Location choice in the cultural and creative industries : a case study of the “big three” record labels in the music industry

(1)

LOCATION CHOICE IN THE CULTURAL AND

CREATIVE INDUSTRIES

A case study of the “big three” record labels in the music industry

Master Thesis

MSc. Business Administration – International Management Supervisor: dr. Johan Lindeque

Second reader: dr. Francesca Ciulli Student: Nikita Wiessner Student ID: 11148160

(2)

Statement of Originality

This document is written by student NIKITA WIESSNER who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating

it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

ABSTRACT

This study studies the role of location within the creative and cultural industries by looking at music industry in specific. Compared to traditional industries, the creative and cultural industries are underrepresented in the international business literature, even though they become increasingly important to the global economy. Via a qualitative multiple case study design, the location strategies of the three biggest MNEs in the music industry are studied, where each case represents one MNE. The focus lies on three embedded units of analysis: location choice, FDI motive and entry mode. The study found that MNEs in the music industry mainly invest with a strategic asset seeking motive and that these investments are done via equity entry modes like acquisitions and joint-ventures. Moreover, the MNEs are almost exclusively located in global cities or cities with features that are important for the music industry. This reflects the MNEs their FDIs that are also done in primarily the same cities. Finally, a new classification for music cities is developed and the strategic asset seeking motive for FDIs is improved to fit the creative and cultural industries.

KEYWORDS: multinational enterprises, creative industries, cultural industries, music industry, location choice, FDI motive, entry mode.

(4)

ACKNOWLEDGEMENTS

I would like to thank my thesis supervisor Dr. Johan Lindeque for his advice and all the useful feedback on my thesis. I would also like to thank him for his flexibility and continuous support throughout the whole process. He made time whenever I needed, even though my thesis took longer to finish than anticipated. It made me want to push myself and resulted in a paper I am very proud of.

Furthermore, I would like to thank my colleagues at mymuesli for their understanding and the flexibility they offered me while writing my thesis.

Finally, and most importantly, I would like to thank my parents for the opportunity to study in different cities. You have always supported me no matter what I wanted to do which I am very grateful for. Without you I would not be the person I am today.

(5)

Index of tables

Table 1: General information on the MNEs of the three cases………..……….….25

Figure 1: Visual map of MNE’s global offices………...……….26

Table 2: Categorization of cities……….28

Table 3: Sample of studied investment of the case’s MNEs………...……….……....29

Table 4: Sample of studied articles for FDIs made by the MNEs……….….…...29

Table 5: Coding scheme…...………...…....30

Table 6: List of Universal Music Group’s global offices……….32

Table 7: UMG’s studied office locations………33

Table 8: Universal Music Group’s FDIs……….36

Table 9: List of Sony Music Entertainment’s global offices………39

Table 10: SME’s studied office locations………...40

Table 11: Sony Music Entertainment’s FDIs………...43

Table 12: List of Warner Music Group’s global offices……….45

Table 13: WMG’s studied office locations……….46

Table 14: Warner Music Group’s FDIs………...48

Table 15: New ranking of music cities……….56

Figure 2 : New FDI motive………....…………..57

(7)

1. INTRODUCTION

Since the beginning of IB literature scholars become more and more interested in the strategies behind an MNE’s location choice – why firms place their activities in certain locations (Dunning, 1998; Rugman & Verbeke, 2004; Goerzen et al., 2013; Enright, 2000; Delgado, Porter & Stern, 2010). Dunning (1998) was one of the first to state that location is a neglected factor in the international business literature when studying multinational enterprises’ (MNEs) foreign direct investment (FDI) activity. However, although a lot of research on location strategies has been done over the past years, the focus seems to be mostly on traditional industries like finance, trade, manufacturing (Yamori, 1998; Chung & Alcácer, 2002; Makino et al., 2002), leaving our understanding of the cultural and creative industries (CCI) underdeveloped.

Over the years, the creative and cultural industries became increasingly important in the global economy (DCMS, 1997; UNCTAD, 2008). In 2015, revenue made in all CCIs accounted for 3% of the world’s GDP exceeding the telecom service industry revenues (World Creative, 2015). It is therefore not that strange that some of the wealthiest and most influential people in society nowadays are artists (Florida & Jackson, 2010). Most of the industries within the creative and cultural industries have specific hubs where all big MNEs are located. In music, cities like Berlin, London or New York are important locations (Krätke, 2003; Bader & Scharenberg, 2010). Same goes for the film or television industry, where for example Los Angeles (Hollywood) and Mumbai (Bollywood) are two of the most important locations (Krätke, 2003).

We know that, over time, these cities developed cultural and creative clusters that became unparalleled. Many studies support this claim of CCIs clustering in cities.However, the existing studies that focus on this relation between location and the creative and cultural sector (Boix et al., 2010: Power and Nielsen, 2010: Capone, 2008; Lazzeretti et al., 2008: De Propris et al., 2009) are merely observatory studies that state that these industries indeed tend to cluster, mostly around inner city regions, but fail to give an explanation on the how and whys. If they do perform a study on a cultural and creative industry like the music industry, it’s scope is limited to the US (Florida & Jackson, 2010). They are also mainly sociological or geographical studies (Sassen, 1991; Scott, 1997). So in spite of the growing significance of these industries, they have not received much attention from international business and management scholars.

(8)

This is why this study explores more in-depth how location plays a role in the CCIs.

Research question

The aim of this paper is to shed a light on the applicability of the existing literature on location strategy to the cultural and creative industries. When looking at the literature on regionalization or global cities and the location choice motives for companies in the cultural and creative industries only little research has been done (Florida & Jackson, 2010; Tomczak, Stachowiak, 2015; Drake, 2003). There is a lot of research on location choice motives, but they mostly focus on location choices within traditional industries like finance, technology, retail, trade etc. (Yamori, 1998; Makino et al., 2002). Since the creative sector operates very differently from traditional industries (Scott, 1997) it should not be assumed that location choice motives for the cultural and creative industries are the same as for these traditional industries. Therefore, this study asks whether Dunning’s (1998) 4 motives for FDI apply to the creative & cultural industries and what kind of location choice strategy they use in this industry. This results in the following research question:

How does location matter to MNEs in the cultural and creative industries?

In order to answer this research question, the study will focus is on the music industry. This is done via a multiple case study, reflecting to Eisenhardt’s (1989) inductive approach. This means that this study is theory building in stead of theory testing. There will be no testing of hypotheses or working propositions to avoid thinking about specific relations between variables (Eisenhardt, 1989). Instead, an open research question is formulated. Furthermore, the theory explained in the literature review is meant to specify possible important variables and is used as a starting point to build new theory (Eisenhardt, 1989). This multiple case study consists of 3 cases, Universal Music Group, Sony Music Entertainment and Warner Music Group, with each case being an MNE in the music industry, and these MNEs being selected because they are critical cases (Flyvbjerg, 2006). Together they control over 70% of the market thus, giving the best representation of the music industry. Each case has 3 embedded units of analysis (EUAs), location choice, FDI motive and entry mode choice which help add a focus to the research question. Documentary data on these EUAs is gathered through multiple data sources like Zephyr, Orbis, industry magazine articles, annual reports and other sources.

