How well are the two chains linked together?

(1)

How well are the two chains linked

together?

A study about the perceived effects of blockchain on transaction costs

within supply chains handling physical goods.

Ludvig Kamlin, Business economics Philip Österström, Informatics

Supervisor: Darek Haftor

Master Thesis, E-level, 30 credits Management, Communication and IT The institution for informatics and media Uppsala University

(2)

First, we would like to thank our supervisor Darek Haftor, for his support, guidance and recommendations while writing this thesis, along with his kindness by letting us contact him even outside of office hours. We would additionally like to thank our friends and family, for their support and cheering along the way, and finally, all the participants of this study and their effort to recommend us to further valuable contacts within their networks.

(3)

Abstract

The purpose of this study is to produce new knowledge concerning perceived effects of the blockchain technology on transaction costs within supply chains handling physical goods through the created and hypothetical TCE-matrix. The study followed an abductive research approach where initial found facts were matched against an appropriate theory in order to be put in context and understood. The study’s data was gathered through semi-structured interviews with experts who were active within the research area and chosen based on their practical experience. The results imply that blockchain holds the possibility to affect five out of nine cells within the TCE-matrix out of the experts’ perception, through its increased trust and transparency between its supply chain actors within the network and the enhanced traceability of physical goods. This study further highlights practical implications concerning the technology’s adoption in practice. It is concluded that the blockchain is perceived to have an effect on transaction costs activities out of the TCE-matrix, which is found to be partly applicable when assessing the blockchains potential effect on transaction cost activities in a general manner, where further studies should apply the matrix in the assessment of one specific transaction to take full advantage of the matrix’ potential.

Keywords_{: Blockchain; Blockchain Supply Chain; Blockchain Supply Chain Transaction Cost} Economics; Transaction Cost Economics.

(4)

1. Introduction

This section presents the introduction, background, wording of the problem, the purpose of the study and the research question along with limitations and a disposition of the paper.

Blockchain is projected to be the latest transformative innovation and is increasingly gaining attention from both academics and practitioners in many different industries (Schmidt & Wagner, 2019). As of 2018, the corporate spendings on blockchain initiatives was estimated to reach $2.1 billion, which is an increase from $947 million in spendings by the end of 2017 (Schmidt & Wagner, 2019), and speaks for the increased interest in the technology. There is more to the technology than cryptocurrencies, whereby the technology has been hyped to add value in several more new areas. The financial crisis of 2008 revealed the difficulty of identifying the correct ownership of an asset, where the problem occurs as a critical aspect in long transaction chains even outside of the financial industry. Even more so, in the supply chains handling physical goods of, e.g. diamonds or food (Nofer, Gomber, Hinz & Schiereck, 2017). As an example, the giant US retailer Walmart has more than 260 million customers a week and are using a blockchain solution provided by HyperLedger, allowing the retailer to track different foods back to their specific batch and origin to prevent outbreaks in case a batch of foods would cause illness to its consumers (Hyperledger.org, 2020) giving them more control of their supply chains. 50% of the world’s supply chains cross a national border, which leads to the increased complexity of the traceability and monitoring of the shipment (Roeck et al., 2019). Therefore, it increases the interest for technologies that facilitate supply chain traceability, such as blockchain. Through the increased transparency of supply chain actors data, which is made possible by the blockchains decentralised structure and function of immutability, meaning that the data that is written onto the blockchain ledger is permanent without the possibility of being manipulated, the technology holds the potential to revolutionise how physical goods and services are transacted (Schmidt & Wagner, 2019). The blockchain may be understood as an organisational technology, reaching over several organisations as well as work processes, which helps supply chain managers intervene and manage the supply chain (Hald & Kinra, 2019).

Transaction cost economics is the theory where transactions of information or goods in between parties are evaluated in terms of their costs (Schmidt & Wagner, 2019). These costs can be evaluated for a firm in order to make the decision of whether to produce a product themselves or buy it off the market (Ketokivi & Mahoney, 2020). These costs will differ depending on where the goods or information are transacted. The costs accrue out of three so called transaction activities, which are _{information and search costs, bargaining costs} _{and enforcement and}

monitoring costs_{(Tate, Dooley & Ellram. 2011), and are therefore suitable to take into account} in order to make the most economical decisions concerning where to acquire a good. Blockchain is believed to enable more transparent and efficient transactions (Schmidt & Wagner, 2019), where the transaction cost economics theory is appropriate to evaluate how the technology is perceived to affect transaction activities and their associated costs within supply chains, and was therefore the theory chosen for the study.

(7)

1.1 Background

Blockchain technology within supply chains holds promises of being able to significantly improve transparency, accountability, trust, security, efficiency, and cost minimisation (Wamba & Queiroz, 2020). It can, therefore, potentially add value in several organisations considering adopting the technology within their supply chains. It is additionally explained that the technology is viewed as a solution to solve supply chain traceability problems, as well as for creating better relationships where trust between parties is increased. Trust is defined as “ The consensual faith that no other part at an exchange takes advantage of the other parts vulnerability”, according to Loebecke, Lueneborg & Niederle (2018). It is further interpreted that trust is necessary in environments prailed by risk and unpredictability, where trustors accept being vulnerable to potential negative outcomes resulting from exploitation of trust by the trustee.

The composition of well-functioning supply chain networks is reliant on understanding transaction costs in order to evaluate many supply chain management related issues (Garfamy, 2012), where Williamson (2008) promotes the applicability of transaction cost economics as a lens when analysing intermediation, meaning third party involvement, and transactions within supply chain management. Transaction cost economics has played an important role when making boundary decisions and is one central aspect of supply chain management (Roeck et al., 2019). Coase (1960) defines transaction costs as “the cost of carrying out market transactions”, which refers to the interactions between firm to firm, or firm to individual and the costs that accrue under different circumstances. In more general and simplified terms, the theory aims to evaluate the costs of, for example, an individual making a purchase of ten apples from a grocery store against what it would cost for the same individual to make the same purchase on the open market. Similarly, these costs could differ when conducting transactions of physical goods in supply chains based on blockchain, versus traditional supply chains.

