The Uber Boundary: Contextualizing the Organizational Boundary of a Digital Platform Organization

(1)

The Uber Boundary

- Contextualizing the Organizational Boundary of a Digital Platform Organization

Master’s Thesis 30 credits

Department of Business Studies Uppsala University

Spring Semester of 2019

Date of Submission: 2019-05-29

Johanna Furulind Olivia Sjöqvist

Supervisor: Stefan Arora-Jonsson

(2)

Acknowledgements

Thank you

Professor Stefan Arora-Jonsson

for challenging us, even at the times we thought we needed it the least.

Our seminar group

for keeping us on track when drifting away.

The former Managing Director of Uber

for sharing your time and providing valuable insights we would not have obtained any other way.

Last, thank you to our families and friends for endless love, support and patience.

Uppsala, 29th of May 2019

__________________ __________________

Johanna Furulind Olivia Sjöqvist

(3)

Abstract

Digital platform organizations challenges the organizational landscape by utilizing technology enabling cost efficient transactions. Uber is a commonly known example of such an organization, where tensions have arisen in various local contexts due to that the organization draw their boundary tightly around the platform to the benefit of organizational efficiency. Uber has as a consequence, been highly questioned in its liability of their primary operations: Is it a technology platform, or a transportation firm?

This thesis sets out to answer: How is the organizational boundary of Uber challenged in various local contexts? The purpose is to provide an explanation towards how the organizational boundary of a digital platform organization could be understood in relation to local context, and how it can or cannot evolve in regards to these contexts. The results show that context matters for the boundary of Uber. In addition, a pattern amongst the context specific events emerged influencing the boundary. These can be categorized into four configurations: Adjusting, Expanding, Withdrawing and Intertwining. Each of the configurations explain the underlying reasons for the varying fluctuations the boundary of Uber show, and thus answer how the organizational boundary of Uber is challenged in various local contexts.

Keywords

Organizational Boundaries, Digital Platform Organizations (DPO), Transaction Cost Economics (TCE), Institutional Theory, Institutional Contexts, Uber, Ride-hailing.

(4)

1 Introduction

Over the last decade, an explosion of digital platform organizations have taken place. Innovative digital technology has introduced this category of firms with ways of organizing that significantly reduce transaction costs while breaking trade barriers and disrupting institutionalized industries. This by excluding components in the value chain to the benefit of organizational efficiency, and instead drawing upon the definition of being solely a digital platform organization. PricewaterhouseCoopers (2016) estimate a growth in transactions of these firms from €28 billion to €570 billion within the next 10 years, and digital platform organizations are furthermore expected to become a major part of the global economy as a whole (Zervas, Proserpio & Byers, 2014; Hamari Sjöklint & Ukkonen, 2016; Yaraghi & Ravi, 2017).

However, while some argue digital platform organizations are only expected to increase, others claim that the fast growth could be the downfall of the digital platform economy due to its ignorance of the basic requirements of a traditional organization. Critics accredit this risk to the rapid expansion in combination with the competitive edge brought on by its tightly drawn organizational boundary compared to the traditional industries, a fact that has come to question the legitimacy of the firm (Heath, 2016). Essentially, the problems stem from the fact that there are institutional expectations that these firms will care about matters like worker rights, taxes to be paid and general industry requirements, but instead they principally attempt to ignore it.

Harvard Professor Edelman (2017) goes as far as stating that some of these firms are finding competitive advantages by founding their business models on lawbreaking, and that it should not be allowed for firms to disrupt industries due to illegal activity.

This is particularly evident when looking at Uber, the unicorn corporation known for its global disruption of the transportation industry (Cannon & Summers, 2014). Ubers idea was both brilliant as subversive. No waiting times, no call centers, no outdated taxi booking systems, no need for cash or payment hassles, no need to guess how far the car was from pick up. Everything was digitized (Fagerström & Johansson, 2016). Instead of calling themselves a transportation firm, Uber consider themselves a technology platform. This is motivated by them solely providing a platform for drivers and consumers to use in order to carry out a service transaction

(7)

(Kenney & Zysman, 2016). By doing so, Uber could initially avoid formalities such as licensing and employing drivers. They also found ways to make use of privately owned cars, enabling them to avoid commercial insurance, commercial plates, vehicle inspections and thorough background checks of drivers, necessities otherwise generally required by law for commercial transportation organizations. The list of cost saving advantages could go on, but the point is that Uber managed to acquire significant cost advantages over its competitors in the traditional transportation industry (Edelman, 2017). They could then use their cost advantage to lower prices offered to consumers, and thereby gain market shares (Edelman, 2017).

In that sense, Uber, like many other digital platform organizations, initially drew their organizational boundary tight around the platform to the benefit of organizational efficiency. As a consequence of this distinction, it has given rise to a tension between Uber and the institutional contexts it operates in. The organization has been challenged in their liability of their primary operations: Is it a technology platform, or a transportation firm? And what does the tension between the distinctions mean for its organizational boundary? Due to the controversy, Uber has been faced with and responded to, varying degrees of bans in various local contexts throughout its global expansion (see Hao, 2017). In line with what is stated above, the bans are directly connected to what institutional contexts expect them to do, but that they are not doing. Building on that fact, this thesis research why and how local contexts impact the organizational boundary of a digital platform organization.

From a theoretical perspective, current organizational boundary research cannot provide us with the answers we are searching for in terms of why and how the boundary is challenged by local context. This is primarily because organizational boundary research usually treat boundaries as a uniform phenomenon which seldom takes local context into account (see; Thompson, 1967;

Pfeffer & Salancik, 1978; Santos & Eisenhardt, 2005). Secondly, such research is often theory driven rather than stemming from empirical observations, which might not necessarily be the most beneficial way of getting closer understandings of modern boundary phenomena (Santos &

Eisenhardt, 2005). The limitations with previous research in relation to digital platform organizations could be summarized by quoting Santos & Eisenhardt (2005, 505):

(8)

“The contemporary landscape is being transformed by new boundary choices and non-traditional settings. Here, while extant theory may apply, it may be wiser to take an open-minded, problem-driven

approach. Such an approach is likely to lead to immediate benefits for practice in terms of firmer grounding of significant corporate and public policy choices. In the longer term, it may lead to fresh

theoretical ideas that are outside of the box of known theory.”