The study found that MNEs in the music industry mainly invest with a strategic asset seeking motive to protect their position in the highly competitive oligopolistic market. These

(9)

investments are done via equity entry modes like acquisitions and joint-ventures to keep control, but also gain from the host company’s FSAs and internalized CSAs (like local knowledge/artists). Moreover, the MNEs are almost exclusively located in global cities, since these cities are highly interconnected and have a cosmopolitan environment, or cities with features that are important for the music industry. This reflects the MNEs their FDIs that are also done in primarily the same cities. Finally, this study introduced a new classification for music cities, and the strategic asset seeking motive for FDIs is improved to fit the creative and cultural industries.

This study is structured as follows. First, the introduction. Here, the aim of this study and its research question was introduced. The next chapter elaborates on relevant literature related to the research question. Then, the proposed research method will be discussed. This is followed by an extensive case analysis. Finally, the case results and a conclusion with a summary of the key findings is presented.

2. CONTEXTUAL BACKGROUND

(of the global music industry) This study is about how location matters to the cultural and creative industries. Since these industries are fundamentally different from the industries that are grounded in traditional location theory, it is important to address what cultural and creative industries are and how they differ from traditional industries. That is why this chapter gives some background information on these industries. One creative industry that is at the core of this study is discussed in particular, namely: the music industry.

2.1.1 What are Creative and Cultural Industries?

Creative and cultural industries are industries that are very knowledge intensive. Examples of the CCIs are: architecture, visual arts, music, movies, gaming, advertising, TV and literature. The most commonly used definition for the cultural and creative industries comes from the Department of Culture, Media and Sports in the UK, they define it as:

“industries which have their origin in individual creativity, skill and talent and which have a potential for wealth and job creation through the generation and exploitation of intellectual property”.

(10)

The united nations later elaborated on this definition and came up with the following:

Creative industries are featured by “cycles of creation, production and distribution of goods and services that use creativity and intellectual capital as primary inputs; constitute a set of knowledge-based activities, focused on but not limited to arts, potentially generating revenues from trade and intellectual property rights; comprise tangible products and intangible intellectual or artistic services with creative content, economic value and market objectives; are at the cross-road among the artisan, services and industrial sectors; and constitute a new dynamic sector in the world trade”

UNCTAD (2008, p.4)

The base of this latter definition, cycles of creation, production and distribution, shows similarity with traditional industries however, the way these products are created, produced and distributed is fundamentally different. Scott (1997) wrote about the cultural industries and distinguished 4 main characteristics:

(1) The labour process in the CCIs uses much more human handiwork compared to other industries and also often complement it with advanced computer technologies.

(2) Production is done in small or medium sized networks that are interdependent of each other. However, large MNEs that have these processes integrated can also be found participating in these networks.

(3) These networks ask for large local labour markets and a big variety in worker skills. Since employment in the CCIs is often project-based, the amount of workers looking for jobs and recruiters looking for workers is high and an ongoing process. This leads to lower risks for both employees as employers.

(4) Within CCIs firms tend to agglomerate to make optimal use of external economies. This is because CCIs, more than other industries, benefit from mutual learning and cultural synergies. Cultural products are rarely a product of an individual and more often a result from interactions between multiple different agents. Agglomeration also leads to the rise of institutional infrastructures making flows of information, mutual trust and cooperation possible.

Scott’s (1997) findings are consistent with the findings of other scholars (Bagwell, 2008; Boix et al., 2010; Mudambi, 2008; Tomczak & Stachowiak, 2015), although other characteristics can

(11)

be added too. For example, CCIs are often location bound and cannot be transferred from country to country to lower labour costs like with manufacturing (Tomczak & Stachowiak, 2015). However, individual art can be made wherever since this does not depend on many physical resources or production facilities (Florida & Jackson, 2010).

Within CCIs demand is impossible to determine beforehand (Caves, 2002). Therefore, whether a firm becomes successful depends largely on their constant access to new information and knowledge. They keep their ears open for “local buzz”, rumours, impressions, recommendations (miss) information that spread around town and are used to make strategic decisions and monitor competitor behaviour (Grabber, 2002: 254). Being located in a creative milieu is also important, here, technology, entrepreneurship and creativity come together. This attracts a diversity of creative people that share their knowledge and ideas (Florida, 2002). This explains why creative industries are much more spatially concentrated than other industries and why you will often find them in urban areas like cities (Boix et al., 2010; Tomczak & Stachowiak, 2015; Scott, 1997; Scott, 2000; Maskell and Lorenzen, 2004). As mentioned before, global cities like New York, London and Tokyo have cosmopolitan environments and are globally orientated which is of extra importance for creative firms. Global cities have another benefit, namely the clustering of people which makes it possible to have many weak ties, which makes the transfer of knowledge, skills and information easier and accessible to everyone (Lorenzen & Maskell, 2004).

2.1.2 Music industry

The music industry is a CCI, consisting of many actors that have different tasks, including: major record labels, smaller, independent labels (often sub labels of the major record labels), and large pool of music publishers, producers, studio engineers, songwriters, agents, managers, and publicists. Record labels take care of promoting and distributing the music of their signed artists while publishing is concerned with the property right on songs and royalties that are collected when these songs are used by radio, TV, Film etcetera. Because of the knowledge intensity that comes naturally with being an CCI all the different actors, the music industry is a complicated industry (Krätke, 2003).

Caves (2002) made a distinction between artistic and non-artistic, or ‘humdrum’, segments of the music industry. The former being the actual artists creating the art. The latter being firms capitalizing this art by sourcing, marketing and selling the music as consumer products.

(12)

According to Watson (2008) there are 3 types of ‘humdrum’ companies at the base of the industry. The first type of humdrum company according to Watson (2008) is the MNE. There are a few huge MNEs that market, promote and distribute the music. The second type is a large pool of medium to large sized, independent, specialized companies involved in production and marketing that often have arrangements with the MNEs since it is hard to vertically integrate all these activities into their own firm. Finally, he mentions the many small & medium enterprises that make up the localized independent networks and have almost no ties to the MNEs. Watson (2008) also found that there are also quite often horizontal relations between the big MNEs. This is attributed to some firms having superior marketing and distribution competencies while other MNEs are more competent in A&R activities (discovering and signing artists).