Tate et al., (2011) divide transaction costs into three separate activities; information and search costs, bargaining costs, and enforcement and monitoring costs_{, whereby a transaction activity} can be placed in one of these categories. Where the costs relate, in the example, to the cost of searching for information about the apples as well as finding a supplier, the cost in terms of time and resources of bargaining for a price and the price to see whether the apples are indeed the same quantity and quality as stated by the salesman when bargaining. These costs will differ depending on where they are purchased, which is necessary to be considered in order for the individual to make the best purchase. The transaction cost economics theory fits well into supply chain management since intra-firm and inter-organisational transactions are explained, as well as their supplied costs and the fitting governance mode (Roeck et al., 2019), which is either firm or market-oriented (Williamson, 1985).

What is of interest is to explore how the blockchain technology could affect transaction costs out of each separate activity, in relation to traditional supply chains, in order to assess its potential economic impact. For example, since blockchain holds the promise to increase transparency within supply chains (Wamba & Queiroz, 2020), it is of interest to see if transaction activity

(8)

costs are increased or decreased as a result. Due to the theory’s broad application of measurement, congruent with Williamsons (1996, p. 5) quote “the measurement of transaction costs poses formidable difficulties”, wherefore the transaction cost economics-matrix was created in order to assess transaction costs out of a qualitative perspective. It is of interest to explore the blockchain technology’s impact on supply chain transactions and costs related to the three above mentioned cost activities, which may very well play an essential role in managerial decisions regarding whether to make use of the blockchain technology in order to reach greater economic efficiency.

1.2 Problem area

Many organisations are today struggling with counterfeit of high-value goods due to the lack of transparency and battling opportunism within their supply chains, where supply chain actors are involved in behaviors such as cheating, lying, and subtle forms of violation of agreements. A behavior that can have a negative organisational impact in terms of monetary losses and inefficient work processes (Pun, Swaminathan & Hou, 2018). Within the original transaction cost economic theory literature, _opportunism _{is defined as the “search for self-interest in the} notion of guile” (Heide & Wathne, 2000). The same authors explain that, in this case, _guile relates to “lies, theft, cheating, the calculated effort to mislead, distort, mask, obscure and in other cases confuse”. For organisations in industries handling physical goods, it is essential to be able to track materials and products throughout the supply chain, from production to sales, in order to ensure that the product is legitimate and of the proper quality as well as optimising the efficiency of the chain through new ways of sharing information. Blockchain is believed to assist in the improvement of transparency, accountability, trust, security, efficiency and cost minimisation (Tönnisen & Teuteberg, 2019).

The importance of trust within the supply chain is critical, since most supply chains today heavily rely on centralised e.g. ERP-systems , where significant trust has to be in place in order 1 to store their sensitive and valuable information (Saberi et al., 2019). The presence of trust makes extensive contracting unnecessary, thus leading to cost savings, as well as closer cooperation (Loebbecke et al., 2018). On the contrary, the lack of trust calls for heavier safeguarding against opportunistic behavior from supply chain actors and results in increased costs. This might have an effect on transaction costs, where blockchain is seen as a solution for supply chain management traceability problems and for generating closer and trustworthy relationships throughout the entire supply chain (Wamba & Queiroz, 2020).

The use of transaction cost economics within supply chains, as a measurement of performance, has earlier been discussed in order to evaluate the costs of e.g. selection, acquisition and maintenance of a specific item (Möller & Horvath, 2004) and can potentially further assist in exploring whether the transaction costs are affected in the supply chains handling physical goods by the use of blockchain, which may speak for blockchain as a potential investment and

(9)

beneficial in organisational terms. No previous studies have analysed the perceived effects of blockchain on supply chains handling physical goods through the assessment of transactional activities and their dimensions. It is, therefore, of interest to explore what potential the blockchain has in the supply chain of physical goods in regards to affected transaction costs through our created matrix.

1.3 Purpose & Research question

The purpose of this study was to obtain new knowledge by exploring experts’ perception of how blockchain affects supply chains handling physical goods, and to evaluate the effects in relation to the transaction cost economics theory. In this case, effects refers to how blockchain’s features may impact transaction activities stated earlier as the transaction activities of; _{information and}

search, bargaining _{and monitoring and enforcement (Tate et al., 2011) of supply chain} operations handling physical goods. The experts refer to individuals with practical expertise within the area, whose perceptions were examined through semi-structured interviews. The goal was, therefore, to obtain knowledge gained through the experts’ practical experience, which may be further elaborated beyond what is presented in current literature since there are considered to be few studies within the area.

Prior research has mainly focused on exploring the technology’s impact on transaction costs in supply chains in a general manner, leaving a shortfall of empirical research (Roeck et. al, 2019), as well as not assessing the costs in terms of the transactions three dimensions, combined with their activities, providing a more granular evaluation of supply chain transaction activities and the blockchain’s economical impact. Therefore, a Transaction Cost Economics (TCE) - matrix was developed out of pre-existing literature and acted as a qualitative framework, intended to evaluate transaction activity costs, resulting in the following research question:

What are the perceived effects of blockchain technology on the transaction activities of organisations within supply chains that handle physical goods, and how would these effects influence costs in conjunction with transaction cost economics?

The study does so by assessing the costs of transaction activities in blockchain-based supply chains from a transaction cost economics perspective, of which, the TCE-matrix will act as a new conceptual transaction cost framework, intended to evaluate transaction activity costs. The matrix will contribute to present future research opportunities to build upon, as well as the opportunity for managers wanting to explore the blockchain technology within their supply chain transaction activities in terms of its economic impact, out of a transaction cost economics perspective, which can justify their investment.

(10)

1.4 Limitations

The comparisons made in this study, between blockchain-based supply chains and supply chains without blockchain technology, does not take the cost of implementation into account, where the actor gets onboard onto the blockchain. Any costs related to instalment and organizational adaptation to blockchain technology are disregarded, which is vital to bear in mind when considering the actual cost of blockchain-based solutions in terms of transaction costs.

Due to the limited availability of active business in Sweden, who are active within a specific industry where the blockchain technology is used in its supply chain, the demarcation has been to primarily focus on the area of supply chains handling physical goods of some value. In this case, value is defined as those supply chains handling goods which calls for governance in terms of fraud, theft, quality assurance or consequences such as consumer health, corporate reputation or monetary losses will accrue.

