Technological Forecasting & Social Change 146 (2019) 94–102

Contents lists available at ScienceDirect

Technological Forecasting & Social Change

journal homepage: www.elsevier.com/locate/techfore

Blockchain-based platforms: Decentralized infrastructures and its boundary T

conditions
Joana Pereiraa, , M. Mahdi Tavalaeib, Hakan Ozalpc
⁎

a
Leeds University Business School, University of Leeds, United Kingdom
b
Surrey Business School, University of Surrey, United Kingdom
c
KIN Center for Digital Innovation, School of Business and Economics, Vrije Universiteit Amsterdam, The Netherlands

ARTICLE INFO ABSTRACT

Keywords: Blockchain technology has been receiving much public attention recently, promising to disintermediate trans-
Blockchain technology actions through decentralized governance and distributed data-infrastructures. However, the majority of the
Platform ecosystems previous studies have focused on the technical aspects, and overlooked blockchain investigation from a man-
Decentralized governance agerial perspective. In this paper, based on platform-ecosystem, transaction cost economics, and open-source
Transaction cost economics
literature, we contrast and compare blockchain-based platforms and centralized platforms; in other words,
Smart contracts
decentralized versus centralized governance modes. We base our conceptual analysis on three dimension-
s—transaction cost, cost of technology, and community involvement—, exploring the conditions under which
blockchain-based platforms are more advantageous than centralized platforms. We first compare gains from
lower opportunism and uncertainty costs thanks to protocols and smart contracts in blockchain technology
versus the costs of higher coordination and complexity of (re)writing those contracts. Second, we compare the
gains from immutability and transparency in blockchain-based platforms versus the technological costs of
verification and storage of a distributed ledger. Finally, we compare intrinsic and extrinsic motivations of the
communities around centralized and blockchain-based platforms in the short and medium term.

1. Introduction void, we build on platform governance, transaction cost economics, and

open source communities literatures to investigate the costs and ben-
Despite the hype around blockchain technology, the main attempts efits of adopting blockchain technology as a decentralized platform
to understand such technology have been mainly restricted to the infrastructure, exploring the boundary conditions and the trade-offs
technical aspects of the blockchain protocols and foundations, or the involved in the adoption of such technology. Blockchain enables
finance of crypto-currencies such as Bitcoin (Risius and Spohrer, 2017). property rights transfer and exchanges built upon decentralized gov-
Nonetheless, implications of the blockchain technology reach far be- ernance and distributed data infrastructure (Catalini and Gans, 2017),
yond the financial system (e.g., De Filippi, 2017; Li et al., 2018). in opposition to centralized platforms that present centralized govern-
Consensus protocols, smart contracts, cryptography, and distributed ance and data infrastructure (e.g. Adner and Kapoor, 2010; Jacobides
ledgers allow for secure, immutable, transparent, and often cheaper et al., 2018; Tiwana et al., 2010). Thereby, we contrast and compare the
transactions, which can be applied to a variety of contexts (Halaburda, centralized versus decentralized governance mechanisms, as in (con-
2018; Tschorsch and Scheuermann, 2016). As a consequence, various ventional) centralized platforms and blockchain-based platforms, re-
digital platforms and start-ups have started adopting blockchain tech- spectively to understand under which conditions are blockchain-based
nology for micropayments, storage system, intellectual property, fi- platforms more advantageous compared to centralized platforms.
nancial and physical assets, supply chain and logistics, social networks, Decentralization of governance and data-infrastructure in block-
media and open science amongst others applications (Davidson et al., chain-based platforms (Halaburda, 2018) can to a certain extent miti-
2018; Li et al., 2018). gate centralized platforms inherent problems, such as high bargaining
A broader understanding of blockchain and its peculiar attributes, power for the platform sponsor, lock-in effects, censorship, data
from organizational and managerial perspective, is less explored leakage, expropriation, and privacy risks (Catalini and Gans, 2017).
(Constantinides et al., 2018; Risius and Spohrer, 2017). Filling this Thereby, we discuss conceptually, how the promising properties of

⁎
Corresponding author.
E-mail addresses: [email protected] (J. Pereira), [email protected] (M.M. Tavalaei), [email protected] (H. Ozalp).

https://doi.org/10.1016/j.techfore.2019.04.030
Received 14 December 2018; Accepted 29 April 2019
Available online 04 June 2019
0040-1625/ © 2019 Elsevier Inc. All rights reserved.
J. Pereira, et al. Technological Forecasting & Social Change 146 (2019) 94–102

