Ethereum 2.0: What You Need to Know

Ryan Watkins and Wilson Withiam, senior research analysts at Messari, explain the nuts and bolts of Ethereum 2.0, including how it will transform ETH as an asset and why they believe it will be Ethereum’s most ambitious upgrade yet. Topics include:

• the different phases of Ethereum 2.0 and the functions they serve
• the problems Ethereum hopes to solve with this upgrade
• the technical requirements needed to ensure the launch can happen
• the new proof of stake consensus and why Ethereum is leaving proof of work behind
• the requirements necessary to maintain a validator node on the Ethereum 2.0 network
• why users won’t be able to use their ETH once it is staked on Ethereum 2.0
• the incentives and services that will allow users staking on Ethereum 2.0 to continue using their ETH on Ethereum 1.0
• whether the appeal of DeFi is a threat to Ethereum 2.0 staking
• how Ethereum 2.0 incentivizes client diversity
• the monetary policy of “minimum necessary issuance” that supports Ethereum 2.0
• how Ethereum 2.0 will allow ETH to achieve the unprecedented combination of a store of value, a capital asset, and a commodity
• what will be done in the short term to help ease the scaling problem while ETH 2.0 is being built
• whether scalability can be efficiently addressed in time to prevent migrations to other blockchains
• and the most significant risks Ethereum 2.0 is facing

Thank you to our sponsor!

Episode links:

Ryan Watkins: https://twitter.com/RyanWatkins_

Wilson Withiam: https://twitter.com/wilsonwithiam

Messari Crypto: https://messari.io/

Messari report on Ethereum 2:0: https://messari.io/road-to-eth2  

Amount of ETH in the Ethereum 2.0 Deposit Contract: https://etherscan.io/address/0x00000000219ab540356cBB839Cbe05303d7705Fa#analytics https://www.stakingrewards.com/earn/ethereum-2-0

ETH issuance based on amount staked: https://docs.ethhub.io/ethereum-basics/monetary-policy/#proof-of-stake-impact

Dan Elitzer article on DETH: https://medium.com/ideo-colab/the-deth-of-ethereum-98553866e81b

Transcript:

Laura Shin:

Hi, everyone. Welcome to Unchained, your no-hype resource for all things crypto. I’m your host, Laura Shin, a journalist with over two decades of experience. I started covering crypto five years ago, and as a senior editor at Forbes, was the first mainstream media reporter to cover cryptocurrency full-time. Subscribe to Unchained on YouTube where you can watch the videos of me and my guests. Go to Youtube.com/c/unchainedpodcast and subscribe today. 

Laura Shin:

Today’s topic is Ethereum 2.0, which launches seven days after 524,288 ETH are claimed in the deposit contract. We will unpack what that means in a moment. Here to discuss are Ryan Watkins and Wilson Withiam, both senior research analysts at Messari. Welcome, Ryan and Wilson. 

Ryan Watkins:

Hey, Laura. Thanks for having us on. 

Wilson Withiam:

Hi, Laura. Excited to talk about this topic so thanks for having us on. 

Laura Shin:

So, let’s just start with a real basic question. What is Ethereum 2.0?

Ryan Watkins:

Yeah. So I think Ethereum 2.0 is a couple of things. Most importantly for Ethereum, it’s kind of its most ambitious upgrade that’s been seven years in the making that will scale Ethereum so that it can actually serve as a globally scalable financial infrastructure. Two, it would also make it more secure and then as well it will also transform ETH as an asset. Those are kind of the big picture ideas of what Ethereum 2.0 brings to Ethereum. 

Laura Shin: 

And what problems does Ethereum hope to resolve with this upgrade?

Ryan Watkins:

Yeah. So as far as Ethereum the blockchain, it’s really a couple of things, and I’ll divide this into proof of stake and then sharding, which are kind of the two major components of Ethereum 2.0. With proof of stake, it’s trying to solve three things. One is making Ethereum more secure. Two is ridding it of centralization risk from miners and then three is making sure that participation in the consensus process is as accessible to the users as possible. So, that’s for proof of stake and then for sharding, that’s the scaling solution for Ethereum. Basically, what that entails is, it’s like splitting up the Ethereum blockchain into subsets called shards and running them in parallel. The reason why you do this is because you want to keep the requirements to run a node as low as possible so that there’s not an excessive burden on people who run these nodes and it is actually accessible to people with basically just consumer hardware. The reason why you do that is because it ensures that you can get the scalability while maintaining that core property of being decentralized. 

Wilson Withiam: 

Yeah. I think I’ll add to that is what we’ve seen since Ethereum launched is that it’s run into a couple issues along the line, specifically during 2017 during the ICO craze and then most recently this past summer when various DeFi liquidity mining programs were up and running that under its current architecture it can get under a significant amount of load and that tends to delay transactions and also increase the transaction cost. So, Ethereum 2.0 was Ethereum developers and researchers recognizing that under Ethereum’s current architecture, it couldn’t scale to meet these characteristics that they were truly desiring out of the platform. 

Laura Shin: 

So, we’re going to dive into each of these in greater depth throughout the episode, but let’s do, first, a quick overview of the different phases of Ethereum 2.0. Can you name them and describe what happens in each? 

Wilson Withiam:

Of course. Yeah. Happy to. So, the first phase is the one that is imminent. The one where users can actually start staking is called phase zero. That simply is the launch of something called a beacon chain and the beacon chain is basically your central core layer of this new network. So, it’s separate from the existing network and it will be this coordination layer that allows people to stake. It coordinates all the different validators or stakers on the platform and then it acts as the coordination layer for all of the additional upgrades that will be added later on. So, all the additional shards it will serve as a reference point to all of those. So, it can actually connect information between all those other chains that eventually get added later on. 

The second phase, phase one, a little bit of a longer timeline. So, I think some of the early estimates are maybe late 2021. That will add the scalability component. So, phase zero, nothing much more than staking and reaching consensus. The phase one will be adding all these different shard layers and that will be the base scalability component for Ethereum 2.0. So really phase one is just kind of reaching consensus with a whole lot of different things, so you’re reaching consensus not only on the beacon chain but also incorporating all those other shard chains. So how do you reference the block data that gets added to the entire network across 65 different chains. 