It is further of importance to recognize that the explored effects are based on the experts’ own perceptions and opinions, and should therefore not be considered as the actual effects of the use of blockchain within this context.

1.5 Disposition

First, a literature review has been conducted in order to evaluate the study area and how far the research has come as of today. Second, we present the development of the transaction cost economics model and present its content, application and purpose. Third, the method used in the study will be presented, and finally, results, analysis and discussion are followed.

2. Literature review

The literature review introduces blockchain, and earlier research related to blockchain within supply chains and transaction cost economics within blockchain-based supply chains. The role of trust is also briefly explained since it fills a vital part in understanding the transaction cost economics theory, blockchain and its contribution within supply chains.

2.1 Blockchain

Polkowski et al. (2018) define the blockchain as a decentralised database for storing and sharing information online. The added information is stored in a chain consisting of several blocks which all are coded and decrypted through algorithms, whereby its name. A new block is connected to the previously added block in the chain through a hash, together with a timestamp for when the

(11)

block was created. Every newly added block is protected against manipulation after that it has been added to the chain. In order to manipulate a particular block in the chain, all the blocks have to be adjusted, therefore making it extremely hard to alter. The blockchain prevents this from happening through advanced encryption. There is no central authority that controls the information which is added to the blockchain since every new block of information has to be approved through the network's users, acting as nodes in the network. Every authorised user has access to the information and therefore the possibility to take part of it at all times. In order for a user to add a new block of information, it has to be verified and accepted by a certain amount of the chain’s other users through consensus. This is usually conducted through a type of mathematical verification that controls the content within the new block as well as the identity of the user that would like to add it to the chain.

What is an architectural property of blockchain technology is its distributed structure. This means that all participants of the blockchain network have access to a copy of the same data, which is distributed over every node in the network (Hald & Kinra, 2019), which differs from a centralised database where the data is stored in one place, and one place only, thus making it less vulnerable to hacker attacks. This allows every party within the supply chain network to have a copy of the same data. _{Fig. 1} _{shows the difference between a centralised and distributed, or} decentralised, blockchain network.

Fig 1. (Source: Deloitte: _{How Blockchain Boost Operational Excellence})

Finally, there are different types of blockchain networks, which Polkowski et al. (2018) separate into three different categories: private, public and hybrids. Demanded information is available to the users of a private network although not necessarily detailed enough in order to reveal a users identity, where only a controlled group of actors have access. Public networks are open to anyone who has an interest in joining the network and is considered to be more distributed against its private counterpart. A hybrid network is a combination of the previously mentioned

(12)

networks. It controls who might have access to the blockchain and uses the public in order to enable a higher degree of participation in the “proof of work”.

2.2 Blockchain within Supply Chains

The supply chain is a network of multiple businesses and relationships, and the management of multiple relationships across the supply chain is what briefly defines supply chain management (Lamberg & Cooper, 2000). Supply chain management deals with total business process excellence and represents ways of managing the business and relationships with other members of the supply chain (Ibid). The role of blockchain within supply chains has gained an increasing amount of interest from both practitioners and scholars over the last few years. Schmidt & Wagner (2019) claims that the technology might substantially affect supply chain management, its relations and governance structure. It also has an essential role in terms of traceability of different goods of value, such as food or diamonds, throughout the production to delivery process to ensure quality and authenticity. Where Treiblmaier (2019) adds and argues that blockchain will contribute with increased visibility, optimisation and forecasting of demand for certain goods and products since increased transparency leads to the more efficient management of unforeseen events.

Additionally, the blockchain’s potential to enhance transparency due to its ability to store historical events and activities, an audit trail may be available, potentially allowing audits of all sorts to be improved and better understood (Hald & Kinra, 2019). Further, it may result in the improved ability to control the supply chains and its activities. Ineffective transactions, fraud, pilferage and poorly designed supply chains lead to more significant inefficiencies in terms of reliance between its parties and less trust (Saberi et al., 2019), which demands better information sharing and verifiability. Blockchain can overbridge this shortcoming; for instance, it can be utilised to localise the origin of faulty products, resulting in lessened costs related to recalling (Behnke & Janssen, 2019). The solution provides data like farm origin, batch numbers, factory and processing data, expiration dates, and shipping data, making it available to every node in the supply chain network, increasing the transparency and ability to locate the origin of a specific food within seconds, contrary to weeks within a traditional supply chain (Hackius & Petersen, 2017).

Saberi et al. (2019) state that competitive issues arise in terms of strategy and reputation for an organisation due to lack of transparency. Walmart uses a blockchain-solution, where 1.1 million items are registered and traced from manufacturer to retailer (Mims, 2018). Maersk, the global shipping company, has teamed up with IBM and are using their blockchain cloud solution to track their shipping containers, making the tolling and customs process significantly more agile and secure (Schmidt & Wagner, 2019). The lack of transparency, makes it impossible to see everything that is going on within the supply chain network, prevent customers and supply chain actors from validating the actual value of an item (Saberi et al., 2019) since most proofs of authenticity documents are paper-based and also easier to alter or manipulate (Hackius & Petersen, 2017), such as the bill of lading. Hackius & Petersen (2017) states an example where

(13)

international container transports are associated with a large amount of paperwork. In some instances, approvals are needed from 30 people, and the involved organisations are required to interact with each other up to 200 times. It is estimated that the cost of handling the paperwork constitutes between 15-50% of the physical transport costs (Hackius & Petersen, 2017). Since blockchain acts like a central ledger where previous information can not be manipulated, it contributes to a new type of transparency that may enhance and simplify activities within today’s supply chains.

Logistic services play an essential role in a business ability to deliver customer value since the emphasis is put on delivering a product in the right condition, in time and to the lowest possible cost. How well this is fulfilled is evaluated against the measurements of quality, time, trust, cost and flexibility (Kshetri, 2018). The lack of traceability within supply chains leaves room for improvement regarding both economic and social factors, concerning working conditions of farmers, as an example. Globalisation, diversified prescriptive policies and different cultural and human behaviour in the supply chain network make it close to impossible to evaluate information and manage risks (Ibid.). Hence, there is room for improvement regarding the ability for actors to trace activity beyond one or two steps up and down the chain, which is often the situation today. Still, this demands that clear standards and internal capabilities are in place (Behnke & Janssen, 2019).