In essence, what we set out to do is to explain how institutional pressures in various local contexts influence the organizational boundary of a digital platform organization. Because, how come Uber is fully banned in Turkey, Portugal, and Denmark, but not in the US, UK and India (see Rhodes, 2017)? Why there are partial bans in France, Italy and Australia, but none at all in Russia (see Hao, 2017)?

1.1 Research Question

How is the organizational boundary of Uber challenged in various local contexts?

1.2 Purpose

The aim is to provide an explanation towards how the organizational boundary of a digital platform organization could be understood in relation to local contexts. It moreover attempts to set the tone for how digital platform organizations can or cannot evolve in regards to local contexts.

(9)

2 Literature Review

The literature review will first provide a short overview of organizational boundary research, and why further research is necessary to answer the posed research question. Then, an explanation of digital platform organizations as a phenomenon follow. In terms of the chosen theoretical tools, Transaction Cost Economics (TCE) will thereafter provide an illuminating understanding towards the ideal boundary of Uber. To shed light on why Uber is met with such controversy, the thesis will then draw upon Institutional Theory which in contrast to TCE consider context a crucial factor for the success of organizations. The chapter will finish with an operationalization section.

2.1 Organizational Boundaries

Organizational boundaries is a field of research which have been given significant amounts of academic attention the last century. Consequently there are numerous theoretical lenses which can be applied to understand boundaries of organizations (see Thompson, 1967; Pfeffer &

Salancik, 1978; Santos & Eisenhardt, 2005). A boundary could in its simplicity be understood by the definition: “A line which marks the limits of an area; a dividing line” (Oxford, 2019). But in organizational studies, such simplicity will not take us far. Instead, researchers have been concerned with all from the social structures’ meaning for boundaries (see Dutton, Dukerich &

Harquail 1994; Kogut, 2000), to boundaries dependency on the resources controlled by a firm (see Helfat, 1997), to extending the boundary definition to the influence an organization has in terms of industry control and power of external forces (see D’Aveni, 2001; Santos & Eisenhardt, 2005). Others have moreover argued that one cannot separate organizations from its environment since the organization itself is made up out of organizations, whereas boundaries really comes down to decision making more than anything else (Ahrne & Brunsson, 2011; Dumez &

Jeunemaître, 2010).

However, the emergence of digital platform organizations have brought substantial societal discussions where such firms and their stakeholders are in major disagreement regarding where the organizational boundary should and can be drawn. At the same time, it is proclaimed

(10)

digitalization and technology is driving the global economy into a reality which is more interconnected than ever before, making boundaries increasingly complex (Stroh, 2014). It is furthermore stated that the phenomenon now include undiscovered mechanisms and that traditional boundaries are dated in regards to the modern economy (Scott & Davis, 2006;

Bernholz, Cordelli & Reich, 2013). Thus, updated theoretical frameworks are necessary to understand organizational boundaries of digital platform organizations. This thesis proposes such an updated framework, but before digging into that, digital platform organizations as a phenomenon will be further explained.

2.2 Digital Platform Organizations

Ever since the internet revolution, digital technology has been utilized to rapidly evolve numerous industries. Ways of constructing and performing business have been reinvented and refined to levels which have made the economy even more efficient (Kenney & Zysman, 2016).

It was through this transformation of the business landscape that digital platform organizations were born. The common denominator for digital platform organizations could be summarized as

“a building block that provides an essential function to a technological system and serve as a foundation upon which complementary products, technologies, or services can be developed”

(Spagnoletti, Resca & Lee, 2015; 1). These types of organizations are based on innovative business models which are translated into digital platforms. As an organizational form, it has extensively captured the interest of c-suites, board rooms, scholars and policy makers (De Falco, Renzi, Orlando & Cucari 2017; Dhar & Stein, 2017; Bailey and Bakos 1997. Digital platform organizations are also commonly predicted to have increasing impact on the global economy in the years to come (Sundararajan, 2014; Burtch, Carnahan and Greenwood, 2018). One of the main benefits digital platform organizations bring from a business perspective, is their ability to take profit orientation to the next level. This is possible due to the fact that these organizations have found methods to offer the same services as many traditional industries, but in a significantly more effective way. Simultaneously, business models of digital platform organizations have not seldom proven to be globally scalable, unlike anything the traditional industries have been able to offer (Kirchner & Schüßler, 2019). Even though these economic benefits poses a few major challenges that will be explained in more detail later, the efficiency of

(11)

digital platform organizations has shifted the idea of regulated transactions known in monopolized markets, to a reality where corresponding services are offered but in an unregulated manner as the institutional environments are lagging behind the competitive development these organizations have brought (Gawner & Cusumano, 2014; Kenney & Zysman, 2016). This moreover means regulators are exposed to challenges in creating new market arrangements due to the emergence of digital platform organizations (Kirchner & Schüßler, 2019).

2.3 Transaction Cost Economics - How to understand the ideal boundary of Uber

Imagine the management of Uber in a conference room right when the organization was about to take form. They are thinking about how to draw the boundary of the organization, but they are not entirely sure of how to best set up the service transaction between them and the drivers.

Ketokivi & Mahoney (2017) explains that naturally, both sides in any transaction want to protect their respective interests while avoiding unnecessary risk and cost. But what is the best way for them to move forward? As resources are constrained it is looked upon as common sense to pursue the solution which is most economically efficient. This ultimately means that when faced with multiple options, the one which requires least resources is generally most preferable (Ketokivi & Mahoney, 2017).

Above section illustrates the foundations of Transaction Cost Economics (TCE), a theory which is concerned about organizational boundaries from an inside-out perspective, building on the most economic efficient solutions. As exemplified in the scenario, it is a theory designed to help understand how individuals or organizations should act to sustain outcomes which are efficient from an economical point of view (Ketokivi & Mahoney, 2015). It is known that not every aspect of an organization could be explained from an efficiency perspective, but a very large number of them can. This is particularly true in cases where transaction cost aspects are emphasized (Williamson, 1981). On a general level, TCE in its purest form provides a tool to make sense of what should be in-house, what should be outsourced or offshored, how things should be contracted and how an organization should be governed (Ketokivi & Mahoney, 2015).

