing during change, UM and BMG clearly appeared to position themselves against change pressures while EMI Music and Warner Music were presented as more inactive actors or even followers. Also Universal and Sony-BMG represented respectively 26% and 21% of the world music market[6] in 2005 which emphasizes their need to protect former institutions. Because of the 2004 merger between BMG and Sony Music which constitutes by itself a revealing strategic response to institutional change, we collected the strategic responses of the BMG entity before the merger and the strategic responses of the Sony-BMGentity after the merger. We systematically and chronologically considered all the articles that revealed a strategic response from UM or BMG toward change during the chosen period (1997-2006) that were found in public sources. This work is based on a significant database (1397 news articles) from a wide range of sources: IFPI (International Federation of the Phonographic Industry), RIAA (Recording Industry Association of America), and finally Sony-BMG and UM’s corporate websites. Screening the database led to the identification of sixty-eight strategic responses from UM or BMG towards one or more sources of institutional change.

In the first stage of the data analysis, each institutionalized practice was considered separately. Therefore we consider institutional change as four interwoven processes in which a specific practice is being questioned. Thus any single strategic decision can potentially be related to more that one process. To illustrate our methodological approach, we present in table 3 the 37 strategic responses to change towards the album/single format collected between 1997 and 2006 – 15 responses from UM, 5 responses from BMG and 17 joint responses from both UM and BMG . Oliver’s (1991) typology has then been used for each process to analyze the nature of each decision. For instance when actors refused to change and worked ardently to maintain the institutions of the traditional era, we interpreted it as a virulent strategic behavior towards pressures (Manipulate and Defy; Oliver, 1991). But when actors were substantially more in favor of change and the emergence of new institutions, we considered that the majors were using passive strategic responses (Acquiesce and Compromise; Oliver, 1991). We chose a double-coding analysis which sometimes led to disagreements between the authors and therefore discussions with actors in the field. Finally we consulted experts in the field to confirm our analysis and results in order to raise the internal validity of our research. We present the analysis of UM’s 32 strategic responses to change towards the album/single format in table 4. Figures 2 to 5 illustrate the repartition of strategic response type (Oliver, 1991) for each practice during the period of change.

Table 3

**Universal and BMG’s Responses to Change Pressures toward the Album/Single Format (1999-2006)**

Table 4

**Strategic Responses from Universal Music (1999-2006) (BMG’s Responses Have Been Excluded). See Table 3 for the detailed responses**

In the second stage of the data analysis, the four processes have been aggregated (table 6) to observe the strategic responses that characterize each stage of institutional change. We used jointly Greenwood and al.’s (2002) framework and Oliver’s (1991) typology to investigate coercive pressures at the field level.

Institutional Practices in the Traditional Music Field

In this part we aim to describe the music field in its traditional setting: the position of actors, their interaction and the institutional context (Figure 1). In the second half of the 20^th century, the music field was highly concentrated with the domination of few majors. Concentration began in the 60’s with the introduction of Warner that radically reinvented the organization of music labels and the distribution networks. Originally based in the movie industry, Warner rapidly played an important role in the music field by implementing a strategy of acquisition of some of the most successful American independent record labels[7]: Atlantic, Elektra and Asylum (Huygens and al., 2001). This strategy became a general tendency in the field; successive acquisitions of independent labels led to the building of important structures that operated globally and were diversified in many music genres and in external activities. The concentration of the industry led to five giants, Universal Music, BMG, EMI, Sony and Warner (Tschmuck, 2006). Rapidly the “Big Five” accounted for more than 70% of retail music sales[8]. Two categories of actors then constituted the field, on one side the majors and on the other the independent labels. The high-level of concentration and the uneven repartition of resources and power within the field naturally shaped interactions between the two types of actors; the majors were looking for new acquisition opportunities in order to enrich their catalogues and independent labels gravitated around the majors to benefit from their resources (Lopes, 1992).

