Among the countless criticisms of crypto that emerged during its rise over the last several years, the environmental footprint of major blockchains like Bitcoin and Ethereum has been a major sticking point.
But for Ethereum, those concerns are no longer valid thanks to a blockchain upgrade called “the Merge” that Ethereum’s global network of developers implemented on September 15, following years of planning and delays.
The Merge changed what security model Ethereum uses to process the transactions on the network. In the past, energy-hungry specialized computers known as miners worked 24/7 to process Ethereum transactions. This model, called proof-of-work, provided security to the system by making it too expensive to alter the ledger of transactions, and it’s still used in Bitcoin. Now, it’s trusted validators who stake tokens on Ethereum—a fancy term for locking up your ETH as a prerequisite to being selected as a validator—that process those transactions and earn a reward.
The transition from proof-of-work (PoW) to this so-called proof-of-stake (PoS) consensus reduced energy consumption by roughly 99.95%, according to the Ethereum Foundation, which supports Ethereum’s development.
But the Merge has also brought about shifts that are challenging the core philosophies of the crypto-faithful. The concentration of power to validate transactions (and profit along the way) in the hands of a few—who may be companies, or individuals—raises questions about the system’s inherent power dynamics. And validators have the ability to decide which transactions make it to the blockchain, straining the principle that the blockchain is a permissionless network where every transaction is treated equally.
Ethereum might finally be green, sidestepping the concerns of its climate-minded detractors, but its journey into a brave new world has just begun.
The Merge, like the blockchain itself, is a successful case of global human coordination. It’s been likened to changing the engine of an airplane mid-flight, and for good reason, with the added wrinkle that it’s the passengers who had to do it themselves.
The blockchain’s loose group of developers, who debate and take decisions through public meetings, have prepared for the Merge for years. They launched in December 2020 a PoS-ready blockchain called Beacon Chain, which ran in parallel to the default Ethereum blockchain called Mainnet. The plan was to merge these two chains, and let Beacon take over. That’s exactly what happened, without any hiccups.
Despite the energy savings—”solving” a problem Ethereum itself created, admittedly—Bitcoiners in particular were skeptical of what power dynamics it might unleash. Under proof-of-work, setting up a mining shop to earn token rewards is an entrepreneurial affair. Anyone, anywhere in the world, can plug in and start mining if they can find energy cheap enough to turn a profit. This has made locations like Kazakhstan somewhat unlikely centers of computer power in the Bitcoin network.
Ethereum’s new proof-of-stake model, detractors said, would usher in a new age of plutocratic control. Because validators must be people or companies who already have a hoard of ETH tokens to stake, then, in the end, it is simply a case of the rich getting richer, the argument went.
“Running servers in developing countries can be expensive and experience network lags”
Post-Merge, this fear has taken shape to a degree. While anyone in the world can become a validator, some in the Ethereum community are acknowledging that proof-of-stake has already concentrated authority and token rewards in the hands of entities that are largely in the Global North.
Data from Nodewatch shows that although there are Ethereum nodes in 80 countries, they are highly concentrated in a few places. Thirty-eight percent of the nodes are in the US, and 14 percent are in Germany alone. All in all, the network of validating nodes is highly concentrated in North America and Europe.
By design, it costs a minimum of 32 ETH to run a validator node. This has given rise to the staking-as-a-service industry, with a few players dominating the market. A company called Lido leads the pack with almost 30 percent of all staked ether deposited through its nodes. Staking services by large centralized exchanges like Coinbase and Kraken follow Lido from a distance.
On its website, Lido acknowledges that the balance of power in proof-of-stake leans toward Europe and the US, but also that it’s dedicated to doing something about the problem. On its company scorecard, it rates its own progress on its validator set being “distributed geographically and jurisdictionally” as being “okay,” but having room for improvement.
“While ~10% of validators are currently with US-based Node Operators, there remains an over-reliance on European based entities,” the Lido website states. “Lido will continue to select Node Operators in the interest of jurisdictional and geographic diversity as it expands the set and will publish this information transparently going forward.”
Vasiliy Shapovalov, co-founder of Lido, told Motherboard that the dominance of the West “is possibly because running servers in developing countries can be expensive and experience network lags.”
“Endgame for Lido is much more numerous node operators and a more diverse crowd of them,” he said.
But for now, Lido doesn’t have Global South node operators on Ethereum, “though we have one or two submissions that could become accepted in the next round of node operator onboarding,” Shapovalov said.
The global distribution problem is a difficult one to solve due to structural reasons, according to Kristof Gazso, an Ethereum researcher.
About 75 percent of nodes are hosted on the cloud, according to NodeWatch’s data. To change the global distribution problem, a wider change needs to happen in the cloud industry first. “For those who stake on the cloud, most providers have few data centers in the Global South. Since most nodes are already on servers in developed countries, it makes logical sense to pick a place close by geographically to minimize your latency and maximize your reliability,” explained Gazso.
And for the 25 percent of users who stake at home without relying on the staking-as-a-service industry, it’s largely about finances. “Staking at home without relying on staking providers is expensive, requiring 32 ETH minimum to become a ‘solo validator’, the crème de la crème of decentralized and trustless staking,” he said. “This is already a massive investment for stakers in more economically developed countries, but it is simply an impossible ask for most in developing countries.
But not all hope is lost. Gazso added that there have been services that allow for what is called “pooled staking”, granting access for potential validators to have to front up a lot less than 32 ETH to start staking. Users cover the rest to earn yield on their ETH without having to go through the hardware set-up. The remainder is covered by other users who want to earn yield on their ETH but without wanting to set up and maintain any hardware for staking.
