Adoption of blockchain in the legal sector : The influence of a firm’s technological, organisational and environmental context on the adoption of blockchain technology

(1)

MSc Business Administration

Track: Digital Business

Master Thesis

Adoption of Blockchain in the Legal Sector

The influence of a firm’s technological, organisational and environmental context on the

adoption of blockchain technology

by

Max Marc Oosenbrug 10173013

22 June 2018

15 ECTS

Research conducted from January 2018 to June 2018

Supervisor/Examiner:

(2)

Statement of originality

This document is written by Max Oosenbrug, student of the Amsterdam Business School, University of Amsterdam (UvA), 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

(3)

Abstract

For the last ten years, blockchain has become subject of discussion across many different fields. Though lots have been written about the technology and its consequences, not every sector is affected as predicted. A qualitative research has been conducted on the adoption and applicability of Blockchain in the legal sector. Factors from the Technological, Organisational and Environmental (TOE) context have been deducted from the TOE framework, on which a semi-structured interview is constructed. A multiple case study was conducted across 5 different law firms. Results show that where the factors derived from literature analysis had either a negative or neutral influence on blockchain adoption at this point in time. Three new factors have been derived from data analysis. Type of work, hype

and image and employee input are factors not yet included in the TOE framework, but were

considered influential. Hype and image are considered of positive influence on adoption. For this sector, type of work is either of positive or negative influence, dependent on the sort of work. Employee input was considered being of negative influence. These factors are added to the proposed research model and may be subject of future research. Two additional factors outside the TOE contexts were perceived as relevant. A service oriented industry, like the legal sector, distinct two types of adoption. Either internal adoption of the technology, or adopt the technology as an advisory product. Finally, the findings of this study may be extended. Apart from the potential the technology may or may not have, blockchain may be seen as a catalyst for process optimisation. Many processes which were formerly obvious and taken for granted, are now being questioned and reformed.

(4)

1 | Introduction

1.1 | Research Problem

Ever since the cryptocurrency Bitcoin was introduced by Nakamoto (2008), the topic has been subject to discussion. Cryptocurrencies like Bitcoin or Ethereum have been described as an unfounded fraud, of which the value is determined by what people are willing to pay for it, and is often compared to the ‘dot-com bubble’ in 1999, or the tulip mania in 1637 (Garber, 1989; Goodnight & Green, 2010). The reputation of Bitcoin, however, may have damaged the reputation of the underlying technique making Bitcoin possible, namely blockchain (Holotiuk, Pisani, & Moormann, 2017), which lately gained popularity across different fields of business. Because Bitcoin and blockchain are two concepts which are highly associated, and are sometimes used simultaneously, blockchain is often considered a hype, derived from the hype surrounding Bitcoin (Bott & Milkau, 2016). Because Bitcoin is merely based on blockchain technology, these two concepts should be approached separately.

Academic research on blockchain started to grow exponentially from 2014 onwards (Google Trend Search, 2018). The first applications are already visible on the market and different sectors are investigating whether and how it may be relevant for their field of business (Accenture, 2016; Deloitte, 2017; IBM, 2015).

Blockchain is most simply and commonly defined as “a shared, immutable ledger that facilitates the process of recording transactions and tracking assets within a network” (Zheng, 2016, p. 6-7) and has the potential to redefine and simplify many devious processes (Grech & Camilleri, 2017). However, several other definitions have been stated as well (Deshpande et al., 2017). The fact that there is not a single definition of a blockchain shows that there is lot of ambiguity about the technology and whether something may be considered a blockchain, when it does not meet all the technological pillars it is based on, but is called a blockchain (Puthal, Malik, Mohanty, Kougianos, & Yang, 2018). This confirms that while many publications have

(7)

been made about the technology and its potential, it is still unclear how the technology may be relevant for several industries. Because there is not a universal definition and not a single appliance of the technology (Deshpande, Steward, Lepetit & Gunashekar, 2017), the next sector gives background of the technology, what it may cause and which sector this research will focus on.

Many statements have been made that blockchain is showing characteristics of the technologies following Gartner’s Hype Cycle for Emerging Technologies (Linden & Fenn, 2003). Numerous technologies from the past two decades have shown the same behaviour as blockchain does nowadays and follows a certain pattern of expectation over time (figure 1).

Figure 1. Position of blockchain (circled) related to other technologies on Gartner's Hype Cycle (Linden & Fenn, 2003, p. 5)

retrieved from: Gartner, 2017, webpage visited march 2018

Most databases are centrally organised. Within these databases, there are one or few locations where all the data is stored. Only at these locations data is modified and shared with

(8)

users. In contrast to a traditional database, blockchain data is not stored at a single location, but data is stored across multiple, also called nodes. At every single node, an exact copy of the databases’ ledger is stored, on which the information about transactions is visible (Kostakis & Giotitsas, 2014). These databases are relevant for real-time data, while the history of the data is stored as well. The way a blockchain is structured, allows the participants to share the ledger that is always up-to-date, through peer-to-peer replication. Whenever a transaction occurs, the ledger is updated at every location. So, every node receives or sends the transaction to or from the other nodes (McReynolds, Lerner, Scott, Roesner, & Kohno, 2015). With blockchain technology, pieces of information are being digitally signed by the different involved parties, without a ‘thrusted third party’ necessarily being involved. Information is then directly saved to the database, it is stored on a secure and unalterable and transparent ledger. In some cases, a transaction may contain a digital currency, in other cases it concerns the exchange of official documents such as contracts, diploma’s or proofs of ownership. These appliances already show that blockchain offers many possibilities for different businesses and industries.

Two different types of blockchain can be distinguished. Public on the one hand, and private on the other. A public blockchain does not require approval to use or join, so anyone may use the network. Within this type, anyone is allowed to see or submit a transaction. To join and use a private blockchain, permission must be granted by a central authority, or must be listed and the individual thus has to be known (Drescher, 2017).