blockchain may, in turn, cause other challenges such as coordination value for the ecosystem and platform users (Ozalp et al., 2018). There is
and complexity problems, and increased cost of verifying and storing also an indirect network effect between the users and complementors;
transactions in absence of a third party. We categorize the benefits and users are better off by a high rate of complementors' participation (thus,
costs arising at different levels pertaining to transaction costs, tech- complements variety), and vice versa (e.g., Parker and Van Alstyne,
nology costs, and community involvement. Based on these three para- 2005).
meters, we propose a framework to circumscribe the boundary condi- Accordingly, a sustained rate of co-specialised innovation and pro-
tions for adopting blockchain-based platforms vis-à-vis centralized duct offerings by complementors intertwined with growth in platform
platforms. We first contrast the lower transaction cost associated to adoption by users are pivotal for the platform success and survival. The
reduced opportunism and uncertainty with the higher coordination and platform sponsor, therefore, should apply appropriate governance me-
complexity costs of changing rules for the blockchain platform when chanisms to motivate third-party firms to join the ecosystem and make
contracts need to be amended (i.e., in situations when uncertainty is investment, orchestrate the innovation process, regulate the access and
higher). Second, we show that while blockchain technology provides interaction amongst users and complementors (Boudreau and Hagiu,
benefits resulting from immutability and transparency (i.e. tamper-re- 2009), enhance the network effect and attract users— in a nutshell, to
sistance, fraud prevention, cost of auditability) it can be too costly at manage the value co-creation and value capture processes within the
both verification and storage levels. Lastly, we also show how block- platform ecosystem (Ceccagnoli et al., 2012).
chain-based platforms leverage on intrinsic and extrinsic (crypto-in- The primary mechanism, which has been studied in a vast body of
centives and reputation effects) benefits to attract participants, in op- literature (e.g., Rochet and Tirole, 2003; Weyl, 2010), is the pricing
position to centralized platforms that mainly leverage on extrinsic structure. The decision about fixed membership fee and/or per-trans-
benefits. action fee and cross-subsidisation pricing strategy are amongst the es-
Our paper contributes to the emerging, yet nascent, body of litera- sential pricing structure decisions to manage the platform ecosystem
ture about the potential and limitations of blockchain technology (see (Rochet and Tirole, 2006).
Risius and Spohrer (2017) for a recent review of the previous studies). An array of non-pricing instruments also exists for platform gov-
In particular, we provide an early answer to the question “how does ernance, such as exclusivity contracts with certain complementors (e.g.,
blockchain technology address misaligned incentive structures and Cennamo and Santalo, 2013), platform sponsor decision to develop in-
trust currently faced by digital platforms?” (Constantinides et al., 2018; house complements, i.e. entry to the complementors market (e.g.,
p. 11). This paper explores the costs and benefits of blockchain-based Gawer and Henderson, 2007), quality assurance and certification for
platforms, drawing the boundary conditions of its applicability. The the complementors and their products, or designing the entry rules. The
adoption of blockchain technology is not only a question of technology last one, perhaps the most-studied topic in non-pricing governance
cost (see Catalini and Gans, 2017), which is rapidly decreasing, but is mechanism, pertains to the platform openness (e.g., Boudreau, 2010;
also a question of governance costs. This paper aims at exploring the Eisenmann et al., 2009; Parker et al., 2017) and the extent to which the
tipping point of the trade-off between the cost and benefits of tech- platform applies exclusion/restriction policies for complementors to
nology and governance modes to answer the question: “under which affiliate with the ecosystem. This openness can occur (vertically) at the
conditions shall transactions be conducted in blockchain-based plat- complementors' and users' level, or (horizontally) at the hardware and
forms in contrast to centralized platforms?” technology interface level. It can also be even deeper at sponsorship and
Before we explore when blockchain-based platforms are more ad- governance level (Eisenmann et al., 2009). For instance, Apple is less
vantageous than centralized platforms, we briefly explore the differ- open than Google concerning the complementors' entry to the app
ences between centralized and blockchain-based platforms across gov- store. Also, while Android is an open platform to various hardware
ernance and data-infrastructure dimensions. developers (such as Samsung, HTC, etc.), iOS has remained closed to
only Apple's iPhone. On the other hand, open software such as Linux are
almost open platforms even to at the governance and design rule level.
2. Theoretical foundations
This is directly related to the concept of proprietary versus shared
platform. The former is when the core functionality of the platform is
2.1. Centralized platform governance
under the control of a single sponsor (as in the case of Apple), while in
the latter the platform sponsorship is shared collectively (e.g., Linux
Platform ecosystems are increasingly dominating the business
open software or Visa owned by an association of several banks). Later,
landscape. The so-called FAANG companies (Facebook, Amazon, Apple,
we discuss more both the regulatory role of the platform and the right
Netflix, and Google), the “motors of S&P 500”,1 are only few infamous
degree of openness in the “The boundary conditions: blockchain-based
examples of platform ecosystem, which is a prevalent model, especially
platforms vis-à-vis centralized platforms” section.
in digital industries, from smartphones, videogame consoles, media-
based and video-on-demand portals, to Internet of Things (IoT) plat-
2.2. Blockchain-based platform
forms and wearable devices. In these ecosystems, usually, a central firm
sponsors the core components and interface upon which third-party
The blockchain technology encompasses the protocol that defines
firms (i.e. complementors) develop and offer their complementary
the main rules that will govern the platform functioning and the data
products (i.e. complements) to the end-users (e.g., Adner and Kapoor,
infrastructure, and smart-contracts, which are self-executing contracts
2010; Jacobides et al., 2018; Tiwana et al., 2010). The platform sponsor
that enable automated transactions (Buterin, 2014; Davidson et al.,
(e.g., Apple), at the core of the ecosystem (e.g., iOS App Store), facil-
2016a). In this section, we will compare Blockchain-based platforms'
itates the interaction between the complementors (e.g., app developers)
governance and data infrastructure with centralized platforms.
and users (e.g., mobile users), for instance by reducing the search cost;
Blockchain-based platforms, either proprietary or non-proprietary,
it also provides complementors with a common set of technology,
tend to present a decentralized decision-making in which the commu-
boundary resources (e.g., APIs and SDKs), and marketing capabilities
nity around the platform not only suggests changes to the code and
(e.g. featured apps and top charts) (Ghazawneh and Henfridsson,
rules of the platform (by committing to codes usually in GitHub) but
2013). Complementors, benefiting from participation and co-speciali-
also decides which of these changes will be implemented through
zation in the ecosystems, build their product offering, hence creating
forums, discussion groups, or voting systems. For example, Satoshi
Nakamoto launched the Bitcoin protocols (which also encompasses the
1
See here: https://www.economist.com/business/2018/08/04/the-tech- blockchain technology itself) to the community, and nowadays the
giants-are-still-in-rude-health. community around Bitcoin maintain the protocol and decide about the

95
J. Pereira, et al. Technological Forecasting & Social Change 146 (2019) 94–102

Table 1
Main dimensions of centralized platforms and blockchain-based platforms.
Platform dimensions Centralized platforms Blockchain-based platforms

Governance Decision-making Centralized Decentralized

Entry rules Always permissioned Permissionless
Verification of transactions Centralized Decentralized
Incentives Pricing mechanisms Crypto-incentives
Data infrastructure Ownership Proprietary Distributed
Accessibility Private access Public access
Examples Facebook, Amazon, Apple, Netflix, and Google Steemit, Bitcoin, Ethereum

directions of the technology through soft and hard forks of the code network effects.
(Böhme et al., 2015). Some proprietary centralized platforms have also Apart from being a new paradigm for governance (i.e. at protocol or
initiated open source movements, as is the case of Android; however, application level), blockchain technology at the data infrastructure
while community members can make suggestions to amend the code, level (mainly ownership and accessibility) is also fundamentally dif-
they are not able to decide which suggestions are actually implemented. ferent from centralized platforms. While in centralized platforms, the
The decisions about the future direction of the platform, which coincide platform sponsor owns and controls the access to data, in blockchain-
with its technology components and interface, are centralized in the based platforms, the ledger of transactions, which stores the history of
proprietary in centralized platforms. all transactions, is stored in many locations simultaneously in a dis-
The entry rules determine who is allowed to participate in the tributed fashion (Nakamoto, 2008). As the distributed ledger is re-
platform, being related to the degree of the platform openness in op- plicated across the network nodes, if there is any attempt of a node to
position to censoring. Centralized platforms often directly regulate the falsify a transaction (e.g., double spend), the moment that this node
access and membership, requiring users and complementors' authenti- ledger is checked against all the other nodes' copies, the falsification is
cation (Boudreau and Hagiu, 2009). Despite recent variations, block- spotted and automatically corrected. This redundancy of information,
chain original conception relies on freely open membership, also known along with the verification mechanism described earlier, ensures se-
as permissionless blockchain. In this sense, blockchain-based platforms curity, immutability, and transparency of transactions (Atzori, 2015;
are both horizontally open (at infrastructure technology and interface Risius and Spohrer, 2017).
level) and vertically open (at complementors and users level) Blockchain-based platforms are based on decentralized governance
(Eisenmann et al., 2009). and data infrastructure, which allows marketplace agents to transact
Blockchain-based platforms also differ from centralized platforms directly with each other without the need for a trusted intermediary
regarding verification processes, which obey to a pre-agreed consensus (Catalini and Gans, 2017; Davidson et al., 2018; Nakamoto, 2008).
mechanism. While in centralized platforms, the platform owner is the Blockchain-based platforms, thus, represent an extreme case of “open-
entity validating transactions and deciding which transactions are valid ness” with decentralized governance and a distributed data infra-
or not; in blockchain-based platforms, an independent pool of valida- structure able to disintermediate transactions. Such disintermediation
tors verify the transactions (known as miners in the Bitcoin blockchain). can reduce transaction costs (Halaburda, 2018) and failures inherent to
These validators need to follow a consensus mechanism based on peer- centralized platforms, such as lack of transparency, corruption, coer-
to-peer cryptographic verification process to be able to validate blocks cion, censorship, and excessive market power (Atzori, 2015; Catalini
of transactions, creating a secure, immutable, transparent, time- and Gans, 2017). Table 1 summarizes the characteristics of centralized
stamped public ledger (Davidson et al., 2018). Validators follow a versus blockchain-based platforms.
verification mechanism that allows reaching consensus about which
transactions are true and eligible to be added to a block of transactions. 3. The boundary conditions: blockchain-based platforms vis-à-vis
Each block of transactions links to the previous block, forming a con- centralized platforms
tinuous chain back to the original first block of all (Davidson et al.,
2018). Blockchain, thus, enables a trustless verification system that Scholars defend that blockchain has the potential to improve effi-
does not require a third party to verify transactions. Instead, it applies a ciency of some economic operations, moving them closer to a peer-to-
verification system to ensure consensus amongst users about the true peer ideal (Davidson et al., 2018). The answer to the question why
state of the ledger, fuelled by crypto-incentive to involve validators in a some transactions occur in blockchain-based platforms rather than in
disintermediated verification process (Davidson et al., 2018). centralized platforms is because blockchain-based platforms can reduce
Decentralization of decision-making and verification processes re- transaction and technology costs, and foster community involvement,
quire high levels of participation of the community around these plat- in comparison to centralized platforms. Yet, these advantages can be
forms either by producing, consuming, voting, coding or verifying offset by some shortcomings such as higher coordination, complexity,
transactions. Crypto-incentives fuels such participation, encompassing verifications and storage costs, and lower intrinsic benefits in the
crypto-tokens and cryptocurrencies. Crypto-incentives are blockchain medium-term. Building on these costs and benefits, we propose three
fungible and tradable assets able to be exchanged inside the platform to main boundary conditions under which adopting blockchain-based
buy complements or converted into other crypto-currencies or fiat platforms is more beneficial than centralized platform. We discuss each
currencies (such as USD or EUR) outside the focal platform. Blockchain of these conditions as follows.
community members can acquire crypto-tokens or cryptocurrencies
through Initial Coin Offers (ICO), crypto exchanges, or instead, earn
3.1. Transaction costs and smart contracts
them by performing some activities inside the platform. The crypto-
incentives fuel participation and verification of transactions, securing
Transaction cost economics (TCE) focuses on “transactions and the
the maintenance of the platform (Davidson et al., 2018). The crypto-
costs that attend completing transactions by one institutional mode
incentives in blockchain-based platforms are the equivalent to pricing
rather than another” (Williamson, 1975: 1–2). In particular, it focuses
structures and non-pricing instruments in centralized platforms, func-
on the relative efficiency of organizing through markets, hybrid forms,
tioning as a coordinating mechanism which is essential to attract users,
or hierarchies, with the main unit of analysis being a transaction
complementors, developers, and validators; hence, boosting the
(Williamson, 1985). This theory predicts that organizations choose the