The third phase is called phase 1.5. That is actually the merge of the existing network with the new network, Ethereum 2.0. Really, it’s just a simple swap of the proof of work consensus layer that Ethereum uses right now over to the proof of stake layer that’s on Ethereum 2.0. So how it’s going to look like to most users and applications it’s not going to change a whole lot. It’s supposed to be pretty seamless but a lot of it is just going to be the underlying layer where it just kind of switches over and all the blocks are being created through a proof of stake instead of by miners. 

And then the final phase, the big one is phase two and that will be when you’re adding execution to these different shards. So, during the first three phases, they actually will not have smart contract execution ability, so you won’t be able to actually port smart contracts on to these individual shards, but there is a bit of a caveat to this whole process and it’s that you might be able to start using some of these shards for scalability means before phase two. 

So there is a possibility that you can get the scalability gains without having to add smart contract execution ability to every single one of these shards so there is a route as we’re going down this roadmap for Ethereum 2.0 where phase two may not be necessary. So, even though it might be farther out on the timeline you can start seeing really the advancements that Eth 2.0 are bringing far sooner than that. 

Laura Shin:

We’re going to spend most of the show talking about phase zero, which is the most imminent change, as you mentioned. Listeners should know that we’re recording this on Monday November 23rd because there is kind of things going on right now. So, by the time this comes out we don’t really know where things will have landed, however, at this moment we’re about 75 percent of the way to meeting the minimum thresholds and that would trigger off a seven-day period before the network could launch. Although, also the earliest day it could launch would be on December 1, so it’s kind of like tomorrow, November 24 really is the first day when people would have the opportunity to stake that minimum amount. So, can you just maybe go through the technical details of what is required to make this launch happen? 

Wilson Withiam:

Yeah. So, there’s really two components to when this chain can launch. Really there is, like you mentioned, a minimum deposit amount and a minimum genesis date. And all you really need to meet is that minimum deposit amount, which is 524,288 ETH and as soon as validators deposit that into this contract, which lives on the current Ethereum network, it will go into the deposit contract. People can start running clients on this new Ethereum 2.0 network now and as a client they read the contract. They’ll recognize it. If it’s a valid deposit, they’ll then tell the new beacon chain to mint 32 ETH for that validator so they can start staking on the network. So, really just kind of a constant contact between the two chains all generated through that single contract. 

The other component, the minimum genesis date is right now listed as December 1, and it must…all that means is that is the earliest date it can launch. So, it can launch any time after that, it just needs to meet that minimum deposit amount. For that deposit amount, once that deposit amount is reached, there is a seven-day delay between when it’s reached and when it’s launched. Like you mentioned, it could be November 24th. That implies a December 1 launch date. It can reach that minimum amount any time after that and then seven days after it reaches that amount then the chain will officially launch. 

Laura Shin:

So, do you guys want to make a bet? We can all either look really stupid or really dumb by the time this comes out because everybody will already know by the time this comes out but we, at this moment in time don’t. Do you think we’re going to get the minimum threshold by tomorrow, November 24th?

Ryan Watkins:

Yeah. I do. 

Wilson Withiam:

I agree. I would put money on it, yeah. 

Laura Shin:

I have a feeling as well because just even in the time that I was researching for this show, the minimum amount was going up very quickly. So today, right when I went to make that little change on what amount we’re at right now, I was surprised to find we’re at 75 percent. So, we were joking before we started recording the show that we think the whales are all back channeling right now. Anyway, all right. So regardless of when the launch actually occurs, let’s discuss one of the most fundamental changes that’s happening, which is this switch to proof of stake from proof of work. Why is Ethereum moving away from proof of work to proof of stake?

Ryan Watkins:

Yeah. So, I touched on this a little bit in the overall design of Ethereum 2.0, and it’s really three things. It is for security, it is for energy efficiency and then it’s to reduce the risk of centralization for miners. It can happen to all three. So on the security front, basically the idea here can be divided into two different categories. One is that you want to make the cost to attack the blockchain very high and then the other one is that in the extreme case where you do get attacked, 51 percent attack, it’s easier to kind of coordinate a new chain that gets rid of the attacker. 

So the way this works in proof of stake versus proof of work is…actually I’ll even rewind to GPU-based proof of work because that’s what Ethereum currently operates on. Now, GPUs are ubiquitous. They’re all over the world. They have ton of different use cases, and if you really wanted to you could rent out GPUs and use those GPUs to attack a proof of work blockchain. We’ve seen that with some blockchains that exist that are susceptible to these attacks. Now, proof of work with ASICs is a little bit different because ASICs are specifically designed to mine the cryptocurrency they’ve been designed for. So, with Bitcoin their ASICs are designed to mine bitcoin. That’s the only thing they can be used for. So, it’s not just about the ability to be able to rent out this hash power from some cloud provider. Proof of work first involves some high initial upfront investment to actually participate in this mining process. 

Now, the difference between proof of work and proof of stake is that with proof of stake there’s two different things with this upfront investment. It’s, one, whereas with an ASIC, an ASIC typically depreciates over, say, about two years and then you have to buy a new one to stay competitive. Your stake is just an internal accounting measure in Ethereum and doesn’t depreciate and then, two, your stake also is not permanently locked. You can actually remove it after a withdrawal period. So really when you go to stake you only face an opportunity cost. The reason why those two things are important not being a depreciating asset and then, two, only facing an opportunity cost is that it makes the attractiveness of putting up more capital at risk a little bit more attractive. The reason why is because if you can just pull out your stake after withdrawal period, if your stake doesn’t depreciate, then you’re probably more willing to accept a lower return for a given amount of capital you put up, so it raises the cost to attacks. So that’s one aspect of the proof of stake component. The other one is how it’s easier to counter coordinate against the 51 percent attack. 