With traditional methods of recording transactions or tracking assets, participants within that network all possess their own ledgers and other records. This system is susceptible to fraud, cyberattacks or human mistakes. Moreover, these methods may be expensive due to the necessary involvement of intermediaries (Reijers & Coeckelbergh, 2016). Whenever buying a house, register a car or apply for a loan, lots of paperwork is involved to ensure the transaction. Blockchain technology may help to transform inefficient processes and make them more

(9)

fraction less. A variety of processes might become unnecessary when people and companies start to organise trough decentralized platforms. Public administration, paperwork and bureaucratic processes may be simplified through permissioned blockchains. In many situations, ranging from agriculture and medical appliances to financial or insurance services, the introduction of blockchain may be reorganising (Nofer, Gomber, Hinz, & Schiereck, 2017). Practical cases in which processes are reformed due to this technology become available. Different sectors are trying to understand what the technology can mean for their area of business (Friedlmaier, Tumasjan & Welpe, 2016).

As well as in the legal sector. Blockchain created a system for digital integrity and trust for data distribution, which can be applied to an industry which is centred around integrity and trust. A big part of legal operations is to consolidate integrity in documents, contracts and signatures. Various possible implications of blockchain have been suggested and arising in this sector. For example: so-called ‘Smart Contracts’ are promising to change the current legal arrangements (Swan, 2015). Where they could be coded as software and being distributed over a blockchain network, leaving the decision making up to the users instead of a select group of executives who make the decisions. This could result in bypassing inefficient and hierarchical processes. Another application is the ‘transfer of ownership’, where people are no longer depending on a ‘trusted’ third party, but where transfers are authenticated via the blockchain (García-Bañuelos, Ponomarev, Dumas, & Weber, 2016; Hamida et al., 2017). Using blockchain, the need for a centralized authority is eliminated. Proof of ownership, proof of existence and proof of integrity may be certified without a third party, like a notary, being involved (Crosby, Pattanayak, Verma, & Kalyanaraman, 2016) by decreasing workload of auditors and regulators. The big advantage and challenge of a blockchain is the interaction between users. It allows the distribution of information between parties which depend on a reliable relationship with each other, to share their data. The transactions are processed within a network of users, to

(10)

form a structure which unanimity is guaranteed. Within the network, everyone is constructing the same shared records. Within a blockchain, control is not centralized around one person, database or authority, but is shared amongst the participants in the chain. The user is only able to edit the part of the blockchain they ‘own’, which can be accessed with a private key. Blockchain makes it easier to review transaction details in cases where third-party supervision is required (Morabito, 2017). However, due to the necessity of different parties to be involved, for many blockchain solutions a form of cooperation is necessary, in which two or more parties which normally compete with each other, have to cooperate in other situations, in this case for creating a blockchain product. This process is described coopetition (Bengtsson & Kock, 2000).

Despite all the above mentioned possible applications mentioned above, ambiguity for this sector remains. Over fifty percent of companies in the legal sectors perceive the new technology as important, and feel the necessity to adopt the technology in order to gain competitive advantage or maintain current market share (Gratzke, Schatsky, & Piscini, 2017). Research conducted by PwC (2017) shows that 41% of Dutch law firms are planning to use blockchain applications for legal transactional services, 31% for high-value legal services and 21% for business support. At the same time, it is reported to be unknown whether and how this change will affect the business and are unsure how to respond to this trend.

Current research predominantly focusses on the subject of ‘what is blockchain’, ‘what are the potential uses’ and the topic of ‘possible encountered technological problems’ (Yli-Huumo et al., 2016). There is a lack of knowledge however, in which area of business this emerging technology will be relevant, on what scale the technology will have impact, and whether and how it should be adopted (Furlonger & Kandaswarmy, 2018). This research mainly focusses on the latter, the adoption of blockchain in the legal sector.

(11)

1.2 | Research Objective

While some utilizations of the technology are already suggested and visible, especially in the financial market, the exact appliances are yet to be discovered. Thus far, it is uncertain in which area/field it might be useful and on what scale the impact will take place. Whether or not the technology will reach the point where it will be adopted on a large scale depends on numerous factors, such as regulation & competing technologies and research shows that often the short-term impact of blockchain is overestimated (Mattila, 2016). To determine whether a company is ready for the adoption, new enterprise readiness research is needed (Furlonger & Kandaswarmy, 2018). Therefore, because of the great current interest in this topic, and the potential threats and opportunities it poses towards different sectors, a company would be wise to gather information about the opportunities and risks for their sector, and explore the option of adopting the technology within the company (Bhatt & Grover, 2005). Henceforth, the main objective that will be researched is:

To identify how a firm’s context factors influence the adoption of blockchain technology in the legal sector.

Four sub objectives relevant for fulfilling the main objective are formulated:

1. To determine in what stage of adoption the sector is currently in 2. To determine the legal firms’ role in blockchain innovation 3. To define which factors are specific for the legal sector

(12)

1.3 | Research Method & Implications

This research aims at contributing to current research by providing an analysis of contextual factors that influence the adoption of blockchain within the legal sector on organisational level. A qualitative research (explanatory, embedded, multiple case study) will be conducted among different Dutch legal firms. This type of case study is the favoured research strategy when: “how or why questions are being posed, when the investigator has little control over events, and when the focus is on a contemporary phenomenon within some real-life context” (Yin, 2006. p. 1). In this case, finding how context-factors from a firm in the legal industry are of influence on the adoption of blockchain.

To fulfil the research objective, a literature research will be conducted to determine the theoretical framework. Different technology adoption models will be compared and weighed against each other to get more insights in the research context. Furthermore, the measurement scales will be determined in order to operationalise the theoretical concepts. From these factors, the appropriate research design for the case study will be derived (Saunders, Lewis, & Thornhill, 2008). When the framework is defined, a semi-structured interview will be designed. The literature research will be tested against the conducted empirical research, which will lead to conclusions and suggestions for theory, practice and future research.

Proposed managerial implications: Because blockchain is still a novel technology, companies are still trying to find ways how to implement this new technology within their company. This research identifies how firm’s context-factors influence the adoption of a new technology within the legal sector. A framework will be presented and tested, which assesses the technology adoption factors for firms within the legal sector. It may contribute to practical viewpoints concerning the potential acceptance or denial of blockchain technology and provide insights on the current contextual state of the legal industry on this matter.

(13)

technology adoption and acceptance for novel technologies. An already validated model will be used with a new technology in a new, not yet researched, context. Because this sector has not been researched before in the context of blockchain, the used adoption model may be revalidated with a contemporary technology. Furthermore, suggestions and additions for future research on this matter are given.