MNEs in the music industry often work on project basis which are inter firm (Maskell & Lorenzen, 2004). For every product or creation of a product a whole team of differently skilled people is used. This ask for a lot of mutual trust and interpersonal relationships since multiple partners are involved. Often a MNE in the music industry has at “least one record company on board, one or more artists, a publisher, an AD provider and often, in addition, media firms and event firms.” (Maskell & Lorenzen, 2004). Artists have temporary contracts most of the time and only have a contract for 1 project (i.e. one CD). The same hold for outside partners, which only participate in one specific part of the project (Maskell & Lorenzen, 2004).

3. CONCEPTUAL FOUNDATION

This study continues with a conceptual foundation of the mainstream literature that is relevant for this study. The concepts discussed here give this inductive study some direction. Three concepts are chosen as they are seen as most relevant for studying the influence of location in CCIs: location choice, FDI motive, and entry mode. First, this chapter elaborates on the underlying theories about how and why MNEs expand abroad and what the reasons for expansion are. Next, multiple location strategies are discussed, like agglomeration and global cities, together with an introduction to location choices in the music industry.

3.1 Foreign direct investments

For many MNEs the expansion to new markets is a tricky process where many variables have to be taken into consideration. Factors like which country to enter and what entry strategy to

(13)

use need to be thought out clearly since they can influence the success of a new market entry (Kogut & Singh, 1988; Buckley & Casson, 1998). The process of expanding to new foreign markets is called internationalization and when a firm internationalizes it it doing a foreign direct investment (FDI). There are two main theories when it comes to internationalization: The Uppsala Model (Johanson & Vahlne, 1977) and The Internalization Theory (Rugman, 1980; Buckley, 1988). In the Uppsala model of internationalization MNEs first start with expanding to spatially close countries with similar country specific advantages (CSAs) before they expand to countries with a higher “psychic distance” (Johanson & Vahlne, 1977). The model follows incremental steps. First MNEs start exporting to a new host location. Next, they establish export channels or open up subsidiaries and finally they can use their learnings to expand to more distant markets (Johanson & Vahlne, 1977). The Internalization theory rest on two concepts: 1) firms choose a location for each activity they perform based on least costs and 2) firms internalize each market if it leads to cost benefits (Buckley, 1988). According to Rugman (1980), the internalization theory explains the reason behind a FDI, namely exploiting market imperfections, which leads to turning country specific advantages into firm specific advantages and external markets into internal markets. The main difference between the two internationalization theories is that the Uppsala model is focussed on expanding to low distance markets, while according to the internalization theory, expansion is done to a country if the firm can benefit from cost reductions related to transaction costs that arise in internal and external markets (Buckley, 1988), no matter if the host country has a high or low distance to the home country.

Both internationalization theories let MNEs slowly learn how to overcome the liability of foreignness (LoF) involved with operating in a new market. LoF was first explained by Zaheer (1995) and builds on Hymer’s (1976) work on comparative disadvantages MNE subsidiaries face. In general, it relates to the underlying costs MNEs experience when doing business abroad. These costs are mainly driven by the difference in culture, politics and economics between the home and host country (Zaheer, 1995). There are four identified sources for LoF: 1) costs related to geographic location and spatial distance between home and host country, 2) firm-specific costs due to unfamiliarity with the host country’s local environment, 3) costs related to the host country, and 4) cost related to the home country (Zaheer, 1995). Foreign firms have a cost disadvantage compared to local firms, leading to lower performance and a lower chance of firm survival (Zaheer, 1995). The LoF seems to be particularly present in market seeking 'horizontal MNEs' (Caves, 1982), MNEs whose subsidiaries have almost

(14)

identical operations and compete on a local-for-local basis. Traditionally, firm specific advantages (FSAs) can overcome LoF, but location choice is also an important factor when making foreign investment decisions (Dunning, 1998; Zaheer, 1994). The degree of LoF is highly influenced by location, being not only different on country-level (Zaheer, 1995), but also on regional-level (Rugman & Verbeke, 2004), in specific clusters or even global cities (Goerzen et al., 2013; Enright, 2000).

3.1.1 Foreign direct investment motives

When it comes to expansion not only the level of LoF important to consider. Most of all, the motive for internationalizing affects where a MNE decides to expand to and Dunning (1998) argues that location is factor that’s wrongfully neglected when looking at MNEs FDI motives. According to him the location of a firm’s activities might be an ownership advantage by itself since this becomes more important to their global competitiveness when the MNE becomes more multinational. He developed a model explaining how different FDI motives affect a MNE’s location strategy. This model distinguishes 4 reasons for FDI: resource seeking, market seeking, efficiency seeking and strategic asset seeking. Dunning (1993) explains these motives as follows: When MNEs have a resource seeking motive for FDI in a certain location they are looking for natural resources that are either simply available, have a better price or are from a better quality compared to the natural resources in their home location. Infrastructure and the presence of local partners that can help exploit knowledge and capital intensive resources are also important factors. MNEs with market seeking FDI motives are looking for new markets where they could operate in (Dunning, 1993). This could be a result of competing firms being already present in that location or because the MNE wants to be closer to its end users. The costs involved with serving markets from a big distance and pursuing localization strategies can also be reasons to invest in foreign locations (Dunning, 1993). MNEs with an efficiency seeking motive want to “take advantage of differences in the availability and costs of traditional factor endowments in different countries”; or “take advantage of the economies of scale and scope and of differences in consumer tastes and supply capabilities” (Dunning & Lundan, 2008, p.72). The final FDI motive, strategic asset seeking, stems from the need to protect or enhance the ownership specific advantages of an MNE by acquiring new assets or forming a partnership with a local firm (Dunning, 1993). The motivation for this type of FDI is “to gain access to knowledge or competences that are not inside the firm” (Franco et Al., 2008, p.9).

(15)

3.1.2 Entry modes

After the MNE decided on the FDI motive, it starts thinking about how to enter the location. There are multiple entry mode strategies divided into two main categories: non-equity-based entry modes and equity-based entry modes, with levels of ownership, control, risk etc. (Pan & Tse, 2000). Each entry mode is shaped by different influences like firm-specific, industry-specific, and country-specific factors (Agarwal & Ramaswami, 1992; Kogut & Singh, 1988; Pan & Tse, 2000; Dikova & Van Witteloostuijn, 2007). Within the equity-based entry mode there are multiple establishment mode choices, referring to the level of commitment and control. It’s the difference between merging with an existing firm, thus starting a joint venture, or owning the whole subsidiary. Joint ventures (JVs) result in less control and also less commitment compared to wholly owned subsidiaries (WOS), e.g. greenfield and acquisitions (Dikova & Van Witteloostuijn, 2007; Pan & Tse, 2000). However, JVs and acquisitions enable an MNE to use the resources and FSAs of the host company, thus lowering the LoF (Meyer et al., 2009). The institutional environment can also play a role when deciding on the right establishment mode. In a weak institutional environment, a JV can be preferred over greenfield or acquisition (Meyer et al., 2009).

