Decentralisation and Economic Outcomes

Decentralisation is usually seen as one of the main benefits of Distributed Ledger Technology (DLT). However, decentralisation can be achieved on several dimensions that are not always connected with each other. In this post, we explore three of the most important dimensions of decentralisation and discuss their impact on economic outcomes in terms of efficiency, market power, coordination failure and externalities.

Introduction

In an early post from 2017, Vitalik Buterin discussed three dimensions of decentralisation.1 Architectural decentralisation refers to a system’s ability to withstand outside attacks, by having its resources as dispersed and as independent as possible. Political decentralisation describes how many independent entities contribute to the decision making. Finally, logical decentralisation concerns a system’s ability to allow the concurrent coexistence of different interfaces and data structures. Permissionless blockchains usually achieve decentralisation in the first two dimensions but not the third one. More generally, there is no logical connection between the three dimensions, so a system can be decentralised in some but not in others. However, the choice of where to decentralise may have implications on the economic outcomes which can be achieved, for example whether we can have efficiency, coordination failures and excessive market power. In what follows, we examine each dimension separately and describe their economic implications.

Architectural Decentralisation

In an architecturally centralised system, all computations are performed by a central server, which means that there is a single point of failure. In a completely decentralised system, the computations are performed independently by many computers, so that even if a significant proportion of them fails or it is attacked, the system can survive. The cost of being resilient to failure is, however, inefficiency, because the same computations are performed many times by many separate computers that need to be in constant communication so that they agree on the current state of the system. Moreover, decentralisation does not necessarily increase as we add more nodes to a network. If they suffer from the same vulnerabilities, then their failure rates are correlated, which implies that the network is not robust. For example, if all computers use the same software or hardware, then the same bug can compromise them all at once. If they reside in one country, then a government can confiscate them. If all developers are employed by the same company, then they might be instructed to steer the software development towards a direction that is not in the interest of the ecosystem as a whole. Architectural decentralisation, specifically in relation to robustness, requires that there are as many nodes as possible and that they are as different to each other as possible, hence it is far more expensive than a similar but centralised system. 

Political Decentralisation

The political dimension of centralisation covers who controls the system and how many independent entities decide on its future direction. On one end of the spectrum, a politically centralised system consists of a single entity that makes all decisions. On the other end, the decision-making power in a completely decentralised system is distributed among all independent entities involved in the ecosystem, each having equal decision-making power.

A corporation is an example of a centralised system. The CEO makes most of the decisions, however their power is constrained by the board of directors and ultimately the shareholders. Moreover, a corporation can achieve different levels of centralisation, by changing its hierarchical structure and delegating decisions to more junior managers. Centralisation is usually beneficial when operations need to scale fast, or when there is a need to standardise production processes, as in the manufacturing sector. If, however, the corporation develops several products or services that cater to different markets and needs to adjust quickly to changing market conditions, then a decentralised structure may be more appropriate. Distributed ledgers face similar tradeoffs, hence we see a variety of designs, from permissionless and decentralised to permissioned and centralised. However, although a blockchain can be completely decentralised in the political dimension, this is not the case for a corporation, because ultimately a majority stakeholder can act as a dictator.

An example of a completely decentralised system is a perfectly competitive market, where there are many producers and consumers, each with insignificant market power. Their individually optimal decisions are aggregated using the laws of demand and supply, leading to an efficient outcome, in the sense that there is no alternative allocation of goods and services that would make everyone weakly better off, and at least one strictly better off. However, an efficient outcome can be guaranteed only if participants have no market power, there are no informational asymmetries and no externalities in production and consumption.

The main effects of political decentralisation are in terms of market power, coordination failures, and externalities. Distributed ledgers decentralise in order to prohibit network participants from colluding and acquiring excessive market power, which can be used to appropriate a bigger size of the surplus. An extreme example of market collusion is a 51% attack, where at least 51% of the nodes in a network coordinate in order to rewrite past transactions and double spend. If some nodes gain significant influence and bargaining power, then they can collude to change the token economics of the distributed ledger in order to benefit themselves. The weaker nodes may be forced to accept these changes because they have already committed significant resources in the ledger, which they would lose if they exited. This is an example of the hold-up problem, which is discussed in a previous post.2 

Collusion and bargaining power are not, however, limitless. If some nodes become too powerful, then the other nodes will find it more profitable to exit and fork to another ledger, or new ledgers may be created by copying the existing open source code. This may destroy the network effects enjoyed by the initial ledger and decrease the value of their token. This constant competition and the implicit threat of a fork can force ledgers to be sufficiently decentralised. This also prevents network owners from unilaterally changing the rules of the game, leading to the holdup problem, where external parties are reluctant to invest in a network as they have limited control of the network and hence the return on their investment. Further, incentivising a decentralised community to undertake the required investment instead of a centralised company has the following benefits:

  • Rapid Scale - Networks can grow faster and across more legal jurisdictions
  • Credible Neutrality solves the holdup problem
  • Collective Ownership creates loyalty, aligns incentives, and drives growth. The shared ownership of the network can generate a virtuous cycle of growth, using tokens to incentivise participation and contribution to the network.

