It’s been a little over a year since Ethereum, the largest smart-contract blockchain, switched out its old energy intensive proof-of-work (PoW) model, used by blockchains like Bitcoin, for proof-of-stake (PoS).
The switch, known as the Merge, introduced “staking,” a new way to add and approve a block of transactions to the blockchain.
This feature is part of Staking Week, presented by Foundry.
Under PoW, miners would compete to add blocks by solving cryptographic puzzles. Now under PoS, Ethereum validators stake 32 ether (~$50,000) on the network, and they are randomly selected to add blocks. Under both models, miners and validators are rewarded some ETH if their block is added to the blockchain.
With staking, Ethereum drastically cut the blockchain’s environmental impact, but it continues to face a slew of challenges around centralized power, censorship, and exploitation from certain infrastructure intermediaries. Here are five takeaways that the Ethereum ecosystem has learned over the last year since the Merge:
Ethereum’s energy consumption has fallen 99.9%
Ethereum’s Merge overhauled the network’s consensus mechanism – the system that a “decentralized” community of network operators use to secure the network and process transactions. The old model, “proof-of-work,” operated using a power-hungry “mining” system, where network operators essentially competed to process blocks (and earn rewards) by expending compute power.
The shift away from crypto mining to staking was expected to drastically reduce Ethereum’s energy consumption – completely retiring the energy-intensive system that the network previously used to produce blocks and secure users.
Ethereum’s energy footprint pre-Merge was around the size of a small country, and the statistics around its energy usage were a main point of contention for critics of early NFTs and DeFi. Bitcoin, which still uses a proof-of-work system to power its network, continues to consume the same amount of energy as Singapore, according to the Cambridge Bitcoin Electricity Consumption Index.
Ethereum’s new emission numbers have dropped drastically
A year on from the Merge, and Ethereum’s new emission numbers have dropped drastically. Ethereum’s new proof-of-stake system consumes 99.9% less energy than its old mining-based system. Whatever the other successes – or failures – of the upgrade, it is now much harder to paint Etheruem as harmful to the environment.
Stake distribution raises centralization questions
In addition to facing criticism for its high energy costs, Ethereum’s old consensus model came under fire for concentrating power into the hands of a small cadre of crypto mining syndicates – entities that had the money, specialized hardware and know-how to build massive crypto-mining facilities. Before the merge, just three mining pools dominated a majority Ethereum’s hashrate – a measure of the collective computing power of all miners
When the Merge transitioned Ethereum to PoS, the network abandoned mining in favor of staking. The new system stripped away the hardware requirements and compute costs of PoW, in part as a way to open the door for more people to pitch in to operate the network..
A year on from the Merge, however, centralization remains one of Ethereum’s biggest challenges. To stake on Ethereum, a validator needs to lock up 32 ETH, or roughly $50,000 with the network – funds that earn a steady stream of interest, but can be revoked if a validator errs or acts dishonestly. Setting up a validator node to stake on the network can also be a complicated task, meaning financial penalties can result if things are set up improperly.
Because of the expense and technical barriers to setting up a node, intermediary services arose – from companies like Coinbase and “decentralized” collectives like Lido – allowing users to pool their ETH together to create 32 ETH for a node. These intermediary entities do most of the heavy lifting: they take ETH from users, stake it on their behalf, and take a cut of the rewards that they earn from operating a validator.
Even before the Merge, some anti-PoS advocates feared staking could increase Ethereum’s centralization – meaning a small number of these intermediaries (or even a single one) might gain disproportionate control over which blocks are added to the network.
That scenario seems to be playing out: currently, the largest staking provider is Lido, the biggest decentralized staking pool. Lido currently accounts for 32.3% of the total share of staked ETH, leading to concerns of centralization as it nears the 33% mark, a threshold that developers say could cause security problems.
MEV and censorship
After the Merge, Ethereum’s validators have managed to net significant extra profits via a practice called maximal extractable value (MEV). This is sometimes seen as an “invisible tax” that validators and builders can collect from users by strategically inserting or reordering transactions before they’re added to the network.
When MEV became an unexpected vector of centralization and censorship on the network, third-parties stepped in to try and address some of the practice’s more pernicious side-effects.