96
J. Pereira, et al. Technological Forecasting & Social Change 146 (2019) 94–102

most efficient (TCE economizing) way of organizing depending on the platforms for the transactions that can be rendered as complete con-
nature of transactions. Guided by two assumptions of uncertainty and tracts, lowering the transaction costs for such activities through irre-
opportunism, three factors regarding the nature of transactions de- versible, transparent, and automated codes of contracts (Davidson
termine the choice of organization in TCE: asset specificity, uncertainty, et al., 2018). Therefore, it is reasonable to admit that we do not need
and frequency. Asset specificity relates to the nature of investments in banks to perform certain types of transactions anymore, because such
the transaction–if some assets that are required for the particular transactions are easy to parameterize and perform through Blockchain
transaction cannot be used elsewhere without loss of (significant) protocols; however, if you look for advice on specific financial assets
productive value, then the asset specificity and the bilateral de- and a customised treatment, which is a difficult transaction to para-
pendency between parties is high, which makes contracting through meterize, one might recur to traditional banks.
markets hazardous. Uncertainty relates to the ex-ante haggling and ex- In sum, blockchain-based platforms, which rely on smart contracts
post bargaining and affects transactions only when there is some non- and publicly available distributed ledgers, can replace centralized
trivial level of asset specificity (David and Han, 2004). In those con- platforms when contracts are (quasi-) complete by reducing transaction
ditions with some asset specificity, as uncertainty rises, markets become costs related to opportunism and uncertainty.
a less economic way of organizing compared to firms. Frequency is the There is, however, an additional set of costs that blockchain incur
last dimension, and it relates to the need of monitoring the transac- relating to coordination (Arruñada and Garicano, 2018) and com-
tions–as frequency increases, more resources are required for mon- plexity. Blockchain transactions are irreversible, being highly inflexible
itoring the transactions in a market, and therefore hierarchical firm and restrictive in their nature. When consumers have a contract with a
represents a better alternative when transaction frequency is high. bank that takes care of their money, they incur the risk of misuse of
Uncertainty and opportunism relate these factors in different ways. their money and data, but they also expect a customised attendance
Asset specificity is generally the stronger element in determining or- based on their needs and some degree of reversibility in case there is a
ganizations' choices, and it is closely connected to opportunism. mistake in a transaction or if the money got stolen. In blockchain, as the
Although not everyone will be opportunistic, there is always the risk of protocol and smart contracts are restricted to a certain amount of op-
opportunism–or as Williamson (1975) puts it forward “self-interest erational conditions that tend to be more standard than customised, it is
seeking with guile”. On the other hand, uncertainty is more related to unlikely that the contract predicts all contingencies, as unintentional
the bounded rationality–the fact that humans have “limited informa- mistakes. Ultimately, as the full code, including protocol and smart
tion, attention, and processing ability” (Simon, 1945), which gives rise contracts, is open source, new conditions could be inserted into ma-
to contractual incompleteness. chine-readable contracts that can indeed be altered in the future.
Blockchain, as a technology is quite relevant to approach from a Davidson et al. (2016b, 2018) argue that the complexity cost of im-
TCE perspective as the technology itself has the purpose to disin- proving or changing contracts would scale linearly, while the transac-
termediate transactions, reducing transaction costs associated to op- tion costs would decrease over time. However, this view ignores that
portunism and uncertainty. Blockchain-based technology encompasses coordination costs increase in a growing open community as everyone
the protocol and self-executable smart contracts, which trigger trans- can suggest changes in the code and have voting rights. Such level of
actions automatically under certain conditions (Iansiti and Lakhani, coordination includes key issues in blockchain-based platforms that
2017). The fact that the protocol and the smart contracts are defined ex- lead communities to split-up through “hard forks” lead by minorities
ante and that smart-contracts are automatically triggered reduces op- and inertia caused by the expectation of split up or not getting the re-
portunism in transactions (Davidson et al., 2018), especially relating to quired number of votes for the change (Arruñada and Garicano, 2018).
ex-post hold-up costs across parties in a transaction. Smart contracts This is the point when communities engage in extensive and intractable
also reduce transactions' uncertainty regarding information problems discussions. Community extensive discussion, split-ups, and inertia lead
(Davidson et al., 2018) as the contract automation guarantees that to inefficient outcomes, which we highlight as an increased coordina-
under certain conditions the output will be the same and irreversible tion cost (Arruñada and Garicano, 2018).
(see Williamson (1973), for sources of transactions uncertainty). In this Contractual changes in blockchain-based platforms increase com-
sense, the degree of uncertainty about the execution conditions and plexity and coordination costs not only at the community level but also
output of a certain transaction decreases, reducing the transaction at a technological level. Davidson et al.'s (2018) argument that open
costs. Additionally, the cost of writing the protocol and smart-contracts databases allow for reduced costs of writing contracts is simply not
is spread by the amount of transactions that on the limit can present an feasible. Increasing complexity, by adding contingencies to the protocol
infinite frequency due to its automated nature (Davidson et al., 2016a). or smart-contract, may cause big issues at the code level, as code bugs
Finally, as the blockchain technology suppresses the need for an in- often lead to security breaches. For example, such problems led De-
termediary—the traditional platform owner—contributes also to miti- centralized Autonomous Organization (DAO) to be hacked, what
gating the hold-up risk caused by intermediary agent itself that could eventually required a “hard fork” on the system, overriding the irre-
intervene in the transaction in order to realize individual gains through versibility principal and causing internal turmoil (Arruñada and
lack of candor or honesty (Davidson et al., 2016a; Williamson, 1973).2 Garicano, 2018). Such increasing code complexity can force transac-
The blockchain, however, generally requires complete contracts as tions to have a lower number of conditions. This then suggests that the
they get executed under certain conditions without intermediaries addition of conditions to protocol and smart contracts exponentially
(therefore require full extent of contingencies), as opposed to compa- increases costs due to an increase of security and uncertainty costs,
nies that exist as a nexus of incomplete contracts (Davidson et al., which we call complexity costs.
2016a; Hart and Moore, 1990; Wright and De Filippi, 2015). Never- In sum, blockchain-based platforms can handle (quasi-)complete
theless, some blockchain-based platforms are able to offer services contracts with lower opportunism and uncertainty costs, therefore
usually performed by traditional firms. One example is Bitcoin that up providing a transactional cost advantage. Yet, blockchain-based plat-
to a certain extent replaces banks, and another is Steemit, which is a forms get costlier when coordination and complexity is an issue (e.g.,
public blockchain-based platform for content generation, equivalent to platforms with a massive number of members), mainly when high in-
Facebook. Blockchain-based platforms are able to substitute centralized completeness requires too many conditions to enact transactions.
P1. When the gains from reduced opportunism and uncertainty costs
2
outweigh the losses from increased costs of coordination and complexity,
However, it is important to note that ex-ante and ex-post bargaining and the blockchain-based platforms are more advantageous than centralized
renegotiation costs may not be eliminated with blockchain (Davidson et al.,
platforms.
2016a)