Hald & Kinra (2019) highlights the need for transparency within supply chains by explaining that it may be used in order to verify ownership of an asset. The primary solution for verifying ownership of an asset is today through intermediation, which is costly and time-consuming (Nofer et al., 2017). Earlier studies are discussing blockchains potential to remove intermediaries and have the potential to lower transaction costs (Tönnissen & Teuteberg, 2019). Where the blockchain potentially will overcome these difficulties through its distributed structure, where each node within the network has the same copy of data present and therefore shares a single source of truth. The problem of uncertainty is addressed, which is based on the difficulties of monitoring the transaction partners’ behaviour as compliance to contracts due to the complexity of evaluating its performance, as well as the related problem of information asymmetry.

Intermediaries possess information about a transaction partner and enter into an agreement with that transaction partner or its intermediary, therefore possibly distorting information leading to the problem of information asymmetry (Roeck et al., 2019). Kamble, Gunasekaran & Sharma (2019) similarly, found that the benefit of disintermediation, enables the integrity of data and no third-party authentication, leading to lower transaction costs and lower settlement lead times. Treiblmaier (2018) describes how blockchain technology has the potential to remove the issue of information asymmetry in supply chains. The issue stems from a lack of transparency between the different actors in the chain. There is currently no guarantee that an actor gets full insight into how business is being conducted further up and down the chain. This creates a relationship balanced around trust, meaning the necessity to assume that another actor is not acting out of self-interest at the expense of oneself. Depending on the relationship between the actors, as well as the amount of transparency provided between them, costly trust mechanisms and control systems might be needed in order to confidently make this assumption, as explained by Jensen &

(14)

Meckling (1976). In a blockchain-based supply chain however, the information flow can be transparent and accessible throughout all of the actors in the chain. Therefore the demand for the trust and control systems are lessened (Kaal, 2017) since the traceability, transparency and immutability of data enhances the level of trust within a blockchain-based supply chain (Treiblmaier, 2018).

Due to blockchains potential to decrease the need of trust mechanisms in some organisations, those who base their business model on a mediation role might find themselves obsolete if the solution is adapted. Within supply chains, these intermediaries include actors that handle transaction validation (Treiblmaier, 2018; Zheng et al., 2017). Roeck et al., (2019) states that the distribution of information that the technology enhances reduces the need for intermediaries as information suppliers, diminishing their control over information and causing them to lose their information monopoly. Additionally, intermediaries become eligible for removal due to their function being filled by the blockchain, which can lead to increased efficiency and cost reduction, congruent with the supply chain management goals (Khestri, 2018).

Fig. 2. Traditional vs. Blockchain based markets. (Tönnisen & Teutenberg, 2019).

Current supply chains mostly rely on centralised ERP -systems within an organisation, which are 2 more vulnerable and possess multiple pitfalls which may have a significant impact on the information management within the supply chain which leaves the system more vulnerable to errors, corruption or attacks (Saberi et al., 2018). One of blockchains, as discussed, prominent features is the immutable and safe information that is written onto the chain. The ledger is therefore, enabling the tracking of data through the chain as well as transactions over time since they are stored in chronological order, in real-time, permanently available and unaltered (Tönnissen & Teuteberg, 2019). Thus, it leads to the creation of new forms of trust based on transparency and traceability in combination with the blockchains technical feature of cryptographic security (Treiblmaier, 2018). It is further elaborated that this potentially minimises errors since they are easier to detect and correct within the supply chain. Hence there is a decrease in supply chain risk in terms of data inconsistency (Hald & Kinra, 2019).

(15)

Kamble et al., (2019) explains that due to the blockchains distributed nature, the susceptibility to manipulation and counterfeit of data within the blockchain network is reduced. Through the transparency and immutability of the blockchain, the data accuracy is high since the network verifies it, or more simply put by the community. What is followed is the validation and trust of the information which is presented by the blockchain, since connected users verify the information that is included in a new block within the chain and have to agree on what is added and removed, the data is virtually impossible to alter (Hald & Kinra, 2019).

Kim and Laskowski (2017) and Kuo, Kim & Ohno-Machado (2017) explains that collaboration between partners plays a key role in making blockchain solutions work in supply chain networks. As a part of this, according to Britchenko, Cherniavska & Cherniavskyi (2018), scepticism between partners may pose a critical issue for implementation since the technology enhances traceability and transparency. Badzar (2016) & Edwards (2017) explains that this is due to partners being varied in their market position and with scepticism comes a fear that increased transparency might adventure their competitive advantage. Thus, partners who are currently at a commanding position could view the technology as a threat. It is therefore crucial that partners can come to terms with what data to share as well as how and when it should be done (Chow, 2018), and that these terms are expressed through clear standards (Behnke & Janssen, 2019). Kolbe (2017) states that the applicability of blockchain is dependent on all the actors in the supply chain to acknowledge the benefits of the technology. Casey & Wong (2017) argues that otherwise the application will lose value for some of the actors, and some participants might not benefit from it whatsoever, which could lead to a reverse effect regarding transparency and trust since some actors feel the need to protect their data from the others, which would harm the supply chain information flow.

2.3 Trust

Trust is defined, according to Loebecke, Lueneborg & Niederle (2018, p. 3), as “The mutual confidence that no other part at an exchange will exploit the other parts vulnerability”. Therefore, trust fills a vital role in order for a business to function correctly (Heide & Wathne, 2000). Blockchain improves trust in transactions (Kshetri, 2017), which can improve supply chain operations. Additionally, the enhanced transparency will enable supply chain actors to gain insight about the supply chain business partners who are all connected to the blockchain. This enables collaboration in real time across wide spaces, in different countries and even on different continents (Hald & Kinra, 2019). Trust is exceptionally important in between supply chain actors, since actors have to ensure that their suppliers abide legal and responsible sourcing practices, as well as ensuring the product’s origin and that the produced goods are authentic (Hastig & Sodi, 2020; Wong, Leong, Hew, Tan & Ooi, 2019). It has been proven beneficial when organisations have trust in their counterparts since it, e.g. facilitates coordination and may decrease transaction costs (Gulati & Nickerson, 2008). If trust is present whenever business partners are entering a contract for an exchange, less formal governance modes are needed, which speaks for that trust enhances exchange performance (Ibid) under the responsibility that they follow what is contracted. It has been discussed how this responsibility is influenced by

(16)

trust and whether trust affects the need for formal control measures in order to secure the presence of responsibility.