3.2 Location strategies

For years, location has mostly been studied on the national level (Dunning, 1998). For example, the concept of LoF is primarily based on country characteristics and not on lower level subnational locations (Qian & Rugman, 2013, Zaheer, 1995). However, with the world becoming more global, sparking the globalization vs. regionalization debate in IB, countries in itself became slightly less important for MNEs when it comes to location strategy and focus started shifting to subnational levels like regional clusters and cities. Therefore, it is important to look into the origins of these developments.

One of the first scholars to write the importance of location strategies on a national level was Dunning (1970). He studied the interaction between location and ownership advantages of firms in terms of FSAs or host country advantages (Cantwell, 2009). However, over the years the interest of IB scholars shifted from macro to micro level, making location less of an interesting topic up until the 1990s (Cantwell, 1980). With the development of knowledge as a value creating asset and advancements in technology and global interconnectedness, scholars started to think about subnational levels of location choice (Cantwell, 2009). From there on, theories on (regional) agglomeration (Ellison & Glaeser, 1999; Ciccone, 2002; Mariotti et al., 2010),

(16)

clusters (Porter, 1998; Enright, 2000; Maskell & Lorenzen, 2004) and cities (Sassen, 1991; Beaverstock et al., 1999) emerged.

3.2.1 Agglomeration

Agglomeration is part of the subnational location theory. Multiple scholars wrote about this theory where MNEs are located in the direct area of the competition (Mariotti et al., 2010; Dunning and Lundan, 2008; Ellison & Glaeser, 1999; Marshall, 1920). Agglomeration can be described as the “spatial concentration of people or economic activity” (Malmberg & Maskell, 2001, p.430). There are three drivers of agglomeration distinguished by Marshall (1920): knowledge spill overs, input–output linkages and labour market pooling. According to Ellison & Glaeser (1999) locations have a natural advantage that accommodate agglomeration. This natural advantage is mostly cost based, e.g. cheap electricity, cheap labour, favourable climate etc. In contrast to earlier theory that agglomeration happens to gain from cost benefits, Mariotti et al. (2010) developed a model proving that MNE’s make location choices based on information externalities and knowledge spill overs. By co-locating, MNEs can learn from other firms that already experienced the problems related to unfamiliarity. This way, information costs about the local environment can be kept at a minimum. Other externalities like knowledge spill overs can enhance a firm’s performance e.g. by making it possible to improve products at little or no costs (Shaver & Flyer, 2000). However, MNEs usually co-locate to learn from others and take advantage of their knowledge and ideas, although these knowledge spill overs do not have to be positive. They can just as easily be negative, possibly resulting in crucial knowledge outflows. Therefore, the location choice of MNEs depend on if they think the overall effect of the spill over is positive (Mariotti et al, 2010).

Clusters are an example of agglomeration. Enright (2000, p.114) defines it as “the development of multiple firms in the same or closely related industries in the same location”. This definition is similar to Maskell & Lorenzen’s (2004) who state clusters are limited in both scope and space. Clusters can be based in regions, inner cities or other locations. Examples are industrial clusters, like the furniture cluster in Italy or the automotive cluster in southern Germany, and knowledge clusters like Hollywood in Los Angeles, California (Rocha & Sternberg, 2005; Scott, 1996). According to Porter (1998, p.80), clusters affect competition in three ways: “first, by increasing the productivity of companies based in the area; second, by driving the direction and pace of innovation, which underpins future productivity growth; and third, by stimulating the formation of new businesses, which expands and strengthens the cluster itself.” There are

(17)

thee types of cluster: independent, dependent and interdependent (Enright, 2000). In the independent model clusters arise through natural advantages certain location have that attract MNE’s. In the dependent model the cluster formed because MNE’s started to invest in a certain area forming the cluster. Enright (2000) developed the interdependent model stating that MNEs not only play an important role in the cluster, the location in the cluster is also vital to the MNE’s strategy. Subsidiaries in specialized clusters can even be of greater strategic importance than subsidiaries not located in clusters (Birkinshaw & Hood, 1998).

3.2.2 Global cities

For this paper, cities are a particular field of interest. Besides clusters, MNE location strategy can also be analysed on the subnational level of global cities. Near the end of the 20th century urbanization became a big trend. People started moving to the cities resulting in exponential growth and the creation of mega cities and global cities. Not only economists showed their interest in this new phenomenon. Sociologist Sassen (1991, 1996) was one of the first to write about cities and is still seen as the expert in this field. The shift of people made cities an integral part of MNE location strategy (Beaverstock et al., 1999). Cities have multiple advantages over rural or regional clustered areas. Despite what a lot of people might think, low-cost labour or real estate are not it. In stead strategic location, local market demand, integration with regional clusters and human resources are the real advantages of cities (Porter, 1995). The difference between global cities and mega cities is that global cities do not necessarily have to be big in size or inhabitants and can also facilitate more than one industry compared to mega cities.

Global cities have a high degree of centrality, are part of global networks, and are important to the global economy and its infrastructure (Sassen, 1991; Wall & van der Knaap, 2011). They are the main ‘command and control points of the global economy’ (Watson, 2008 p.14). Examples of such global cities are: Berlin, London, New York, Los Angeles, Tokyo & Sydney. Goerzen et al. (2013), found that MNEs tend to locate more in these cities because they have a lower LoF. There is less uncertainty, less discrimination and less complexity when it comes to doing business compared to other foreign locations. Global cities are characterized by 3 attributes: “a high degree of interconnectedness to local and global markets; a cosmopolitan environment; and high levels of advanced producer services.” (Goerzen et al., 2013, p.430). The three global city characteristics are described by them as followed: Interconnectedness makes the transfer of inflows and outflows easier; these can be flows of knowledge, capital, people, goods or information, as well as other global flows. It also lowers LoF, because there

(18)

is less information asymmetry about the local environment. Interconnectedness facilitates a cosmopolitan environment since foreign human resources flow easily in and out of the city creating a large, diverse pool of national cultures. Advanced producer services mean business service providers of MNEs are located closely to the MNE. This lets MNEs work with the same service providers globally reducing the need to import these services which results in lower coordination costs (Goerzen et al., 2013).

When it comes to the relation between cities and FDI motives Goerzen et al. (2013) also give a good analysis. Their study found that it is more likely for MNEs to locate in global cities when they have a market seeking FDI motive. In this scenario, MNEs exploit the global city’s infrastructure and access to media and branding channels to attract new customers in the host country. Their subsidiaries are therefore more demand driven with a focus on market servicing (Cantwell & Mudambi, 2011). In contrast, the other 3 FDI motives, resource-, strategic asset- and efficiency seeking tend to result in locating outside of global cities (Goerzen et al., 2013).