(12)

As Williamson (1981; 549) himself put it: “This can be thought of as developing the criteria for and defining the “efficient boundaries” of an operating unit.” In essence, TCE aspires to convey how waste can be minimized while value is maximized in transactions connected to the firm, also known as discriminating alignment (Ketokivi & Mahoney, 2017).

Uber could be considered an example of TCE embodied in real life as their organizational efficiency is designed to be superior to the one of the traditional transportation industry as technology enabled them to redefine the most efficient way to organize their boundary from an economic perspective, which facilitated for them to come up with a comparatively superior business model. However, TCE’s extensive economic perspective fail to take other relevant aspects into account. If TCE would be functioning perfectly according to how the theory is designed, Uber would not be faced with the challenges it has stemming from institutional forces, and instead likely would conquer the market as the model is minimizing waste and maximizing value to greater lengths than the transportation industry it is so infamously challenging.

2.4 Institutional Theory – Putting Uber into context

Previously we explored a theoretical idea which could motivate why the ideal boundary of Uber is so narrow. Now we dive into another theory which put Uber into context and explains how local pressures has an influence on any organization – institutional theory. In contrast to TCE which favors economic efficiency, institutional theory on the other hand suggests that context is a make or break factor for any organization (Meyer & Rowan, 1977; Furusten, 2013). Moreover, Scott (2014) emphasize how institutions sets limits for how organizations can or cannot act and hence determine whether or not an organization survives within a context. In the very same way, institutional theory, in combination with TCE could explain why a digital platform organization is being fully or partially banned within certain local contexts.

In Meyer & Rowan (1977) it is argued organizations often adapt to prevailing structures within the institutional environment in order for the organization to gain legitimacy and increase its chances of survival. Legitimacy is thereby a central concept within institutional theory and symbolizes the importance of being acknowledged as a credible actor in order to survive as a

(13)

business. If an organization violate the demands the institutional environment impose on them, it will be at the cost of their legitimacy and thereby its operations (Scott, 2014; Furusten, 2013). To illustrate further, not obtaining legitimacy could mean that neither customers, suppliers nor other stakeholders would want to be associated with the organization, which means that the rules and expectations of the specific context must be followed (Meyer & Rowan, 1977; Furusten, 2013).

In contrast, it is also argued organizations often shape the institutional contexts they act within.

This points to two important considerations since adapting to institutional environments often conflicts with the organizations' efficiency criteria and consequently also implies that if an organization adheres to conformity to contexts, institutionalized legitimacy is at the expense of the organization's efficiency and vice versa. “Organizations do often adapt to their institutional environment, but they often play active roles in shaping those contexts” (Dowling and Pfeffer, 1975; Parsons, 1956; Perrow, 1970; Thompson, 1967). In the former, adapting to institutional environments, Meyer & Rowan (1977) argue organizations incorporate institutionalized elements into the existing internal structure, which results in an increased commitment from both external and internal stakeholders. To be seen as part of the social system, rather than a separate independent system, can make organizations seen successful from a social perspective and give legitimacy that can help the organization avoid failure, or as in this thesis, a ban. In the latter, changing institutional environments, Meyer & Rowan (1977) argue organizations can maintain a so-called ceremonial conformity by making organizations loosely coupled from their formal structure and activities. Here one can also choose to "seek charters from collective authorities and manage to institutionalize their goals and structures in the rules of such authorities." A telling example of how that could be explained by the cite below:

“Efforts to mold institutional environments proceed along two dimensions. First, powerful organizations force their immediate relational networks to adapt to their structures and relations. For instance, automobile producers help create demands for particular kinds of roads, transportation systems, and fuels that make automobiles virtual necessities; competitive forms of transportation have to adapt to the

existing relational context. But second, powerful organizations attempt to build their goals and procedures directly into society as institutional rules. Automobile producers, for instance, attempt to

create the standards in public opinion defining desirable cars, to influence legal standards defining satisfactory cars, to affect judicial rules defining cars adequate enough to avoid manufacturer liability, and to force agents of the collectivity to purchase only their cars. Rivals must then compete both in social

networks or markets and in contexts of institutional rules which are defined by extant organizations. In

(14)

this fashion, given organizational forms perpetuate themselves by becoming institutionalized rules.”

(Meyer & Rowan, 1977, 348)

It is not an unreasonable assumption that the underlying reasons for Ubers’ different welcomings in various contexts comes down to that it one the one hand is part of a new organizational form which has not deserved its legitimacy yet, and that it is a global firm which originates from a specific institutional context. And on the other, that the firm might not have paid too much attention to institutional forces initially in general. But regardless, if an organization does not conform to the structures at hand to be accepted as a serious and legitimate actor, there are according to Furusten (2013) three options to choose from; 1) Surrender to the rules and structures anyway; 2) Try to refuse the rules and structures. This is a risky option as the odds will not be in favor of the rebellious organization and will likely lead to loss of credibility and trustworthiness; 3) Try to change the requirements of the institutional context. If an organization would go for option three, both patience and systematic work is necessary. However, some levels of legitimacy is still required to be able to have the demands respected and heard. To be successful, it is also important to mobilize with other actors to increase the chances of potential change.

Another way of describing how organizations go about strengthening their legitimacy within a context is by drawing on organizational isomorphism. Organizational isomorphism offer understandings in how firms with time will start to imitate each other in order to strengthen their legitimacy (Scott, 2008). The concept of isomorphism could be defined as “a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions” (Pugh, 2007; 439). This means organizations exposed to a given institutional context will conform their characteristics in a direction which is in accordance with the environmental characteristics. But maybe even more relevant for the ambition of this thesis is one specific form of isomorphism; coercive isomorphism. Coercive isomorphism relates to the formal and informal pressures an organization is put under by other organizations in the same field. This in particular by other organizations which another is dependent on, hand in hand with the cultural expectations the societal context at hand provides in relation to the field. Such pressures comes in many shapes and forms, and can be forced ones, stemming from lobbying or other types of persuasion or simply invitations for collusion. Obviously, governmental pressures

(15)

and enforced laws and rules is included (Pugh, 2007). In addition, Aldrich conveyed already in 1979 that one of the most crucial factors for organizations to consider is other organizations. The core of his message was that firms are not only concerned with competing for resources and customers, but also for political power, social and economic strength as well as for institutional legitimacy.