1.4 | Thesis Structure

In the next chapters the subjects discussed above will be reviewed. At first, the main concepts and models, and literature will be discussed, along with background information on the topics introduced above. Subsequently, the research design and methodology will be described to provide information on how the research has been conducted, followed by data collection and analysis. Hereafter, the results and conclusions will be presented. These includes an outcome of performed analysis and an explanation of the research objectives and main findings. Finally, the discussion will be presented. Both the theoretical and managerial contributions will be drafted and suggestions for future research are presented.

(14)

2 | Literature Review

This chapter provides an overview of literature on the life cycle of emerging technologies and blockchain as new technology, followed by technology adoption theories. Finally, the most appropriate model will be chosen as a basis towards the conceptual framework.

2.1 | Blockchain as a really new innovation

As mentioned before there is a big potential in using blockchain technology. Apart from the general potential benefits originating from the adoption of innovations, due to the possible disruptive characteristics of these services, blockchain technologies can be considered really new innovations, that come with risks of failure for a company (Chao, Reid, & Mavondo, 2013). Many researches make a distinction between radical innovations on the one side and incremental product innovations on the other. These categories are based on the newness of the innovation to the market. Products, services or techniques of an incremental nature depend on less organisational change and thus make it easier to adopt compared to radical innovations (Reid & De Brentani, 2004). Radical innovations rely on a higher degree of technological adaptation and are paired with more uncertainty for a company (Verleye & De Marez, 2005). According to research by Garcia and Calantone, (2002) the lion’s share of innovations are of an incremental nature. When we speak of ‘really new products’, we consider an innovation we cannot classify it under either incremental or radical, but of innovations that “include either a market discontinuity or a technological discontinuity to both customer and companies” (Garcia & Calantone, 2002, p. 127).

A really new innovation has several characteristics: it focusses on a technology which has never been used in the industry before, it causes an impact in a whole industry and it causes significant adjustments in the whole sector and is unique in a new market (Song & Montoya-weiss, 1998). The definitions of blockchain correspond to this classification, where services

(15)

within a market are potentially discontinued, and new technological systems and ways of working affect both companies and customers.

Blockchain technology systems can thus be seen as a really new innovation. However, at this point in time, it is still going through the technology development cycle, and is not yet ready to be adopted on a large scale (Wang, Chen, & Xu, 2016), but is adopted by first by innovators and early adopters (Rogers, 2010). Research by PwC (2017) on blockchain adoption in the legal sector shows that less than 1% of Dutch legal firms have implemented a blockchain solution (figure 2).

Figure 2. Emerging technologies within the legal industry (PwC, 2017, p. 21)

2.2 | Emerging technologies

A new technology may be relevant for a sector when it can offer several advantages over current systems, or not yet implemented systems. These advantages can be acquired through the adoption of technologies. Examples are gaining competitive advantage within an industry, reduce costs or streamlining already existing processes (Bhatt & Grover, 2005). Whenever a new technology may offer such an advantage, it is necessary to identify these opportunities and determine whether or not to adopt the new technology. For the current stage of blockchain, it is

(16)

uncertain to tell whether and on which scale the technology may be implemented or whether it will offer competitive advantage, reduce costs or is able to streamline processes (Meijer, 2017).

Because the subject is said to be thriving in many different sectors, large companies cannot ignore it, because of the risk non-adoption brings (Frambach & Schillewaert, 2002). When looking back on the emergence of the world wide web, back in the nineties, companies now consider blockchain the biggest digital innovation since the arrival of internet (Mougayar, 2016). How and whether to adopt the technology in order to gain competitive advantage is valuable knowledge for firms. New emerging technology often follow a process described as follows: “a technology will inevitably progress through the pattern of overenthusiasm and disillusionment” (Linden & Fenn, 2003, p. 5). This process concludes that companies should keep the emerging technologies in mind, but should not invest in the technology just because of a hype.

Because of a company’s environmental context, like competition, external knowledge and regulations cannot be ignored. Within industries, prominent parties affect other parties in the decision to innovate. As consequence of the hype, more and more firms are showing interest in blockchain technology (Kamath & Liker, 1994).

2.3 | New technologies in lagging industries

The rate of the adoption of an innovation follows an S-Shaped curve (Sahin, 2006). After origin of the new technology the adoption starts slowly, few people or companies adopt the technology, called the early adaptors. When there is more awareness around the technology, in the subsequent period, the technology gets adopted in a higher pace. In the final period the pace slows down again until the adoption is saturated (figure 3).

(17)

Figure 3. The Innovation Diffusion Curve (Rogers, 2010, p. 38)

The Diffusion model uncovers the general principles of a diffusion process (Rogers, 2010). Five different categories of the adoption process are differentiated: Innovators, Early Adopters, Early Majority, Late Majority & Laggards. Rogers (2010) describes laggards to be cautious & isolated when it comes to the adoption of innovations. This should not necessarily be seen as negative. The described characteristics are considered traditional and suspicious to change.

Legal Enterprises are often considered laggards when it comes to technology adoption, both internally as externally (Poje, 2013). Though the legal sector has shown a quick rate of technological advancements last decade, it is still a sector which is considered to be conservative and a laggard when it comes to technology adoption. Research by the American Bar Association (2013) on this matter shows that less than 60% of the firms had a budget for technological upscaling and less than 40% of lawyers considered IT training important. Over 40% still uses printed versions for services that can be accessed more easily digitally. The reason for this comes forth from the advising role firms often fulfil. Only if other companies are

(18)

in need of advice, firms come into the picture, and thus have a following instead of a foregoing role.

When a technology is perceived as new, or if a technology is paired with uncertainty, innovative decision making is characterized. As soon the ‘prior conditions’ of the innovation decision process are satisfied, the model (Rogers, 1995) describes the five steps from first acquaintance with a technology, towards adoption and confirmation of the adoption decision (figure 4). These prior conditions are satisfied when a decision-making unit perceives the current situation as potentially problematic, has certain needs or wants to comply with social norms (Miranda, Farias, Schwartz, Almeida, 2016).