3.2.3 Location choice in the music industry

Within the music industry, face-to-face meetings and interpersonal relationships are much more important compared to any other traditional industry (Watson, 2008). That is why you will often find all music related companies located closely to one another. Baker (2012) found that media industries, like the music industry, benefit most of agglomeration effects. This is similar to knowledge intensive industries that are also famous for their agglomeration (Mudambi, 2008). For example, in the UK the productivity elasticity, the increase in natural productivity if a district increases 1% in size, in the music industry is almost two times as big as the economy average.

In the music industry, location can also be part of a certain brand image. That is why you will find different music genres being popular in different areas or cities, resulting in multiple locations where the industry is agglomerated (Bakker, 2012). For example, New York, Los Angeles and Nashville in The US. However, there is not one main music city. In smaller locations like Nashville, heritage plays an extra big role (Florida & Jackson, 2010). For example, because of the artists or specific music genre that originated from the city. The reason why these cities are attractive for the music industry is firstly, because large cities have large populations which means a big pool of diverse, potential customers (Ellis and Beresford 1994). Global cities are also more likely to have creative clusters (Wu, 2005; Florida, 2005). Most of

(19)

the high ranked art schools and conservatories are located in cities, even in small, homogenous countries, like The Netherlands and Denmark (Maskell and Lorenzen, 2004). Another reason for creative cluster to arise in cities is that artists tend to move there. Cities in general have better employment opportunities, in the sense that all top art critics, representatives from global studios, and recording companies are located there (Krätke, 2003), a higher rate of diversity and more tolerance towards diversity and artistic expression (Florida, 2002). That is why “even though musicians can come from anywhere, they migrate over time.” (Florida & Jackson, 2010 p.310). Cairncross (2001) predicted that location would become less important and that it would be taken over by other factors. Gibson (2002) supported this by saying that, with technological advances like the internet and digital distribution, location becomes irrelevant. Artists and firms could start making their location decisions based on their lifestyle. However, we currently still see firms in the music industry migrate to global cities. For example, Sony recently (2016) moved its Dutch subsidiary away from the area of Het Gooi to Amsterdam. This has to do with the fact that there are more advantages for musicians and firms to cluster in the same area.

These clusters are useful for artists to find inspiration, learn from peers and experience what other artists make (Lubbren, 2001). Which is similar to the reasons MNEs in traditional inudstries agglomerate: to learn from others and take advantage of their knowledge and ideas (Mariotti et al., 2010). In this sense, artists are resource and efficiency seeking. These resources can be: finding topics to write about by being surrounded by likeminded people or having other industry professionals around that can record, produce, market, and distribute your music. The later reason being in turn also an efficiency solution if the case is that the other party can do the task faster and better. Over time, city clusters and metropolitan areas rather played a big role in commercializing and popularizing music that has been made somewhere else (Florida & Jackson, 2010). Thus, these music clusters in cities became independent model clusters that exploit natural advantages the city has (Enright, 2000).

For this specific study, the focus is on the MNEs, the three major record labels: Universal Music, Sony Music and Warner Music. These MNEs are large, globalized operating firms that have subsidiaries all over world, mostly in cities where music clusters can be found. Each city can be seen as a local node within the global network of the MNE (Krätke, 2003). Music MNEs often acquire other firms to open up a market and increase their market share. Setting up subsidiaries enable MNEs in music to make quicker and more direct links to potential customers in that specific area. It also allows for some local integration since these subsidiaries are closer

(20)

to small specialists and service providers (Krätke, 2003). Although the record labels participate in clusters, they also have to deal with activities outside of this cluster. Since the companies and their artists are often operating globally, they have to arrange global sales and distributions. This makes them gatekeepers for relations outside of the cluster (Power and Hallencreutz, 2002).

3.3 Conclusion

This study explores how location choice plays a role in the creative and cultural industries. A solid foundation is build by covering the important international strategy and location choice theories in the IB literature and explaining what the CCIs are, how the music industry fits the label of CCI and how location choice in the music industry has been studied before. From this point, new theory will be developed by this study.

The next chapter discusses the research design of this study and introduces the 3 cases this study is build around. Furthermore, the methods used to gather and analyse the data are discussed.

4. RESEARCH DESIGN

The third section of this paper covers the methodology behind this study. It starts off with the explanation of the research philosophy for the research project and why ontology and epistemology are important. Then, the basics of qualitative research and why a multiple case study design is chosen are discussed, followed by the quality criteria related to this design. Next, the cases that are examined in this study are introduced, followed by explaining the data collection process. This section ends with a presentation of the used data analysis methods.

4.1 Research philosophies and approaches

Research philosophy concerns the manner in which data is being gathered, analysed and used to generate new knowledge (Saunders et al., 2012). In this study the post-positivist research philosophy applies.

There are two important concepts in research philosophy: ontology and epistemology (Saunders et al., 2012). A researchers’ ontological and epistemological perspectives are important, because they determine what the researcher will consider as valid contributions to theory (Peter & Olsen, 1983). Post-positivism assumes an objective ontological view. Ontology is the way a

(21)

researcher sees the world and what his point of view on reality is. An objective ontological view means that the researcher sees reality as it is (it is external to the researcher) and where the focus of the research project is not influenced by a researchers’ thoughts or beliefs, but are reflected in theory (Gephart, 2004). But, while positivists believed that the goal of science is to only study what can be observed and measured and nothing else, post-positivists do not rely on one single method (Clark, 1998). However, post-positivists recognize that humans are biased by theory, which is an important fact in regards of the inductive nature of this study, thus never reach 100% objectivity. Epistemology is about knowledge. It tries to answer questions like what can I know and how do I know it, and what is the relationship between a researcher and the knowledge (Saunders et al., 2012). A post-positivist epistemology assumes that it is not possible to know the exact truth or reality, only to a certain probability, because, for example, observations can be imperfect making all theory revisable (Creswell, 2013). This stance also fits the inductive approach of this study.

4.2 Qualitative research: multiple case study design

The design used in this study is qualitative. Qualitative research is different from quantitative research in that it analyses text and interviews to explore or explain certain phenomenon (Auerbach and Silverstein, 2003). Qualitative research is descriptive in nature and is about “who

said what to whom as well as how and why” (Rynes & Gephart, 2004; 455). It has a more textual

focus opposed to quantitative research which is based around statistical and mathematical knowledge (Rynes & Gephart, 2004). Furthermore, qualitative research often studies phenomena in their natural environments in which they naturally occur (Denzin & Lincoln, 1994).