Now, the idea with that is that when you…and granted, it’s very difficult to 51 percent attack a proof of work blockchain like Bitcoin at this point. You’re talking about billions of dollars of investment and much more. But in the case that someone was able to 51 percent attack Bitcoin there’s basically a couple of things you could do. You could basically create a fork and then switch the mining algorithm to something else so that the attacker can’t continue to use their ASICs to attack the chain. But the problem when you do that is now you have to bootstrap hash power for this new chain. Probably the way you’re going to do that is by switching to an algorithm that could be mined using, say, GPUs. 

Then some of the things I mentioned previously is that you can just rent hash power for GPUs on the internet and then you can kind of just maybe say if the person attacking, well we can just keep on forking but it’s just kind of this never-ending cycle of keeping having to fork and someone being able to attack it. Now, the difference with proof of stake is that in the event there’s a 51 percent attack, and granted like I said, it would be extremely difficult because where are you going to buy…let’s say like right now there’s 10 percent ETH participating in proof of stake and ETH is worth 60 billion dollar market cap. That’s six billion dollars’ worth of ETH you need to buy. I mean I don’t know where you’re going to buy six billion dollars’ worth of ETH but if you could do that, the thing is if you’re launching your 51 percent attack on ETH, Ethereum can create a new fork. Granted, forking is definitely hard but they create a new fork and just delete your stake. So basically, the equivalent that many Ethereum researchers have…a great analogy is that it’s almost like the equivalent of burning someone’s ASIC farm when you can slash someone’s stake for behaving maliciously. It’s a way of kind of permanently ending this person’s ability to attack the proof of stake blockchain. So, that was a long portion on security. 

The other two are much quicker on mining centralization and energy efficiency. Energy efficiency, the basic idea is Bitcoin uses a tremendous amount of energy to secure itself, as much as a small country. You can debate whether that’s good or bad. It’s doing something useful but end of the day you’re burning enormous amounts of electricity to make this thing work, so if you could do it more efficiently is probably a good thing. 

Then last one on reducing mining centralization risks, that’s reducing the risk across a couple of different dimensions. I mean one is geographical. So, if you think about…this is not some conspiracy theory against Bitcoin but 65 percent of hash power is in China. So, it’s kind of one risk you can get rid of with proof of stake. There’s also the risk of manufacturer centralization. So, I want to say about 90 percent of all ASICs are manufactured amongst four firms in Asia. So there’s kind of that risk there. Another thing with proof of work or ASICs is that there’s economies of scale. So the larger you become the more cheap it becomes for you to mine with Bitcoin because you negotiate bulk orders for ASICs and it’s just easier to scale it, whereas with proof of stake, in theory anyone who has 32 ETH can quite easily participate in the consensus process using commodity hardware. So that’s the basic idea, like mining centralization risk is kind of getting rid of those risks, although maybe not that likely, just getting rid of the fact that they exist. 

Laura Shin:

One thing I want to dive into a little bit is how staking works exactly because I just wonder, is it going to be one of those situations where more everyday people can become validators but then I don’t know what it takes to be running a validating node and keeping the uptime such that you won’t get penalized or slashed. I just wonder, will a lot of these end up being on centralized services anyway where they’re kind of paying somebody else to do that part for them? 

Ryan Watkins:

Yeah. So as far as how can you stake, there’s probably three different ways. One would be doing it through a centralized provider like an exchange or staking as a service provider who will just do all the work for you. The benefit of that is not only convenience but the fact that you can do it with less than 32 ETH, so if you hold one ETH you can now participate in this. 

The second way is to do it yourself. If you’re doing it yourself, it’s actually not that difficult. You can do it on a pretty standard laptop and the way that the incentive systems of rewards and penalties was designed is that it’s very forgiving. So even if your computer is down 20 percent of the year, you’ll likely still be able to be profitable. So that’s what I mean it was designed to be forgiving so that even if you’re just not good, for up time you can still be profitable or at the very least, not lose your money. 

Then kind of third would be these kind of decentralized staking pools that are almost like two-sided marketplaces where on one end you have people who want a stake, who submit their ETH to this contract and this contract then goes in, allocates that stake to different validators in this ecosystem and that’s another way of just bidding and staking. 

Laura Shin:

All right. We’re going to talk a little bit about kind of the opportunity cost between staking and doing other things but let’s just make one thing clear, once people move their Ethereum to Ethereum 2.0 in order to stake they’re not able to use it anymore. Why is that? 

Wilson Withiam:

At this point, there are two different chains and so you don’t want to have two different chains that represent the same native asset having assets that can be tradable and running on each because they you start getting some price deviation between the two, and it can complicate matters in case any issues do pop up as ETH 2.0 starts to evolve. So for instance, a bug runs out, maybe they need to find a way to resolve any differences between the staking amount and what’s being read on the current Ethereum network to make sure those are completely equal because you start running any sort of differences between the two that’s going to make it far more complicated when they do finally merge. So right now, keeping them separate and making sure everything on Ethereum 2.0 is untransactable can prevent any sort of issues like that running down the road but once they do merge then essentially Ethereum 1.0 and 2.0 become the same network. It’s essentially Ethereum at that point and all of the accounts and balances will be transferred on to Ethereum 2.0 and they can be read the same as everything that’s on the beacon chain which would allow everyone to start removing the staked amount that they have as well as any rewards that they may have earned during that time. So it’s more of a safety measure at this point, which makes sense because going forward you’re only going to be able to use Ethereum 2.0 to a limited extent so really you’re just looking forward to when that functionality starts to unlock that’s when you’ll be about to access everything. 

Laura Shin:

So, because obviously people realize that or the designers of Ethereum 2.0 realize that there’s a lot of things people could be doing with their ETH so they probably need to incentivize them in some way to move the ETH over to Ethereum 2.0 without losing too much opportunity cost. So, what are some of the examples out there of new services that have popped up that still enable people to use ETH on ETH 1, once they have started staking?

