The myth and the reality
Represented by the Nakamoto coefficient, decentralization is supposed to be the core attribute of blockchain technology. However, the reality is more complex.
How decentralized are large blockchains?
The decentralization characteristic of blockchain is what makes it unique among other technologies. But are the large blockchains we know really decentralized? Let’s take a closer look at this.
Blockchain decentralization aims to eliminate the trust of the members and prevent them from destroying the efficiency of the network with their authority or commands. On this topic, you may have heard some discussions about the Nakamoto coefficient, formally introduced in 2017 by former Coinbase CTO Balaji Srinivasan.
Using the Nakamoto coefficient as a measure of decentralization, you can determine the minimum number of validator nodes required to disrupt the blockchain network. The higher the Nakamoto coefficient, the more decentralized the blockchain.
The Nakamoto coefficient is a way to quantify the decentralization of a blockchain or other decentralized system. That’s the number of devices you need to compromise at least one important subsystem.
Original article: https://t.co/CjVgfy2T0K
— Balaji Srinivasan (@balajis) 3 December 2020
Srinivasan suggests that a blockchain comprises six subsystems: mining, clients, developers, exchanges, nodes and owners. Each of these subsystems has its own set of statistical data that must be considered when measuring the Nakamoto coefficient:
- Mining: The rewards users receive for mining within a specific time period.
- Clients: Number of users for each client
- Developers: The number of commitments developers stock
- Exchanges: The volume of exchanges made within a specified time period
- Nodes: The node distribution across countries
- Owners: The distribution across individual addresses
Let’s look at some blockchain networks where they stand regarding the Nakamoto coefficient, using data from Nakaflow, Crosstower and Blockwork.
Bitcoin
Nakamoto coefficient: 7,349
Area Highlight: Developer, Owner and Validator metrics score high.
Validator Node Count: 14,409
Nakamoto scores tend to be the highest when it comes to Bitcoin. In general, Bitcoin is one of the most decentralized blockchains.
Ethereum Beacon chain
Nakamoto coefficient: Unknown
Area Highlight: Ethereum scores well in Node distribution. For developer and owner decentralization, Ethereum scores low to moderate.
Validator Node Count: 300,000+
Ethereum has such a large network size that its total number of validators cannot be determined.
BNB chain
Nakamoto coefficient: 7
Area Highlight: Low number of validators.
Validator Node Count: 21
The Nakamoto coefficient of the BNB chain is 7, around the median for large blockchains. By expanding the validator set with inactive validators, BNB Chain could ensure greater security and network reliability.
Solana
Nakamoto coefficient: 30
Area Highlight: Solana scores well when it comes to mining pools. However, Solana scores poorly for decentralization of nodes and owners.
Validator Node Count: 1,875
It was Solana who popularized the idea of the Nakamoto coefficient. The Nakamoto coefficient for Solana is strong, thanks to its 1875 validators.
Avalanche
Nakamoto coefficient: 30
Area Highlight: Avalanche has one healthy active validator count as well as optimal node decentralization.
Validator node count: 1,267
Decentralization has always been a priority for Avalanche. As the Nakamoto Coefficient shows, Avalanche is the most decentralized Proof of Stake (PoS) blockchain along with Solana.
Despite its usefulness, the Nakamoto coefficient has some flaws. For example, a blockchain may score well for one type of decentralization, and another of its more important systems may be centralized. In addition, in some cases, Nakamoto scores are calculated over a short period of time or with a huge number of users, which reduces their reliability. Even Bitcoin, the most decentralized network, is dominated by one few large mining pools.
In addition to the Nakamoto coefficient, many researchers also use Gini coefficients and Shannon entropies to estimate blockchain decentralization. However, these metrics largely agree with Nakamoto scores for Bitcoin and Ethereum.
Centralization before decentralization?
Early in a blockchain’s lifespan, centralization often precedes true decentralization. This development may be necessary for the next phase of decentralized networks to emerge. While centralization can bring stability to a young blockchain, there are sometimes downsides.
In a recent Twitter threadcrypto influencer TyLucky called out Cronos Chain and parent company Crypto.com for appearing to favor certain protocols when other protocols were also performing.
Why centralization under the guise of decentralization matters – 🧵
To be crystal clear, no FUD intended, I do not have a competing product or ulterior motive. I just genuinely care #crofam and for better or for worse, I speak up when I see a problem. Many can confirm.
— TyLucky | Crypto YouTuber and NFT Collector (@TyLuckyOfficial) 11 August 2022
Consequently, more projects seem to want to bridge out of Cronos than bridge into it, he said.
I’m just a guy, this is just one guy’s opinion, but in the last few months I’ve noticed significantly more projects wanting to bridge out of Cronos than wanting to bridge into Cronos. Why? Not just one reason to be clear, nor are the people I have mentioned alone responsible.
— TyLucky | Crypto YouTuber and NFT Collector (@TyLuckyOfficial) 11 August 2022
As we can see, many established blockchains are still not as decentralized as they could be. However, it is important to note that blockchain is still a relatively new technology. Most blockchain protocols are still in their infancy, and many must continue to trade decentralization for network stability.
As networks mature, they may become more decentralized or define their specific degree of decentralization.