This qualitative research is conducted via a multiple case study design. Case studies usually answers “how” or “why” questions (Yin, 2014). Relatively little is know about the CCIs, making a case study is the best option. This is because they are the best unit of research when exploring certain fields that are relatively unknown since they allow the researcher to understand observed complex social phenomena within a certain context and emphasise the importance of showing a holistic and real-world-perspective (Yin, 2014). This study follows an inductive approach as explained by Eisenhardt (1989) with the aim to generate new insights that can be used to propose new theories. An inductive study is characterized by an open-ended research question and a bottom up approach. This means that data is collected first, then it is analysed to look for patterns, and finally developing some theories (Eisenhardt, 1989). A theory

(22)

building approach ideally starts off without any theory under consideration or hypothesis to test (Eisenhardt, 1989). However, in this study, some important concepts are explained beforehand in the conceptual foundation to help readers understand the topic and give some direction data collection without being specific about the relationships between the variables and the theory. According to Eisenhardt (1989), another important characteristic of theory building is that the concepts or theory that emerges from the data analysis is compared to the existing literature to see what similarities and differences there are. Tying emergent and existing theory to one another strengthens the internal validity (Eisenhardt, 1989).

Yin (2014) describes how multiple case studies can use literal replication or theoretical replication to strengthen the external validity, or generalizability, of the research project. Choosing a literal or theoretical replication logic helps to manage the expectation of a study’s outcome and help to make sense of the insights generated in the cross-case analysis (Yin, 2014). This multiple case study will use literal replication since the cases are carefully selected because they are similar in context, which makes it believable that each case will generate similar outcomes across cases. In this multiple case study design, the object (Thomas, 2011) is location strategy and the subject is major labels in the music industry. The multiple case study consists of 3 cases within the same context: Universal Music Group, Sony Music Entertainment & Warner Music Group are all MNEs in the music industry with similar operations and markets. Within the 3 cases there are three embedded units of analysis: location choice, FDI motive and entry mode choice. Having multiple units of analysis makes generalizability of the case easier and therefore strengthen the external validity (Eisenhardt, 1989). The within-case analysis studies all three cases individually. Patterns are looked for within each case and and compared to all three cases in the cross-case analysis to gain new insights and possibly form new theory. Pattern matching strengthens the internal validity of this study.

4.3 Quality criteria

Case studies receive a lot of critique, because they are ought to not be rigorous enough, not generalizable and biased (Yin, 2014). These critiques can be overcome by building the case study according to four quality criteria: construct validity, internal validity, external validity and reliability. In the next few paragraphs these concepts are explained and elaborated on how they are applied to this study.

(23)

The definition of construct validity is: “identifying correct operational measures for the concepts being studied” (Yin, 2014, p.46). This means: how are you making sure that you are really studying the concept you say you are studying. In other words, how well does your case represent the claims you are making. There are several ways to overcome the critiques associated with the construct validity in case study designs. Firstly, one can ensure triangulation. One form of triangulation is to use multiple sources of evidence. This is called data triangulation (Yin, 2014). By having multiple data sources, it is possible to “constantly cross check information and data from different sources to increase the reliability and accuracy of [their] explanations” (Frynas, Mellahi, & Pigman, 2006; p.237). This specific case study shows a high level of data triangulation by using several sources of data (e.g. annual reports, newspaper articles, multiple databases). Secondly, making sure there is a chain of evidence to make replication possible (Yin, 2014). A chain of evidence is present (see appendix 2).

Internal validity, or credibility, is defined as: “seeking to establish a causal relationship, whereby certain conditions are believed to lead to other conditions, as distinguished from spurious relationships” (Yin, 2014, p.46). There can be several critiques the internal validity of a study. Yin (2014) wrote about a few ways to overcome these critiques. The first one is by making sure there is explanation building, where one makes iterative causal links and revises them throughout the cases. This study makes initial links in the within-case analysis and revises them in the discussion, if necessary. The second way to address internal validity is through pattern matching. By analysing data from multiple cases and looking for similarities across cases, the internal validity gets higher (Gibbert & Ruigrok, 2010). In this study pattern matching is used. There are 3 cases that are individually analysed in the within-case analysis after which the results of these cases will be laid side by side in the cross-case analysis to look for patterns. The emerged theory from the analysis is also compared to existing literature in the discussion to strengthen the internal validity.

External validity, or transferability, is: “defining the domain to which the study’s findings can be generalized” (Yin, 2014, p.46). This concerns if the results apply to situation beyond the case itself too. Ways to make sure external validity of a case study is sufficient is by using theory to build your case and by using replication logic when doing a multiple case study (Yin, 2014). Since this case study is based on companies that hold almost 70% of the market, it is very likely that the findings will be generalizable for the whole industry. Furthermore, the findings will be compared to relevant literature to strengthen generalizability (Yin, 2014). By

(24)

having 3 cases with similar operations and target markets it is possible to do logic replications which also strengthens generalizability (Yin, 2014).

Finally, it is also important to make sure the study is reliable. Reliability is defined as: “Demonstrating that the operations of a study – such as the data collection procedures [ analysis etc.] – can be repeated, with the same results” (Yin, 2014, p.46). The reasoning behind this quality criterion is to reduce mistakes. Gibbert & Ruigrok (2010) gave 3 tips to ensure this. 1) walk the talk. Make a case study protocol and database, and 2) focus on construct validity and internal validity over external validity and 3) be real and report issues and redesigns making it more believable. In this study a case protocol is made and all data is stored in a case database. Furthermore, all important steps in the data collection and data analysis are described in sections 3.5 and 3.6 of this study.

4.4 Case selection

This study is focused on the cultural and creative industries, with a special interest in the music industry. The study is a multiple case study based on the major record labels (“The Big 3”) in the music industry: Universal Music Group, Sony Music Entertainment & Warner Music Group (see Table 1). These cases are selected based on an information orientated selection of a few critical cases (Flyybjerg, 2006). They are also the only large MNEs in the music industry, making them the most interesting option to study the role location in this industry. Each MNE is globally dispersed with offices on every continent (see Figure 1). Together these MNEs make up +-70% of the whole music market, thus being extremely representative. The large influence of these MNEs also make the results more generalizable (Yin, 2014). Each of the MNEs in the cases has its ownership originating from different parts of the world/cultures, making it interesting to see if this is influencing decision making within each MNE. Since the cases are similar in their both operational locations (see Figure 1) and operational activities (recording & publishing) it is likely to assume that the cases will predict similar results making it possible for this case study design to do a literal replication.

