In the early hours of September 15th, the Ethereum Merge took place in what could be described as no less than the ideal blockchain implementation upgrade. It happened at 8:43 AM Berlin time, where the Ethereum Foundation was gathered for the occasion. The night before, members of the Ethereum community celebrated in cities around the world, excitedly watching blocks inch closer to the much-anticipated block 15537394. Our head of R&D, Robert Drost, even stayed up in California to watch it live with his daughter.
As a follow-up to our article on what to expect from the Merge, we wanted to check in with the same contributors to see how/if their thinking has evolved. Robert Drost, Patrick Li and Praneeth Srikanti are respectively members of the MESH and Ethereal Ventures with years of technical experience in the industry (full bios at the end). Here’s what they had to say.
#1 Let’s start with a question for Robert. As you predicted, there has been some price volatility with the forked ETHW token. Do you still think holders will be incentivized to continue to sell off this token?
RD: We’ve definitely seen a lot of price volatility on the forked POW chain, with a sustained drop on the first day post-Merge. It's largely a positive reflection of the success and smoothness of the Merge for ETH PoS. This smoothness means there’s a lack of a sustained narrative around Ethereum having any systematic issues with proof of stake. Plus, this newly forked ETHW chain has competition from Ethereum’s original Ethereum Classic (ETC) fork, which is also an ongoing proof of work version of Ethereum. ETC has even had some shine from Vitalik, who in July pointed proof of work die-hards to consider moving their hash power and transactions to ETC post-Merge, calling it a “totally fine chain”. For anyone interested in this alt-ETH world, one useful indicator of health can be seen via hash rates. While ETC hashing has been stable at around 180 TH/s, ETHW started post-Merge at about 80TH/s and has dropped to around 30TH/s. Keep in mind that the chain security and double-spend prevention from long-range reorgs is proportional to the hash rate, so tread carefully if you’re cashing out any ETHW!
#2 It seems like the Merge took place relatively smoothly. Even the ETHW fork came as no surprise to those of us who had been closely following developments. Were there any elements of the Merge that you think could have been executed better, or any elements you didn’t see coming?
PS: The Merge really showcased the capabilities of new consensus and finality mechanisms to an extent that was previously unthought of. The fact that there have been no block re-orgs, no form of uncle blocks (thereby saving computational cost and bandwidth), and highly consistent block times and gas utilization have all been incredible to see! One of the more interesting aspects to watch post-Merge has to do with the interplay between inflationary mining rewards and the burn mechanics imposed by EIP-1559.
Two other interesting observations post-Merge:
i) the relatively low number of blocks with the fee recipient set to the null address. As a quick reminder, a suggested fee recipient is set by beacon chain validators in order to collect transaction fees and tips from user transactions. Validators, who use BLS keys and have no ‘presence’ on the execution chain since these keys correspond to Ethereum addresses, have the option to nominate an address to collect fees on the execution chain. The fact that we’re seeing only a few of these transactions reveals that validators are: i) paying attention to potential rewards through tips by accepting transactions from public mempools and ii) cautious about upgrading to MEV-Boost and accepting pre-built blocks from sophisticated builders to capture block extractable value (the latter value is currently at 28% of the validator network).
ii) A magnitude lower sandwich MEV losses (from 34% to 3.6% of all MEV). Sandwich attacks are a consequence of trusted block building and including miner-generated transactions before and after a user’s trade on a decentralized exchange, leading to value capture by miners. This certainly is an interesting phenomenon and may be reflective of the self-building of blocks or choosing to connect to ‘ethical’ relays.
PL: The smooth Merge comes from thoughtful planning, rigorous testing and full dedication from developers and communities. It’s worth noting that, according to Ethernodes, 25.2% of node operators hadn’t upgraded their clients to Merge-ready version and caused the hiking of the “missed block rate” after Bellatrix upgradation, but this number has quickly come down to less than 1% at the time of writing. This is one example of how Ethereum communities are aligned with the Merge. So I would say the Merge has accomplished its predefined task, although people will, of course, want more from the Merge, such as the unlocking and withdrawal of staked ETH. This feature is likely to be addressed in the next Shanghai upgrade.