Regardless, for any type of isomorphism to occur, the involved organizations primarily need to be part of the same organizational field. Organizational fields as a concept is defined by Pugh (2007; 437) as “those organizations that, in the aggregate, constitute a recognized area of institutional life: key suppliers, resource and product consumers, regulatory agencies, and other organizations that produce similar products or services”. It should further be noted, organizational fields are not limited only to competitors, but rather captures the totality of relevant actors (Pugh, 2007). DiMaggio (1982) explains that organizational fields only can be defined as a result of empirical investigation as they are dependent of the institutional definition at hand. In Pugh (2007), it is moreover explained that organizational fields in the early stages of their life cycle very well can be quite different both in terms of approach and form. But even if so, once a field has matured and settled into its environment, the entities who are part of that field will start to homogenize as a way of acquiring legitimacy. Research further concludes, this will also be the case of new entrants to a field (Pugh, 2007). Keeping this in mind, organizational fields allow for the thesis to combine Uber, a prominent actor in the community of digital platform organizations, with the highly established transportation industry as they are vastly relevant to each other - and thereby also exposed to the same isomorphic forces.

(16)

2.5 Operationalization

The literature review has shown that organizational boundary research is constrained in relation to the posed research question of this thesis. To find answers, the theoretical tools are instead constituted by TCE in combination with Institutional Theory. In order to analyze the contextual adaptations of Ubers’ organizational boundary the thesis needs to clearly define a starting point from which the adaptations are measured. The most natural starting point for this purpose is what could be described as Uber’s ideal organizational boundary. TCE serves its purpose by defining such ideal boundary and thereby allows for an analysis to be made of how the boundary is changing and not. It should further be noted that the general discussions about digital platform organizations descend from a hypothetical environment with idealistic conditions in which TCE could fully explain any organizations boundary. However, TCE does not paint the whole picture of the problematics such organizations face as its weak spot is what is going on outside of the organization. It simply does not concern the outside perspective. Institutional Theory on the other hand firmly argue external environment is everything, and thereby add further important perspectives which explains how Uber is differently challenged in different local contexts.

Together, these two theories in combination serve as a sufficient tool with which the research question can be properly analyzed and answered.

(17)

3 Method

3.1 The case of Uber & why it is interesting

Uber has during the past decade found ways to break trade barriers, disrupt an established industry and managed to reduce transaction costs in a way the organizational landscape have not seen before. This gives us two important presumptions. Firstly, Uber could be regarded a commonly known organization which has paved the way for similar platform organizations.

Secondly, its extensive local presence globally and due to the fact of Uber being one of the world's largest digital platform organizations, Uber composes a representative case as it could be argued to exemplify a digital platform organization. In line with this, Eisenhardt & Graebner (2007, 25) argues that “case studies emphasize the rich, real-world context in which the phenomena occur”. Furthermore Yin (2007) state that single case study designs with benefit could be used with studies of representative cases to where the purpose is to explain something that can represent a commonly occurring phenomenon. Moreover, Bryman & Bell (2015) argues that the researchers should choose a case where the deepest learning could take place. It is thus arguable that the usage of Uber will provide a deeper insight into digital platform organizations and hence also contribute to a deeper level of knowledge.

3.2 Interpretive Reasoning & Qualitative Approach

Santos & Eisenhardt (2005) concluded that the vast majority of previous science has been theory driven and that there are indications towards a more contemporary landscape where the organizational boundaries ends up in nontraditional settings which results in a need of including alternative perspectives. Simultaneously, Scott & Davis (2006), Bernholz et al. (2003) and Stroh (2014) highlight how organizational boundaries also increases in complexity due to an interconnectedness by digitalization and the modern economy. As a result, it is arguable that through solely use a theory driven approach, important elements that such modern digital platform organization which seem based on transaction cost economics contains, where the significance of context also has been given another meaning, could be lost. Drawing on our theoretical knowledge in combination with the empirical problem at hand, interpretive research

(18)

has been the most appropriate approach in order to not become limited, neither theoretically nor empirically. An interpretive research enables according to Bryman & Bell (2015) an interplay between theory and practice and an appropriate selection of theories to the empirical findings and vice versa. Additionally, Alvesson & Sköldberg (2008) emphasizes that an interpretive research is preferable when studying complex phenomena since it seeks to explain the underlying patterns. Moreover, to understand the organizational boundary of a digital platform organization through the theoretical framework of transaction costs economics and institutional theories we have, due to the complexity of the phenomenon chosen a qualitative approach. A qualitative method could be argued to be the most appropriate approach since that, in contrast to quantitative, seek to understand complex phenomena and increases the chances of a deeper understanding of the phenomenon studied (see Bryman & Bell, 2015).

3.3 Data Selection

In order to capture the essence of how the organizational boundary of Uber is challenged in various local contexts, we constructed the data selection around contexts in the world where Uber have been faced with any type of full or partial ban. Thereby, the data is not limited to specific bans, nor are they distinguished in terms of timeframe. Instead, we are simply interested to see how the organizational boundary of Uber have been challenged within contexts, and how Uber has responded. In that sense, the data selection is not constructed with the purpose to make exhaustive comparisons between local contexts. The selection of locations rather give the thesis a maximum variation in the sample which according to Eisenhardt & Graebner (2007) could be explained when choosing samples of extreme cases, very high and very low performing data.

This enables the researchers to observe contrasts and patterns that would have been otherwise difficult to detect and thus leads to a clearer pattern of the phenomenon.

(19)

The sample size is constituted by 25 local contexts, situated in North America, Europe, Asia and Australia. As Uber to date offer multiple product lines, it should furthermore be noted this study is primarily concentrated to drivers on demand. Based on above, we argue that since Uber constitutes the case, and we are interested in how Ubers boundary has been challenged in various local contexts, the choice of including all type of bans, the thesis has greater potential to bring forward a more nuanced understanding of how the organizational boundary of a digital platform organization could be understood from a deeper perspective in contrast to if we rather chose to focus on a specific context or part of the world, which would have been beneficial if the purpose would have been to analyze a specific type of ban, and not the organizational boundary of Uber.