Figure 1

**Institutionalized Practices and Change Pressures**

For years, the “traditional business model” remained efficient and appropriate despite technological disruptions (radio, magnetic tape, CD) and was taken-for-granted by artists, independent labels, distributors, media and customers. For these reasons, a dominant business model acts as a cognitive template structuring the field; we consider the business model a key institution of the pre-digital era. A business model is defined as “a description of the roles and relationships among a firm’s consumers, customers, allies, and suppliers that identifies the major flows of product, information, and money, and the major benefits to participants” (Weill and Vitale, 2001). Business models capture the organization of the field by describing how the value chain is organized around the focal firm (Amit and Zott, 2001; Linder and Cantrell, 2000; Zott and Amit, 2008) but also how the firm generates revenue. To describe this dominant business model, we disassociated its two main components, the first is the value chain organization (Component 1 in Figure 1) and the second is related to the value proposition (Component 2 in Figure 1). Both of these components can also be split into two more specific institutionalized practices (Practice 1 and 2 in Component 1, then Practices 3 and 4 in Component 2).

The first institutionalized practice is control of distribution (P1) which captures the organization of distribution networks. As the music labels were concentrating during the traditional era, the music distributors also followed the same path, going from small record shops to the domination of hypermarkets and big specialty stores such as Virgin Megastore or Fnac in France. The majors set up strong ties with these retailers in order to distribute their wide music catalogues. Also most independent labels used the majors’ distribution network in order to benefit from their worldwide networks. By controlling the distribution of almost every artist, the majors had a powerful and privileged intermediary position between artists and consumers (Lopes, 1992).

The majors have also established strong ties with media in order to promote their artists. The second practice is control of promotion (P2) which captures the organization of music promotion. The majors have sufficient resources to develop important marketing campaigns using radio and TV channels in order to reach a large public. Their international structures also enable them to promote artists worldwide. In opposition, independent labels use specific media because they often focus on local markets and particular segments of customers. Indeed, artists who are affiliated to one of the majors have access to more resources and benefit from their distribution and promotion networks that will give them more opportunities to succeed and to develop internationally (Tschmuck, 2006).

The album and singleformat were the two commercial standards of the traditional era in which the price of music was set for a limited number of tracks; this revenue model is considered the third institutionalized practice (P3). Music labels generalized this delivery format through different technological support: the vinyl record, the analogical tape and the Compact Disc.

Finally, copy control (P4) is a key practice for the traditional business model’s stability by enabling labels to capture value. Because consumers could not duplicate music, music labels were capturing revenue out of the music duplication process, especially with the Compact Disc that required less production costs.

Change over the Last Decade

After describing the field during its traditional era, we will now present chronologically how the different institutionalized practices have been challenged between 1997 and 2006. Some technological jolts have precipitated change and allowed the introduction of new actors from tangent fields that developed innovative business models. The Internet and digital technology have been tremendous innovations that cannot be isolated from the process we are studying, but their application and their impacts on the music field really started with the development of P2P networks and the spread of the mp3 encoding format. Because these two innovations appeared in 1997, we consider this the starting point of the change period.

The mp3 format which reduced the amount of data required to encode music rapidly became the standard for audio file on the Internet. For the first time, the mp3 technology allowed users to copy music while minimizing quality losses in comparison to analogical copy (magnetic tape, for example). With the introduction of the mp3 format, music labels were no longer exclusively in control of duplicating music. Also in 1997 the first CD burners became available for personal use, which allowed consumers to transfer digital music to a material support and duplicate it. Based on these technological innovations, new actors started developing alternative models that threatened the majors’ traditional business model, their central position and the structure of the field. In 1997, Michael Robertson introduced a revolutionary model, the website mp3.com, which was a new way to listen to and share music. It is considered the first “P2P model” of music distribution (Alderman, 2001; Knopper, 2009). Mp3.com could be used by any artist to promote and distribute independently their music creations using the mp3 format and therefore challenged simultaneously the four institutionalized practices. Rapidly mp3.com users had online access to thousands of songs.