(25)

Table 1: general information on the MNEs of the three cases

* Source: Annual reports of respective companies for the fiscal year of 2016

** Source: https://musicindustryblog.wordpress.com/2017/02/26/global-recorded-market-music-market-shares-2016/ *** WOS only. Source: Company website for each respective company

Case _CompanyParent _RegionHome HQ _{(in billion $)}Revenue* Share** Market (%) # of countries present*** (offices) # of international offices studied

Vivendi _(France)Europe

Santa Monica, California (The USA)

5.57 28.9 60 (47) 7

Sony (Japan) Asia

New York, New York (The USA) 5.8 22,4 50 (47) 7 Access Industries North America (The USA) New York, New York (The USA) 3.24 17,4 50 (42) 7

(26)

(27)

4.5 Data collection

To be able to answer the research question, both qualitative and quantitative data is collected from various sources. By using multiple data collection methods, the results become more generalizable because of triangulation (Yin, 2014). First, a list is made of all subsidiaries of MNEs and their city location. Next, relevant data is gathered on these cities’ characteristics. Finally, data on FDI activity of each MNE will be gathered.

Data on each MNE’s current subsidiaries and their locations is gathered through company websites and the Orbis database. The search terms used are: “universal music”; “universal music group”; “sony music”; “sony music entertainment”; “warner music”; “warner music group”. The results are scanned for relevance since the list compiled of Orbis data often consist of double entries in the form of companies in the publishing division, sub labels with their own entity, companies with another industry classification or, as in Sony’s case, many subsidiaries of the parent company. Eventually this leads to a list of subsidiaries around the world for every MNE. Next, this data is matched to the data on global offices available on the company website of these MNEs to create an ultimate list compiling all relevant subsidiaries and their location on country and city level for each of the three MNEs. Matching the data from Orbis to the data on the company website strengthens construct validity and also secures that no outdated data from Orbis is used and to check if the global offices are officially wholly owned subsidiaries of the MNE. In the case of Universal Music Group, additional data from IFPI is used since they do not give any information on their global offices on the company website. They only mention that they have offices in almost 60 countries, but it’s unclear which countries exactly.

The following step is to gather data about the compiled list of cities (see Table 2). The data consists of general city characteristics, like population, if the city is a capital, and its rankings on the Global City and Student City index, but also music specific characteristics, like number of music venues, recording studios, music festivals and the presence of higher music education in the city. The choice for these arguments is explained in the next paragraph.

According to Wu (2005) and Florida (2005), creative clusters are often located in global cities. Therefore, AT Kearney’s global city ranking is used to categorize all studied cities. The ranking is based on Sassen’s (1991, 1996) ‘global city’ concept where interconnectedness, cosmopolitan environment and advanced producer services are present. Beaverstock (1999) also made a ranking of world cities. However, this ranking is not seen as appropriate because it

(28)

is based on advanced producer services only, while AT Kearney’s ranking also considers the cultural and political environment and human capital among others. The original index ranks from 1-128. For this study a special mapping is made. The cities 1-10 are seen as Alpha cities, 11-30 Beta cities, 31-50 Gamma cities and everything above rank 50 are Other.

Capital and population are basic demographic characteristics that indicate city size and if the city is a capital or not, while Student City index rates how attractive a city is for students. The index is based on different measures like: student mix, desirability, affordability, employer activity and student view. Students are important for the music industry, because this group spends the most money on music (IFPI, 2015). Furthermore, a vibrant live music scene is important for cities to help new talent grow and engage with artists. Music festivals are a great way to discover new artists and by having many live music venues in different sizes (small, but also major concert halls) the artist has the opportunity to grow and play to a bigger audience when its career progresses (IFPI, 2015).

Finally, an advanced supportive infrastructure is necessary to create a successful music city (IFPI, 2015). Music schools attract new artists and the presence of recording studios keeps them in the city, since it gives them everything they need to create music.

Table 2: Categorization of cities

City characteristics Description Classification

Global City type AT Kearney Global City _{index 2017} The more global a city is, the more attractive to the _{music industry} Capital Capital of the country

Capitals are often the biggest or most important city in a country, making it possibly more attractive to

the music industry

Population Number of inhabitants _{audience which attracts artists and music MNEs}A higher population suggests a larger possible Student City Rank QS Best Student City

ranking 2017

Higher ranking indicates the city is more attractive to students, which is important for the music

industry (as explained on p.27) Live music venues Total number of live music

venues

Higher number indicates livelier music scene and opportunity for artists to grow

Major concert halls Number of concert halls; capacity > 5000 people

More major concert halls allow popular artists to come back to a city once they’ve become succesfull.

Keeping the music scene lively in the city Recording studios Number of recording studios

in the city

Higher number indicates livelier music scene; more opportunities for artists to make music. Music Education Number of higher music

education institutions

Higher number indicates more local talent, since music schools attract artists

(29)

Next, data on FDI activity of the sampled MNEs in the Triad is gathered through publicly available data sources (Zephyr, industry magazine articles, and company websites). The extended Triad (Rugman & Verbeke, 2004) is chosen because these are the most important regions in the music industry. The Zephyr database is used to gain access to data about mergers and acquisitions (M&As) performed by the sampled MNEs in the sample period from 2000 to 2015. This period is chosen, because it shows both economically difficult periods as flourishing times, making it interesting to see how location strategy is effected by the world economy. Furthermore, this period is long enough to ensure enough data is gathered. Finally, this timeframe is quite recent, making the outcomes relevant for managers in the present day. The search is focussed on the following deal types: acquisitions, mergers, and joint-ventures. Ownership in all these cases has to be 50% minimum to make sure the MNE is in control. The Zephyr results are scanned for relevance. M&A Entries from parent companies are removed. In addition, the company websites from every MNE are used to look for greenfield investments done in the sample period.

Case

Zephyr investments (M&A)

Company website investments

(Greenfield) Investments

studied

Total Studied Total Studied

Universal Music Group 47 18 - - 18

Sony Music Entertainment 46 11 - - 11

Warner Music Group 39 19 - - 19

Table 3: sample of studied investment of the case’s MNEs

An overview of the studied investments is found in table 3. Next, news articles on all investments are collected to analyse the FDI motives behind the investments. The articles were found on the company website or in industry magazine articles like Billboard, Music Business Worldwide, Music Week and more. The LexisNexis database is not used, because it was not possible to search specifically on FDIs. Results would contain too many articles about artist signings, board appointments and other non relevant news. By having multiple data sources data triangulation is secured to increase construct validity. Not for every investment a news article was found. For this reason, only investments that had at least one relevant article available were analysed (see table 4).

Case Relevant articles Articles used

Universal Music Group 46 33

Sony Music Entertainment 25 15

Warner Music Group 51 26

(30)

4.6 Data analysis

Since the data we need to answer the research question is mostly qualitative, thematic coding is needed (data on FDI and entry modes) to be able to understand it to part of the data. The data on all 3 MNEs subsidiaries and their city location is of both qualitative and quantitative nature, but does not need thematic coding since the data does not come from archival or documentary data, like newspaper articles or texts.