97
J. Pereira, et al. Technological Forecasting & Social Change 146 (2019) 94–102

3.2. Cost of technology and verification mechanism complementors to develop. Conversely, in the blockchain-based plat-
forms, as a distributed ledger, the records of all transactions are stored
Not only can blockchain reduce the transaction cost, but it may also on all the nodes of the network (i.e. users of the platform). In other
lessen the initial building and ongoing activity costs directly related to words, the core of the platform is not owned by a single sponsor but
the technology. Efficiency also comes from deleting layers of activity shared and distributed across the users. As a copy of every piece of
that are no longer needed because a trusted third party is not required information is available on each node, the data manipulation and
anymore (Davidson et al., 2018). A central intermediary platform for failure become even more difficult (on top of difficulties due to ver-
securing the transaction, generating trust, and maintaining the data can ification and consensus mechanism) and easily detectable. The double
open up the risk of data breach, privacy risks, and censorship risk storage of data also provides transparency to the nodes of the network.
(Catalini and Gans, 2017). Blockchain, by alleviating these risks and On the one hand, the more nodes have a trace of the data; the data
vulnerabilities, can enhance productivity. Yet, like other new technol- becomes more tamper-proof. On the other hand, scaling up the network
ogies, blockchain, while beneficial in some dimensions and materi- means participation of more users with access to the transaction data
alizing previously non-existent opportunities, may suffer from draw- and ability to (dis)validate transactions. The blockchain protocol as a
backs and inefficiencies on some other dimensions. “trust machine”3 does not rely on the trustworthiness of the users;
We classify the cost (dis)advantages of blockchain technology in however, the possibility of misconduct and fraudulent attempts cannot
two categories; first, at the protocol and application level. For example, be excluded. Scaling up, for example, being an entirely public platform
blockchain enhances the efficiency of international money transfer via without any entry rules, can increase the likelihood of misconduct,
omitting the cost of intermediation, process, and verification (Catalini which creates diminishing returns to trustworthiness or deterioration of
and Gans, 2017). Each transaction (be it financial as in Bitcoin or other “peripheral trust” (Evans et al., 2016). These two opposite forces
types of transaction such as property transfers) to be added to the chain foreground a scalability trade-off for the platform—it is more difficult
of existing blocks needs to go through a verification process and con- to fool many; yet, the probability of cheating increases with being open
sensus mechanism (such as the proof-of-work). The proof-of-work in- to many.
volves solving a randomized mathematical puzzle, which is compli- Additionally, distributed ledger brings cost savings by replacing the
cated to solve but easy to verify by other nodes of the peer-to-peer central servers and infrastructures with a peer-to-peer network. Yet,
network. The validators of the blockchain (also called operators or duplicating all records and updates of the data can make the re-
miners) compete with each other to solve this puzzle which generates a conciliation and integrity of the ledgers slower and more costly, which
number, called a hash, to encrypt and seal the blocks of the recent deteriorate exponentially as the size of the network increases, which
transactions. Upon verification of the hash value by other nodes, the puts additional restriction to the scalability of blockchain-based plat-
new block will be attached to the blockchain. Each block contains the forms.
hash value of its own as well as that of the previous block; hence, the Finally, the cost advantage of blockchain technology is accentuated
blocks are linked securely to each other. if the assets in transfer are purely digital, an ideal example of which is
Manipulating a single piece of information in the blockchain not the Bitcoin. However, when the transactions represent some offline or
only requires to generate a new hash for the given block; it also needs to physical entities (such as the usage of blockchain in supply chain or real
alter all the consequent blocks´ hashes, so that no one can detect the estate sector) both the verification and storage become more costly, i.e.
break in the chain, a task which is nearly impossible in practice (e.g., less technological cost advantage at both protocol/application and in-
Böhme et al., 2015). This protocol makes the information stored on the frastructure stacks. There should be a reliable link between the digital
blockchain tamper-proof and immutable without any need for a central record and the corresponding event in the physical world, which ne-
intermediary or a trustable third-party. This disintermediation, there- cessities the existence of some trustable parties (which are called ora-
fore, cuts the fee that would have been charged by the intermediary for cles), without which the blocks may be a secure and immutable record
conducting the transactions and operating costs of such platforms, and of merely some fake assets/incidents (Catalini and Gans, 2017).
mitigates the risk of double-spending, data manipulation, and cost of Keeping a strong link between the two and solving this gateway pro-
auditing, amongst others, all of which has been called as cost of ver- blem (Halaburda, 2018) increases the cost of verification (i.e. multiple
ification by Catalini and Gans (2017). They argue that blockchain parties and agreed rules to verify the data entry and authenticity of the
technology makes a costless verification possible. link) and/or the cost of infrastructure and storage (i.e. hardware de-
However, the reality is different than the idealized case portrayed vices such as GPS, RFID, or internet of things to substantiate and store
above. The consensus and verification mechanisms, which guarantee the online record of the offline world).
the immutability and transparency of the stored transactions, hence
P2. P2.When the gains from immutability and transparency of transactions
generating trust via protocol and codes without a need to a trustable
outweigh the losses from increased technology cost of verification and
party, inherit some restrictions. For instance, conducting transactions
storage, the blockchain-based platforms are more advantageous than
via Bitcoin network is still slower and less efficient than Visa or PayPal
centralized platforms.
(Davidson et al., 2018). In fact, the potential throughput in the Bitcoin
is up to seven transactions per second, compared to two thousand
transactions per second in Visa (Yli-Huumo et al., 2016). Moreover, the
3.3. Community involvement and crypto-incentives
proof-of-work is indeed an energy consuming and capital-intensive
task. It can cost “approximately $178 million per year at average US
We have not yet discussed the role of the collective of individuals
residential electricity prices” (Böhme et al., 2015: 218). In fact, the
that build and sustain blockchain-based platforms. Similarly to other
effort required to keep the immutability of the blockchain (time, en-
online communities, specifically open source communities, in block-
ergy, computing power, etc.) through proof-of-work should remain
chain-based platforms, the community encompasses individuals who
difficult and costly enough, despite any feasible increase of computing
communicate, interact, and develop relationships, in order to collec-
power of the validators. Thus, the technological cost of verification and
tively attain a common goal through an IT-supported virtual space (Lee
immutability of the blockchain are intertwined.
et al., 2002; Preece, 2000; Tardini and Cantoni, 2005). The main dis-
The second category of technology cost of blockchain-based plat-
tinction between an open source community and a blockchain
forms is related to the infrastructure (i.e. the bottom stack upon which
protocols, tokens, and applications are built). In the case of centralized-
platforms, a single entity sponsors and owns the infrastructure stack or 3
See here: https://www.economist.com/leaders/2015/10/31/the-trust-
core of the platform while keeping the components to the machine.