On the one hand, there exists an understanding that the presence of trust allows more informal control measures and governance structures. The other understanding is that trust is considered as a compliment and should not act as the foundation upon which the decision for a specific governance mechanism is taken. Regardless of what relation that exists between trust and governance, the implications of the presence of trust make the counterpart more prone to manage potential problems and disputes, which thereby contributes to mitigating governance costs (Gulati & Nickerson, 2008).

2.4 Transaction cost economics in blockchain-based supply chains

Transaction cost economics theory could aid in terms of assessing supply chain management efficiency, more specifically in terms of “make-or-buy” decisions for firms (Ketokivi & Mahoney, 2020) where supply chain management goals such as price, cost and time (Khestri, 2019) possibly could be more easily obtained. The theory evaluates the different costs of transacting, which could aid in an organisation to decide whether to produce a product themselves or buy it from the market.

Concerning blockchain, the theory of transaction cost economics helps in the assessment of whether the blockchain will affect the costs of transacting physical goods compared to a traditional supply chain, therefore justifying its investment. Few previous articles address the blockchains effect on supply chain transaction costs, whereby previous articles by Schmidt & Wagner (2019) discuss the blockchains effect on supply chain relations out of a transaction cost economics perspective and Roeck et al., (2019) address the effect on transaction cost in a broad sense from the use of distributed ledger technology . The application of transaction cost3

economics in supply chains based on blockchain seems appropriate because blockchain works as a ledger for transactions, and a fundamental construct within supply chain relations is the transactions and their costs (Roeck et al., 2019). Schmidt & Wagner (2019) found that there is potential for the technology to reduce information asymmetry and contractual incompleteness as well as the reduced cost of gathering information, drafting and negotiating contracts, and enforcing and monitoring agreements. Where Roeck et al. (2019) found that due to the increased transparency, primarily the related activities of information and search and bargaining were affected. Information and search through the collection, storing and sharing of information and bargaining through revealing actors historical activities, causing a powershift where the loss of bargaining power may occur.

3_{Distributed ledger technology (DLT), is the hypernym of blockchain technology (}_{Roeck et al., 2019)}_{, it allows for} storing of data in a distributed, decentralised network and is based on peer-to-peer validation (Hawlitschek,

(17)

For instance, Lumineau et al. (2012) explain that blockchain provides the option to omit third parties that work as trust enhancers in traditional exchange systems, which would allow the fee of the third party actors to be spent on other resources, enabling reduced transaction costs. Moreover, due to the immutable ledger’s ability to track the activities of the actors on its network, it shows potential for facilitating performance control and measurement of contract fulfilment. Schmidt & Wagner (2019) has found that the blockchain can save costs related to controlling and monitoring current and potential suppliers and ensuring product or service quality.

Generally speaking, in an economy based around blockchain, the increased transparency of the technology may influence more stakeholders to take part in the governance of transactions, including those who currently occur under relatively restrictive conditions. Blockchain has the potential to produce a more considerable amount of short-term relationships and ad-hoc partnerships, which eventually may challenge some of today’s assumptions of supply chain structures and processes since, according to Schmidt & Wagner (2019), long-term relationships build trust, and may, therefore, reduce the level of opportunism. Roeck et al. (2019) highlight that trust is a contrary assumption that negates the assumption of opportunism. Wherever there is trust present to a certain arbitrary level, the need to choose a governance structure in regards to safeguarding against opportunistic behaviour is no longer of relevance.

In parallel with the trust, the risk for a counterpart to act opportunistically exists (Ha, et al., 2011). That is an individual or organisation that actively seeks what is in its prime economic interest at the expense of his counterpart, where the theory also aims to reduce this risk through choosing the appropriate governance structure (Ketokivi & Mahoney, 2020). An example of the opportunistic behaviour of companies in supply chain management context can be their effort to make suppliers highly dependent on them, and take advantage of this power to make pressure on them (Saberi et al., 2019, s. 2127). The exploitation of opportunism brings several consequences, which vary depending on the industry in which the business works within. The lack of traceability of ethical production, according to sanctions and product safety within supply chains is today a highly relevant problem area in the supply chains handling physical goods (Hastig & Sodhi, 2019), which many organisations are investing vast resources in to solve. As mentioned, the resources spent in this aspect produces a significant amount of opportunity costs through potential deals that get lost, in which this time and resources otherwise could have been spent (Christopher, 2016). This kind of opportunism can be manifested through intentional misinterpretation at the forming of relations as well as in violations at the initiation of a relation. If the risk of opportunism is high within a specific relation, considerable resources have to be spent on governance and monitoring, which potentially could be spent more productively in other purposes. Roeck et al. (2019) state that blockchain, within this context, can reduce transaction costs as well as limits opportunistic behaviour and uncertainty, which contribute to the accumulation of transaction costs.

Cocco et al. (2017) state that blockchain will change costs related to intra and interorganizational transactions and that it may affect organisational borders. These transaction costs can be both

(18)

internal, including, for example, bureaucratic costs that take place within an organisation, or external through exchanges between different organisations. The implementation of blockchain has the potential to affect both internal and external transaction costs simultaneously. Schmidt & Wagner (2019) further argue that governance structures are at the core of any supply chain relation where the setting for buyer and supplier relationships aspire to create value, and that blockchain holds the potential to reduce governance costs and shift to the most optimal governance structure under given conditions. Roeck et al., (2019) assert that the design of active supply chain networks is dependent on understanding transaction costs, and therefore the effects of emerging technologies, whereby blockchain comes into the context.