The sample selection is however, limited in the sense that there may be other institutional pressures than those identified in the thesis, which did not result in a significant ban, but which may still have challenged the organizational boundary of Uber. Having stated that, we are humble in that the data cannot provide a complete overview. On the other hand, from a research perspective, the linkage between the ban and the boundary can be argued to enable a more accurate analysis to be able to draw relevant and distinct conclusions out of the data.

(20)

3.4 Data Collection Documents

To be able to collect the distinctive events of respective bans, secondary data in the means of documents deriving from public and official documents have been collected. More specifically, news articles from globally renowned newspapers as well as local ones were used along with court- and legal documents as well as press releases and blog entries made directly by Uber which primarily were used to validate information found in news articles. The gathering of documents was all read in its original in order not to extract information out of its context and thus avoid misinterpretations. We chose to review at least three documents from various sources in each context and focused on concrete and practical arguments of the ban as well as public statements from Uber addressing the information at hand. This enabled us to continuously evaluate and secure the quality of the data from two independent, and opponent sources and with that confirm the quality of the ban in relation to how it had an impact on Uber's organizational boundary. Inevitable, there is a risk of a potential loss of documents that was not included in the thesis much due to the fact that a ban in itself is of such complexity that it would not fit in one thesis alone, but again, the purpose is to gather a greater understanding for the organizational boundary of Uber.

3.5 Data Analysis Documents

The local contexts were initially structured according to the level of ban and secondly divided into 1) event causing ban and 2) Ubers response to ban. The data were further analyzed through a configurational analysis, which Fiss (2009) points out is unique in case studies in that the totality could only be understood by the combination of the different parts or elements. Similarly Meyer

& Tsui (1993) argues how configurations build on conceptually distinct characteristics occurring together while the specific parts then form the configuration. In this thesis, the configurational analysis took a taxonomic approach in order to structure the data according to different events causing the ban in relation to Ubers response to the ban. This furthermore enabled us to derive four different configurations: 1) Adjusting, 2) Expanding, 3) Withdrawal and 4) Intertwining, which facilitated an understanding how the various parts or elements, in combination, resulted in a certain outcome for the organizational boundary of Uber (see Appendix Table 1, 2, 3 & 4).

Therefore in this phase, these four will further on be viewed as separate institutional contexts.

(21)

This will enable the analysis to treat various configurations in accordance with boundary outcome, rather than shifting back and forth between national contexts. This is motivated by the fact that an institutional context does not necessarily need to coincide with national border.

3.5.1 Coding Rules

To enable the analysis of the collected data, we bundled the context specific events into broader configurations which define the organizational boundary: 1) Adjusting 2) Expanding 3) Withdrawing and 3) Intertwining. To ensure a logical and consistent mapping of context specific events to respective configuration, a number of mapping rules were set up. Starting with the Adjusting configuration, we have analyzed events where Uber has attuned their boundary by making minor adjustments to their platform, but still offer the initial core services. The

“adjusting” events have been analyzed context by context, and then reorganized into broader categories; adjusting to legislation, adjusting the application and adjusting worker right offerings.

The Expanding configuration is defined by events where Uber remain active in the context, but have pulled out one or more core services. The Expanding events have been analyzed context by context, and then reorganized into broader categories; expanding driver requirements (no unlicensed drivers), and expanding service offerings due to unfair competition. The consideration whether an event was mapped expanding driver requirements versus expanding service offerings due to unfair competition lies mainly if the principle argument used by regulators relates to formalities, or to protect the established domestic transportation industry.

For the Withdrawing configuration, events were analyzed in contexts where Uber have exited entirely from a context. Withdrawing events are defined by Uber ceasing its entire operations under its own brand in a context where it has entered. The Withdrawal events have been analyzed context by context, and then reorganized into broader categories; withdrawal due to business model protection, withdrawal due to unfair competition and withdrawal due to mergers.

The consideration whether an event was mapped to withdrawal due to unfair competition versus withdrawal due to business model protection lies mainly with the principle argument used by the

(22)

regulators. In the context specific events mapped to unfair competition, this argument was cited explicitly by the regulators. In events mapped to protecting the business model the principal argument has rather been related to specific requirements. The Intertwining configuration is defined by events where Uber has joined forces with a competitor and started a new ride-hailing business. This is a unique event only situated in the context of Russia.

3.6 Data Collection Interview

An expert interview, with a former national Managing Director of Uber (Uber MD), holding a position at the same time the organization faced a ban, was furthermore chosen strategically in line with the research problem. This enabled us to further extend the data collection and to ensure expert input from the case firm itself. The Uber MD could be argued to possess relevant knowledge and direct information of a ban which increases our chances of a more to the point analysis. Various types of data could according to Bryman & Bell (2015) be used advantageously as they can be seen strengthen and build on each other and give a deeper understanding of the research problem. An interview guide (see Appendix Table 5) was designed based on the emerging themes and key findings identified in the initial document collection as well as from the theoretical framework. The interview guide consisted of questions related to the themes and while some questions were more direct, others were more open to add more flexibility around the topics. Bryman & Bell (2015) refer to such an interview as semi-structured and typical for a qualitative method. The interview was furthermore recorded with consent and the very same day transcribed word-by-word to decrease the risk of misinterpretation.

3.7 Data Analysis Interview

The expert interview with the Uber MD was analyzed with inspiration from Corley & Gioias (2004) coding process. Its purpose was to cross check the findings from the document studies, as well as bring potentially new perspectives to the empirical data. This resulted into a coding scheme (see Appendix Table 6) with seven themes distilled into key findings to lastly show our interpretations.

(23)

3.8 Ethical Considerations

The perspective of solely including contexts where Uber have been subject to bans could be perceived as a negative angle towards Uber as a brand, however that is not the ambition with this thesis. The choice to include Ubers own response regarding each of the specific bans in the thesis will hopefully grant the recipient with a more gradate picture of the research problem at hand and therefore we argue that this contributes to not damage either Uber's nor the local contexts reputation. Regarding the interview, Bryman & Bell (2015) state that there are four main areas to consider when reflecting upon ethical principles of business studies. 1. Potential harm for participants 2. Potential lack of informed consent 3. Potential invasion of privacy, and 4.