The thematic coding is done via CAQDAS (Computer Assisted Qualitative Data Software). This makes it easier to create codes/themes and identify those in the data. Thematic coding allows a researcher to identify patterns and starts of with creating themes based on the theoretical framework and then ranked accordingly to one’s importance (Ryan & Bernard, 2003). As mentioned in section 3.1 this study uses inductive techniques as described by Eisenhardt (1989). This means that most codes will be made after the data analysis of the collected data. However, a set of pre-set categories and pre-set codes is also used. These pre set categories and codes are based on the theoretical foundation of this study and developed during the collection of the data (Yin, 2014). A coding scheme can be found in table 5. The data is analysed by linking the data to some pre set codes. For this analysis of the articles NVivo is used. All relevant articles on FDI motives and entry modes of the 3 MNEs are scanned on relevance and uploaded to the program.

Codes (Parent nodes)

Sub-codes

(Child nodes) Definition

Entry mode Greenfield FDI where the MNE enters the host country via an _{independent subsidiary} Acquisition FDI where the MNE enters the host country via an _{acquisition of the subsidiary} Joint venture FDI where the MNE enters the host country by setting

op a subsidiary with a partner and split ownership Foreign direct

investment motive Resource seeking Reason for FDI is to seek and exploit resources Market seeking Reason for FDI is to seek a new market Efficiency seeking Reason for FDI is to exploit efficiency gains Strategic asset seeking Reason for FDI is to strengthen ownership advantages MNEs Universal Music Group

Sony Music Entertainment Warner Music Group Table 5: coding scheme

After the data collection and thematic coding, the data is analysed via via within-case and cross-case analyses. The goal of within-cross-case analysis is to find patterns in each stand-alone cross-case which

(31)

could later be used to generalize across all cases in the cross-case analysis (Eisenhardt, 1989). During the cross-case analysis, found patterns are also matched with the theoretical foundation (pattern-matching) to create new insights. This increases the internal validity too. All the data that has been gathered for this study has been analysed via the aforementioned methods. The next section goes further in depth and presents the data.

5. RESULTS

This chapter gives an overview of the results based on the data analysis of the multiple case studies. First, all three cases are discussed individually in the within-case analysis. Then, the three cases are compared to one another in the cross-case analysis to look for patterns.

5.1 Within-case analysis

In this section the findings of the data analysis are presented. It includes three within-case analyses for each MNE. As mentioned in the research design, each case has three embedded units of analysis: location choice, FDI motive and entry mode. Within each case, the three embedded units of analysis will be discussed separately.

5.1.1 Case 1 – Universal Music Group

Universal Music Group (UMG) is the biggest company in the music industry, with a market share of 28,9% (see Table 1). Their 3 main activities concern the distribution of recorded music, music publishing (Universal Music Publishing Group) and merchandising (Bravado). Since 2000, UMG is a subsidiary of Vivendi, a large media conglomerate based in France. UMG’s revenue in 2016 was 5.57 billion U.S. dollar. UMG is active in approximately 60 countries, including the industry’s biggest markets like the United States, the United Kingdom, France and Germany. Together with Australia, Brazil, Japan, The Netherlands and South Africa, these countries account for 81% of UMG’s total revenue (Vivendi, 2015). Furthermore, according to the 2015 annual report of Vivendi, over 65% of UMG’s sales come from local artists in their own countries. The biggest signed artists of UMG are: Taylor Swift, Lady Gaga, Justin Bieber.

UMG’s location choice: global offices in cities

UMG has approximately forty offices worldwide and operates in over sixty countries. A complete list of the global offices of UMG can be found in Table 6 together with a visual representation of the global offices for all the cases’ MNEs (as seen in Figure 1).

(32)

Company name Country City Region Global city _type

Universal Music Argentina SA Argentina Buenos Aires SA Beta Universal Music Australia PTY Limited Australia Sydney AP Beta

Universal Music GmbH Austria Vienna EUR Beta

Universal Music Belgium Brussels EUR Beta

Universal Music LTDA Brazil Rio de

Janeiro SA Other

Universal Music Group Canada Toronto NA Beta

Universal Music Chile SA Chile Santiago SA Other

Universal Music Colombia SAS Colombia Bogota SA Other

Universal Music de Centro America SA Costa Rica San Jose SA N/A

Universal Music D.O.O. Croatia Zagreb EUR N/A

Universal Music S.R.O. Czech republic Prague EUR Gamma

Universal Music A/S Denmark Copenhagen EUR Gamma

Universal Music OY Finland Helsinki EUR N/A

Universal Music France France Paris EUR Alpha

Universal Music GmbH Germany Berlin EUR Beta

Universal Music Limited Hong Kong Hong Kong AP Alpha

Universal Music Hanglemezkiado

Korlatolt Felelossegu Tarsasag Hungary Budapest EUR Other

Universal Music India Private Limited India Mumbai AP Gamma

Universal Music Indonesia PT Indonesia Jakarta AP Other

Universal Music Ireland Ltd. Ireland Dublin EUR Gamma

Universal Music Italia S.R.L. Italy Milan EUR Gamma

Universal Music LLC Japan Tokyo AP Alpha

Universal Music SDN BHD Malaysia _LumpurKuala AP _Gamma

Universal Music Mexico SA DE CV Mexico Mexico-City SA Gamma

Universal Music B.V. Netherlands Baarn EUR N/A

Universal Music New Zealand Limited New Zealand Auckland AP N/A

Universal Music AS Norway Oslo EUR N/A

MCA Universal Philippines Manila AP Other

Universal Music Polska SP. Z O.O. Poland Warsaw EUR Other

Universal Music Portugal SA Portugal Lisbon EUR N/A

Universal Music Romania SRL Romania Bucharest EUR N/A

Universal Music Russia Moscow EUR Beta

Universal Music Publishing PTE Ltd. Singapore Singapore AP Alpha

Universal Music Slovenia Slovenia Ljubljana EUR N/A

Universal Music PTY Ltd. South Africa Johannesburg Africa Other

Universal Music LTD. South Korea Seoul AP Beta

Universal Music Spain SL Spain Madrid EUR Beta

Universal Music Group Sweden AB Sweden Stockholm EUR Gamma

Universal Music GmbH Switzerland Zürich EUR Gamma

Universal Music Ltd. Taiwan Taipei AP Gamma

Universal Music Thailand CO LTD Thailand Mumbai AP Gamma

Universal Music Taxim Edition Turkey Istanbul ME Beta

Universal Music MENA

Middle East / North Africa United Emirates Dubai ME Gamma Universal Music Group International

Limited

United

Kingdom London EUR Alpha

Universal Music - Latin America United States Miami NA Beta Universal Music Group Nashville United States Nashville NA N/A

Universal Music Group United States New York NA Alpha

Location choice in the cultural and creative industries : a case study of the “big three” record labels in the music industry