3. Development of the transaction cost economics matrix

The section presents the development of the conceptual transaction cost economics matrix and the theory of transaction cost economics. Initially, the overall theory is explained. Further, the purpose, structure, content and application of the matrix are described and put in relation to supply chain transactional activities. Additional sections will gradually elaborate and explain the model’s components, concludingly presenting the transaction cost economics matrix.

3.1 Transaction cost economics

Transaction cost economics theory, hereafter referred to as TCE, originates from several perspectives and fields of use. TCE is an interdisciplinary theory which originates from law, economics and organisational theory. TCE is a useful theory in regards to choosing a proper governance structure to minimise total costs of different transactions (Schmidt & Wagner, 2019). The optimal organisation structure is, therefore, one that minimises the costs of an exchange, thus enabling economic efficiency (Williamson, 2007). Schmidt & Wagner (2019) explain that a transaction is considered as a unit of analysis within the theory and can be seen as an exchange of information, goods or services between different stages within a production line. It can, therefore, be the exchange of information as well as the exchange of physical goods. The costs themselves occur through the gathering and processing of information, contract negotiation, monitoring and arrangements of contracts and agreements and the management and maintenance of relationships.

A transaction is described, according to Williamson (1981), to consist of three different characteristics; uncertainty, frequency and investment idiosyncrasy. Uncertainty refers to the risk associated with a specific transaction (Williamson, 1981), frequency refers to whether the transaction that takes place is of one-time, occasional and recurrent occurrence (Williamson, 1979). Finally, the investment idiosyncrasy or termed as _{asset specificity} _{is, according to} Williamson (1981), the most crucial dimension for describing a specific transaction as well as the most neglected by previous studies of organisation. The specificity of an asset is described as whether an investment is specialised in conducting a specific transaction. More general use

(19)

assets are not as much of an investment risk since the parties can turn to the market in order to sell it by ease. The higher is the risk of higher asset specificity, where the investments are put into the conduction of a specific transaction between two parties, and for that transaction alone. An example of high asset specificity is high purpose technology, intended for a highly specific use case.

According to the TCE-theory, a firm is viewed as a governance structure (Grover, Varun, Malhotra & Manoj, 2003) where different transactions are supervised through hierarchies. The authors explain that the basic premise of the theory is to assess whether the cost of transactions, in other words, different types of exchanges, are within reasonable limits under the given conditions. It is therefore of interest to determine how these transactions ought to be governed. Grover & Malhotra (2003); Schmidt & Wagner (2019) explains that there are two key assumptions about the theory which can be summarised under the titles of; _{bounded rationality} and _{opportunism. It is explained that bounded rationality is what limits the individual in terms of} being able to receive, store, retrieve and communicate information without error. At the same time, Williamson (1975) defines it as the cognitive limitations of humans with regards to receiving and storing information. Put in context, Schmidt & Wagner (2019) explains it as even though an individual wish to act rational, the inability to process all available information leads to a limited rational decision. Bounded rationality is viewed as a problem, in times of uncertainty, where the contingencies of a future exchange are impossible to consider and are therefore harder to safeguard against. Grover & Malhotra (2003) explain that it directly relates to the difficulty of specifying the conditions in a contract regarding a specific exchange. Ketokivi & Mahoney (2020) adds that due to bounded rationality and information asymmetry, no other party can fully guarantee perfect execution of a transaction, despite the belief of being able to trust the other transacting party. According to Wathne & Heide (2000), opportunism arises through contracted and long term relations. These relations are controlled through contracts, forbidding or demanding specific actions, and are of great importance in a supply relation. These regulations may, e.g. control where and who a supplier might be allowed to sell components or goods to. Wathne & Heide (2000) explains that infringements in this aspect are relatively common. Infringements have a direct impact on the actors which the contract has been established between, since for example the sale of a product, can be to a competitor in the same market. Thus, when this type of transaction succumbs to opportunistic behaviour, it may indirectly be connected to financial losses as well as the loss of market position.

TCE intends to evaluate which governance structure, which can be either hierarchical or market-oriented, is appropriate in conducting a specific transaction and minimise transaction costs through safeguarding against opportunistic behaviour (Williamson, 1979). Contracts are one way to safeguard against opportunistic behaviour, under the assumption that the counterpart will stick to its terms and conditions (Loebbecke et al., 2020) which, as mentioned, is rarely the case since a contract never can be considered as genuinely complete due to bounded rationality (Grover & Malhotra, 2003) and hence leave room for opportunistic behaviour. According to Grover et al. (2003), the governance structure should, therefore, be chosen based on the level of opportunism present. Depending on the risk of opportunism, the need for governance commitment costs will vary. The TCE theory shares a common foundation with game theory,

(20)

where the parties of a contract are assumed to have an understanding for the strategic situation they are in, and position themselves accordingly (Williamson, 2007), where the two key goals are to maximise revenue and minimise financial losses.

Foss & Klein (2010) explains that the dimensions of transactions, apart from uncertainty, that primarily determine the cost are their frequency and asset specificity. Asset specificity is present when one or several parties are heavily invested in an asset required to conduct a transaction, which makes them more vulnerable to financial losses if the transaction does not carry through as planned. Furthermore, since a contract never can be considered complete, due to bounded rationality, it often needs to be renegotiated when uncertainties present themselves over time. If the renegotiation takes place under the knowledge that one actor is already heavily invested in the trade, it allows the other party to exploit their position, thus opening up for opportunistic behaviour.

3.2 Purpose of the matrix

Within this section, the purpose of the matrix will be presented, which combines the transactional activities with the transactional dimensions, creating nine separate cells open for an assessment concerning supply chain transactions and their related transaction activities. Along with this, further arguments are presented to justify its area of application as well as how the costs are measured, dimensions defined and last, which transactional instances that are included within the matrix.