Potential deception. This study concerns one participant; the Uber MD. We have since the initial contact given the respondent transparent information, provided the opportunity to review the used material prior to submission, as well as provided the opportunity to be anonymous and delete information. The respondent was further informed that no question was mandatory. It was decided to not publish the name of the respondent but rather state the business title as such information would be enough to grant the reader with understandings of the respondent’s insight in Uber.

(24)

4 Analysis

4.1 Why is the boundary of Uber challenged?

When Uber was first born as an official organization, it was known by the name UberCab.

However, it did not take long for the firm to realize this would cause problems as the transportation industry and regulators were quick to apply their institutional rules and regulations on them. Therefore, by the end of 2010, Uber attempted to detach themselves from the transportation industry and its regulations by changing their name to just “Uber”. Thereby, the firm also attempted to hold their ground as a sole technology platform with significantly narrowed organizational boundary (Uber, 2019). From that point forward, Uber has continued to challenge existing incumbent industries by disrupting how these industries have been functioning for decades. Uber and digital platform organizations in general could thereby be seen as an entirely new organizational field. Pugh (2007) defines organizational fields as a group of organizations which have become institutionalized with key characteristics in common while providing similar products or services. But given the youth of digital platform organizations, no institutional rules and playing fields had yet been defined. It could be argued this is what enabled Uber to define their business model as they did, and then to aggressively roll it out throughout the globe without paying notable attention to the rules of the local contexts they were entering.

But the transportation industries Uber set out to disrupt were and is however highly institutionalized, and undoubtedly part of an inherent organizational field. And as the service itself enabled by Uber in essence is the same as the function of taxis, have caused significant levels of resistance from regulators and the transportation industry. As Pugh (2007) also have stated, the concept of organizational fields is not limited to direct competitors, but rather captures the totality of relevant actors. Seeing the development from that perspective, it is clear one could argue for Uber not being part of an entirely new organizational field - but rather belonging to already institutionalized one as they would fall in the category of a relevant actor. In that case, Uber would be considered belonging to the organizational field of the transportation industry, which exposes them to a set of institutionalized forces difficult to escape from. And as the empirical material will show, they have been unable to fully escape as they have been forced to

(25)

adapt to certain extents. This in itself serves as part of an explanation as to why the boundary of Uber is being challenged.

As above reasoning provide understandings of how the resistance towards Uber can be understood in general, it does not explain how their boundary has been affected in particular. To gain a further understanding of what effect the institutional pressures have had on the boundary of Uber, various contexts have been thoroughly analyzed. As a result, four sets of categorized, boundary-influencing, configurations emerged which sheds light on the nuances in how context matter for the boundary; Adjusting, Expanding, Withdrawing and Intertwining.

4.2 How is the boundary of Uber challenged?

The map provides an overview of the empirical findings of this thesis. When presented with any of these, we want you to do the following: Imagine a white paper with a circle in the middle that represents Uber and its tightly drawn boundary right around the circle. Close outside of this circle there is another, wavy, line that sometimes moves closer to Uber in its ideal form, and sometimes it moves slightly away from it. This is the Adjusting configuration. This specific context represents events where Uber has made shallow adjustments to its boundary in order to remain operative depending on the context at hand. Further outside of these circles there is

(26)

another one, but it is instead being pushed away further from Uber and its ideal form. This is the Expanding configuration and illustrates events where Uber has been compelled to move their boundary beyond its ideal form and isomorph into the specific demands of institutionalized contexts. The Withdrawing configuration adds yet another circle on the paper, but it moves back towards the ideal form in the center. This category entails events where Uber has tried to adapt to a context, but been forced to make a full exit and thereby regress back to the ideal boundary.

Finally, the very last circle is the one furthest out from the ideal boundary, and it is braided together with the boundary of an external actor. This is the Intertwining configuration and represents the rare events when Uber and a competitor together have started a third business to remain in a context.

4.2.1 Adjusting

In the adjusting context, three sets of categories were identified: 1) Adjusting to legislation 2) Adjusting the application, and 3) Adjusting to worker rights. All these have in common that they represent events that stem from institutional pressures that have had direct impact on the organizational boundary of Uber. This section will explain what events made Uber adjust their boundary in order to become compliant with the local context.

(27)

Meyer & Rowan (1977) and Scott (2014) emphasizes that organizations often conform to the institutional context at hand, while they at the same time also can pose as influential actors of change for a context. In Ubers case, it is evident how they initially aimed for the latter, to keep their boundary tight rather than adjusting too much. The former Managing Director described it by the term principled confrontation, which has become almost an ideological belief. Uber is convinced what they are doing is what they believe in, and that it is right and morally justifiable.

In essence Uber claim to have found a way to help citizens travel in a smarter and better way than what outdated regulators can facilitate, and that they won’t stop until their work is 100%

completed (Uber MD, 2019).

The aggressive expansion strategy of principled confrontation enabled Uber to enter markets without giving notice to the institutional context at hand. This, in combination with the usage of Greyball, a software used to circumvent government officials looking to investigate Uber, facilitated Uber to protect and decrease insights from authorities which led to that Uber could maintain their operations undisturbed by avoiding institutional pressure for the time being. By doing so, their boundary remained unaffected to a great extent.

“Uber has for years engaged in a worldwide program to deceive the authorities in markets where its low- cost ride-hailing service was resisted by law enforcement or, in some instances, had been banned. The program, involving a tool called Greyball, uses data collected from the Uber app and other techniques to

identify and circumvent officials who were trying to clamp down on the ride-hailing service. Uber used these methods to evade the authorities in cities like Boston, Paris and Las Vegas, and in countries like

Australia, China and South Korea.”