Ryan Watkins:

Yeah. So, one of the biggest hesitancies many people have is that there’s going to be this kind of lockup period on your ETH. I mean it will be unlocked by phase 1.5 but who knows really when that comes. It could be 12 months from now. It could be 36 months from now. For all practical purposes it’s indefinite. So in the meantime you need to be able to access some liquidity on your ETH. There is basically two major classes of solutions. One would be kind of like what stake is doing. So basically, what they allow you to do is if you stake through them they will allow you to borrow UCC on your ETH. So, it’s like getting a kind of collateralized loan. It’s kind of this larger class solution for staking derivatives. These class of solutions are kind of a little bit less developed at this time than I would say the former, but the basic idea is that you have either a centralized service provider like an exchange or you’ll have a decentralized protocol, say like Rocketpool or Sapphire you will be able to deposit your ETH with them to stake and then in exchange they’ll give you kind of an ETH derivative, which was called DETH for now instead of death. The idea is that you can use this like DETH on Ethereum to access everything in Ethereum. So, there you could send it to Compound and borrow DAI against it. You could mint DAI against it in Maker or whatever opportunities you want to do in Ethereum, you get to do with this kind of derivative of ETH. 

Laura Shin:

Well, one thing, despite all these developments and allowing people to both stake and do other things with their Ethereum, there is this concern that because there’s a lot of financially appealing opportunities in DeFi on Ethereum that people won’t be super interested in staking. How much of a threat or concern do you think the appeal of DeFi is to staking on Ethereum 2.0?

Ryan Watkins:

I actually don’t think it’s a major concern. I think the biggest thing between staking and DeFi just boils down to how validator returns are designed. So, if there was this event where, say, yields in DeFi were very high and then some people who were staking, this is actually significantly more important in phase 1.5 and beyond when you can actually withdraw your ETH and use it in DeFi, although I will touch on the phase zero kind of after this thought. 

If there was a case where ETH was being withdrawn from staking to go for use in DeFi, the way that the return to validators are designed is that if there’s less people staking, the rewards will be higher and if there’s more people staking there will be…rewards will be lower. At a certain point, the rewards become more attractive again so it’s kind of a minimum amount how much people would want to leave for DeFi for staking. Then kind of another related point is that staking and DeFi are just two different products. Staking is you’re just getting kind of a native yield on your ETH for providing Ethereum security and basically the two main risks that you assume…well one, the Ethereum blockchain failing, which is just systemic risk that you get from holding any asset on Ethereum. Then two is like validator risk. The risk that your validator just doesn’t perform or you behave maliciously and even that can actually be diversified away if you were to go through kind of like a decentralized staking protocol that kind of diversifies your validator risk across multiple validators. 

Laura Shin:

But can you just lay out what those penalties or what happens if you get slashed, just so people know? 

Ryan Watkins:

Yeah. So part of the idea is that if you behave maliciously, let’s say you double sign blocks you can lose some of, if not, all of your stake for doing so. Basically, what that means is that 32 ETH that you posted as basic collateral to stake gets deleted. 

Laura Shin:

Wow. Okay. Yeah. That’s a little bit harsh but I guess there have been plenty of people in DeFi this summer who got totally wiped out by various things that happened. 

Ryan Watkins:

One hundred percent. 

Laura Shin:

So one other thing that I wanted to talk about is this incident that happened on the testnet where one particular client for Ethereum 2.0 had a design flaw and that caused participation on the network to go from 80 percent down to five percent because pretty much all those validators got wiped out. So how is client diversity, which is like a principle really that is important to Ethereum and weirdly not at all to Bitcoin or rather they just have completely different perspectives on it, which is super interesting to watch. But coin diversity is something that Ethereum really values and yet here we have certain clients that are really just becoming the most popular out of the gate. How is Ethereum working to mitigate the risks around that?

Wilson Withiam:

Yeah. I’m happy you brought that up. It’s kind of very interesting to look at the two different philosophies. So right now you have Bitcoin where over 98 percent of clients run on Bitcoin, run Bitcoin Core. On the other side, you have Ethereum where about 80 percent are running Geth at the moment but there is a bit of diversity into that 20 percent of what clients are people running there. So, there’s definitely a push to move towards a more diverse client base so that you don’t have one client that’s overreaching or having too much control over the network. 

Laura Shin:

Yeah. But actually one thing I’d say is that they’ve been trying to do this since the time that they launched and it’s always been that there’s been one dominant one, which is also very interesting. But anyway, keep going. 

Wilson Withiam:

Yes. I completely agree and I think they blame a lot…I don’t want to say blame but I think a lot of the reason behind that was they didn’t really have too many clients to start with when they first launched the chain. So, you have this first mover advantage where Geth becomes the popular client to use. It gains a lot of traction, gains a lot of trust and all of a sudden it’s kind of hard to eat into that market share. So that’s why, for Ethereum 2.0, the Ethereum Foundation put a lot of money via grants, into building out a variety of clients. I know they, at one point, were working on eight to ten different clients. At the moment right now, you have four viable ones going into the beacon chain. 

Even with that said, there has been one client named Prism that has become the favorite amongst most validators. Like you said, on the testnet it had a bug and that issue reared its ugly head. That one bug caused the entire network not go down but to drop to a very low percentage of validators actually working on the client so that means they couldn’t finalize blocks. They could keep building and adding blocks to the network  they couldn’t finalize them, which just means that at some point it’s possible that those could be reverted. 

So how do you prevent that? So there’s a pretty unique design within Ethereum that if more than 33 percent…if you can’t finalize you need 66 percent of validators actively working and voting on blocks to finalize the chain. If you can’t do that then Ethereum will automatically start penalizing validators that are not online. So, it’s called an inactivity leak. If you’re inactive it’ll start gradually and then actually exponentially reducing your staked amount either to get you to a point where you’re dropped off the network or to actually incentivize people to come back and join the network and do what they’re supposed to do. So, eventually it gets back to that 66 percent minimum. What that means in the terms of client diversity is if you were running a majority client and say that is somewhere between 40 and 60 percent of the network and it has a bug you are automatically under risk if all of a sudden those clients go under. So you are, in some sense, incentivized to run a minority client and run one that may not make up 33 percent or more of the network because then you are not at risk for those bugs. 

Laura Shin:

Yeah. It’s almost like surge pricing or something but applied in a completely different context. 