However, extreme decentralisation can also create problems. For example, it can lead to coordination failure, as every node acts in order to maximize their self-interest, without taking into consideration the positive or negative externalities they impose on everyone else.

The prototypical example of coordination failure and negative externalities in economics is the tragedy of the commons. A community of people have access to a public resource, such as a lake full of fish. Because no one owns the lake and there is no coordination, each member of the community fishes up to the point where his marginal cost equals his marginal benefit, ignoring the social marginal cost that takes into consideration the fish reproduction rate. By ignoring the negative externality of their actions, everyone catches more fish than what is socially optimal, leading to a complete collapse of the available resources. This negative outcome can be avoided if private ownership is introduced, so that people’s incentives are aligned with the survival of the lake, or if there are clear regulations on what everyone can or should do. The most successful blockchains, such as Bitcoin and Ethereum, have been able to avoid the tragedy of the commons, by allowing some `benevolent’ collusion and coordination between the miners and the largest token holders. At the same time, they discourage bad coordination by being decentralised, while having a cryptocurrency which captures and prices in the positive or negative externalities of the actions of individuals.

Logical Decentralisation

A system is logically centralised if its interface and data structures operate as one, so that it cannot be partitioned into several independent parts. Distributed ledgers are logically centralised because all nodes in the network need to agree on what is the correct order of transition and who will validate the next transaction. If the nodes disagree on what is the correct order of transactions, then we have a hard fork and a new ledger is created.3 Logical centralisation is independent of architectural and political decentralisation. However, logical centralisation may confer some market power to the majority nodes that have the biggest influence on how the distributed ledger operates and evolves. The reason is that if a minority of nodes disagree with the proposed changes, logical centralisation implies that they can implement their own views only by creating a hard fork. This is risky, because network effects imply that the users will most probably choose the already established ledger, which is what happened with Bitcoin and its various forks, as well as the hard fork of Ethereum and Ethereum Classic. As a result, in most cases the minority nodes are forced to stay with the initial ledger, an outcome which essentially confers market power to the majority nodes.

Conclusion

Decentralisation can be achieved in at least three independent dimensions: architectural, political, and logical. Each of these dimensions has implications in terms of economic outcomes. Architectural decentralisation implies that the system is more expensive to run and less efficient, however it is more resilient against attacks. Political decentralisation can inhibit collusion and the accumulation of market power; however, it can lead to coordination failures and the tragedy of the commons. Logical centralisation confers some market power to the majority nodes, as it makes it much harder for minority nodes to implement their own changes and coexist within the same blockchain.
Finally, it is important to emphasise that the main innovation of distributed ledgers is not decentralisation per se, but the way that they achieve it. Instead of having a set of rules that are enforceable by law or by some authority, they create a native token which promotes and enforces decentralisation by acting as a governance tool, an incentivisation mechanism and a currency. The possible economic designs are endless and as the most efficient for each use case will prevail, we are bound to a see a more democratic and inclusive form of capitalism.

Footnotes

1 The post can be accessed at https://medium.com/@VitalikButerin/the-meaning-of-decentralization-a0c92b76a274
2 See https://en.aaro.capital/Article?ID=DLT%20and%20the%20Hold-up%20Problem for more information.
3 Bitcoin has been hard forked several times and the most successful for was in 2017 for Bitcoin Cash. A history of all Bitcoin hard forks can be accessed at https://www.investopedia.com/tech/history-bitcoin-hard-forks/

Disclaimer

Aaro Capital is the trading name of Aaro Capital Limited (“Aaro”), a private limited company, registered in England and Wales with number 11419585, whose registered office is at 5th Floor 14-16 Dowgate Hill, London, United Kingdom, EC4R 2SU. Aaro is not authorised or regulated by the Malta Financial Services Authority ("MFSA") or any other financial regulator.

The material provided in this article is being provided for general informational purposes. Aaro does not provide, and does not hold itself out as providing, investment advice and the information provided in this article should not be relied upon or form the basis of any investment decision nor for the potential suitability of any particular investment. The figures shown in this article refer to the past or are provided as examples only. Past performance is not reliable indicator of future results.

This article may contain information about cryptoassets. Cryptoassets are at a developmental stage and anyone thinking about investing into these types of assets should be cautious and take appropriate advice in relation to the risks associated with these assets including (without limitation) volatility, total capital loss, and lack of regulation over certain market participants. While the directors of Aaro have used their reasonable endeavours to ensure the accuracy of the information contained in this article, neither Aaro Capital Limited nor its directors give any warranty or guarantee as to the accuracy and completeness of such information.

Please be sure to consult your own appropriately qualified financial advisor when making decisions regarding your own investments.