In 1999 a more sophisticated kind of P2P is introduced, Napster. Its immediate success greatly destabilized the music field and also challenged different institutional practices. Napster swiftly became the biggest music catalogue access in the world, offering much more diversity than any other traditional retailer. Its success greatly contributed to establishing the mp3 format as the new standard of digital music on the Internet. In the beginning of the 21^st century, music sales declined massively while the number of P2P networks (Emule, Gnutella, Grokster, Imesh, Kazaa, Limewire…) and users continued growing. In 2003, P2P networks counted more that 60 million users in the USA while Kazaa, one of Napster’s clones, claimed 9.5 million users in Europe who could download music tracks from a 180 million-track catalogue (Dubosson-Torbay and al., 2004). The music industry accused P2P networks of copyright infringement and lawsuits were made against them.

In 2000, E-music reacted to the P2P revolution and introduced the first legal subscription model that offered “unlimited” independent music for a monthly fee. In 2002 and 2003, MusicNow and Streamwaves developed streaming[9] models also offering unlimited access to music directly on the web. Although consumers could listen to any music they wanted, they couldn’t store or share it.

A new generation of business models focusing on community building and generating revenues from advertising appeared in 2004 with the emergence of the so-called web 2.0[10]. For instance, MySpace established a social network that artists could use to share and promote directly to specific communities without involving any intermediate music label. In 2005, YouTube developed a similar concept specialized in the diffusion of videos introducing the opportunity for users to interact creatively with music by posting a remix of a song they like or their own interpretation of the music video.

These new business models revealed consumers’ interest for creative music offers even though these models were still struggling to convert traffic into cash. Whereas the industry evolution used to be driven by a quality need (from the gramophone until the CD format), consumers are now sensitive to accessibility and flexibility. The discourse of Jerry Pierce, vice president of technologies in Universal Pictures, illustrates perfectly this new paradigm: “In the music industry, consumers were offered, with the SACD[11], a brand new format with much higher quality and very restrictive DRM rules. And consumers chose mp3 as their solution. They could have had higher quality, but they chose flexibility and usability”[12]. In 2005, the Internet service company Yahoo inaugurated its new unlimited music subscription model, Yahoo Unlimited. The following year, MySpace recorded its 100-millionth subscription and YouTube counted more than 100 million users daily watching some of their 6.1 million videos. The huge success of community websites caught the attention of large groups. For instance Rupert Murdoch’s media conglomerate company News Corp bought MySpace in July 2005, Google bought YouTube in October 2006 while Microsoft launched its own Internet video service, SoapBox.

In 2007, the music field was facing a debate regarding the interoperability issue of digital music protected with DRM. Not only does DRM restrict the duplication of music files, it also limits the way music can be used. Unlike DRM-free music from P2P networks, DRM music sold through legitimate channels was often not compatible with some electronic devices. This very specific aspect provided a competitive advantage to P2P networks that distributed a wide range of music from different labels without any constraints. Some legitimate actors rapidly understood that it was a central issue for consumers and launched DRM-free music services.

We presented chronologically the main change pressures (referred hereafter as CP) that have been challenging each institutional practice that was described earlier (Figure 1). For example, models that allow music distribution without the majors’ involvement (P2P or platforms specialized in independent music) are considered as pressures toward control of distribution (CP1, pressures toward institutional practice 1). During the change period, the emergence of alternative promotion tools such as web 2.0 sites, which provided another outlet for independent artists to reach their audience, are considered as pressures towards control of promotion (CP2). Then change pressures toward format (CP3) are related to alternative revenue models, those models include P2P offers, subscription offers and also streaming websites. Finally, change pressures towards copy control imposed on music (PC4) were represented by innovations such as mp3 technology as well as by actors that sold DRM-free music. In the next chapter, we describe UM and BMG’s strategic responses to these four types of change pressure between 1997 and 2006.