Wilson Withiam:

Exactly. Yeah. 

Laura Shin:

All right. So in one moment we’re going to talk about how ETH, the asset, will change but first a quick word from the sponsors who make the show possible. 

Laura Shin:

Back to my conversation with Ryan Watkins and Wilson Withiam. Yields for staking start at 30 percent if the number of ETH staked is around the 524 thousand necessary to launch the beacon chain. But if it ever reaches 16 million ETH staked, the max yield would be about four percent a year and if it goes higher like to 100 million then it would drop below two percent. So in general, when you look at this, does it feel like this is a good monetary policy that would keep the chain secure, etcetera, or are there any concerns that you have about it?

Ryan Watkins:

Yeah. I think this is a good opportunity to talk about kind of the philosophy of Ethereum’s monetary policy. So, Ethereum’s monetary policy can be defined as minimum necessary issuance. The basic idea is that Ethereum will always issue an amount of ETH that will keep the network secure over time. The difference between this philosophy and, say, Bitcoin, which is deterministic issuance and fixed supply is that where as with Bitcoin it’s kind of optimizing for monetary idealism in the sense that you have people that created this that wanted to create this extremely scarce big supply of currency that was resistant to debasement. Whereas Ethereum says, hey, yes we want our currency to be scarce but what’s more important is that Ethereum blockchain stay secure. So, that’s kind of like the monetary policy for Ethereum versus Bitcoin, optimize for security over this monetary idealism. Each has their own tradeoffs. So, the basic idea is that there’s tradeoffs with these two different monetary policies. For Bitcoin, the tradeoff is that in a sense you kind of set your security budget arbitrarily in the beginning because you just said, hey, we’re going to have halvings every four years and we’re going to have a fixed supply and this fee market for block space is suppose to pick up. That’s what’s supposed to secure Bitcoin in the long-term. 

Then the tradeoff for Ethereum is that Ethereum’s monetary policy, I mean sure you can say that it’s minimum necessary issuance but who determines that? Then what does it actually mean, practically speaking? Because with Bitcoin it’s very easy to understand, yeah, every year there’s going to be X amount issued. This is going to be your total supply in 2056. It’s all deterministic, it’s easy to understand but whereas with Ethereum it’s just like I get that. It still has to meet a minimum necessary to be secure but it’s a little bit hard to wrap your head around what that means and how it works. There’s downsides to both but I think there’s room for experimentation. They each have their own benefits that each community believes are important. 

Laura Shin:

One of the most interesting parts of your analysis around Ethereum 2.0 was about ETH, the asset, and how that will change. You said that under Ethereum 2.0, ETH will become a store value, a capital asset, and a commodity and that this is an unprecedented combination. So how does it fit the definition of all three and how do you know that this is the first time in history that this has happened? 

Ryan Watkins:

Yeah. So there’s…Chris Burniske I think back in 2017 introduced this paper from an academic called, I think his name was Robert Greer about the three super classes of assets. The idea is that every asset in the world that has ever existed can be classified into these three different categories. One being stores of value with the idea being that these are assets that maintain their purchasing power throughout time. Two being capital asset with the idea being that these are assets that produce or generate an income. And then three being commodities, which are assets that can be consumed or they are transformable into something else. So examples of those would be for a capital asset would be a stock or a bond in Apple. It’s kind of this income-producing asset. An example of commodity would be oil. You can use oil for a ton of different things to power your car. It can turn into different end products and then for a store of value it could be debatably the US dollar, although some people would kind of scoff at that in this space. Or gold, which actually is both a commodity and a store of value. So that’s kind of the high-level idea of these asset classes. 

Now, Ethereum and Ethereum 1.0 is a commodity and a store of value. It’s a commodity because it’s used as gas to pay for transactions. I think this analogy will be especially powerful when a new EIP is introduced called EIP 1559, where the majority of transaction fees will actually be burned instead of being paid for miners. So quite literally be consumed by the Ethereum blockchain. Then for store value, that’s Ethereum’s use as an asset in DeFi to store value, to send transactions, like that’s the idea. So that’s Ethereum as it exists today. 

Now, with Ethereum 2.0 it introduces staking, and what staking allows you to do is you can post your ETH as collateral to become a validator on Ethereum 2.0. Now you can actually start generating yields on your ETH. I mean the amount of yield you will get varies on the amount of ETH that’s being staked. So 524,288 ETH, kind of like you said before, the rate that you will be getting is very high. It’ll be about 23, 24 percent and then at about 10 million or 16 million ETH staked, it’ll be somewhere between four and six percent. But the basic idea is that now you’re getting a native yield on ETH. 

So, when you combine those three things, it’s like you have all these different sources of demand for ETH the asset, and ETH is being used for all these different things. I’m a little bit hesitant to say that it’s unprecedented because I’ve not literally explored every asset in history. But from what I have seen, there’s nothing really like this. Something that is a non-sovereign store of value just like Bitcoin that is also used to power this globally scalable computer and also offers a native yield almost similar to a kind of sovereign bond in a sense. The combination of the three just makes ETH super interesting as an asset. 

Laura Shin:

When you mentioned EIP 1559 and about how the fees will be burned, how do you think that will affect the issuance of ETH or the net issuance I should say and what it will look like in terms of a store value? 

Ryan Watkins:

Yeah. I think some helpful context as well is that when the beacon chain launches hopefully eight days from now on December 1st, all the issuance from beacon chain will actually be incremental to the issuance on Ethereum 1.0. So over the next year-and-a-half, Ethereum will issue about 3.8 percent worth…its annual inflation rate will be about 3.9 percent. Then on the beacon chain you’ll get an incremental issuance depending on how much ETH is staked of anywhere from .10 percent to .8 percent in most aggressive scenarios where there’s 30 million staking. 