98
J. Pereira, et al. Technological Forecasting & Social Change 146 (2019) 94–102

community is that the latter comprehends a broader range of roles for manifest concerns with intellectual property and value appropriation,
the individuals. Precisely, communities around blockchain-based plat- what may lead to diminishing incentives to participate on open source
forms do not only include the end-users that consume complements and projects (Boudreau and Lakhani, 2015). At the same time, it is indeed
the producers of those complements (complementors), but also the difficult to measure and value members' contributions to meet their
developers, which contribute through code and commits to the main- extrinsic motivation, as through pecuniary incentives. Measuring
tenance of the platform interface and components, and validators, which members' contributions is difficult because the process of finding and
verify transactions and register them on the distributed ledger. Ad- negotiating a price for each contribution and protecting and licensing
ditionally, all individuals across these different groups can vote or de- intellectual property could induce prohibitive transaction costs (Franke
cide for or against implementing changes in the platform in a decen- and Shah, 2003).
tralized and “democratic” fashion. This rationale is aligned with Contrary to open source communities, centralized proprietary
Davidson et al.'s (2016a) suggestion that blockchain forms constitutional platforms rely on pricing structures based on membership fees, cross-
communities around the platform. subsidisation pricing strategy, rent appropriation and revenue sharing
The idea of collectives of people coordinating to achieve common to foster members' participation by leveraging on extrinsic benefits
goals beyond the boundaries of the firm is not new; indeed, there is a (Boudreau and Hagiu, 2009). However, researchers have already
long history of users and communities as important drivers of in- showed that strong extrinsic rewards for engaging in an activity might
novative activity and new organizational forms (O'Mahony and decrease individuals' intrinsic motivation, hence, negatively affecting
Lakhani, 2011; Rao et al., 2000). The example of Apache, Linux, and the nature of interpersonal interactions and creativity (Amabile, 1985;
Wikipedia, amongst others, show that collectives that communicate and Franke and Shah, 2003). While open source communities leverage
engage in repeated interactions can efficiently coordinate to create mainly on intrinsic rewards, and centralized proprietary platforms rely
socio-economic value (Benkler, 2017). Those individuals share not only mostly on extrinsic benefits, blockchain-based platforms harness in-
similar needs, concerns, passion and interests, acting collectively in centive mechanisms anchored on both intrinsic and extrinsic benefits,
order to meet these needs (see Felin et al., 2017; Preece, 2000; Tardini which, however, vary over time (Davidson et al., 2018).
and Cantoni, 2005; Wenger, 2011), but they also share values and be- In blockchain-based platforms, it is possible to distribute value
liefs (Preece, 2000; Tardini and Cantoni, 2005), which guide the way amongst community members through protocols, smart-contracts, and
activities are conducted in those organizations. Indeed, many crowd crypto-incentives (crypto-tokens and cryptocurrencies) with much
and community-based organizations are considered social movement lower transaction costs. The crypto-incentives are embedded in block-
entities (Felin et al., 2017) that present strong political, social and even chain-based platforms functioning, being a key piece of their protocols
revolutionary ideological aspirations where goals, values and beliefs are and smart-contracts. Such protocols and smart-contracts link commu-
intermingled (Stallman, 2002). For example, on Steemit, a blockchain- nity members' actions to respective rewards. For example, when end-
based public content platform, community members have the common users consume the complements, they can spend tokens, but if they
need to consume or produce content in a certain topic, sharing the provide reviews, or promote or curate content, they can receive tokens
belief that community members must appropriate the value of their for their contributions. Similarly, producers receive tokens in exchange
contributions in producing, promoting, and curating content for others. for their complements, and validators of verifying transactions. As
These values and beliefs oppose the ones of similar platforms like members' number of tokens is registered on the ledger, which is dis-
Reddit, Facebook, and Twitter (Steemit Whitepaper, 2018), in which tributed and publicly available; the number of tokens earned function
the platform owner appropriates the value created by users and pro- as a reputation indicator—members with more tokens earned are the
ducers. members that contribute the most to the community. Additionally, to-
Research on open source communities reveals that participants' kens have a convertible value in crypto-currencies and fiat money.
motivation intrinsically relates to the nature of the community activ- Therefore, all community members (users, complementors, validators,
ities. For example, activities that satisfy a need, fulfill values and be- developers, and investors) that possess tokens, either by buying or
liefs, and have a self-rewarding nature, which encompasses intellectual earning them through contributions, have strong incentive to con-
stimulation, new skills development, and making a positive difference tribute and collaborate towards network growth and interface and
(Lerner and Tirole, 2002; Villarroel and Tucci, 2010). Such intrinsic components quality improvement as such factors can positively impact
motivation feeds volunteers participation, cooperation, and coordina- the value of their tokens.
tion around a project and a common goal, sharing their knowledge For example, on Steemit, a public blockchain-based platform for
without, in most of the cases, subjacent direct pecuniary incentives content development, complementors are the people that create con-
(Amabile, 1983). Having said that, in addition to intrinsic motivation, tent, for which they receive a reward in STEEM Power (one of the
open source communities also enact reputation mechanisms, through platform's native crypto-tokens) based on the number of votes that they
which members get recognition, respect, and status amongst peers, collect from the audience. End-users, i.e. readers of the content, vote for
what can translate in future rents in terms of improved job opportu- or against the content, promoting and curating such content. Steemit
nities (Dahlander and Magnusson, 2005; Franke and Shah, 2003; operates on the basis of one-token for one-vote, which means that
Lakhani and Von Hippel, 2003; Lee and Cole, 2003; von Hippel and von STEEM Power owners (end-users, complementors, and validators) that
Krogh, 2003). contribute the most to the platform, as measured by their account
The debate about the underlying nature of community members' balance, have the most influence over how contributions are scored.
motivations is an ongoing conversation. While some argue that intrinsic Steemit validators, called witnesses, are the ones that create and sign
motivation continues to play a significant role in fostering participa- blocks of transactions. Outside of Steemit platform, the STEEM token
tion, others argue that all of these activates simply represent a type of (which can be obtained through STEEM Power) can be bought or sold
selfish market logic, where members still seek for rents; yet deferred on exchanges, as well as transferred to other users as a form of payment
into the future (Lerner and Tirole, 2002). Indeed, empirical studies (Steemit Whitepaper, 2018). While in the beginning, members tend to
reveal that a few number of communities are successful at retaining receive a higher number of tokens in order to foster community parti-
their members and fostering members' repeated contributions (Ma and cipation, these tokens tend to have low value in the beginning, and the
Agarwal, 2007). Most of the communities are unable to attract a con- opposite happens on the long-term.
siderable number of members or because self-selected members do not Blockchain community members' benefits result from a balance
have the right set of skills or are not engaged enough to generate an between intrinsic benefits that they extract from using, complementing,
interesting amount of content and interactions (Ma and Agarwal, developing the platform, and the potential extrinsic benefits from the
2007). One possible reason is that members, through time, start to future value of the tokens and reputation effects. However, as the