Stages of Institutional Change and the Majors’ Strategic Responses to Change Pressures

Analysis at the actor-level, focusing on the majors’ strategic responses

We have used Oliver’s (1991) typology to analyse the strategic behaviour of UM and BMG and observe the variation of their responses to change pressures over several years. Their strategic responses to pressures towards each institutionalized practice are presented in separate figures (figures 2 to 5). Tactic sub-categories do not appear in the figures because we chose to focus on the main strategic categories. To better understand the encoding of data we provide an example of two representative events that occurred during the change period and explain the way we interpreted them.

Figure 2

**Strategic Responses to Change Pressure toward Control of Distribution**

Figure 3

**Strategic Responses to Change Pressure toward Control of Promotion**

Figure 4

**Strategic Responses to Change Pressure toward the Album/Single Format**

Figure 5

**Strategic Responses to Change Pressure toward Copy Control**

Since 1998, P2P networks had been a significant threat to the majors and especially to the traditional distribution practice (P1). As a result, in 2001 BMG acquired the P2P network Napster that was an independent distribution network on the Internet. This costly acquisition was considered as a control tactic which refers to a manipulation strategy (Oliver, 1991). Also in response to the increasing pressures linked to Internet distribution, UM and BMG decided to heavily develop music distribution for mobile phones in 2002. On this specific market, the majors were able to maintain high margins on diverse music products for mobile phones. In this case, UM and BMG adopted an escape tactic which refers to an avoidance strategy (Oliver, 1991).

For the purpose of our approach, we decided to look separately at the four institutionalized practices to study change and the incumbents’ strategic responses. These four institutional change processes are not synchronized which has led to different time frames in the presentation of the results. For instance, strategic responses to change pressures towards copy control started in 1997 when UM and BMG intentionally ignored the mp3 technology. However figures 2, 3 and 4 start in 1999 because we didn’t observe any strategic responses related to distribution, promotion or format practices during the previous years. On the actor-level, the results demonstrate the complexity and the evolution of incumbents’ strategic responses over time.

First, we observed a wide range of strategic responses that have been used simultaneously by the majors. This result is far different from Lawrence and Suddaby’s (2006) research on institutional work which suggests a homogeneous perspective on agency. On the contrary, the majors adopted concomitantly different types of strategic responses to threats on a single institutionalized practice. This observation can be interpreted as a poor strategic vision from the majors that were facing a high level of uncertainty in the music field. However adopting a more qualitative look at the data, we realized that the type of responses depends on both the nature of the alternative practice and the source of institutional pressures which is consistent with Oliver’s (1991) perspective. For instance in 2003, UM and BMG chose a compromise strategy regarding Apple’s alternative distribution practice by licensing their entire catalogue and even accepting the $.99 pricing policy. At the same time, they filed a lawsuit against two of the most important P2P providers, Kazaa and Grokster, which also introduced alternative distribution practices. Not only did the majors try to protect former practices but they also participated in the selection process of new alternative practices. The majors selected the alternative distribution practice from Apple considering its legitimacy in the computer field but also the opportunity to capture revenues participating in Itunes’ business model.

Second, the majors’ responses appeared to be more and more passive throughout the process of institutional change. During the first years, they adopted mostly manipulation and defy strategies towards new entrants and experimental practices. But while new practices spread and gained legitimacy within the field, we identified more and more passive strategic responses (avoid, compromise, and even acquiesce strategies) by the end of the process.

Analysis at the field-level, focusing on the different stages of institutional change

At the field level, institutional change affecting the traditional practices we are studying can be observed through the scope of Greenwood and al.’s (2002) framework (Table 5). We consider the four institutionalized practices as independent processes that result from common precipitating jolts. Precipitating jolts were both technological, i.e. the Internet or the mp3 technology, as well as social with a rise in P2P users and an increasing demand for unlimited access to entertainment goods. Although they result from the same precipitating jolts, the processes are not synchronous and they did not all reach the same stage of institutional change at the end of our study in 2006.