Figure 4. Five steps in the adoption decision process (Rogers, 1995, p. 165)

The five stages in the model each have their own characteristics. In the knowledge phase, the party is first exposed to the technologies existence and gains some conceptual comprehension about the technology, but has not enough information to make a proper adoption decision. In the persuasion stage, an attitude towards the technology is formed, the user is open to the idea of innovation and is actively looking for information to support his or her decision. During the decisions phase, the party is performing actions which lead to adoption or rejection. Advantages and disadvantages are being weighted. During the implementation phase the technology is put into use and information is looked up to give confirms the adoption decisions.

(19)

During the confirmation stage, the decision of innovation is evaluated and the final decisions towards abandoning or keeping the product is made (Rogers, 1995).

2.4 | Technology Adoption Models

As discussed, blockchain has many potential implications. Successful implication of a technology can lead to better productivity, competitive advantage and maintaining the position of incumbents, where adoption failure may lead to a negative cycle of non-adoption (Straub, 2009). New technologies can only be proven to be effective when they are actually used. Technologies which can potentially increase the performance of an individual or a company may sometimes be underutilized (Mathieson, 1991). Without exactly knowing what a technology could mean for the individual or the company, many companies show great interest towards these technologies (Meijer, 2017).

When planning to adopt a new technology, it is important to predict whether the new applications/services/systems will be acceptable to the users and are compatible with current systems. In order to understand whether a technology will be adopted within an organisation, the company has to understand which factors affect the adoption process. To predict this, certain models are typically used; the Technology Acceptance Model (TAM), the Unified Theory of acceptance and usage of technology (UTAUT), the Diffusion of innovations (DoI) and Technology-Organisation-Environment (TOE) Framework are the most common (Straub, 2009; Hashim, Hassan & Hashim, 2015).

The TAM was introduced by Bagozzi, Davis and Warshaw (1989). Within this model two constructs are central in predicting the acceptance of a new technology on an individual level. The perceived usefulness and perceived ease of use. Both have influence on the intention of accepting the new technology (Davis et al., 1989; Wu, Li, & Lin, 2010). As an extension to

(20)

the TAM, the UTAUT, was proposed by Venkatesh (2003) to aggregate all difference acceptance models into one unified model.

To describe how innovations spread in a group, the DoI was introduced by Rogers (1995). The theory is explaining the several stages from first experience with a product or technology to final adoption. Diffusion represents a group appearance, describing how the innovation spreads (Sahin, 2006).

Finally the TOE framework was introduced by Tornatzky, Eveland and Fleischer (1990). The TOE framework categorizes the three different components of a company’s context, the Technological, Organisational and the environmental context. These three contexts determine the technological adoption decision on firm level.

Where the TAM and the UTAUT describe technology adoption or acceptance from an individual level, the DoI and the TOE describe technology adoption from the organisational perspective (Oliveira & Martins, 2011). Since this research focusses on the contextual factors of firms within the industry, the TOE framework and DoI are considered most appropriate. The first to study the factors influencing adoption, the second to determine in which stage the adoption process is currently in.

2.5 | Technology-Organisation-Environment (TOE) Framework

Tornatzky, Eveland and Fleischer (1990) explain the development of an innovation to the adoption of end-users. The TOE framework takes one part of this process explaining the firm’s context and how it influences the adoption of a new technology. It consists out of three blocks, the technological context, organisational context and environmental context. These contexts consist of both the formulation of opportunities and restrictions (Borgman, Bahli, Heier, & Schewski, 2013). The technological context includes the relevant applications of the technology, both internally as externally. It determines whether the technological landscape of

(21)

the firm is perceived as suitable enough to adopt the technology. The organisational context refers to the characteristics and resources of the firm. The environmental context describes the conditions outside the firm in which a firm conducts its business. To measure factors which influence the adoption of new technologies, the TOE framework has proved itself to be effective in different fields of business (Bosch-Rekveldt, Jongkind, Mooi, Bakker, & Verbraeck, 2011). Figure 5 illustrates the fundamental model with its primary components. These components will be put apart below and relevant factors assigned to each context will be explained.

Figure 5. The Technology-Organisation-Environmental Framework (Tornatzky et al., 1990, p. 77-99)

Technological context

The technological context factors concerning the new technology are explained in the model as the perceived innovation characteristics (Tornatzky et al., 1990). Decisions on the adoption are determined by what technologies already exists, and to what extent the technology will fit the

(22)

current technological context. Due to the novelty of blockchain, the factors influencing the adoption are not yet determined. Therefore, a wide variety of factors have to be examined. For the technological context different perceived technological factors are considered relevant when explaining the adoption decision (Rogers, 1995). In context of the TOE framework several perceived technological factors have been proven to be of influence in either a positive or negative way: maturity, characteristics, relative advantage, compatibility, complexity and observability.

Maturity describes to what extent the technology is perceived as being ready for adoption, the more mature a technology is, the sooner it will be adopted (Smith, 2005). Technological characteristics specify the requirements a technology has to meet in order for adoption to be considered (Chau & Tam, 1999). Relative advantage is defined as “the degree to which an innovation is perceived better than the idea it supersedes” (Rogers, 1995, p. 213). It is often explained in terms of monetary profitability or obtained status and of positive influence on adoption. Compatibility describes in what way the technology coheres with “existing values, past experiences and needs of the adopter” (Rogers, 1995, p. 223). It is determined by its compatibility with social norms and former suggested ideas. Complexity explains the extent to which the technology is hard to use and understand, it has a negative relationship with the adoption (Premkumar & Roberts, 1999). Observability is defined as the degree the effects of a technology are already visible by others. Lastly, availability is defined to what extent the technology may be already tested and used. The more visible a technology is, the more positive it will influence the adoption. (Rogers, 2010).

Organisational context

The organisational context is explained as the firm’s properties and assets. Decisions on adoption are influenced by factors from this context. This includes within-company

(23)

communication style, company size, amount of resources available and organisational structure (Baker, 2011). To test the organisational context within a company the following factors have been determined for the TOE framework: firm size, top-management support, IT-skills/knowledge of non-IT employees.