99
J. Pereira, et al. Technological Forecasting & Social Change 146 (2019) 94–102

community evolve in terms of the number of members, contributions, more conditions are added to protocol and smart contracts in order to
and interactions, member's benefits from these different sources may improve contracts completeness.
vary throughout time. Specifically, in the beginning, the members who Coordination costs can also relate heavily to the evolution of gov-
join the community are the ones that present higher intrinsic motiva- ernance modes and experiences in blockchain-based platforms.
tion. Therefore, the high intrinsic benefits compensate the yet low ex- Experimentation through application cases will help to delineate the
trinsic benefits (as the tokens have low value and there are low re- most efficient and effective contracts according to the nature of the
putation effects) in the short-term. Members' intrinsic motivation application. For example, DAO was an extremely important experiment
increases as the level of interactions and contributions start to increase. on decentralized autonomous governance modes, which has failed due
The intrinsic benefits reach an optimal point when the number of to a security breach in the code but opened an avenue for novel gov-
participants is big enough to generate interesting and stimulating levels ernance solutions that might also have contributed to the emergence of
of contributions and interactions (Baldwin and von Hippel, 2011), but it blockchain technology. Such evolution is similar to centralized platform
is not too big that the sense of community in terms of shared needs, models that result from years of institutional support and well-known
values and beliefs dilutes. If the number of community members in- organizing routines (Arruñada and Garicano, 2018). Blockchain tech-
creases too much, the sense of community slowly vanishes, coordina- nology evolution throughout forthcoming years might change drama-
tion and complexity increase, discussions become extensive and in- tically the set of transactional gains and costs proposed in this paper.
tractable, and at this point, intrinsic motivation starts to decrease. We hint that in the further years, the gains from adopting blockchain
However, extrinsic benefits (reputation and crypto-incentives) show a will increase; implying that such decentralized “market” platforms will
slightly different trend. While in the beginning, extrinsic benefits have be more suitable for an increasing number of applications, in contrast to
little value as the community is small, hence the gain from reputation is centralized platforms, which in the future might be advantageous for a
not substantial, and the value of the token is minimal. As the commu- narrower number of cases (Davidson et al., 2018).
nity grows in the number of participants, contributions, and interac- Blockchain verification process and consensus mechanism can also
tions, it tends to increase the value of the token and the significance of prevent the network from failure and fraud without any need for a
reputation. In this sense, whereas in the short-term, the intrinsic ben- central regulator or even a trustable third party. This “trustless” and
efits compensate the low extrinsic benefits; in the medium-term, as- tamper-proof governance system, however, is costly to implement and
suming that the community grows, the increasing extrinsic benefits may can be inefficient vis-à-vis centralized platforms. The high and ever-
compensate the decrease in intrinsic benefits. increasing cost of verification via proof-of-work is a prime example, as
described earlier. As a blockchain-based platform network and the
P3a. When the gains from intrinsic benefits outweigh the low extrinsic
number of validators (miners of Bitcoin for instance) grow, the higher
benefits in the short term, blockchain-based platforms are more
computing power and more massive energy are needed to solve the
advantageous than centralized platforms.
mathematical puzzle and create a new block. Hence, there is no clear
P3b. When the gains from extrinsic benefits outweigh the low intrinsic prospect for a reduction of this cost of technology in the future unless
benefits in the medium-term, blockchain-based platforms are more the consensus mechanisms changes.
advantageous than centralized platforms. There are alternative verification and consensus mechanisms that
try to find a remedy for the problems of proof-of-work protocol. Most
importantly in the so-called proof-of-stake verification process, which
4. Discussion and conclusion has been introduced by Ethereum platform (Davidson et al., 2016a). In
such a consensus mechanism, the computing power is replaced by the
This paper explores the distinction between centralized and block- stake of validators—i.e. the amount of native crypto-currency that users
chain-based platforms, where the latter represents an extreme case of possess. This process reduces verification costs, regarding computer
“open” and decentralized platform. We compare and contrast these power and energy consumption; however, it may boost costs and
platforms' governance typologies across three dimensions, namely drawbacks in other dimensions. For instance, entitling the wealthiest
transactions, technology, and community involvement. The contribu- validators to validate transactions and create new blocks endangers the
tion of this paper lies in the identification of the main benefits and costs decentralized nature of blockchain-based platforms. For instance, if
of each platform governance type, drawing on the conditions under most of the stake is in hands of few people; technically the (dis)ver-
which blockchain-based platforms are more advantageous than cen- ification of the transactions would be skewed towards few nodes,
tralized platforms. around which the governance of the platform will be centralized.
In previous sections, we highlighted that from a transactional per- Therefore, proof-of-stake while reducing the verification cost of tech-
spective blockchain-based platforms are preferable over centralized nology, it may open of the risk of opportunism; hence, higher trans-
platforms when reductions in the transaction costs–mostly in opportu- action cost. Further research shall investigate the main distinguishing
nism, and partially, uncertainty costs–are higher than increases in co- dimensions amongst different consensus protocols, relative benefits and
ordination and complexity costs. We can unpack each of these gains and costs, and under which conditions certain consensus mechanisms are
costs related to blockchain to see further patterns of improvement, and preferable over the others.
consequent increases in the use of decentralized platforms compared to We also discuss that blockchain-based platforms are more ad-
centralized ones. A first set of potential changes relates to the increased vantageous compared to centralized platforms when the community
potential of adaptation given the current upsides and downsides of the members' intrinsic benefits outweigh their low extrinsic benefits in the
blockchain. We are currently observing an adaptation process with short-term. We also defend that through time and as the community
many competing start-ups and “product” or “business model” innova- grows, intrinsic benefits would decrease, as extrinsic ones would in-
tions are competing to take advantage of the blockchain technology–we crease. As long as intrinsic and extrinsic benefits balance each other,
see a similar evolution in AI technologies, for example. The adoption of blockchain-based platforms would be the most beneficial form of or-
blockchain and its impact will be more evident as advances in both ganizing. This proposition might holds even when there are “hard
blockchain protocols and accompanying complementary innovations forks” in the community. As the community increases, higher is the
improve the technology infrastructure (Brynjolfsson et al., 2017). In- likelihood that dissonant voices start to emerge, leading to the emer-
creased adoption with such innovation complementarities between gence of minorities that differ in their needs, values and beliefs, what
applications (or use cases) and the blockchain technology itself eventually will lead to a community split-up. For example, Bitcoin XT,
(Bresnahan, 2010) will make the technology even better in its benefits, Bitcoin Unlimited, and Bitcoin Cash are Bitcoin protocol forks that aim
for example, allowing decreased complexity and coordination costs as at increasing block sizes, which allows more transactions per second,