(Isaac, 2017)

Naturally, this gave rise to a substantial institutional resistance which came to question the legitimacy of Uber in the UK as well as in the US. Uber had simply been operating in parallel to the existing institutional framework by deceiving the authorities and somewhat operating “in the dark”. However, while the principled confrontation technique came to jeopardize the organization's legitimacy, it moreover facilitated Uber's growth since they had already established their operations within the local contexts. It should though further be noted that the usage of Greyball did not directly affect the boundary of Uber, more than its credibility was diluted in the eyes of the regulators. Even though Uber did not strengthen their legitimacy nor

(28)

complied with existing law, Uber succeeded with time, to maintain a large portion of its flexibility connected to the digital platform as they already had been established within the local contexts. This gave Uber legitimacy among the greater community and Uber became institutionalized and “taken for granted” which enabled the organization to adjust its platform without having to compromise its ideal boundary too much. Such an example is evident in the US, UK and India, where Uber's response to the institutional pressure, were uttered by Uber incorporating technical features within the platform, such as an emergency button, in order to meet the institutional requirements on safety measures (Millward, 2015; Isaac, 2017; Transport for London, 2017).

Uber thus slightly adjusted their boundary based on what the platform could technically handle without compromising its core. Meyer & Rowan (1977) argues that organizations are driven to incorporate the practices and procedures defined by rationalized concepts within institutional contexts. Organizations who do so also increase its legitimacy and survival opportunities, regardless of the immediate effectiveness of such concepts. It is therefore evident within the adjusting context how Uber on the one hand influenced change within the institutionalized contexts, while they on the other hand evolved, or conformed to the prevailing concepts of such context and thereby adjusted their boundary accordingly. However, doing so by drawing on the benefits of technology which thus contributed to a greater degree of flexibility regardless of pressure, as long as the platform could carry it. Kirchner & Schüßler (2019) argue that digital platform organizations pose new challenges for institutional organizers as they create a new type of order within the context. The adjustments of the organizational boundary contributes to Uber to some degree had to abandon the idea of being solely a tech company because when these technical features are added, metaphorically, it is clear how the boundary rather moves out into local context, which means that Uber is forced to take a larger overall approach than only considering individual transactions. In addition, it is clear that institutional pressures in India pushed Uber to adapt to employee rights by incorporating UberCare, an insurance intermediary service for drivers (see Kalra, 2014). Even though Uber did not directly refer to it as a matter of employee rights, it is still evident that they further adjusted their boundary beyond its ideal form.

Instead it could be argued this example illustrates how Uber is performing what Meyer & Rowan

(29)

(1977) would call ceremonial activity - actions which will promote acceptance within an institutional field.

It is evident that Uber is faced with events that require them to adjust their organizational boundary in order to remain present in the local context. When these institutional requirements are not too compromising Uber seem to be able to adapt their boundary accordingly. It is moreover possible to see that the organizational boundary changed, from initially reducing transparency and control from authorities, which can be perceived as highly "illegitimate" within institutional contexts, to being forced to adjust by the institutional requirements of conformity in such contexts.

4.2.2 Expanding

In the expanding context two sets of categories were identified 1. Expanding driver requirements (no unlicensed drivers), and 2. Expanding service offerings due to unfair competition. Both derive from institutional pressures in the local contexts that have had a significant impact on the boundary of Uber. This part of the analysis will illustrate how Uber will be forced to become increasingly alike the industry it set out to disrupt by having their boundary isomorph into the ones of the traditional industries.

(30)

Kenney & Zysman (2016) have familiarly stated that digital platform organizations have reinvented and refined ways of performing business in a manner which has made the economy even more efficient. In line with that, Uber has previously been referred to as a commonly known example of an innovative firm which enables the same services as the established transportation industry, but in a much more efficient way by simply providing the platform for sellers and buyers to meet. Ketokivi & Mahoney (2017, 2) similarly refer to such phenomena as discriminating alignment, “the organizational response offering the feasible least-cost solution to govern a transaction”. While this might be true, it is also evident that novelty is not necessarily equivalent with pure success if the innovating organization is competing with an industry that is highly institutionalized. Or as Furusten (2013) and Meyer & Rowan (1977) would put it, it is impossible to disregard the institutional environment. If an organization violates the institutionalized demands put on them, it will be at the cost of their credibility, and ultimately its operations (Scott, 2014; Furusten, 2013). When asked if Uber ideally wants to operate the same way throughout the globe, the Uber MD (2019) answered they do. It was conveyed there are lots of costs that arise from the complexity it brings to operate in specific markets. A lot of politicians think that they influence Uber, but it is rather the other way around. Especially in minor countries that are not even the equivalent of a medium-sized Chinese city, they do not have the power to have Uber change the platform. Instead they will have to change their legislation. But there is a clash there because countries think it is a deeply arrogant attitude to enter with. Especially when your name is Uber, and the logo is black. That has probably been a negative for Uber. Instead the former Uber MD (2019) made a case it would most likely have been beneficial for them to choose something “pink and fluffy”, like Lyft did.

Even though this quote may mirror the reasoning and ambition behind Ubers’ expansion, it is evident from looking at different contexts that Uber had to conform by expanding their boundary in line with what theory suggests. This brings us to the first expanding category – expanding driver requirements (no unlicensed drivers). The very first service Uber developed was UberPOP, allowing any individual with a driver’s license and access to a car not older than a 2005 model to pick up paying customers and drive them to their destination (Casinge, 2015).

However, the fact that these drivers did not attain any formal licenses contributed to an institutional resistance, most distinguished in Europe (see Eddy, 2015; Davies, 2016; Hern, 2017;

Hope, 2019; Transportstyrelsen, 2019).

(31)

In an extraction from the European Court of Justice (2017) the following verdict poses:

“The service provided by Uber connecting individuals with non-professional drivers is covered by services in the field of transport. Member States can therefore regulate the conditions for providing that

service.”