Table 5

**The Four Processes of Institutional Change in the Music Field between 1998 and 2006**

Greenwood and al. (2002) describe the deinstitutionalization stage as “the entry of new players, the ascendance of existing actors, or local entrepreneurship. Their effect is to disturb the socially constructed field-level consensus by introducing new ideas and thus the possibility of change” (p.60). During this stage, new entrants develop alternative practices inspired by new technologies that are available in the field. Sometimes they even purposely work for change in order to seek a dominant position. The deinstitutionalization stage in the music industry is linked to two events, the increasing number of illegitimate P2P networks that attract more and more users and the adoption of the mp3 format that progressively became the standard for music files on the Internet. During the blooming of the Internet, customers, new entrants and some artists perceived the opportunity to benefit from alternative practices and questioned the traditional ones as well as the central position of the majors.

As traditional practices are being questioned, the legitimization of new practices and the building of new institutions are more problematic. In creative industry, legitimacy is mostly related to copyright concerns (Lampel and al., 2008). Because labels owned the intellectual property rights of music creation (Dubosson-Torbay and al., 2004), new entrants struggled to gain legitimacy and alternative organization of the value chain did not spread. Alternatives were either illegal in terms of copyright infringement which led to legal actions (P2P networks) or concerned exclusively independent music catalogues (E-music, Airtist, Jamendo) which represented a small proportion of the market in terms of revenues. Consequently, distribution and promotion alternative practices did not reach the preinstitutionalization stage, however traditional practices kept being frequently challenged until 2006. Recently, Youtube and Dailymotion introduced creative promotion and distribution practices based on video sharing and interactive contents but they also faced copyright issues.

Regarding the two other processes, we observed that alternative practices related to the value proposition reached further stages of institutionalization. During the deinstitutionalization stage, the alternative format and copy control practices (mp3 format and unlimited offers) were mostly initiated by P2P networks. Later legitimate actors like Yahoo, Music Now or Jamendo developed new value propositions based on these alternative practices leading to a new paradigm in the way music would be sold and consumed (Tschmuck, 2006). The diffusion of this new paradigm demonstrates the preinstitutionalization stage of change. By the end of our observations, the majors developed new offers based on these alternative practices by slowly giving up the use of DRM technologies for specific offers and also experimenting unlimited ad-supported offers. Incumbents’ interest in alternative practices shows the beginning of the theorization stage that started in the last couple years of our empirical study. The institutionalization of these two practices related to the music offer did not end in 2006. In 2008, the majors even accepted to generalize DRM-free offers on main distribution platforms (Itunes, Fnac Music, Virgin Music). In the meanwhile, they also partnered with several firms from the telecommunication field (Vodaphone, Neuf Telecom), the banking field (Société Générale) or the phone manufacturers field (Nokia) to develop joint offers based on unlimited access.

Discussion

Observing the process of institutional change that took place in the music field over the last decade helped us to better understand the relationship between actors and change. Through a multi-level approach, we emphasized the interactions between the individual and the collective level. We illustrated how UM and BMG, two traditional and central actors in the music business, actively participated to institutional change in order to preserve their central position in the field and the traditional business model. Given the characteristics of the research setting and of the method implemented, we think that our findings may give insight to understand institutional change in other fields, particularly where technology change may affect consumers practices, for example in most cultural fields as well as in information related industries such as computer, software or pharmaceutical industries in which new technologies have created the emergence of new practices, creating controversies among actors within each field. The outcomes of this research lead to a better understanding of the interactions between the incumbents’ action on the field level, agency, and the overall institutional change process at the field level, coercive pressures (Figure 6).