100
J. Pereira, et al. Technological Forecasting & Social Change 146 (2019) 94–102

overcoming one of the scalability problems of Bitcoin (Cryptocurrency This paper has the goal to inspire discussion and further research on
facts, 2018; Gervais et al., 2008). Communities “hard forks” mean that blockchain-based platforms relative benefits and costs, offering a more
the community will split in two (or more). Members that keep adhering contingent perspective on this new emerging technology. The three
to the original community format, they will increase their intrinsic main sources of costs and benefits explored in the paper show key
motivations, as the dissonant voices left and the community got more points for further empirical research that surely needs to test our pro-
homogeneous. However, as a part of the members left the community, positions. Not only our propositions are individually showing some
the community size decreases, which will negatively impact the ex- expected effects, but also taken together, they originate relevant in-
trinsic benefits (token valuation and reputation effects). In this regard, teractive effects for further consideration. This paper is of theoretical
the community would survive a “hard-fork” as long as the increase in and empirical relevance, as many blockchain based consortiums and
intrinsic benefits outweigh the drop in extrinsic benefits. start-ups are trying to uncover and build a “killer” decentralized plat-
The blockchain-based platforms tokenization mechanisms are useful form to compete with centralized ones.
not only to foster extrinsic benefits to enhance cooperation and colla-
boration but also to dynamically distribute authority amongst members Acknowledgement
in a meritocratic system (Davidson et al., 2018). The biggest challenge
of such system is to design a protocol that score individual reputations This paper was as part of a collaborative research project funded by
and incentives that most of the community members consider fair and Leeds University Business School and Surrey Business School.
attractive. Simultaneously, such algorithms also need to be resistant to
intentional manipulation. If there is a wide abuse of the incentive sys- References
tems, the community members “lose faith” in the platform, decreasing
participation. Therefore, incentive systems need to reach a balance Adner, R., Kapoor, R., 2010. Value creation in innovation ecosystems: how the structure
between creating mechanisms to avoid abuse and opportunistic beha- of technological interdependence affects firm performance in new technology gen-
erations. Strateg. Manag. J. 31 (3), 306–333.
viour, but it also needs to be simple and clear, so members have a clear Amabile, T.M., 1983. The social psychology of creativity: a componential con-
expectation about their rewards. Reaching this equilibrium between ceptualization. J. Pers. Soc. Psychol. 45 (2), 357.
simplicity and robustness might be difficult to achieve. Indeed, nowa- Amabile, T.M., 1985. Motivation and creativity: effects of motivational orientation on
creative writers. J. Pers. Soc. Psychol. 48 (2), 393.
days, most tokenization models comprehend multiple tokens with dif- Arruñada, B., Garicano, L., 2018. Blockchain: The Birth of Decentralized Governance
ferent natures, and extremely complex incentive systems, hard to un- (April 10, 2018). Pompeu Fabra University, Economics and Business Working Paper
derstand and predict expected rewards. Such complexity denotes the Seriespp. 1608. Available at SSRN: https://ssrn.com/abstract=3160070.
Atzori, M., 2015. Blockchain technology and decentralized governance: is the state still
lack of knowledge in selecting the most efficient incentives systems
necessary? Available at SSRN: https://ssrn.com/abstract=2709713.
accordingly with the nature of the blockchain activity and goals. Baldwin, C., von Hippel, E., 2011. Modeling a paradigm shift: from producer innovation
Quantitative and experimental research is much needed on this field in to user and open collaborative innovation. Organ. Sci. 22 (6), 1399–1417.
Benkler, Y., 2017. Peer production, the commons, and the future of the firm. Strateg.
order to understand community members' adherence to different in-
Organ. 15 (2), 264–274.
centive systems. Böhme, R., Christin, N., Edelman, B., Moore, T., 2015. Bitcoin: economics, technology,
It is worthwhile to mention that while we discuss the costs and and governance. J. Econ. Perspect. 29 (2), 213–238.
benefits of blockchain based platforms in three separate dimensions Boudreau, K.J., 2010. Open platform strategies and innovation: granting access vs. de-
volving control. Manag. Sci. 56 (10), 1849–1872.
(i.e. transaction cost, cost of technology, and community involvement), Boudreau, K.J., Hagiu, A., 2009. Platform rules: Multi-sided platforms as regulators. In:
in principle, all three dimensions are intertwined. Improving in one Gawer, A. (Ed.), Platforms, Markets and Innovation. Edward Elgar, Northampton, pp.
area and bending the boundary condition in favour of blockchain-based 163–191.
Boudreau, K.J., Lakhani, K.R., 2015. “Open” disclosure of innovations, incentives and
platforms may be disadvantageous in another dimension. For instance, follow-on reuse: theory on processes of cumulative innovation and a field experiment
as mentioned earlier, applying proof-of-stake consensus mechanism in computational biology. Res. Policy 44 (1), 4–19.
may overcome the massive cost of verification problem but make the Bresnahan, T., 2010. General purpose technologies. In: Handbook of the Economics of
Innovation. Vol. 2. pp. 761–791 (North-Holland).
platform more exposed to the opportunism of a few wealthy validators. Brynjolfsson, E., Rock, D., Syverson, C., 2017. Artificial intelligence and the modern
We hope that our paper provides a simplified, but clear, theoretical productivity paradox: A clash of expectations and statistics. In: Economics of
framework to assess the “optimal” points of adoption of a decentralized Artificial Intelligence. University of Chicago Press.
Buterin, Vitalik, 2014. A next-generation smart contract and decentralized application
platform compared to a centralized one. Future research can provide a platform. White paper.
more holistic picture by linking these three areas together and poten- Catalini, C., Gans, J.S., 2017. Some Simple Economics of the Blockchain. Rotman School
tially exploring different dimensions not investigated in this piece. of Management Working Paper No. 2874598. Available at SSRN: https://ssrn.com/
abstract=2874598.
Nowadays, we already witness several variations amongst block-
Ceccagnoli, M., Forman, C., Huang, P., Wu, D.J., 2012. Co-creation of value in a platform
chain-based platforms, where some are proprietary but keep decen- ecosystem: the case of enterprise software. MIS Q. 36 (1), 263–290.
tralized governance and distributed data infrastructures, or others that Cennamo, C., Santalo, J., 2013. Platform competition: strategic trade-offs in platform
are non-proprietary but the code and data infrastructure are closed and markets. Strateg. Manag. J. 34 (11), 1331–1350.
Constantinides, P., Henfridsson, O., Parker, G., 2018. Introduction—Platforms and
permissioned. We are passing through an experimentation period when Infrastructures in the Digital Age. (Forthcoming).
“a thousand flowers are blooming” and we might be far from standar- Cryptocurrency facts, 2018. A list of upcoming bitcoin forks and past forks. Available at:
dization. On the top of this variance amongst governance and data in- https://cryptocurrencyfacts.com/a-list-of-upcoming-bitcoin-forks-and-past-forks/.
Dahlander, L., Magnusson, M.G., 2005. Relationships between open source software
frastructure dimensions, we can also find differences in the nature of companies and communities: observations from Nordic firms. Res. Policy 34 (4),
the activities of blockchain-based platforms. For example, while some 481–493.
are purely transactional, as is the case of cryptos like Bitcoin, whose goal David, R.J., Han, S., 2004. A systematic assessment of the empirical support for trans-
action cost economics. Strat. Mgmt. J. 25, 39–58.
is to exchange and store value; others use crypto-tokens associated to Davidson, S., De Filippi, P., Potts, J., 2016a. Economics of blockchain. Available at SSRN:
services or products, called as utility-tokens, such as Steemit platform. https://ssrn.com/abstract=2744751.
This paper intentionally under explores such differences, focusing on Davidson, S., De Filippi, P., Potts, J., 2016b. Disrupting governance: the new institutional
economics of distributed ledger technology. Available at SSRN: https://ssrn.com/
the conditions under which blockchain-based platforms with both de- abstract=2811995.
centralized governance and distributed data infrastructure are likely to Davidson, S., De Filippi, P., Potts, J., 2018. Blockchains and the economic institutions of
be more advantageous than centralized platforms. However, we en- capitalism. J. Inst. Econ. 1–20.
De Filippi, P., 2017. What blockchain means for the sharing economy. In: Harvard
courage that further research focuses not only on the main dimensions
Business Review Digital Articles, pp. 2017 Retrieved on March, 15.
that distinguish different blockchain-based platforms, but also on under Eisenmann, T., Parker, G., Van Alstyne, M., 2009. Opening platforms: How, when and
which conditions our propositions about transaction costs, technology why? In: Gawer, A. (Ed.), Platforms, Markets and Innovation. Edward Elgar,
costs, and community involvement have higher or lower traction. Northampton, pp. 131–162.