Another fact I would also like to address is that the Merge will make ETH deflationary. Before the Merge, ~3,000 ETH was issued to miners and ~1,600 ETH to validators daily. After the Merge, ETH will only need to issue rewards for validation, which will reduce total supply inflation from ~ 4-5% a year to ~0.5%. But the burning side/transactions haven’t caught up yet. According to the Ethereum Foundation, we will need more than16 gwei so the burn mechanism can net out ETH issued to validators. Currently, we’re still sitting less than this. Therefore, the total ETH supply increased by 2,350 ETH at the time of writing per https://ultrasound.money/. But after the on-chain activities pick up, I do expect deflation to come.
RD: The Merge day was smooth as glass, and the whole Ethereum community can see it as a crowning achievement of so much hard work and preparation by so many. It was a great demonstration of that old saying that good things come to those who wait. Doing a secure upgrade to PoS while heavily balancing towards maximal decentralization has always cost Ethereum some measure of development speed. Ethereum is playing the long game, and its community continues to show patience and support of that end goal.
Praneeth and Patrick highlighted all the major points I had in mind. One thing I’ll add to Patrick pointing out ultrasound.money is on the discussion about the burn to issuance balance. Ultrasound.money is a great continuation of the now sunset watchtheburn.com (shoutout to them and thanks!). It has a POW simulator option which shows the strong effect of ETH’s PoS issuance reduction combined with the burning of transaction fees. While ETH POW would have had ongoing inflation of around 4%, and Bitcoin is currently inflating at 1.7%, even given the current very low gas fees in this lull of crypto macro activity, ETH PoS is presently only inflating by 0.2%, which is already 8x less than Bitcoin! When crypto macro activity returns, we can expect the burn rate to overwhelm Ethereum staking issuance, and we’ll net burn. Looking at historical burn fees during the last crypto bull run, ETH’s yearly burn rate was well over $10B/year, so the effect of deflationary economics will certainly be an interesting facet of Ethereum to watch closely!
#3 Within hours of the Merge, SEC chairman Gary Gensler announced that staked cryptocurrencies may be subject to federal security regulations. This aligns with what you predicted regarding the resurgence of attention to censorship issues in US regulations. How do you see the relationship between the Ethereum community and regulators evolving, and what do you think we could be doing better to improve relations?
RD: While we’d like to be wrong on the quick resurgence of censorship worries and regulatory discussions around the Merge, it’s important to be aware that both are central and recurring themes in web3. Crypto is an evolving space that is destined to be shaped by the yin-yang of two interconnected opposing (but net-net very creative) forces: the individual-centric forces of freedom and libertarian style ideals and the society-centric forces of regulation and protectionism. We’re probably past the beginning of this evolution as crypto is starting to approach scales that affect global financial systems even in larger economies like the US and Europe, but we still have a long way to go.
In the U.S., there’s active discussion around fitting in the unique properties of crypto base token assets and DeFi layered constructs into commodity, security or debt frameworks. Federal agencies like the SEC and CFTC will rightly work to understand and meld the new financial instruments in crypto systems into the current frameworks. And there are likewise efforts in most jurisdictions around the world. But in addition to regulatory agencies, we will likely see increased legislative attention that aims to address the public needs and benefits around crypto. Democratic systems balance out the regulatory actions of the executive branches with the legislative branches' attention to voter core interests. And, of course, judicial oversight adds further checks and balances.
In the larger picture, the yin-yang mental model for crypto helps us see that either extreme of freedom or regulation causes imbalance. In the past year, the excesses of risk-taking in flawed economic designs like Terra/Luna, as well as outright fraud and pure scams, were a black mark on the crypto peak but not different from the excesses of prior peaks. Furthermore, hacks grew this cycle to total billions of USD on a broad swatch of crypto protocols. This showed business growth priorities getting too far ahead of security practices and common sense. And all this was on top of the overshoot from the irrational exuberance of new investors who join the crypto space on each new up cycle.
Now in the calm of a crypto lull, we are seeing positive renewed attention to the security of protocols both in their mathematical design and from general cyber technology threats well known in web2. We are also seeing great work in the security space with ample technical capability and passionate individuals and companies working overdrive to get ahead of these problems for the next crypto cycle upswing. If anyone’s interested in getting a current view of that, it is more than worth your time to take a look at the talks and discussion at the first DeFi security summit at Stanford last month. If you work in the tech security space and but not yet in crypto, check out that workshop and see if now might be a good time to pivot into learning more about security infrastructure for the blockchain space. The growth trajectory for security in crypto will be massive and highly rewarding to those who pick up its calling!
