Re -staking is a method of reusing equity that allows ETH holders to earn additional income independent of network activity. Similar to traditional finance, rehypothecation applications and AVS are similar to verifiable corporate bonds , but the difference is that default events can be punished through cryptographic incentives. Ethereum is the only sovereign entity capable of supporting such a repo market, but users will need a new understanding of the risks . The risks brought by re-staking include slashing, liquidity and centralization risks, but the risks can be mitigated through parameter optimization and liquidity of re-staking tokens, allowing investors to obtain greater benefits.
What is ReStaking?
People find restaking interesting but confusing when they first hear about it because it's not clear where the additional risk comes from. Hopefully we can try to explain here some of what those risks are, and what the different mitigation methods are, and generally make the risks a little less scary. We have heard a lot of discussions related to re-pledge today, including its application in corresponding scenarios.
I want to look at it from the broadest perspective. Next, I'm going to briefly discuss something similar to traditional finance, and at the same time talk about risk, and I'm going to make sure not to write down mathematical equations, and we're going to see some graphs, and then talk about how to mitigate those risks.
This is Vitalik's definition. Note that he uses a hyphen, unlike other usages. But really, the idea is that someone has a way to reuse their stake, with additional slashing rules, completely unrelated to network activity and only for security purposes .
As a high-level diagram, you can substitute that you are an ETH holder, and you participate in the network through LST (liquid staking, hereinafter referred to as LST) or directly run a validator, and between the fees and penalty income Create a feedback loop. This is the situation when there is no re-pledge.
By introducing re-hypothecation, we have these new networks such as MEV Auction and FHE (Fully Homomorphic Encryption). They have introduced a new slashing process when handling fees . These links are triggered by the re-pledge delegation contract, which deposits funds into the pledge contract. During the re-pledge process, the income of ETH holders is the principal plus L1 handling fee and re-pledge handling fee, minus any L1 and re-pledge violation penalties.
Suppose an ETH holder owns 100 ETH and wants to earn 10% by staking. They deposit their ETH into a staking contract, such as Lido, and agree to Ethereum's slashing rules, such as sending conflicting messages (equivocation) and double signing. This way, they support the Lido validator and receive 10%, but may also face slashing.
When the holder withdraws, they receive 107 ETH. If they re-stake 100 ETH, they not only receive 10% from staking ETH, but also receive an additional 5% from the re-staking application. However, at the same time, they also face new risks.
Now, there are three possibilities for a slashing event. For example, people can get slashed on Layer 1, rollup, Layer 1 and rollup. This suggests that users and validators are choosing different risk paths. Compared with the past where there was only a fixed set of L1 slashing events, the current state space has expanded and users need to have a new understanding of risks.
Re-pledge versus...traditional finance?!
This might be a less interesting-sounding concept than tying rehypothecation to traditional finance, but why not?
Re-pledge applications and AVS (Actively Validated Service) are similar to verifiable corporate bonds to a certain extent. Emerging networks seek Layer 1 security, which is a bit like a company using another company's national financial system to issue bonds to seek shelter for its assets. Corporates and sovereigns issue bonds based on the highest possible safety and liquidity.
Argentina, for example, still chooses to issue bonds denominated in U.S. dollars because the U.S. dollar market offers the highest liquidity and the widest range of collateral-buying opportunities. From this perspective, if we think of ETH Layer1 as this type of sovereign currency, then the re-pledge application is borrowing from this sovereign and then repaying the principal and interest.
So, people don't issue bonds everywhere. If you look at the history of the corporate bond market, you'll see that it's very concentrated in a handful of jurisdictions. People are willing to cross their own legal jurisdiction to issue bonds somewhere because the liquidity and exit opportunities are much better there.
In the repo market, people borrow money against government debt or bonds: with a collateralized bond, someone lends you money; you pay it back over time (such as through profits from running a company). Airbus or Boeing, for example, park much of their cash in sovereign bonds and then borrow against the bonds when needed to build factories or other projects.
In 2023, Ethereum is the only sovereign entity capable of supporting such a repo market. But I would argue that one of the differences between restaking and traditional finance types is verifiability, which is the default event, which is when someone doesn't pay, which is actually one of the default slashing rules that you trigger, which is incentivized by cryptography and can be passed Layer1 independent verification.