Agency

Whereas neo-institutionalism traditionally underestimates the role of actors within institutional phenomenon or, on the contrary, overly focuses on the influence of institutional entrepreneurs that are working in favour of change, we have highlighted that agency can be more complex. The first contribution helps to better understand the impact of incumbents’ action on the institutional context and the evolution of institutional change at the field level. This argument is indeed consistent with recent contributions that reintroduce agency in neo-institutional approaches by using concepts such as institutional work (Lawrence and Suddaby, 2006). Beyond the concept of institutional entrepreneur, we enrich the understanding of agency by identifying three different ways actors can interfere with institutional processes depending on their position towards institutions: alternative practicesselection, alternative practices modification and process duration.

Alternative practicesselection. Facing institutional change, incumbents adapt their strategic responses depending on the alternative practices’ perceived opportunities and their legitimacy. The use of heterogeneous responses interferes with the selection process between competitive alternative practices. For instance, the majors responded negatively to Napster by engaging lawsuits or to E-music by not giving them the authorization to distribute their catalogue but at the same time the majors signed a distribution agreement with Apple. Incumbents facing institutional change are neither totally passive neither totally stubborn to change. They seem to rationally measure the possible outcomes of each alternative practice and then develop appropriate strategic responses.

Alternative practices modification. Not only did the majors seem to choose between the different alternative practices but they also modified the content of new practices. When Youtube launched a new video streaming service online, customers enjoyed the possibility to share musical content but also to creatively interact with it. Based on web 2.0 possibilities, they could easily modify the contents. Facing a great interest from customers UM perceived this as an opportunity to promote their musical catalogue and decided to co-opt with Youtube. But this partnership involved a practice modification while Universal broadcasted its catalogue on Youtube; private users were no longer allowed to interact with it. The alternative practice of promotion is now about broadcasting and no longer about creativity.

Process duration. Dynamic between incumbents’ action and entrepreneurs’ institutional work hindered the stabilization of the field and increased the length of the process of change. Considering previous innovative disruptions, like the introduction of the gramophone, the radio, the magnetic recording technologies or the CD, these disruptions did not challenge the central position of the majors in the field. Indeed new practices replaced former practices in a smooth and rapid change process. The contemporary change process we studied is a lot different; because incumbents have developed active strategies in order to preserve former practices, the field remains unstable and alternative practices to distribution and promotion have not yet been institutionalized ten years after the introduction of mp3, mp3.com and the first CD burner.

Coercive pressures

During the period of institutional change, some new entrants and the alternative practices they defended gained legitimacy (Apple’s Itunes, E-Music) while others disappeared (Napster, mp3.com). The evolving set of institutional pressures also had an impact on the incumbents’ strategic behaviour; we observed three different ways it impacted the incumbents’ responses: strategic adjustment, traditional practices modification and legitimization.

Strategic adjustment. For instance, the progressive pacification of the majors’ responses along the institutional process demonstrates coercive pressures at the field level. Regarding the value proposition change process, new practices gained substantial legitimacy especially through a high rate of adoption from consumers and firms evolving in adjacent fields like the computer industry or electronic devices industry. Indeed, the majors adopted passive responses towards change by slowly accepting new ways to look at the music business. Initiatives like Buzzmusic, the partnership with Spiralfrog or the later renunciation of DRM technologies in 2008, demonstrate a strategic adjustment from incumbents facing a new paradigm in the music field. In order to present the co-evolution between agency and institutional change, Oliver’s (1991) typology and Greenwood and al.’s (2002) framework have been used jointly (Table 6).

Table 6

**UM and BMG’s Strategic Responses during the Stages of Institutional Change**

Traditional practices modification. Beyond the strategic adjustment, incumbents also reconsidered the content of traditional practices while facing a new institutional context. For instance, the majors tried to increase the customers’ willingness-to-pay for the physical support by adding video bonus (live performance, interviews, videos) or privileged access to internet content. This response is particularly interesting because the majors perceived a decrease in the value perception of the traditional value proposition.