But back to the question. In the longer term, re-establishing balance between the opposing forces of individual freedoms and regulations requires effort on both sides. Both need to understand the negative impact of excesses and imbalances and work to find the middle ground. We’re seeing that in the crypto space with a lot more introspection and work to understand the underlying useful aims of regulation and map those into viable attempts at self-regulation. Smart protocol teams and companies are also increasingly willing to reach out to or address incoming concerns from regulatory agencies. Similarly, we’re seeing both regulatory and legislative bodies making efforts to have open and ongoing dialogues with the crypto community.
In terms of specific regulatory statements, right after the Merge, there was discussion of ETH and its evaluation under the Howey test with Proof of Stake. While this ongoing news creates a lot of short-term volatility, we can view statements like this as just more ongoing steps in the overall evolutionary process. The immediate reaction and ongoing softness in the crypto space demonstrate the volatility it causes for shorter-term investors, but it's not a systematic existential risk.
#4 Vitalik tweeted that the Merge would reduce global carbon emissions by 0.2%. This effectively makes it one of the largest decarbonization efforts of all time, and we are now seeing increased pressure on the Bitcoin community from environmental groups to also switch to Proof of Stake. Do you think this will play a significant role in the long-term relationship between Bitcoin and Ethereum?
PS: The Merge is a very strong signal that almost all of the fundamental research and new innovations are strongly bound with the Ethereum ecosystem - and that Ethereum could continue to grow out and secure its place as the ultimate settlement layer for all digital assets. The notion of reducing 20bp of global energy usage will factor strongly in the radar of ESG-focused groups and tilt the perception of Ethereum as an environmentally friendly coordination mechanism. There continue to be interesting research (and implementation) efforts towards the utilization of Merge-mining techniques that could continue making use of an extrinsic-security model underpinning Bitcoin and use the same towards securing certain forms of settlement (and maybe use this as an alternative form of redundant security). We do, however, feel that the notion of parametric security and modular design could only come about with the ability to align economic incentives as part of slashing guarantees on the Ethereum network (which, personally, represent one of the greatest incentive designs) - and this could bring about a lot more innovation on the block and network level.
PL: The Merge effectively makes Ethereum ESG-compliant, which is a necessary condition for Ethereum to get further mainstream adoption. Bitcoin, on the other hand, still produces a massive carbon footprint despite supporters claiming it largely uses renewable energy. At the very least, this will make a difference with large investors with ESG concerns, such as pension funds and insurance companies. “Green ETH” narratives are also likely to set a high ESG standard for the whole crypto industry, and any new blockchains will have to take ESG into consideration.
RD: The carbon footprint of everything is top of mind, rightly so given the existential risks of climate change! The environmental impact will be an ongoing central point of discussion in the crypto space and will specifically be a main discussion point anytime Ethereum and Bitcoin are talked about together. There are things that Bitcoin could do to address it while avoiding Proof of Stake, which is antithetical to its ethos because proof of stake removes the anonymous mining feature of POW. However, Bitcoin could look to learn from chains like Filecoin, Arweave or Subspace, or even Chia that uses data storage rather than compute hashing in their anonymous proof of “effort” (in this case, storage) mining protocols. Those are all much more environmentally positive as they avoid energy-intensive computation hashing and instead proof storage of data over time. All of these I mentioned, except Chia, also deliver useful web3 decentralized file storage (akin to but more decentralized than AWS’ public cloud storage). Thus they demonstrate the possibility for Bitcoin to balance the mining anonymity of Nakamoto’s original POW with a method that doesn’t wreck the Earth.
Without a change like this, right now, Bitcoin’s energy consumption and carbon emissions are roughly proportional to its market capitalization. So Bitcoin prices going to a million USD would not only represent a 50x increase in price from $20k, but also cause a roughly 50x increase in its carbon emissions. Barring the Bitcoin community moving its PoW to something more environmentally friendly, unfortunately, the only way to prevent ever-increasing energy and carbon emissions for Bitcoin is to prevent its price from increasing that much. Whether that happens will depend on what investors and DeFi systems that choose what to build on and use. We’ve seen investors and consumers vote with their dollars on contentious societal issues like smoking or hydrocarbon versus renewable energy, so this is a very important topic for everyone in web3 to consider.
#5 The Merge was a hugely pivotal step in making the Ethereum network an efficient settlement layer. Are there any projects you are most excited about that are building in the zk rollups, minimized trust assumptions, and PBS space? It seems like there is a bit of a race on at the moment to build the most efficient scaling solutions for Ethereum.