The question of winners and losers aside, another implication of the Merge is validators having greater control over which transactions hit the blockchain.
So far, this has mostly looked like validators using their privileged position to effectively de-prioritize transactions sent to addresses that are sanctioned by the US government. That list includes any wallet address that’s interacted with privacy mixer Tornado Cash on Ethereum, the developer of which is currently jailed in the Netherlands. To some in the libertarian-minded crypto community, that looks a lot like censorship. “Ethereum today is experiencing so-called weak censorship,” Justin Drake, a researcher at the Ethereum Foundation, told Motherboard. “This is a mild form of censorship where some transactions get included on-chain with longer-than-usual inclusion times.”
In blockchain speak, a transaction can mean blockchain activities like an NFT purchase or a token swap. Transactions involving OFAC-sanctioned addresses are delayed by five times, taking one minute to go through instead of the usual 12 seconds. (But that’s still 10 times faster than Bitcoin’s 10-minute inclusion time, said Drake.)
The weak-censorship trend, which involves delays, accelerated after the Merge and has most recently flatlined, according to data from tracker mevwatch.info. On September 16, one day after the Merge, only 9 percent of blocks enforced OFAC compliance, according to the site. But in the last week of November, that proportion of OFAC-compliant blocks jumped to 72 percent.
Not everyone is equally concerned about this shift. “So some transactions might be delayed a bit, but I have no concerns about protocol-level censorship,” A. Rev., the pseudonymous founder of tracker site ethereumpools.info, told Motherboard. “Note that this delay is in the order of magnitude of dozens of seconds. Even with 99 percent OFAC-compliant, we are talking about delays of a few minutes.”
“Still better than our legacy financial system,” he added.
What’s most perplexing about all this is the fact that the underlying cause of the rising trend isn’t so much Ethereum’s validators becoming more controlling and authoritarian.
Ethereum’s “weak censorship” occurs due to validators battling a curious phenomenon that lets the technically-savvy earn more profits. That phenomenon is called maximal extractable value, or MEV for short.
What MEV boils down to is Ethereum users searching for transactions waiting to be processed in a “waiting room” called the mempool, and then effectively paying validators to order transactions in a block so that they benefit. For example, if you buy a particular token, an MEV searcher could order transactions to buy it before you and also sell after you. MEV isn’t always so sinister, however, and can often look like a typical arbitrage play, taking advantage of momentary differences in price between platforms.
This is all a big deal. Searchers have extracted $6.2 million worth of MEV in the last 30 days, as of this writing. The total amount of extracted MEV so far is estimated to be worth $685 million.
To combat this, validators have turned to services like Flashbots, which helps transactions bypass the mempool. Services like Flashbots are called “relays,” and validators use them to earn more ETH. Flashbots in particular dominates the market by 78 percent, and the service is OFAC-compliant. The end result is that OFAC-compliant transactions sail through to the blockchain, while those that may violate sanctions get held up until they’re processed by the network.
“Protocol-level censorship is a concern in Ethereum”
“There are two pieces of good news,” Drake said. “Relays are temporary pieces of infrastructure. An update to Ethereum, so-called ‘enshrined proposer-builder separation’, will eliminate the need for relays, and therefore eliminate relay censorship”
And secondly, Drake said, “anyone can operate a relay and I expect a migration away from the Flashbots relay towards credibly-neutral relays.”
Today’s weak censorship that befell Ethereum is only a temporary problem, according to Ethereum developers like Drake optimistic about a change in relay dynamics.
But that’s not the end of worries. There are larger and more long-term concerns due to the concentration of Ethereum staking among a few big players.
“Protocol-level censorship is a concern in Ethereum,” Gazso told Motherboard. “My main source of concern stems from the continued centralization of staking on Ethereum into pools.”
On the one hand, he said, these pools help casual Ethereum holders to contribute to the network’s security without the whole hardware setup. “On the other hand, it gives the pool operators a lot of power that they can use to, for example, censor transaction[s],” he said.
But the root of the problem isn’t so much about the fact that so many nodes are based in the US, Gazso said. “Although that is not ideal, but rather that most of the staking is done by large institutions that can be easily be made to bend to the will of governments. If more stakers were staking at home, this would be less of an issue,” he said.
“The more people stake at home, the more resilient the network becomes to outside pressure,” he added.
In the face of these concerns, some ardent Ethereum users have heeded the advice of Ethereum researchers like Gazso and spun up their own validator nodes. But like pretty much everything in Ethereum, that’s a costly endeavor.
Andreas Tsamados, a student in the United Kingdom, told Motherboard that he used a grant to purchase “three advanced, ready-made, and easy-to-set-up nodes” worth €1,875 ($2,000). “Decentralization is not a given,” Tsamados told Motherboard to explain his motivation. “It is not something that just comes for the technology underlying the chain but the values and social coordination of its community.”
“Running your own nodes breaks your dependence on convenient but centralized services that help you interact with the chain. That means you preserve your privacy, you will be able to voice your opinion in the case of a fork or attack on the network, you reduce risks of censorship, and you improve the resilience and decentralization of the network,” Tsamados said.
There is a long way to go for Ethereum.
At the Ethereum Community Conference in France in July, Ethereum’s creator Vitalik Buterin said that Ethereum, as a project, is only 40 percent complete. And the Merge would bring it to 55 percent.
Despite the lack of global diversity that irks many of Ethereum’s core community, and provides fuel for its detractors, the Merge remains a positive step for the world’s second-biggest cryptocurrency. It allowed Ethereum to solve its energy consumption problem and demonstrated how a group of developers could coordinate globally to accomplish a technical undertaking of such enormity.