101
J. Pereira, et al. Technological Forecasting & Social Change 146 (2019) 94–102

Evans, P., Aré, L., Forth, P., Harlé, N., Portincaso, M., 2016. A Strategic Perspective on (6), 385–409.
Blockchain and Digital Tokens. Boston Consulting Group Available at: https://www. Rochet, J.C., Tirole, J., 2003. Platform competition in two-sided markets. J. Eur. Econ.
bcg.com/en-gb/publications/2016/blockchain-thinking-outside-the-blocks.aspx. Assoc. 1 (4), 990–1029.
Felin, T., Lakhani, K.R., Tushman, M.L., 2017. Firms, crowds, and innovation. Strateg. Rochet, J.C., Tirole, J., 2006. Two-sided markets: a progress report. Rand J. Econ. 37 (3),
Organ. 15 (2), 119–140. 645–667.
Franke, N., Shah, S., 2003. How communities support innovative activities: an explora- Simon, H.A., 1945. Administrative Behavior. The Free Press, New York.
tion of assistance and sharing among end-users. Res. Policy 32 (1), 157–178. Stallman, R. 2002. Free Software, Free Society: Selected Essays of Richard M. Stallman.
Gawer, A., Henderson, R., 2007. Platform owner entry and innovation in complementary Lulu. com.
markets: Evidence from Intel. J. Econ. Manag. Strateg. 16 (1), 1–34. Steemit Whitepaper, 2018. An incentivized, blockchain-based, public content platform.
Gervais, A., Ritzdorf, H., Karame, G., Capkun, S., 2008. Tampering with the delivery of Assessed at: https://steem.com/wp-content/uploads/2018/10/steem-whitepaper.
blocks and transactions in bitcoin. In: Proceedings of the 22nd ACM SIGSAC pdf.
Conference on Computer and Communications Security, pp. 692–705. Tardini, S., Cantoni, L., 2005. A semiotic approach to online communities: Belonging,
Ghazawneh, A., Henfridsson, O., 2013. Balancing platform control and external con- interest and identity in Websites' and video games' communities. In: IADIS
tribution in third-party development: the boundary resources model. Inf. Syst. J. 23 International Conference e-Society, pp. 371–378.
(2), 173–192. Tiwana, A., Konsynski, B., Bush, A.A., 2010. Research commentary—platform evolution:
Halaburda, H., 2018. Blockchain Revolution without the Blockchain. Available at SSRN: coevolution of platform architecture, governance, and environmental dynamics. Inf.
https://ssrn.com/abstract=3133313. Syst. Res. 21 (4), 675–687.
Hart, O., Moore, J., 1990. Property rights and the nature of the firm. J. Polit. Econ. 98, Tschorsch, F., Scheuermann, B., 2016. Bitcoin and beyond: a technical survey on de-
1119–1158. centralized digital currencies. IEEE Commun. Surv. Tutor. 18 (3), 2084–2123.
Iansiti, M., Lakhani, K.R., 2017. The truth about blockchain. Harv. Bus. Rev. 95 (1), Villarroel, J.A., Tucci, C.L., 2010. Motivating firm sponsored e-collective work. MIT Sloan
118–127. research pa per no. 4767-10. http://papers.ssrn.com/sol3/papers.cfm? (abstract_id=
Jacobides, M., Cennamo, C., Gawer, A., 2018. Towards a theory of ecosystems. Strateg. 1536209).
Manag. J. 39, 2255–2276. von Hippel, E., von Krogh, G., 2003. Open source software and the “private-collective”
Lakhani, K.R., Von Hippel, E., 2003. How open source software works: “free” user-to-user innovation model: issues for organization science. Organ. Sci. 14 (2), 209–223.
assistance. Res. Policy 32 (6), 923–943. Wenger, E., 2011. Communities of Practice: A Brief Introduction.
Lee, F.S., Vogel, D., Limayem, M., 2002. Virtual community informatics: What we know Weyl, G., 2010. A price theory of multi-sided platforms. Am. Econ. Rev. 100 (4),
and what we need to know. In: System Sciences, 2002. HICSS. Proceedings of the 1642–1672.
35th Annual Hawaii International Conference. IEEE, pp. 2863–2872. Williamson, O.E., 1973. Markets and hierarchies: some elementary considerations. Am.
Lee, G.K., Cole, R.E., 2003. From a firm-based to a community-based model of develop- Econ. Rev. 63, 316–325.
ment from a firm-based to a community-based model of knowledge creation : the case Williamson, O.E., 1975. Markets and Hierarchies: Analysis and Antitrust Implications.
of the Linux kernel development. Organ. Sci. 14 (6), 633–649. Free Press, New York.
Lerner, J., Tirole, J., 2002. Some simple economics of open source. J. Ind. Econ. 50, Williamson, O.E., 1985. The Economic Institutions of Capitalism. Free Press, New York.
197–234. Wright, A., De Filippi, P., 2015. Decentralized blockchain technology and the rise of lex
Li, M., Weng, J., Yang, A., Lu, W., Zhang, Y., Hou, L., Jia-Nan, L., Xiang, Y., Deng, R., cryptographia. Available at SSRN: https://ssrn.com/abstract=2580664.
2018. Crowdbc: a blockchain-based decentralized framework for crowdsourcing. Yli-Huumo, J., Ko, D., Choi, S., Park, S., Smolander, K., 2016. Where is current research
IEEE Trans. Parallel Distrib. Syst. 30 (6), 1251–1266. on blockchain technology?—a systematic review. PLoS One 11 (10).
Ma, M., Agarwal, R., 2007. Through a glass darkly: information technology design,
identity verification, and knowledge contribution in online communities. Inf. Syst. Joana Pereira is a Lecturer (Assistant Professor) in the Strategy and Organization group
Res. 18 (1), 42–67. and a member of the Centre for Technology, Innovation and Engagement (C-Tie) at Leed
Nakamoto, S., 2008. Bitcoin: a peer-to-peer electronic cash system. Assessed at: https:// University Business School, University of Leeds, UK. Her research investigates crowds and
bitcoin.org/bitcoin.pdf. communities based organizations that engage in production and innovation activities.
O'Mahony, S., Lakhani, K.R., 2011. Organizations in the shadow of communities. In:
Communities and Organizations. Emerald Group Publishing Limited, pp. 3–36.
M. Mahdi Tavalaei is a Lecturer (Assistant Professor) at Surrey Business School and a
Ozalp, H., Cennamo, C., Gawer, A., 2018. Disruption in platform-based ecosystems. J.
Manag. Stud. 55, 1203–1241. member of Centre of Digital Economy (CoDE) at University of Surrey in the UK. His
Parker, G., Van Alstyne, M., Jiang, X., 2017. Platform ecosystems: how developers lnvert research focuses mainly on competitive strategy, particularly in multisided markets and
the firm. MIS Q. 41 (1), 255–266. platform ecosystems.
Parker, G.G., Van Alstyne, M.W., 2005. Two-sided network effects: a theory of informa-
tion product design. Manag. Sci. 51 (10), 1494–1504. Hakan Ozalp is an Assistant Professor of Strategy and Innovation in the KIN Research
Preece, J., 2000. Online Communities: Designing Usability, Supporting Sociability. Wiley, Group, School of Business and Economics at Vrije Universiteit Amsterdam, The
Chichester. Netherlands. He holds his PhD in Business Administration and Management from Bocconi
Rao, H., Morrill, C., Zald, M.N., 2000. Power Plays: How Social Movements and Collective University. His research explores the impact of industrial and technological change on
Action Create New Organizational Forms. vol. 22. pp. 239–282. platform ecosystems and firms, as well as the antecedents of these changes.
Risius, M., Spohrer, K., 2017. A blockchain research framework. Bus. Inf. Syst. Eng. 59

102