Un seul compte Saxo suffit pour accéder à nos plateformes.
One year later: How proof of stake has changed Ethereum
Mads Eberhardt
Analyste des cryptomonnaies
Résumé: In September of last year, Ethereum replaced its proof of work consensus mechanism with proof of stake in a highly anticipated upgrade known as the merge. Since then, Ethereum has seen its circulating supply decrease by nearly 300,000 Ether, while consuming 99.95% less electricity. However, the network nears a potential point of no return that threatens its decentralization for years to come.
On September 15, 2022, Ethereum successfully changed its consensus mechanism by its transition away from proof of work to proof of stake. This transition was known as the merge. The latter was a goodbye to miners in verifying transactions but a welcome to stakers. Ethereum has now been running on proof of stake for about a year, so in this piece, we look back at tangible factors brought forward by the merge rather than describing the transition in detail. If you are interested in learning more about the transition, you can find a more in-depth analysis here, published a few months prior to the merge.
A financially sustainable cryptocurrency
During the proof of work era, Ethereum issued about 5.4mn new Ether yearly to reward miners for validating transactions. This is also known as the security cost, as it ensures sufficient security by a cost funded by inflation in the native cryptocurrency. To fund this security cost, Ethereum had a yearly inflation of about 3 - 4%. However, any cryptocurrency derives much greater security at a much lower cost by proof of stake compared to proof of work. This has allowed Ethereum to slash its yearly issuance to 816,000 Ether at the current staking ratio. As the latter is set to increase, Ethereum is likely to hit a yearly issuance of 1,000,000 Ether at some point, but that is still significantly below the pre-merge issuance of about 5.4mn yearly.
As the majority of Ether paid in transaction fees are burned, hence removed from the supply, Ethereum has seen a nearly 300,000 Ether reduction in its supply since the merge. This is in stark contrast to a supply increase of nearly 3.9mn Ether in the past year provided that the merge had not occurred. Not only has Ethereum been deflationary since the merge, but its issuance has also been rewarded to holders of Ether if staked rather than miners with no duty to hold Ether. This means that Ethereum has truly turned into a financially sustainable cryptocurrency, rewarding holders with a negative supply change, and staking rewards if they choose to stake. Barely any cryptocurrency can brag about this, particularly none even close to the size of Ethereum.
Energy consumption has been reduced by 99.95%
In proof of work, miners participated in a never-ending rivalry to be the ones to process a block. This rivalry was about computing a string below a target set by the network. The latter was computed using a mathematical function to mostly go through trial and error. This called for immense computational power, in Ethereum’s case, GPUs, using a great amount of electricity. Now, the network randomly chooses a validator every 12 seconds to process a block, so there are no battles for the blocks. The bottom line is that Ethereum has reduced its energy consumption by about 99.95%. This great reduction makes Ethereum much more appealing in a never-satisfied ESG world.
Another advantage of proof of stake is the consistency of 12-second blocks, so they do not fluctuate similar to during Ethereum’s proof of work era. This makes the network more predictable and easier to utilize for various protocols. The shift from average blocks of 13 seconds during proof of work to consistent 12-second blocks has also made Ethereum slightly more scalable, as block sizes have stayed the same.
Is Ethereum now more centralized?
This is a discussion largely beyond the scope of this piece; however, we will shortly touch upon the most critical considerations. Leaving the matter of decentralization aside for a second, Ethereum has certainly archived greater security by proof of stake. The network now reaches block finalization about 15 minutes after a block has been proposed. This is an extra layer of security not found in proof of work. It says that for a block to be altered or removed after reaching finalization at least 33% of the total staked Ether must be burned, effectively imposing a cost of at least $13bn for doing that. On top of that, the network can slash validators by burning some of their Ether in case they behave dishonestly or against the objectives of the network. In other words, by having collateral in something of great value from validators, in this case, Ether, the network can enforce much stricter rules for strong security.
Decentralization is another matter. This highly depends on who you ask. In our view, it is roughly the same as before the merge, as it is now. The challenge of the present staking distribution is that the largest staker is the liquid-staking provider Lido with around 32.5% network penetration, but still growing slowly. The Ether staked by Lido belongs to thousands of holders and is spread across 31 independent validators by smart contracts, so it is considerably better than if it was a single entity. The challenge, however, is that Lido may capture the consensus layer of Ethereum if it continues to grow, so the governance token of Lido suddenly decides much of the future of Ethereum. You had a somewhat similar situation during proof of work, as miners mingled their computational power to form so-called mining pools to enhance their chances of creating a block.
In both instances, the market operates with strong economies of scale. For mining pools, there are scale economies, as the capital flows to few but large mining pools to boost the chances of proposing a block. In proof of stake, there are no such advantages, but instead, large liquid-staking providers attain an upper hand in terms of liquidity. This enhances the ability of users to buy and sell the given liquid staking token without much price impact. The fear is that if Lido is not limited sooner rather than later, then it will continue to devour Ethereum with ever-increasing advantages for users by choosing Lido over other much smaller liquid staking providers.
Another challenge facing Ethereum in proof of stake is that its staking ratio continues to rise, meaning more and more Ether is staked. Most critically, this may lead to fewer Ether than optimal to be left to be paid in transaction fees to maintain the ecosystem, while liquid-staking providers, particularly Lido, archive damaging network penetration. Ethereum’s core developers decided last week to slow down the acceptance of new validators in Ethereum’s next upgrade to allow the community more time to figure out a proper solution.