PS: One highly interesting aspect that people have not yet fully grasped is the magnitude of the economic security underpinning Ethereum and the very broad design space around not only putting this to use to secure blockspace but also powering incremental innovation and a whole host of modular middleware services. EigenLayer (founded by Sreeram Kannan) is one such project which is built on the notion of using restaking as a primitive - and this could enable validators to start offering services such as data availability, RPC services and also things as diverse as censorship-resistant block production frameworks.
We are also very excited about the immense amount of research that is being dedicated to core cryptographic primitives - and extending their capabilities to bring about the notion of a fully private state (which could also reduce trust assumptions on the sequencer networks and reduce MEV capabilities). Founded by world-class cryptographers, Aztec Network has started offering its capabilities to existing DeFi projects to use a rollup without fragmenting liquidity and is a project we will be watching closely.
PL: Besides the race in Layer 2 rollups, I would like to mention some developments in the so-called Layer 3 as well. As L2 extends scalability for L1/mainnet, L3 will also have the ability to further customize L2, either on the functionality or scalability side. According to Vitalik's latest tweet, he thinks the following three L3 will be needed:
(a) L2 for scaling, L3 for customized functionality (for example privacy)
(b) L2 for general-purpose scaling, L3 for customized scaling
(c) L2 for trustless scaling (rollups), L3 for weakly-trusted scaling (validiums).
StarkWare (founded by Eli Ben-Sasson and Uri Kolodny) is a leading scaling solution provider active on this front. Their L3 solution, “Fractal Scaling,” is set to deliver hyper-scalability, better control of the tech stack, and privacy. Basically, after the Merge, the purpose of L2s shouldn't be purely limited to scalability. Instead, L2s need to expand what they’re offering and further extending to L3s is a likely route.
RD: Praneeth nicely covered the important topics of modular blockchains, where the consensus, execution, data availability and even more granular aspects like transaction ingress network, ordering and block building, relaying, and execution can be built separately and composed into different wholes with different features (akin to using different materials like brick, wood, and steel in architecting structures). Patrick dovetailed this with his comments on what some of these modular structures may aim to accomplish as L3s in the Ethereum space, building on top of L2 rollup + L1 mainnet combinations.
So in the spirit of this question about building scale on top of Ethereum, I’ll add some comments back at the foundational base L1 layer that all this modular construction relies on. Like the anecdotal answer an architect gives when asked how tall a skyscraper they can build, we can build structures only as tall as their foundations are strong.
For Ethereum, which aims for maximal decentralization, this aim of a strong foundation translates to very decentralized participation of stakers and validators. Post-Merge, these core Ethereum participants now adjudicate what blocks are being built in the network and ensure the rapid and reliable dissemination of the permissionless and censorless transactions they contain. Two aspects that pervade the mindset of Ethereum and hence drive protocol work are keeping the requirements as accessible and simple as possible for the hardware and network that are needed to run the base nodes of the network, and making sure that economically this version of staking and validating is financially competitive to less decentralized systems like pooled staking in Rocketpool, Lido and others. As Ethereum progresses post-Merge, individual protocol advancements developed and deployed may aim for protocol-builder separation (PBS), MEV-resistance blind ordering, data availability, state history pruning, or stateless validation. However, if you look at the underlying core motivation for these projects, you’ll see they really are about this core ideal of maximal decentralization of the core layer. With this solid core foundation of Ethereum’s L1, we can expect to see ever more amazing multi-layer structures being securely built.
Robert Drost is the head of R&D at MESH. He holds his Ph.D. EECS from Stanford and brings experience leading R&D in large Silicon Valley tech companies. He is a multiple startups Founder and CxO and brings an extensive array of expertise to Mesh’s R&D strategy and projects.
Patrick Li is a research analyst at MESH, where he mainly covers various research initiatives related to internal treasury management and ad-hoc analysis for portfolio companies and secondary markets. He has over 6 years of experience trading fixed income instruments in investment banking and pension funds.
Praneeth Srikanti is an Investment Partner and co-founder at Ethereal Ventures. He has a background in enterprise networking at Microsoft and Oracle and has worked on investment teams of Consensys Mesh along with other European and US VCs. He obtained his MBA from ESADE and the Chicago Booth School of Business, and is a CS graduate from IIT Bombay.
Note that the opinions reflected in this article are proprietary to those sharing them and do not reflect an official MESH stance on the Merge.