Firm size is predominantly positively associated with adoption, because large organisations often possess more resources for new technologies and are more capable of bear mistakes. On the other hand, smaller companies may be more flexible to incorporate new systems within their organisations (Rogers, 2010). Top management support may contribute to adoption by allocating resources and by creating a supporting environment in which the company is operating (Jeyaraj et al., 2006). IT-skills of non-IT employees are important where non-IT employees are often responsible for making the adoption decisions. More IT knowledge under employees is positively associated with adoption (Levenburg et al., 2006). Formal/informal Linking structures and hierarchy explain how the company is divided in terms of departments, employee influence channels, links to other departments and other mechanisms (Baker, 2011).

Environmental context

The environmental context consists of external factors influencing the adoption decision within a firm. Incentives for adoption may be creating competitive advantages offered by a technology or creating a strategic connection between the business and IT. It may refer to the external businesses’ industry, competitors, access to external resources supplied by others, and legislative (governmental) aspects (Tornatzky et al., 1990). From TOE related literature, the following factors are subtracted: competition, regulatory environment, industry characteristics and market structure, and customers.

(24)

Competition is the degree to which the adoption is influenced by the perceived competitors pressure. The pressure and the adoption stage of the competitive environment is of positive influence on adoption of new technologies (Gatignon & Robertson, 1989). Regulatory environment or government regulation can work both ways, legislation may be a constraint for adoption of new innovation, but may also be beneficial by the introduction of for example, tax advantages (Grembergen & de Haes, 2008). Industry characteristics and market structure is defined as the degree to which the industry is usually performing when adopting new technologies, and to what extent the presence of skilled labour or suppliers of technology are accessible. Service industry, like the legal sector, often rely on IT for the processing of information. Specific industry characteristics may be of influence on the adoption (Raymond, 1994). Finally, a customer question within a service industry is considered leading for organisational change. Because the service industry is mainly working commissioned by its customers, customer influence is positively associated with the adoption on new technologies (Oliveira & Martins, 2011).

(25)

3 | Research Design

3.1 | Conceptual model

This research provides factors influencing the adoption of blockchain technology in the legal sector. A technology which is perceived as potentially useful, but uncertain in which area/field, within a sector in which the adoption of technology is rather slow, and lagging behind. The theoretical goal is to give an addition to the TOE model by testing it in an industry, with a relevant, current technology which requires exploration. Relevant factors from the TOE contexts are derived from literature, to determine whether and how they impact the adoption of blockchain technology within the legal sector (figure 6).

(26)

3.2 | Methodological choice

A literature review has been conducted towards blockchain and technology adoption models, whereafter the conceptual framework has been determined, the TOE framework has been widely accepted, but little is known about its use in context of blockchain and the legal sector (Bosch-Rekveldt, Jongkind, Mooi, Bakker, & Verbraeck, 2011). Therefore, this industry is selected to revalidate the model with a relevant new technology, in a scarcely researched industry.

To fulfil the research objective, desk research in combination with a case study is selected. This research acquires understanding of a social contemporary phenomenon in a real-life context. ‘How’ and ‘why’ questions are being asked (Yin, 2006) in order to fulfil the research objective. Qualitative research will be conducted to answer how and why the contexts influence the adoption of blockchain in the legal sector. In order to be able to replicate findings across cases, a multiple case study is chosen (Baxter & Jack, 2008), in which different law firms are selected.

Research can be either exploratory, descriptive or explanatory (Saunders et al., 2008). This research is of an explanatory nature, most appropriate when an answer is sought to a presumed, predetermined link (Yin, 2006). The causal relationships between the factors from the TOE contexts and blockchain adoption are studied. The research will be cross-sectional, in which every case will be interviewed at one point in time within a short time span, and will be compared to each other afterwards (Saunders et al., 2008).

A combination of deductive and inductive research design has been chosen. The deductive research approach is described as: “to adopt clear theoretical position that you will test through the data collection”, the inductive research approach as: “to explore a topic and develop a theoretical explanation as the data are collected and analysed” (Saunders et al., 2008, p. 48). The conceptual framework has been deducted from the TOE framework

(27)

(Tornatzky & Fleischer, 1990), on which the interview is based. Because research on blockchain adoption in the legal sector is not yet conducted, an inductive approach is selected to explore the topic of blockchain adoption in the legal sector, to complement or revise the TOE framework.

3.3 | Sample & Sample size

To find a representative group of participants, able to provide the solution to the research objective, a combination of quota sampling and self-selection sampling method has been chosen. A predetermined selection of participants (approximately 30 contacts) was approached by colleagues as a means of purposive sampling, which they presumed to be representative for the whole population. After having listed the replies of who were willing to participate (self-selection), the final participant pool was determined, whereafter 10 appointments were made. Finally, a personalized e-mail was sent as briefing and appointment confirmation (Appendix III). As means of triangulation, from each case, different hierarchy-levels (associate to partner) were selected from different branches (notary-office, lawyers & IT-professionals).

To fulfil the objective, ten case studies were performed across five different Dutch legal firms from different sizes with different levels of employees. A description of participants is displayed in table 1. Different types of law firms may be distinguished varying from big international firms, big national firms, medium sized national firms and small/niche firms. As a form of triangulation, each category was interviewed at least once. Another form of triangulation was achieved through the different levels of knowledge on blockchain interviewees had. Levels varied between minimal (n=3), conceptual (n=3) and expert (n=4).

(28)

Table 1

Participant information

# Gender Company characteristics Employees function Date

1 F (1979) Medium sized law-firm Staff (data analyst) 17/04 2 M (1991) Medium sized law-firm Lawyer (Junior associate) 18/04 3 M (1985) Medium sized law-firm Lawyer (Senior associate) 18/04 4 M (1984) Small sized law-firm Lawyer (Senior associate) 19/04 5 M (1986) Big sized law-firm Lawyer (Candidate Partner) 23/04

6 M (1969) Medium sized law-firm Lawyer (Partner) 24/04

7 M (1981) Small sized legal tech-firm Lawyer (CEO) 04/05 8 M (1976) Medium sized law-firm Candidate Notary (Senior associate) 04/05 9 M (1984) Small sized legal tech firm Lawyer (CEO) 07/05 10 F (1977) Big sized law-firm Lawyer (Partner) & Chairwoman of

the Dutch legal blockchain coalition

04/06

3.4 | Interview construction

To construct the framework, used for the interview, guidelines by Brinkman (2009) were used. Following this method, a scheme of themes and key questions could be constructed, which were underlying the subjects to be covered. This made it possible to assign the questions under the right context in a structured way, so that a clear interview could be constructed. As the objective was to find how the technological, organisational and environmental context factors influence the adoption process, the questions were based on indicators underlying the dimensions derived from literature analysis (Appendix I). Because this interview researches a complex concept, first the construct was defined, whereafter the construct was divided in contexts, contexts in factors and factors into measurable items. The contexts that are assumed to influence the adoption process, as well as the factors loading on these contexts, are derived from literature. By following these steps, the final interview-framework was constructed (Appendix II).

