What is Decentralisation?
In Satoshi’s white paper outlining Bitcoin, the word decentralisation is not mentioned once. However, the inventor of Bitcoin alludes to the idea, by insisting one should not rely on a ‘central entity’ and that Bitcoin allows transactions ‘without relying on trust’. Here lies the crux of decentralisation, a trustless and permissionless system. One need not rely on the character of a middle man and anyone can utilise the asset if they have the means.
Bitcoin and most crypto assets are not inherently decentralised, this normally happens over time and as a product of their consensus mechanisms. For example, one can imagine a time in 2009 when Bitcoin was only just established, where Satoshi was the sole miner. By one entity controlling over 51% of the network’s hash rate, Bitcoin was a centralised asset and it was only through the addition of extra miners that Bitcoin evolved to become the trustless asset that is seen today. By the very fact that Bitcoin was once centralised, one can see that decentralisation is not a necessary consequence of Bitcoin’s code. Currently, Bitcoin has an estimated Nakamoto Coefficient of 4, meaning 4 entities would have to collude to obtain 51% of the hash rate – allowing the entities to double-spend and change the ledger. Although an imperfect metric, the coefficient does provide ample confidence that Bitcoin is truly trustless.
Although one might have an incredibly decentralised consensus mechanism (i.e. many miners or stakers share the voting power), there are other areas into which centralisation can creep. One key area of centralisation is through the usage of cloud providers. For example, AWS provides around a third of all cloud computing services worldwide, meaning many projects are reliant on their infrastructure. When AWS experienced an outage in December 2021, dYdX, a (supposedly) decentralised exchange with over $700m daily volume, also went down. This raises a number of questions. Firstly, how can an exchange claim to be decentralised if it relies on a centralised service? How can one quantify or compare decentralisation when there are exogenous (external, non-blockchain specific) sources of centralisation. If an asset relies on a centralised cloud computing provider, is this a potential attack vector? Why wasn’t dYdX’s centralisation noted by investors before this event? Decentralisation, although an integral aspect of any crypto project, is still a nebulous concept that needs to be standardised across the community.
Another aspect of decentralisation that is often overlooked, is the role that the founder plays in the project’s development. Ethereum and Cardano have both drawn criticism due to the prominence that their founders, Vitalik Buterin and Charles Hoskinson, have within their respective projects. It is argued that the level of influence each man possesses undermines the notion of true decentralisation. In almost a direct contrast, Bitcoin’s founder has not been heard from, nor made a Bitcoin transfer, since 12th December 2010. Bitcoin enthusiasts argue that the lack of a key public figure brings the asset closer to true decentralisation. Whether the lack of a main frontman is needed for decentralisation is also not clear, but is an interesting debate, regardless.
Why is Decentralisation so Important?
Without decentralisation, the purpose of utilising blockchain technology for transactions comes under question. Crypto assets have limited throughput, with assets known for high throughput still having relatively low transactions per second (TPS). For example, Algorand has a maximum TPS of around 1000 and XRP can currently manage around 1500 TPS. Compare these figures to, say, the maximum throughput and actual 2018 output of VISA (65000+ and 5600+ respectively) one begins to appreciate the limited nature of crypto assets. Of course, this is not to say that VISA’s technology is superior to crypto assets’, it is to say that by utilising crypto one is sacrificing throughput for trustlessness.
Every function of crypto assets and DeFi can be achieved just as effectively through a centralised manner: VISA or PayPal can be used for payments, gold/commodities as a store of value, bonds and deposit accounts for lending/savings, and NFTs can be replaced with ownership papers. Thus, one must conclude that the only competitive advantage of cryptocurrency is its decentralisation. Without naming any projects, certain assets are emerging with high throughputs and incredible applications, yet they are centralised. It begs the question, what is the purpose of such inventions? It seems that one could achieve the same functionality with a high-speed centralised database, rather than masquerading as a pseudo-decentralised system.
Underlying this discussion is the Blockchain Trilemma. Popularised by Vitalik Buterin, founder of Ethereum, the trilemma states that for one to have a functioning blockchain project, there must be a sacrifice of one of: decentralisation, scalability or security. I.e. one can pick two areas at which to excel, to the detriment of the third characteristic. For example, Bitcoin and Ethereum are arguably very secure and decentralised, while their scalability is limited, as exhibited by low throughput. The trilemma has yet to be resolved, with many projects that seemingly have the potential to meet all three criteria, are yet to become decentralised, or more often, yet to show they can scale beyond the test net. Until an asset overcomes the shackles of the trilemma, projects should prioritise decentralisation, realising the intention behind Satoshi’s vision.