So in total, Ethereum launched their policy between four and five percent over the next one to two years. Now, where this changes is that if EIP 1559 gets implemented on ETH 1 where the majority of the transactions will be taking place until the ETH 1-ETH 2 merger is that you will actually be able to reduce the issuance in the meantime. So, if you think about over the summer…so I’ll actually back up again. So that’s kind of what the monetary policy looks like for the next year-and-a-half. I think where it’ll be particularly interesting is once the ETH 1 merger happens and ETH 1 is now merged into ETH 2 as a shard, well that once incremental issuance from the beacon chain, like I said of .1 percent to .8 percent is now the only issuance for Ethereum. 

This is actually where it gets interesting because I believe it’s 20 million ETH staked per day Ethereum, and this is in Ethereum 2.0. A 20 million ETH stake per day Ethereum will be issuing about two thousand ETH per day to pay to validators. If you look at how much ETH per day people were paying over the summer in DeFi on average, it was about five thousand ETH. So if Ethereum 2.0 is launched today and people are paying the same amount of transaction fees, net issuance would be way negative. 

No, of course, there’s some nuance here and the big nuance is that Ethereum 2.0 is going to increase transaction scalability. So, in theory, it’ll actually alleviate the fee pressure so that it’s not where you have to pay 40 dollars to go and send your friend some money or exchange an asset on Uniswap. But according to some of the analysis that we did that we based off the economic model is that consensus put together it looks like the likely issuance rate for Ethereum 2.0 will probably be like let’s say net issuance rates, net of the burns will be anywhere from say negative 0.5 percent to 0.5 percent. It really just depends on how many validators are participating and ultimately how much ETH is being burned. The big idea here is that issuance for Ethereum 2.0 will be extremely low, if not negative. 

Laura Shin:

One other thing I wanted to ask about was the DETH that we mentioned earlier, these ETH derivatives that will be minted so that people can stake but also then continue to play around in the Ethereum 1 ecosystem. So, let’s say I stake my ETH and I get back some DETH and then I use that to mint DAI on MakerDAO but then my vault gets liquidated. So hopefully people understood that. I think most of my listeners have been following my DeFi podcasts but essentially that’s like the money that you put up as collateral and now you’ve lost it but supposedly you own this ETH on Ethereum 2.0. So, if that happens then who can claim the ETH that I’ve staked on Ethereum 2.0?

Ryan Watkins:

Yeah. So I guess this is basically before phase 1.5 when ETH 1 merging to ETH 2 and you’ve got these derivatives where you can’t actually claim the underlying ETH. So, in this scenario where the MakerDAO system votes in kind of DETH, this derivative ETH that is collateral and then people are minting DAI against this collateral, the liquidation process would be the same. The collateralization ratio falls below the target amount, the DETH gets liquidated and then whoever is the liquidator now has this DETH. Now, it depends on the staking solution that created this DETH but let’s say it was one of these decentralized protocols that minted this DETH. Well, yes, eventually you would be able to claim the underlying ETH by having this DETH derivative. Basically, what it is, is for these decentralized staking protocols is that this DETH is a right to redeem a certain amount of underlying ETH. So if you hold 32 DETH and the exchange rate is one-to-one you’ll be able to redeem the 32 ETH once it is redeemable. So, I guess the important concept here is that, yes, all these ETH derivatives are fungible when they’re issued from the same provider. 

Laura Shin:

All right. So now let’s talk about the fact that this whole process to moving to Ethereum 2.0 is going to take years, and meanwhile Ethereum, as we’ve been talking about, has kind of been bumping up against its max capacity, in particular because of DeFi, which is a pretty active ecosystem. One of the issues with DeFi is that the different protocols in DeFi are composable, which is the way that people describe the fact that you can do a transaction that involves multiple of these protocols within one block. That’s why they call DeFi money Legos. So, because that would kind of want…all the DeFi protocols would then sort of want to stay with each other to keep in the same ecosystem. What do you think is most likely to happen in the meantime to solve this fairly urgent problem that we already have around scaling?

Wilson Withiam:

You’re absolutely right and a lot of the Ethereum researchers recognize this. Scalability is a need now and frankly what Ethereum 2.0 can bring from a scalability perspective is years away, maybe a year, maybe a little bit over that. So in the short-term, the solution that a lot like Vitalik and some of the other teams have been working on for years is transitioning over to a layer two solution. Specifically the one everyone is talking about is called rollups. There are various flavors of them but in general you can kind of look at them as rollup contracts. Essentially, all it is, is taking a lot of the transactions that DeFi is handling today, moving them to a layer that’s above Ethereum right now, so kind of taking them off chain but having security ties back into the chain. So, it would periodically actually transfer these transactions, batch them together and submit them to the base chain. So, in a way, it’s providing similar security guarantees. 

So I think the perspective is that even if any issues happen on these chains, on these other layer two solutions you would still, at any chance, be able to claim the assets on the main chain. So right now it’s kind of moving towards that layer two solution. Like you brought up, the issue is as an application moves from the base chain up to one of those layers, it removes those composability bonds. No longer will I be able to run a transaction that calls different applications in the same block. It would have to be done in an asynchronous manner or it would have to take multiple blocks. So that one issue with rollups is that and one of the reasons why it has to be asynchronous is to ensure security there is a lengthy withdrawal period. 

So I go up into a layer two protocol. I can start using an application that deployed a contract there. I want to get back out and use some of the other contracts. It’s about, I’ve heard, anywhere between three to four days to a week to get out. So there is this issue of, one, getting liquidity out to trying to call transactions between two different applications across these two chains. There aren’t any limitations to composability within a single rollup so that might mean a lot of applications go and build on the same rollup and eventually one of the issues is if everyone moves onto the same rollup you start running into the same scalability issues that are present today. 

So that’s why you kind of have this relationship where you’re trying to figure out what is the best way to scale Ethereum but not lose what really makes it special, which are some of these composability bonds. So there are some solutions in place to try and help out on the liquidity side to minimize the issues of getting out. So, one could potentially be a cross-chain liquidity solution like THORChain, for example, where you’d be able to actually move between chains without waiting for that lengthy withdrawal time. That’s a decentralized option. You can also go through a centralized market maker to accomplish that. There isn’t a great solution to completely fix the composability issues at this moment. There are some solutions out there, one in particular from TenX Network called Space Folds, which would allow two different layer two networks to communicate and transact. But right now it’s still working towards what is that solution that would all cross-chain composability and that’s definitely going to be one of those issues going forward. 