(29)

4 | Data collection

4.1 | Data collection instruments

Two ways of primary data collection will be carried out as a form of triangulation, semi-structured interviews and document analysis, resulting in a multi-method, qualitative study. Documents will be consulted through company websites, mailings and personal sources. Organisational documents containing useful data will be organised through content analysis. Major themes, categories and examples will be derived (Bowen, 2009), to analyse as objectively as possible. Each interview is recorded, transcribed, coded and analysed. During the interviews notes were taken as well, to emphasize or summarize certain statements made by participants (Devers, Richard & Frankel, 2000).

4.2 | Interview conduction

After having constructed the first interview set-up, a pilot test was conducted among two legal professionals to develop a TOE based interview, relevant for the legal sector. Because no TOE related research was conducted in the legal industry, pilot testing controlled whether the variables deducted from literature were also relevant for the legal firms. Hereafter, two pre-tests were conducted to test the structure, time and quality (i.e. suggestive/clearly formulated) of the interview (Yin, 2006). Based on these tests, several changes were made to the interview protocol to make it more relevant for the legal sector. For example, the factor room-to-innovate was added, meaning to what extent the employee had freedom and resources to innovate and describes in what way employees feel like they are given the opportunity to be involved in innovative projects, e.g.: “To what extent do you get room-to-innovate next to your daily

activities?”. Industry characteristic questions were added or altered. For example: “In what way

(30)

Two days in advance of every interview, a briefing was send by e-mail (Appendix III). The interviews took approximately 45 minutes and followed the determined semi-structured set-up. Throughout the interview process, questions that are considered more important by the participants, received more attention in later interviews. Though questions that are less relevant are rechecked and examined why they were less relevant, or how to better formulate them. For example, the question: “do you see any consequences of blockchain within the legal sector?” was not asked anymore after it became clear that there were no impactful blockchain products available yet, within the legal sector.

During the interview process, several questions were added, altered or deleted, for example: “To what extent is the legal sector front- or back runner in innovation?” was perceived to be formulated incorrectly, because law never innovates and always has to follow innovation (Kaal & Farris, 2017), moreover, the legal sector is seen as a service industry, in which innovation is traditionally not a part.

(31)

5 | Data Analysis

5.1 | Analysis process

According to Miles and Huberman (1994), data should be “Valid, mutually exclusive

and exhaustive” (p. 155). To realise this, a combination of open-, axial- and selective coding

was performed (Walker & Myrick, 2006). The transcribed text was uploaded in QRS Nvivo 12 to make the analysis process easier and more structured. The analysis of the transcribed data followed three different phases: data reduction, data display and conclusions drawing (Miles & Huberman, 1994).

The data was reduced by organising the relevant parts under both predetermined codes as codes deducted from the data set. When a statement was relevant for the research objective, but could not be listed under a predetermined factor, a new code/node was created. Irrelevant data was discarded, but still accessible through the master data set.

To be able to draw conclusions, data was systematically displayed in a cross-case table, in which the participants’ statements were summarized and displayed orderly (Appendix IV).

To draw conclusions the regularities, patterns and explanations were derived from the cross-case table whereafter preliminary conclusions could be drawn. To interpret these textual statements, for analysis, the textual statements were replaced by a value symbol (e.g. + or -). To give a direction of the statement (Table 2, results)

Since the questionnaire alongside its overlying factors was deducted from the TOE framework, the main factors influencing adoption were predetermined. During the analysis, several factors were added, which could not be placed among the predetermined factors. After analysis, the codes were revised and hierarchical levels in nodes were determined. During the process, several overlapping answers were given to questions categorized under different factors. For example, answers that were given to maturity and observable effects, could be categorized under both maturity as under availability. Therefore, these categories were merged.

(32)

5.2 | Reliability & Validity

Reliability

To ensure reliability of the test the different types of bias, such as the researcher bias needs to be minimalized. Therefore, consistency in measurement and analysis is a requirement (Saunders et al., 2008). Interviews were conducted in a quiet one-on-one setting to set the same stage for everyone. Researcher-bias is minimalized by following the predetermined interview structure (Appendix II).

Construct validity

Riege (2003) explains that a chain of evidence should be created in order to compare theoretical data to empirical data. Construct validity will be controlled through literature research, pilot testing and multiple data sources. Whether the results determine the TOE contexts and whether the factors load on these contexts, depends on a clear theoretical demarcation between the constructs, factors and relationships (Yin, 2006). The three contexts were systematically put apart through a method described by Brinkman (2009), so that the factors could be translated into measurable items.

Internal validity

To control for internal validity in qualitative research, the conclusions should be representative for the concerning research group as a whole. This research looks for causal relationships, in which internal validity may be a concern (Yin, 2006), since external factors may be of influence which were accounted for. Every participant was systematically asked whether other factors could be of influence as well, to maximize internal validity. Other factors which could have been of influence were the locations and timespan in which the interviews were conducted. The interviews were conducted over a period of three months.

(33)

External validity

External validity shows to what extent conclusions could be generalized, partly dependent on participant representability (Saunders et al., 2008). The ongoing case-study, external validity has been controlled on through sampling. In consultation with legal professionals before contacting the participants, a research population group was determined presumed to be representative for the Dutch legal sector. Around thirty potential participants were chosen, and approached. As a means of convenience sampling, an appointment was made with the ones that positively responded the quickest. Though the spread across different firms was accounted for, convenience sampling oppresses the external validity, making generalizability to other sectors harder.

(34)

6 | Results

This chapter gives an overview of the results derived from the conducted research among ten cases across different firms within the legal sector. The results are divided through factors underlying the three contexts: technological, environmental and organisational. Also, the sector’s position within the decision innovation process is reviewed.