The transaction cost economics theory is complicated and rather broad in its area of application. Therefore, a conceptual transaction cost economics matrix was created. The purpose of the model is to concretize, based on transaction activities and the dimensions of a transaction, transaction costs within supply chains that give rise to different costs, providing a qualitative framework through the lens of transaction cost economics for the assessment of supply chain activities. The matrix can contribute to the decision whether to make use of the blockchain technology, through a comparison between an assessment of the traditional supply chain and a blockchain supported supply chain against the framework. It may further bring practical value for managers seeking to make use of the blockchain technology within their supply chains handling physical goods and can aid to locate new effects and gain new insights in this regard, which otherwise would not have been possible.

3.2.1 Measuring transaction costs

As stated in earlier sections of the study, three activities have been found and categorized due to their characteristics. These individual activities relate to different introduced costs within the transaction cost economics theory. According to Eggertsson (1990, pp. 14-15); “When information is costly, various activities related to the exchange of property rights between

(21)

individuals give rise to transaction costs. These activities include:”

● The search for information about the distribution of price and quality of commodities and

labor inputs, and the search for potential buyers and sellers and for relevant information about their behavior and circumstances.

● The bargaining that is needed to find the exact position of buyers and sellers when prices

are endogenous.

● The making of contracts.

● The monitoring of contractual partners to see whether they abide by the forms of

contract.

● The enforcement of a contract and the collection of damages when partners fail to

observe their contractual obligations.

● The protection of property rights against third-party encroachment – for example,

protection against pirates or even against the government in the case of illegal trade.

Möller & Horváth (2004) explains that the measurement of a transaction cost is usually based on a comparison. It is not easy to stand-alone measure the transaction cost of a specific case without having a similar one to benchmark the underlying factors affecting the transaction against. This type of comparison is, therefore, suitable when evaluating different organizational arrangements or between alternative business partners. The three constituent dimensions of factors affecting the transaction costs are, according to Ketokivi & Mahoney (2020) & Williamson (1979) sources of differential governance costs.

In line with the study’s aim, only the transactional factor and cost categories are of importance. The discussed factors are; _{opportunism, bounded rationality, uncertainty, asset specificity,} _and

frequency. _{Ketokivi & Mahoney (2020) focuses more primarily on assessing the transactional} factors; uncertainty, specificity and frequency of transactions in order to determine their costs and consequently include the results when making operationalization decisions. Consistent with prior authors, Rindfleisch & Heide (1997) concludes that the transactions are characterized by the dimensions of uncertainty and asset specificity where Ketokivi & Mahoney (2020) & Schmidt & Wagner (2019) makes use of frequency as a third dimension. There is thus a consensus about the characteristics of a transaction in terms of dimensions that have to be evaluated when making use of transaction cost economics in the evaluation of supply chain operations.

Followed by its dimensions, there is a less shared view of the transactional activities that have to be evaluated and measured. Tate, Dooley & Ellram (2011) discusses the three activities, or also understood as three types of costs, which are _{information and search cost, bargaining cost}_and

enforcement costs, _{under which the different costs emerge. Similarly, McCann, Colby, Easter,}

Kasterine & Kuperan (2005) addresses these three types of transaction costs amongst others, instead referring to them as; _{research and information, contracting}_{and monitoring/detection and}

prosecution/enforcement_{. While other authors, such as Wacker, Yang & Sheu (2011) direct their} focus away from the more specific activities and pay attention to the attributes of the transaction

(22)

itself, such as asset specificity and uncertainty. Although the transaction itself being what is analyzed, several authors are rationalizing their measurement through specification in terms of different activities. Garfamy (2012), on the other hand, points out the factors producing transactional difficulties, which include bargaining, loss of resource control, and information asymmetry to mention a few. They are not explained as transactional activities but rather factors, which generate different costs and are further measured as a whole in order to state the transaction cost. Rindfleisch & Heide (1997), as mentioned, are also discussing the factors concerning the transactions in terms of the original three dimensions, albeit leaving out the more general explanation of the costs.

Schmidt & Wagner (2019) are exploring the blockchains effect on transaction costs within the supply chain relations network where the authors are addressing the activities of information and search cost while referring to it as the _{gathering and processing of information, bargaining by} talking about the cost of _{drafting and negotiating contracts, and lastly monitoring and enforcing}

agreements_{, which is synonymous with the activity of enforcement and monitoring. Macher &} Richman (2008) have examined several businesses within different sectors and discussed how the cost of contracting, which can be related to the cost of bargaining, correlated to the measures of frequency, uncertainty and asset specificity of that specific transaction. There is thus a common, but a somewhat discrete consensus about the use of transaction cost analysis in terms of dimensions and activities, solely defined and used differently in several parts of the literature. It is therefore of interest to specify the costs more appropriately for the evaluation of supply chain transactions and in a more precise manner, where every transaction activity is measured out of the three dimensions.

3.2.2 Dimensions

In this section the different dimensions of a transaction are briefly explained.

Uncertainty - Uncertainty refers to the risk associated with a specific transaction. Transactions conducted under high certainty are rather uninteresting and any governance structure will do. Uncertainty refers to the predictability of the number of modifications by performance properties like quality, time, price or volume during a transaction (Williamson, 1985). The more these modifications vary in time, the higher the uncertainty. The uncertainty may be based on future events or changes in the environment (Macher & Richman, 2008), which have a direct effect on the transaction itself.

Asset specificity - _The investment idiosyncrasy or termed as _{asset specificity} _{is, according to} Williamson (1981), the most important dimension for describing a transaction as well as the most neglected by previous studies of organization. The specificity of an asset is described as whether an investment is specialized in conducting a specific transaction. More general use assets are not as much of an investment risk since the parties can turn to the market in order to sell it by ease. The higher is the risk of high asset specificity, where the investments are put into the conduction of a specific transaction between two parties, and for that transaction alone. An example of a high asset specificity is high purpose technology (Ibid.), where a lock-in effect is

(23)

created where a buyer and a seller are highly dependent on each other in order to conduct the transaction. According to Wever, Wognum & Trienekens (2010), when high levels of asset specificity are present, the actor making these investments tend to seek more hierarchical governance structures due to the risk of opportunism.

Frequency - _Refers to whether the transaction that takes place is of one-time, occasional or recurrent occurrence, in other words how frequent. Transaction frequency has received the least attention in empirical literature out of the three dimensions (Macher & Richman, 2008).