Flashbots, an Ethereum research and development firm, invented MEV-boost, a piece of software that validators can run to reduce negative side-effects of MEV. Flashbots’ solution to the MEV problem is a controversial one, however. While some think MEV should be eradicated altogether, Flashbots’ introduced MEV-Boost to make the practice ubiquitous.
Currently, about 90% blocks on Ethereum go through MEV-Boost, which optimizes how transactions are organized into blocks in order to extract the maximum profit for validators.
The popularity of MEV-boost has become a point of contention for the network. As mentioned, MEV is viewed by some as an unfair tax on users. Flashbots’ central role in Ethereum’s MEV market has come under fire: most blocks assembled via Flashbots’ software are “relayed” – or delivered to validators – via Flashbots itself.
This kind of centralization has been viewed by some as a potential vector for censorship: when the U.S. Treasury Department sanctioned some Ethereum addresses associated with Tornado Cash, a mixer program, Flashbots stopped adding those transactions to the blocks it sends to the validator. This move was anathema to Ethereum builders who think that the infrastructure level occupied by Flashbots should be completely neutral – lest the entire network become more similar to centralized payment processors like Visa.
Since the early days of the Merge, the Ethereum community has made efforts to reduce censorship by configuring MEV-Boost to use non-Flashbots relays. Currently, 17.3% of blocks rely on Flashbots’ relay to extract MEV, and censorship is down to 35%, a tremendous reversal compared to its highpoint of 78% in November 2022.
Liquid staking tokens have taken over the ETH market
Following the Merge, the emergence of liquid staking has risen in the Ethereum ecosystem.
Anyone can earn rewards and participate in Ethereum’s security system via the process of staking, which involves locking up ETH tokens in an address on the Ethereum blockchain in exchange for a steady stream of interest. But there’s one problem: Once tokens are staked, they can’t be bought, sold, or used in DeFi (eg. as collateral for loans) — limiting staking’s appeal for investors interested in maximizing the value they earn from their investments.
Liquid staking services from third-parties present an alternative to traditional staking. Users who stake through services like Lido — rather than stake directly with Ethereum — earn a kind of derivative ETH token representing their staked assets: liquid staking tokens, or “LSTs” for short.
LSTs earn interest just like regular staked ETH, but they can be bought and sold like any other crypto — which makes them an extremely appealing investment for DeFi traders who want easy exposure to ETH staking. As an added bonus, LSTs offer users exposure to staking without the requirement that they put up 32 ETH, the minimum required for staking oneself.
Staked ETH was impossible for stakers to withdraw before the Shapella upgrade in April 2023, so people initially turned to liquid staking to earn staking yields without the risk of locking up tokens for an unknown amount of time. Once it became possible to withdraw staked ETH — removing one of the key risks of staking, but eroding one of the value-adds of LSTs — some thought the liquid staking market might shrink in favor of conventional staking. That’s not what happened.
Macroeconomic factors have probably had the larger impact than supply changes in the near-term
Currently, the liquid staking market is worth almost $20 billion, and it is growing rapidly — largely due to the ubiquity of LSTs in DeFi and the accessibility of LSTs compared to conventional staking. Lido’s token, stETH, represents the largest share in the LST market, with about 72.24% of the total share of LSTs.
Net supply of ETH is down
The Merge update came with some tweaks to ether’s tokenomics – the rules underpinning the blockchain’s native token.
Most notably, the upgrade made ETH “deflationary” for the first time, meaning that the overall supply of the token is now decreasing rather than increasing. The circulating supply of ETH today is .24% lower than it was a year ago. The decrease in supply stemmed in part from EIP-1559, a network upgrade that preceded the Merge by about a year. That upgrade began “burning” some ETH with every transaction on the network, but ETH didn’t become net-deflationary until the Merge made additional cuts to the rate at which new ETH is issued.
When ETH’s supply was growing year over year, some investors fear their share of tokens will become devalued over time. Some hoped that deflation would help make ETH more valuable. So far, it’s hard to say whether that’s happened. Ether’s price hasn’t changed much in the months since the Merge, and macroeconomic factors have probably had the larger impact than supply changes in the near-term.