During analysis, it became clear that for the legal sector, two types of adoption may be distinguished. The adoption of the technology for the own industry on the one side, using the technology as an advisory product on the other. Participants indicated that for these two implications different answers would be applicable. Because the intention of this research focussed on the technological adoption for the legal sector, analyses were based on that purpose and conclusions were drawn on that level.

6.1 | Decision innovation process

To identify in what perceived stage of the adoption of blockchain is currently in, a question based on Rogers’ (1995) decision of innovation stages was asked. The perceived stage is described as being in transition from 1 to 2 (figure 7). This is characterized by gaining understanding of the technology, but not having enough information to make a profound decision, yet actively looking for information and being open to the idea of innovation.

(35)

A typical statement supporting the position in the model was:

6.2 | Proposed Model

To arrive at the proposed research model, a cross-case analysis was conducted through (Yin, 2006) the extensive word table. This table was created to analyse the factors across cases and to see how they relate (Appendix IV). Overarching themes could be interpreted and read in what way the factors influence the adoption of blockchain (Ryan, 2012).

Statements were predominantly of a negative influence on the adoption of blockchain at this point in time. Some factors were perceived as either positive or negative influence, depending on factor specific characteristics. Some statements were perceived as being of positive influence on the adoption. For each category, an example quote will be given, to specify how these findings were derived:

Factor: Relevance & Urgency (Negative influence) -

Factor: Expectations (Positive influence) +

Factor: Firm size (Either positive or negative) +/-

“Whenever business is good, as it is nowadays, we don’t need a new market. We just keep doing our thing and we don’t feel the urge to adjust.”

“High expectations for the office of notary, or every group which builds on trust, the expectations across the sector is contributing to look into the possibility to adopt”

“In big firms, you have to convince more people, that makes it slower. That’s a disadvantage I think smaller firms are able to change direction much quicker.”

“There are projects already in the implementation stage, but shockingly uninteresting and often have nothing to do with blockchain, realistically in our sector, between 1 and 2”

(36)

Positive (+), neutral (0) and negative (-) categorizations were made from the cross-factor analysis, so that the data was interpreted and became visible (Table 2).

Table 2

Outcome directions per context per case*

Case à 1 2 3 4 5 6 7 8 9 10

Technological Context

Availability/maturity - - - N.a. - N.a. - N.a. - -

Characteristics - - - +/- - - N.a. -

Complexity - - - 0 - N.a. 0 N.a. 0 0

Expectations - - - N.a. +/- + - - N.a. -

Observable effects - - N.a. - N.a. N.a. N.a. N.a. N.a. N.a.

Relevance & Urgency - - + - - N.a. - N.a. + -

Environmental Context

Clients or customers - N.a. - - - +/- N.a. -

Competition - - - - N.a. 0 N.a. - N.a. -

Industry Characteristics - - - - -- - - -

Legislation & Regulation N.a. - 0 - +/- N.a. - N.a. - -/0

Organisational Context

Organisational structure - - - N.a. - N.a. N.a.

Firm Size +/- - +/- +/- +/- +/- N.a. +/- N.a. N.a.

IT-skills of non-IT employees - - - N.a. - - N.a. - N.a. -

Employee input - - - - N.a. - N.a. N.a. N.a. -

Top management support + N.a. - - 0 -- N.a. N.a. +/- -

Factors derived from

analysis

Adoption distinction +/- +/- +/- +/- +/- N.a. N.a. N.a. N.a. +/-

Marketing N.a. ++ + + + + + N.a. + +

Costs & Benefits (T) N.a. N.a. - - - N.a.

Hype & Image (T) N.a. - - - -

Type of work (O) N.a. +/- N.a. N.a. N.a. +/- +/- +/- +/- +/- *N.a. stands for ‘No answer’. 0 stands for no effect. +/- is dependent on factor characteristics, whether

positive or negative.

From the table, the TOE framework could be revised for this particular sector and concluded how the individual factors are of influence on blockchain adoption (Figure 8). These

(37)

Figure 8. Factors influencing Blockchain adoption. Factors deducted from data-analysis and pretesting (employee input) displayed in italics, followed by either positive, neutral or negative influence on adoption.

6.3 | Technological Context

Availability, Maturity & Observable effects (-)

Availability, observable effects and maturity were factors that were perceived as the same, answers on the questions of these categories showed much overlap. Blockchain was perceived as not yet mature enough and therefore not yet available or visible. Some use-cases were presented, but at the same time were described irrelevant due to other factors influencing the adoption (cross-factor analysis, chapter 6.7). Due to the unavailability of tangible products and the perceived immaturity of the technology, the influence on the intention to adopt the technology was negative.

(38)

Many predictions are made towards the use of the technology in this particular sector, but no observable effects have been reported. Blockchain products that have been launched are seen as gimmicks and no more than a normal database with no characteristics of a blockchain, where visible products were said to miss critical blockchain determinants such as decentralization, tokenization, encryption or immutability. At this point, no observable effects are visible, and it negatively influences the adoption. The technology has to be proven and become visible.

Characteristics (-)

Several characteristics may be distinguished at this point in time as being specific for blockchain and being of negative influence on the adoption. The technology has to have added business value. There is still ambiguity is what the added value is for this technology, which is of negative influence of the adoption. Added business value should come in terms of making work easier, faster, cheaper or better. Moreover, it should not lead to extra work in terms of keeping more systems up-to-date.

Another characteristic refers to the principle of ‘crap-in-crap-out’. As long as the technology depends on human input, there cannot be given an assurance for the technology to deliver what is promises.

A final characteristic of blockchain should be its backwards- and forwards compatibility. The decentralized character of the database may not lead to two different database locations in which new data has to be stored ‘on the blockchain’ and old data has to be accessed in conventional ways.

The potential of blockchain or innovation in a broader sense is hard to value. Blockchain is not yet seen as something which may be beneficial at this moment. The costs and benefits of a

(39)

product are also still ambiguous. As soon something may be earned from it, the industry will start to adopt it.

Complexity (-/0)

Perceived technological complexity towards blockchain does not have impact on its adoption. Every technology nowadays is not easily understood by its users. Same goes for blockchain. Not many legal employees are able to tell how the technology works and what its technical components are. By experts, however, blockchain is perceived as not complex at all. In a few hours, it is possible to construct a simple, working, blockchain-based product. For the technology, it is important for it to make it user-friendly, so not complex in its use. The technology itself may be complex, but on the background.