EU thus concluded that Uber should be considered a transportation company but handed over the responsibility to respective member state, to decide the premises for how Uber could, or could not exist. EU thereby limits its own authority and puts the power to decide about the development for digital platform organizations in the hands of the local contexts. With this as a base, it becomes interesting to reflect upon the reason for each ban and its response, and the fact that they have been very similar within this configuration. Given this fact, Uber had to conform and thereby expanded their boundary by developing a service more similar to the incumbent industry, UberX. UberX is the same as UberPOP, but with the meaningful difference that they are required to obtain a registered taxi license in order to operate (Casinge, 2015). In addition, there is also UberBLACK which is the service offering most alike the established transportation industry. This service only entail professional drivers who are fully employed and have standardized worker benefits accordingly (Casinge, 2015). Essentially, it could again be argued Uber has initiated the process of imitating actors within the organizational field to acquire legitimacy. Scott (2014) denote this process as organizational isomorphism. Or more specifically, “a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions” (Pugh, 2017; 439). As Uber seem to be responding to pressures they are put under by other organizations in the same field as well as legislation, it is also an illustrative example of coercive isomorphism as Pugh (2007) would define it. Essentially, the fact that Uber got UberPOP banned in most of these contexts, and instead implemented a service which incorporates drivers with a registered taxi license, it could be argued that Uber is forced to expand its organizational boundary and thus approaching the features institutionalized within the transportation industry.

However, two of the identified contexts offer interesting observations in contrast to the clash between institutionalized contexts and the endeavors of Uber. Even though Finland and Australia both have had their reservations towards Uber, they still have expressed enthusiasm towards the newcomer in a, as far as this study can tell, unique way. In essence, Uber has in these two cases

(32)

contributed in shaping the institutionalized environments to its own advantage even if they have been forced to expand their boundary to fit in with the contexts as well (see Browne, 2018;

Khan, 2018). Furusten (2013) state that when an organization does not approve of the institutional structures at hand, it can choose to try to change the requirements of the institutionalized context. In order to achieve this, the organization needs to work systematically and remain patient (Furusten, 2013). In Finland, Uber was able to re-enter the context after updated legislation came to force which enabled for them to operate legally with UberX and UberBLACK (Ilie, 2018). In Australia, ABC News (2016) reported:

“The New South Wales Government also legalized ride-booking services last year, with Uber drivers being required to have criminal and car-safety checks. As a result of the change, traditional cab drivers

were given a $250 million "industry adjustment package. Taxis still have exclusive access to cab ranks and "hail jobs" — where taxis are hailed on the street rather than booked in advance.”

This illustrates how new entrants in an organizational field also withholds the possibility to shape the institutional context at hand. Australia proves an example were the government will even go to the length of supporting the incumbent industry with supporting funds in order to catch up with the technological development and innovation firms like Uber brings (ABC News, 2016). It should further be noted that Uber have proven throughout the identified contexts its willingness to carry dialogue to improve its possibilities by updated and adapted legislations that will support technological innovation. An Uber spokesperson stated:

“There is opposition because we are coming and shaking up an old, deeply entrenched industry. Some countries are protecting the taxi industry to the detriment of consumers. If politicians would focus on what’s good for consumers, rather than what’s good for the taxi industry or Uber, then we would be in a

different place.”

&

“The big opportunity here is, if taxi laws are deregulated enough to allow competition, you can actually generate full-time jobs so that interested drivers could be self-employed on a full time basis. We have

created 25,000 such jobs in London alone, 10,000 in Paris.”

(Casinge, 2015)

(33)

This part of the analysis have shown that the boundary of Uber is expanding in certain contexts, becoming more alike the attributes of the established transportation industry. Even though it often seem to be the case that the institutional forces relevant to this study demonstrates an unwillingness towards innovation and rationalizing in the same manner as Uber, there are also possibilities for Uber to contribute in shaping the organizational field. But nonetheless, it is clear that Uber regardless if they want to or not, have been forced to expand their boundary from its ideal form and moved in a direction increasingly alike the transportation industry.

4.2.3 Withdrawing

In the Withdrawing context three categories were identified. 1. Withdrawal due to business model protection 2. Withdrawal due to unfair competition, and 3. Withdrawal due to mergers.

All three have in common that they represent events that stem from local institutional pressures and have forced Uber to exit from the respective contexts, and thereby leave the boundary intact.

In institutional theory, it is argued that institutional context will make or break an organization (Meyer & Rowan, 1977; Furusten, 2013). This would be an example of the latter as the observed events have hindered Uber from establishing their organization in the contexts. However, the organizational boundary of Uber could still be regarded as intact in their ideal form as they have

(34)

chosen to pull out entirely from a context, as opposed to making adaptations and further attempts to remain in the context.

In both Denmark and The United States (Texas) Uber has faced difficulties as they have been presented with requirements which they have been unable to conform to (see Dickinson 2018;

Kelly, 2016). These requirements derive from institutional forces in the shape of regulators, and suggest an unwillingness from Uber to expand its boundary at any cost. In Denmark, Uber was deemed a transportation organization and thereby they were automatically forced to obey under the same requirements as the established transportation organizations. The central issue came down to specific requirements in the shapes of mandatory fare meters and occupancy sensors.

Uber deemed these requirements would compromise the business model too significantly and therefore opted out of the context (Dickinson, 2018). In Texas, regulators along with state citizens required third party fingerprint screenings of Uber drivers, a requirement Uber could not conform to as it would slow down its flexibility (Kelly, 2016). Accordingly, Furusten (2013) argue that when an organization does not approve of the institutional forces at hand, one of the options is to surrender to the rules and structures anyway. Considering the fact that Uber left, that also represents a form of surrendering and examples of their boundary remaining close to the ideal form.

A particularly clear pattern that emerged when analyzing the Withdrawing configuration concern unfair competition. This is interesting from an analytical point of view considering how Uber could be explained from a TCE perspective. Ketokivi & Mahoney (2017) explain that TCE seeks to identify comparatively superior organizational arrangements in terms of economic efficiency.

As stated, this thesis argue this is the reasoning behind the business model of Uber. If it is accepted Uber has found a way to disruptively innovate the transportation industry in a way which is far more efficient than what the traditional industry can offer, it appears that very fact is also what is having Uber banned. Keeping that in mind, superior efficiency is not equal to success. In multiple contexts (Hungary, Turkey, Bulgaria and Portugal), it is evident regulators have outlawed Uber, either by suspending their operations as a result of opposition from drivers from the incumbent transportation industry - or simply because Uber is perceived a threat against

The Uber Boundary: Contextualizing the Organizational Boundary of a Digital Platform Organization