3.2.3 Activities

Within this section, the three different types of transaction activities are presented. The activities are considered categories and are sorted by their character and purpose.

Information and search cost - _{According to Tate et al., (2011) are information, or also known as} search costs, the costs that arise whenever information and data are gathered in order to determine the suppliers best course of action. The gathering of information is preemptive since it decreases the level of uncertainty as well as decreasing the risk of encountering moral hazards, as well as proactive due to better understanding a party’s genuine commitment to the contract. These types of costs could be the process of gaining knowledge of new technology, new ideas or acquiring a new competency. In times of uncertainty, information and search costs can become significant and are therefore of importance to evaluate. The assessment of information and search costs could be of great importance when balancing the information asymmetry that leads to transaction risks (Grover & Malhotra, 2003). For instance, if a buyer requests something from a supplier, the buyer probably has more knowledge about environmental practices than the supplier. In this situation, the supplier is at a disadvantage because of information asymmetry (Lee & Klassen, 2008). According to Tate et al. (2011), information and search costs can become significant in the face of uncertainty since the higher uncertainty, further information has to be gathered in order to make a decision, which generates costs in terms of spent time and resources.

Bargaining - _{The cost of bargaining is defined as “the direct cost of negotiating, documenting} and enforcing an agreement” (Knez & Simester, 2000, 3). The costs are mainly based on the time and effort it takes to negotiate an agreement. In times of uncertainty, bargaining is more time consuming, and renegotiation is more likely to take place (Tate et al., 2011). The cost may act as a way to decrease the risk of opportunism which instead makes the cost seen more of an investment since the risk of exploitation or deviance from the agreement by a counterparty is less likely to take place.

Enforcement and monitoring - _{Tate et al. (2011) explain that the costs of monitoring or enforcing} are related explicitly to reducing opportunism from counterparties. Opportunism is likely to occur when performance is not monitored or when behaviour is challenging to define as well as not appropriately monitored. It is therefore of interest, in some instances, to design specific

(24)

incentives, which consequently makes the performance and behaviour more easily measured. Some suppliers might need specific systems in order to measure its performance or to corroborate the buyer’s assessment of its performance. The activity of enforcement and monitoring refers to the ex-post coordination cost once an agreement is entered (Roeck et al., 2019).

3.3 Transaction cost economics matrix

In this section, the dimensions and activities are cross-matched in the TCE-matrix, creating nine different cost aspects open for an assessment of supply chain transaction activities. Every cell is a unique combination of activity and dimension.

According to Williamson (1975, 1985), a transaction is the exchange of information, goods or services between different steps in the line of production. The transaction itself can be, according to Tate et al., (2011) divided into three separate cost categories, also explained as activities. Information and search is the cost that arises whenever resources and time are invested in the retrieval of information to determine the suppliers best course of action. Bargaining is the process and resources spent negotiating and renegotiating contracts, and enforcement & monitoring, the process of making sure that the counterpart is compliant with the terms of the contract. Each transaction can be assessed based out of its three dimensions; _{uncertainty, asset}

specificity _{and frequency (Williamson, 1985). The matrix combines the stated activities with the}

dimensions of a transaction, embodying the nine different areas of assessment for the context of supply chain transactions.

Costs / Dimensions Uncertainty (1) Asset Specificity (2) Frequency (3) Information and

search (A)

The predictability of the number of modifications in the process of information gathering (A1)

The level of specificity in information gathering assets (A2)

The frequency of which certain information is requested (A3)

Bargaining (B) The level of uncertainty when bargaining with suppliers (B1)

The specificity of one or more parties investments to conduct a transaction (B2)

How often the bargaining for an agreement takes place (B3)

Enforcement and monitoring (C)

The level of uncertainty in an entered agreement (C1)

The chosen monitoring structure after investment specificity (C2)

How often the activity of enforcement and monitoring takes place (C3)

(25)

Table 1 describes the three different activities of transaction costs, stated by Tate et al., (2011) about the three dimensions characterizing a transaction by Williamson (1985). Based on the discussion in 4.2.1, most empirical literature discusses the measurement in several, independent ways, although the theory lacks a common understanding. The matrix is combining the transaction activities with the transaction dimensions, making it practical for the qualitative assessment of supply chain transaction management. Following is an explanation of each cell in the TCE-matrix.

3.3.1 Information and search costs dimensions

A1. _{In terms of uncertainty, when acquiring a specific type of information within the supply} chain, the first cell _{(information and search/uncertainty)} _{relates to the process of information} search. The cell addresses the level of certainty about who in the supply chain to contact or request information from, e.g. requested information concerning a transaction from a supplier. Uncertainty is low when the party requesting the information are sure where to turn to in order to retrieve the information. High uncertainty is when the party requesting the information has restricted or no knowledge of where or from whom to retrieve the specific information of interest.

A2. _{Concerning asset specificity, the second cell (information and search/asset specificity) refers}

to the specificity of assets used in order to conduct the process of information gathering. In the supply chain, it is of utmost importance to be able to communicate with the suppliers in the chain and to maintain an efficient flow of information (Lambert & Cooper, 2000). The cell addresses the idiosyncrasy of the assets carrying through the information, such as different communication platforms or systems. Low levels of asset specificity in this regard refers to the asset that carries through the stream of information and communication is not specialised in carrying through this specific transaction. In contrast, high asset specificity is where the communication asset is intended specifically for that transaction, and that transaction alone.

A3. _{Frequency, the third cell (information and search/frequency) refers to the occurrence of an} information transaction that takes place. Explained how frequently certain information is requested. Low frequency is transactions that are of less frequent occurrence, where the high frequency is, e.g. standard and regular transactions such as weekly reports from suppliers or status updates.

3.3.2 Bargaining dimensions

B1. _{The activity of bargaining in terms of uncertainty, the fourth cell (bargaining/uncertainty),} refers to the degree of familiarity with another actor’s business processes and behaviour. If uncertainty is low, the bargaining process is potentially shorter, since the actors involved know what to expect from the interaction to a higher degree. Conversely, if uncertainty is high, the bargaining process tends to be more complicated or not to take place at all. Uncertainty indicates

How well are the two chains linked together?