Another theme concerning complexity is not the technology which is perceived complex, but the environment in which the technology has to work (cross-factors analysis, chapter 6.7). Since different parties have to work together, the implementation in terms of cooperation of a blockchain product is perceived as more complex than with other product.

Expectations (-)

On short term, expectations are low. Not many applications are imaginable. Most opportunities are seen in cooperation between different external parties or companies. Possibilities are seen within the notary sector, a sector which builds on trust. If you are to replace the need for a notary by a decentralized ledger in which notary deeds are stored or decentralize shareholders registers these processes would become more efficient. However, transformations on this scale will not happen overnight and are dependent on many other factors (Cross-factor analysis, chapter 6.7).

(40)

Relevance & Urgency (+/-)

Several factors determine the relevance and urgency for a blockchain product. At first, in law firms ‘business trumps innovation’. Whenever business is good, time is spent on delivering conventional services, when business is less good, time is spent on acquiring new cases.

Secondly, since the sector is perceived as a technological laggard, the perceived fit of a distributed ledger technology service is minimal, since the focus often is on replacing old systems to modernize current outdated ones. Thereby, the current implications that are on the market are seen as irrelevant, not solving any problems and not novel.

Great opportunities and relevancy is described for the future, but highly dependent on other factors (Cross-factor analysis, chapter 6.7).

Hype & Image (-)

The technology may harm the business, which causes reluctance by this sector to invest in a product since it can potentially replace certain aspects or processes from their work. This negatively influences the adoption within the sector. Another factor is the link with Bitcoin. Because blockchain and Bitcoin often get used together in the same context, Bitcoins image harms blockchain. Finally, the hype around blockchain makes that it is projected as much more promising than it is in practice. The so-called disruptive nature of the technology is perceived as not as influential as many proclaim in the legal sector. The hype causes firms to look into it more often, but at the same time for them to see the true value of the technology.

6.4 | Environmental Context

Clients or customers (-)

For a service provider, like a law-firm, a client’s demand is always leading. Whenever the client asks for something, the firm will deliver. As for the subject of blockchain, no client demand for

(41)

adoption has been documented. Small requests for legal aspects on blockchain innovations in other sectors have been reported, but none for legal blockchain products. The reason why companies publish about the subject and proclaim to have a product is to attract new customers or clients in order to be able to advise them in a traditional way on the subject.

Competition (+/-/0)

Different firms have published articles claiming a blockchain product has been adopted. Because of this, other firms were triggered to investigate the subject as well. Concluded was that the ‘adopted’ innovation was not more than an advisory team with some knowledge of the technology. If something disturbing would arise, it would come not from one of the established law firms, but from an unexpected or new party. Since firms often look into the ‘back mirror’, they expect not much of their competition concerning blockchain technology. Competition in the form of new, not yet established parties are seen as more realistic competitors or influencers.

Industry Characteristics (-)

The legal sector was perceived as a laggard when it comes to technology adoption. The fact that some documents are still sent by fax is an example of this. Reasons for this are two-folded. On the one hand, it is not the firm’s business to innovate. Products are not brought to market by law firms. On the other hand, it is the sector’s nature to work in a conservative way. Technologies are adopted slowly, and processes are done in conventional ways. The sector is back in line when it comes to innovation, a so-called following authority, facilitating and following in service of the client. It comes in the picture when there is a problem or opportunity, and cannot take the lead as an outsider. These factors make adoption of this technology slower and perhaps irrelevant.

(42)

Legislation & Regulation (-)

The biggest promises of blockchain are currently prohibited by law. At this point in time, for example the Chambre of Commerce (Dutch: kamer van koophandel) has a monopoly position by law. A flawless decentralized, immutable and encrypted ledger, may be created, but that does not free anyone of still registering or requesting documents at this party. Disintermediation may work, but law obliges people or parties to still work in conventional ways. For smaller, more local blockchain related products regulation will be of less influence, but for the disruptive, large scale implications, regulation often is prohibiting innovation.

It is said that law always follows innovation. In the case of blockchain, laws and established parties inhibits the incentive to innovate in this area.

A final, more complex aspect, that comes forth from constitutional concerns are ethics. There are deep grounded ideas and reasons why certain laws and parties are included in the law. The legislator cannot always be aware of everything, that is why legal firms are often considered of being important for advisory and make governmental organisation aware of certain tendencies which arise.

6.5 | Organisational Context

Organisational structure (+/-)

Law firms are often structured in teams of different areas of expertise, working independent from other teams, which causes fragmentation of knowledge and skills. Especially in the bigger firms, it may well be that different teams have knowledge on blockchain without employees being aware of it. The structure of a typical law-firm can be described as a wide, flat pyramid with few people on top making decisions for the large amount of people on the bottom. Smaller law-firms however, are more agile in adapting and adopting with a less clear structure (Cross-factor analysis, chapter 6.7).

(43)

Firm Size (0)

Small and large offices can be roughly categorized on two different continuums for influencing adoption; available resources and adaptiveness. For small non-radical decisions, a large organisation helps in terms of number of full-time employees and amount of money to be spend. Since a big firm is less agile, radical changes take much longer to adopt. The predicted scale of blockchain makes that smaller and younger firms will be able to adopt the technology faster. Because smaller firms often focus on more commoditized work this sector will be affected sooner than big firms (Cross-factor analysis, chapter 6.7).

IT-skills or knowledge of non-IT employees (-)

Since legal professionals do not have any education on subjects of technology, not enough knowledge of blockchain and how to use it for the sector is available. This is raised as an important matter for future potential. Because legal professionals are not provided with training on new technologies, it is seen that the amount of knowledge of blockchain and how to use it is perceived as deficient.

Employee input (-)

After pilot testing the factor room-to-innovate was added. During analysis, it turned out that innovate showed overlap with the factor top management support. The factor room-to-innovate was changed to employee input, since this factor aimed to test the how employees input affected the adoption.

Lawyers are usually traditional and conservative in their way of working. Due to the nature of their work, innovation is not something which is part of the day-to-day activities of a legal professional. Little initiatives come from lawyers. A lawyer is being judged by its billable