Laura Shin:

What’s your sense, do you think that they’re going to address the scalability issues quickly enough or do you think that we’re going to start to see significant migration over to some other blockchains? I’ve heard some people saying they’re looking at Polkadot, Solana, NEAR…What’s your sense?

Wilson Withiam:

It’s realistic that some applications, there’s going to be some leakage. I think a lot of what’s going on in DeFi right now is Ethereum centric and moving off is not part of their near-term plan and if it’s going to be, they’re going to move up to a rollup solution. Also, a lot of these other networks aren’t necessarily production ready. They’re really on kind of the same timeline as rollups. So, the rollup solutions that we’re talking about, some of the bigger name ones like Optimism and Offchain Labs’ solution, they might be coming out sometime within the next four to five months and you might see the same sort of maturity trajectory timelines for these other chains to allow applications actually start building on some of these. So I think some use cases, particularly maybe non-financial use cases might move over to some of these other chains but I think when it comes to DeFi applications, Ethereum is where the action is and that’s where a lot of these platforms that are building off of Uniswap, building off Compound, they’re not going to want to move away from that and those base layer platforms aren’t going to want to move from Ethereum because a lot of what they are doing is based on that and it’s tied into what Ethereum has been building. So, if you want to really get the advantages of what Ethereum 2.0 can bring down the line, staying within that ecosystem is really going to be a part of their roadmaps going forward. 

Laura Shin:

All right. Well, we’ll have to see what happens. So we’re running out of time a little bit but I do still want to dive a little bit more into phase one and at least phase 1.5 we can discuss how deeply to go into 2.0. One thing I wanted to understand about phase 1 was actually…I should do this question all at once because as you know, Vitalik wrote a post where he said because the movement toward rollups is kind of happening at a pretty quick pace and there seems to be some consensus around that as a good way to scale and frankly if they make it work then that will actually provide more transactions per second than the originally planned Ethereum 2.0. So, he said that there is a scenario where Ethereum would move to Ethereum 1.5 and then be done. I wanted to know what your thoughts were in terms of whether or not or how likely you thought that would be to happen but then also when I was researching for the show that confused me a little bit is I read that in phase one that’s where they introduce the different shards. But then I didn’t really know how the shards work with the rollups. Are we still going to have all these little mini blockchains and then also the rollups? If you could help me understand and the listeners that would be great. 

Wilson Withiam:

Yes. Okay. So one of the benefits that can happen when phase one launches is that you have all these various shard chains. They’re acting as, they call them, data availability layers. Really, they can just act as kind of storage chains. At the moment, on their own, somewhat useless mainly just trying to get the network to come to consensus but there is an option for rollups to actually use these networks, that they can plug into these networks and use them as data storage layers. So that would actually get…if the rollup plan works and you see applications moving up to another layer, they can actually plug into one of these various 64 shard chains and use them as kind of this data storage layer and that would actually give the opportunity to use Ethereum 2.0 scalability earlier than phase 1.5 or phase two. 

Laura Shin:

So in a way it’s not that different from Ethereum 2.0’s final vision in the sense that there would be kind of like a layer for execution, there would be a layer for storage. It’s just that in Ethereum 1.5 and done what with happen is that the execution would happen on rollups and then the storage would still be in the shards that are introduced in phase one. Is that it? 

Wilson Withiam:

That’s precisely correct. Exactly. The scalability, everything that’s happening on the base layer, that would all be taken care of by shard chains that plug into the beacon chain. All the execution where the applications live, where users actually interact with that’s going to all happen on a rollup layer. 

Laura Shin:

So what do you think the odds are that we end up with Ethereum 1.5 and done versus Ethereum 2.0 as originally planned?

Wilson Withiam:

I think a lot of applications are going to move up to rollups. This is all dependent on how much adoption rollups get and right now you’re starting to see Synthetix be very bullish on the fact that rollups are really going to be the near to mid-term scaling solution for Ethereum. There’s a good chance that Uniswap has done some integrations with Optimism as well, which is a rollup solution and as soon as you start to see some of these big DeFi players move up I just suspect that a lot of other applications will move up as well because they’re very reliant on those. They plug into them and they, like we said, are composable with each other. If this rollup solution works out, which I think it will, I think there’s a very good chance that you’ll either see no phase two or very limited phase two. So maybe instead of putting execution layers on every shard chain maybe it’s four to eight of the shard chains, a very small percentage, which kind of limits the complexity and limits development time so you can actually get there sooner. But I think there is a very strong chance that either phase two gets dialed back or there’s really very little need to go that far. So, you’ll be able to actually use Ethereum 2.0 for its scalability needs a lot sooner than initially projected. 

Laura Shin:

Great. 

Ryan Watkins:

I was going to say, one thing I would add is that with sharding, phase one sharding using these shards of data availability layers and rollups, you can probably get about 100 thousand transactions per second. The benefit of this is that this could potentially come in the next 12 to 24 months and then you have in Ethereum 2.0 which can scale Ethereum orders of magnitude. So, if the ecosystem adapts to this world then we get scalability way sooner than we would if we were to wait for phase two. Then at that point, the prospect of doing phase two becomes less attractive because you don’t actually get any incremental scalability from enabling computation on these shards. 

Laura Shin:

Yeah. I already know we made a bet in this episode, but I think if I were to make another bet I would also bet on this assuming that everything continues going as its been going. All right. So, we’re going to do two last quick questions. One is just a general question but one is for people who might be interested in staking. What do you think is the biggest risk that Ethereum 2.0 faces? If you were an Ethereum researcher or developer or whatever, what would you be thinking about most right now or working on most right now? Where do you see potentially the biggest pitfalls in the game plan moving forward?