Legitimization The stalemate of the change process and the selection of alternative practices offers opportunity for incumbents to position themselves in the new institutional context. Because the majors owned the property rights of music contents (Lampel and al., 2008), they were able to negotiate and reach several agreements with new entrants that developed successful alternative practices. These agreements with copyright holders helped new entrants to gain legitimacy while sharing revenues with incumbents. For example, the agreement that UM reached with Microsoft in order to receive $1 from each music player sold illustrates the way new entrants “buy” legitimacy. Similar agreements have since been settled with start-up companies (Youtube), phone manufacturers (Nokia), Internet providers (Neuf Telecom) or even banks (Société Générale).

The deep turbulences that occurred in the music field over the last decade illustrate the interwoven relationship between incumbents’ actions and the evolution of the change process, and emphasize the dynamic of the field. Unlike the traditional interpretation of the “iron cage” from DiMaggio and Powell (1983) that illustrates the coercive pressure of institutions and the isomorphic mechanisms which obstruct actors from alternative behaviours, our results demonstrate that actors perceive alternative practices since the early stages of institutional change. This contribution also questions more recent neo-institutional research focusing on agency. We contend that new approaches introducing the concept of the institutional entrepreneur, that tend to emphasize the “heroic” figure of the challenger, or institutional work, that offers a rigid typology, suggest simplified representation of agency in which actors have a specific role and use homogeneous responses in favour or against institutional processes. Our research suggests that the understanding of agency is more subtle by considering the different strategic responses towards institutional change.

This dynamic representation of institutional change shows the limits of static approach such as the typology of Oliver (1991) or Greenwood and al. (2002) which may fail to fully depict the rich mechanisms of institutional process. By overly focusing on the actor level or on the contrary on the field level, these approaches suggest a linear and yet stereotyped perspective of both agency and institutional change that remain permanently connected and interdependent. Agency is greatly regulated by the overall institutional context that evolves over time; also deinstitutionalisation and re-institutionalisation stages rely on the actions of many individuals. These actions are not only in favour or against change; they also contribute to the construction of new institutions through a complex social process. In other words, research on institutional change requires a clear understanding of institutions.

This paper raises also a methodological debate. Stereotyped representation of agency during the institutional change process may be directly related to methodological choices. Most research on institutional change studies an empirical field retrospectively which we believe leads to bias. Indeed, researchers observe and interpret processes that occurred in the past and of which they know the outcome. We believe that depending on the achievement or not of the institutional processes, strategic responses can be interpreted differently which leads to simplified representation of agency and a teleological analysis of actions. Unlike most research, we chose an in situ observation of the institutional change process in the music industry. Thus by the end of the empirical work and data analysis period, the institutional change process is not yet complete which does not allow us to take into consideration the outcomes and the institutional context that results from change. This methodological approach has its own limits especially to study institutions that take place and are characterized by their persistence over a long period of time. However, insitu observations appear to be relevant to better understand institutional process and agency in order to disconnect the process from the institutional context that result from it. Indeed, we believe that further research on this topic should focus on contemporary phenomenon.

Because of the methodological characteristic that we just mentioned, the following stages of the institutional change process, diffusion and re-institutionalization have not yet been reached in the specific case of the music field but they could be the focus of further research in upcoming years. In order to pursue this work and give a transversal understanding of the process in a multi-level perspective, observations of the music field and institutional change may be pursued until the field reaches stability. Upcoming research should not only focus on the majors but also include heterogeneous actors that perceive institutions differently. For example, the strategic responses from challengers such as Amazon, Apple or Yahoo towards institutional change would bring interesting insights on the way institutions are socially constructed or deconstructed. We could also consider organizations that are involved in the preservation of traditional institutions but with other motivations (for instance, organizations in charge of property rights regimes within the music field). Taking into account different types of actors and their work on institutions in a dynamic perspective would enrich the overall understanding of the process of institutional change and the factors that trigger each of its stages.

Co-evolution between Stages of Institutionalization and Agency: The Case of the Music Industry’s Business Model

Abstract

Résumé

Resumen

Agency and Stages of Institutional Change