If it's a government transaction (like I bought a bunch of Treasury bonds and then used them as collateral for a loan, but then I didn't pay back the loan) that person is unlikely to go to the U.S. government to complain. He could try to file a lawsuit, but they have no way to directly prove that I didn't pay the loan, and they can't say that my credit should be taken away or there should be immediate penalties. Therefore, the economic incentives of bond issuers and borrowers are not aligned.
In the case of AVS, this is not the case . In a way, this makes the process less like Defi lending and more like bonds and bond markets.
What risks does re-pledge bring?
Okay, the next question is, what is the risk itself from re-hypothecation ? Of course, we skip smart contracts and operator risks here.
Among the three financial risks of re-pledge, the most serious is the risk of slashing, which is the only way you can directly lose your principal. The second is liquidity risk. Many re-pledge protocols have locked LST (liquidity re-pledge token). Now if a large part of LST is locked in the re-pledge pool, the loss of liquidity means that the price of LST is relatively As ETH is more volatile, the security of AVS is measured in LST, so the implied volatility is higher for end users. This liquidity risk occurs when a certain type of LST is too concentrated in AVS.
Finally, there is the risk of centralization. Taking the DAO hacker attack as an example, assuming that one-third of ETH is in a single AVS, it exceeds the traditional BFT security threshold. Now assume that one-third of the ETH can be forfeited through rules that are not ETH consensus. For example, I did not submit a fraud certificate and I was forfeited, not because of double signing or anything like that. So, in a sense, centralization also means that the two systems are coupled.
Visualizing rehypothecation risk
These charts show the value of a position over time, where the Y-axis is the value of the position and the X-axis is time. The red line represents the point at which default is triggered, which in all of these cases is the worst-case scenario where the principal is completely reduced to zero. We will compare a range of scenarios for different applications and observe their defaults.
In on-chain lending, a default is a style where the indicator function jumps to zero, you have a position with a value that goes below a certain liquidation threshold and then jumps to zero. But this jump is single and a random time, so it's a random stop time based on the actual process, one reset to zero.
Then let's consider the case of perpetual contracts. Perpetual components, whether on-chain or off-chain, have this periodic funding rate update , but it may not be easy to see when these funds are paid. The net payout between longs and shorts, or shorts to longs, results in a change in the value of the position, so you have this periodic jump at times that are almost close to default.
Now consider ordinary staking, so with enough isolation in a sufficiently decentralized network, such as many different nodes, many different overlay networks, different data centers, different houses, the slashing event should be IID (Independent and identically distributed (independently identically distributed), for example, a crash somewhere in AVS will not cause everyone to be forfeited at the same time. Of course, whether it is actually IID is an empirical question.
But in the staking of this model, it is decentralized enough, and these losses caused by slashing are IID events. So the idea is that my locked principal value usually goes up and you can see that I have a slashing event and it goes down, so there are instances where my principal value jumps to zero, but they're independent and It's random.
The final question is, what does re-staking look like?
And in restaking you now have this correlation jump, and the idea here is that when you think about restaking, you can't think of it as being in a siled world the same way you think about lending or perpetual options.
Because in lending, you really only care about a fixed threshold and a price, say the point in time when the purple line crosses the red line; in perpetual options, all you really care about is the maximum deviation of the price within these periodic intervals. . In ordinary staking, if it is IID, you are only relatively safe when the time scale is very long relative to the value scale, but it is not that simple here. You need to carefully consider the interaction between these two factors.
So the interesting thing is that in re-staking, you can actually copy all the previous returns, plus some returns that you can't copy.
How do we mitigate these risks?
So, that leads to the natural final question, which is what exactly are you supposed to do to mitigate these risks, what can you do to make sure that these correlated events aren't too correlated, and that these jumps aren't too big.
In fact, you have two tools at your disposal. The first is parameter optimization. Each AVS has some parameters that control the security of the AVS, whether it is the TVL upper limit or the choice of slashing rules.
As you notice in the picture, if you notice that the AVS slashes are all roughly the same size, but you can construct some slashing rules that may gradually increase or decrease under certain conditions, you are choosing to slash There's more freedom in terms of the L1 consensus layer, you don't have as much (freedom) in terms of the L1 consensus layer.
For another example, if you agree with the analogy I proposed that AVS is actually a corporate bond, then the Liquidity Rehyping Token (LRT) is a bond fund. The question is how to rebalance this bond fund, how Decisions were made to reallocate funds among AVSs.
In effect, both of these things are shaping the picture to make it look more favorable (to investors).
That's it, thank you all.