Ryan Watkins:

I think one is around ETH derivatives. So there is this idea that…I guess maybe just to kind of zoom out a little bit higher level is that there’s always this tradeoff in cryptocurrency specifically that people care about between decentralization and convenience. People oftentimes are comfortable custodying their cryptocurrencies on centralized exchanges or custodians or in the case of staking this comfortable staking through a service provider rather than doing it themselves. The reason why this is important is because there could be a scenario, although I think this is unlikely because of the existence of decentralized staking options and the fact that validating is very accessible to the average user. But there is a scenario where a lot of people end up staking with say like a Coinbase or a Binance of the world and Coinbase or Binance issues their own ETH derivative and then that ETH derivative becomes extremely liquid in secondary markets. They start to get a lot of integrations in DeFi, maybe even CeFi as well so now you can borrow and lend against it. This kind of liquidity creates a network effect. 

Dan Elitzer wrote a great piece on this where it almost comes like a USDT …like a tether kind of thing where if people want to go and stake they’re probably going to go do it with centralized service provider because, yes, it’s convenient but most importantly they get liquidity and utility that’s not available with these other options. So, there’s a scenario where staking could end up centralized for this reason. Like I said, I don’t think this risk is likely, but it is a possibility. 

Wilson Withiam:

I was going to echo that. I think staking centralization is one of the biggest issues that some people present as a problem with staking because it’s very real. People usually pick convenience over sovereignty. So you could see a lot of people staking with exchanges very easily, but I will preface this with the design around ETH 2.0 somewhat disincentivizes that, because like we said, if validators are running the same clients and they all happen to go down at once, those ones will be more likely to be the ones that get slashed. Because another one of the designs around ETH 2.0, we didn’t mention this, is there are slashing penalties. The slashing penalties get ramped up if more validators are offline or behaving maliciously at the same time. So, if you’re running all these validators on the same services and they’re kind of doing the same thing, they all fall offline at the same time, they’ll all get slashed a much higher amount. So, if you’re staking with a single service, you’re at a much higher risk unless they’re using some sort of decentralized validator service on the backend. 

Ryan Watkins:

I was going to say that kind of brings up a good point that’s kind of related to the ETH derivatives as well is that if a centralized service provider, say Coinbase, were to issue their own ETH derivative, the risk of that ETH derivative are not the same as the risk of a decentralized staking pool issuing its own derivative. The key difference is that with Coinbase, the risk is that…I mean beyond the custodial risk but the risk for that derivative is that Coinbase’s validators just don’t perform or they behave maliciously. With the ETH derivative that a centralized staking protocol issues, the validator risks are spread across this entire kind of network of validators in this protocol. I mean I think actually one good analogy…hopefully it doesn’t bring memories of the financial crisis when you think about this but it’s the difference between an individual mortgage and a mortgage-backed security that kind of pools all these individual mortgages into one security to diversify the risk. 

Laura Shin:

Yeah. It’s like a reverse analogy but anyway. So then this is the question for people who are potentially interested in participating. What do you think is riskier, putting your money in DeFi or trying to stake on Ethereum 2.0 at this point?

Ryan Watkins:

So I think it depends on what you’re doing in DeFi. If you are putting your…basically, any time you put your money into DeFi there are a bunch of different risks that you assume. There is smart contract risk, there is composability risks because of all these different protocols interacting, then there’s also the risk of just Ethereum as a blockchain failing, which I think of it as the systemic risk of DeFi whereas with staking there’s really just two risks. One being validator risk, which you can diversify away with these decentralized staking solutions and then just kind of systemic risk of Ethereum. So those are kind of differences in risk. Now, within DeFi there’s a bunch of things you can do. I mean you could deposit your funds into Compound, which is being heavily audited and known and reputable team or you could go in, put your money into a new yield farm that popped up yesterday that some pseudonymous developers created and you risk getting rob pooled because there’s a lot of differences here. So I would say yes, they are competitive but it’s just there’s kind of different things. 

Laura Shin:

Yeah. I don’t know every last person who listens to my show, but I’m pretty sure if they’re listening to my show then a lot of them wouldn’t be going into an unaudited DeFi contract by an anonymous developer and getting web pooled. So I’m sure people will write in now and be like that happened to me. Wilson, do you have a take?

Wilson Withiam:

Yeah. I think obviously it comes down, it’s different personas are attracted to the various forms of risk. So, there’s just going to be a different user type that is willing to dive into these unaudited contracts that are proposing high-yield returns. But it’s going to depend on what applications you’re going to use and what is your investment horizon. If you’re looking at on a long-term, staking is going to be relatively safe as long as you’re doing your research on what validator provider that you end up using and what service that you’re going to go forward with that because like we said, it’s very difficult to not make money on the staking part. So, you’re really going to have to work hard to not have a good uptime or use a validating service that may not be as trustworthy to not make some sort of return or lose your money on it. Whereas with DeFi of course there are some applications that are very well-audited that have done a lot of work on that service and are trustworthy in the whole ecosystem that would be just as safe to work with. Then obviously there’s a whole other part to it where the madness comes in, where these flash loan attacks are very relevant. So, it’s just something to keep an eye out for that there are just more risks involved with DeFi at this point. 

Laura Shin:

All right. Well, this has been super-fascinating to dive into this huge, momentous change happening on one of the biggest blockchains. Where can people learn more about each of you and Messari?

Ryan Watkins:

You can check out both Wilson and I work on Messari.io and then you can follow me on Twitter @RyanWatkins_.

Wilson Withiam:

And you can find me on Twitter @WilsonWithiam. Not as active as the other guys but feel free to jump on. I retweet them constantly. 

Laura Shin:

All right. Great. Well, thank you, both, so much for coming on Unchained. 

Ryan Watkins:

Yeah. Thanks, Laura. 

Wilson Withiam:

Thank you for having us. This was a lot of fun. 

Laura Shin:

Thanks so much for joining us today. To learn more about Ryan Wilson and Ethereum 2.0 check out the show notes for this episode. Don’t forget, you can now watch video recordings of the show on the Unchained YouTube channel. Go to YouTube.com/c/unchainedpodcast and subscribe today.

Unchained is produced by me, Laura Shin, with help Anthony Yoon, Daniel Nuss, Bossy Baker, Shashank, and the team at CLK Transcription. Thanks for listening.