RToken insurance and the role of Reserve "unborn" DAO

Unchained · December 26, 2021, 5:07pm

I like the vision of the Reserve team of bringing stability to countries with unstable currencies. However, I don’t fully agree on how they plan on executing such a vision especially in regards to the insurance of the RTokens.

When a user holds an RToken, let’s say pegged to a stablecoin like USDT, there is an implicite assumption that they really hold 1 USD. Users don’t need to know (and some of them won’t understand) how such a stablecoin is structured, if it is fully collateralized or who governs what they are holding.

In reality, the RToken could lose up to 100% of its value if :

There is a bug in the smart contract or if it gets hacked
The Rtoken Governors have the ability to withdraw the collateral backing the RToken
There is a problem with the protocols where the collateral is deposited (this is Defi, everyday some protocol gets hacked even the big players are not that safe)
In case the RToken (like RSV) is using a centralised stablecoin, the Reserve address could be blacklisted by the centralised issuer rendering any collateral in this stablecoin unusable.

The Reserve team are planning on including RSR staking on RTokens for those who wish to get an extra yield for providing insurance. I don’t think this is enough, because once the RSR used for insurance is depleted, the USERS of the RToken pay for the rest. So some farmer in Venezuela holding 1 RToken assuming they are holding (let’s say) 1 USD, the next day some protocol on the internet gets hacked, and his money is now worth 50% less or so … where is the stability in this ?

I know that the Reserve team are aware of such risks but they still plan on rolling the Reserve protocol anyway … As I said I like the vision and respect the work that has been done, but I think users
protection should be a PRIORITY.

Following the current model, the Reserve protocol will be creating a new form of instability while trying to solve one.

The issue could be solved if the RSR token is used as a last resort for the Rtokens that could be voted and backed by the governance.

RTokens would continue to function as intended by the Reserve team, but the RSR Governance could have the possibility of adding some of them under its umbrella and provide last resort insurance by minting and selling new RSRs.

For example there are RToken1, RToken2, RToken3 … the DAO Governance could vote on backing RToken1 and act as last resort in case the staked RSR is not enough. The other RTokens could continue to function as normal but outside of the DAO.

There are two scenarios:

Either have RSR staking per RToken + A new staking module for all RTokens supported by Governance
Or have only one staking module for all RTokens supported by Governance with a slashing % (like Aave) before new tokens are minted.

The RToken revenues will be allocated to the different players in the system.

Without a last resort in the design of the system, I don’t really see the stability that the team is aiming for. The risks are very high and users shouldn’t pay for them. Any revenues generated by the RTokens could go to the insurers as a payment for their service. The RSR DAO would have to manage the risks to the system (by using part of the revenues to buy insurance for example or put ceilings on how much RTokens could be issued as a function of the marketcap of RSR).

Sinatra · December 27, 2021, 7:31pm

Hello there, Unchained.

It seems like you found a way to these (old) Reserve forums, which we are currently rebuilding towards the DAO discussion forum for the new stablecoin Reserve will be releasing on mainnet launch - that’s why it might seem a little empty here.

Thank you for your critical thinking about the Reserve protocol, we very much welcome people challenging our idea for what the protocol might look like.

That being said, I personally do not agree with your statement about how the protocol’s design leaves a large risk for instability. Your arguments about the risks of the protocol are mostly correct, but I feel like your analysis wasn’t done on a deep enough level to understand the potential of those risks.

If I understand your point correctly, your main point is that - if the amount of staked RSR on a certain RToken runs out - the RToken will no longer be able to protect its peg, and thus the value of the RToken could suddenly be significantly less.

While the statement above is true, it helps trying to calculate how big the risk of the staked RSR running out is, and what kind of setup an RToken would need to have in order to make this risk as low as possible.

Let’s take the new stablecoin Reserve will launch at mainnet as an example. Let’s imagine that it will initially be backed by 3 DeFi-yield bearing tokens (cUSDC, aDAI, crvUSDT). The chance of two of these DeFi protocols being hacked in such a way that entirely empties the pool the stablecoin relies on is, in my opinion, very low. If you agree on that, that would mean that - at each point in time - the stablecoin would need a minimum of 1/3 of the circulating supply of the RToken as RSR insurance.

But even if you wouldn’t agree on that, if you think logically about an early stage situation of that RToken, you’ll come to the conclusion that it will look somewhat like the image below. Since the market cap of the RToken will initially be low, but there is already much more RSR in circulation, you will most probably end up with a situation where more RSR is being staked than the market cap of the RToken. In this scenario, even if all the DeFi protocols would get hacked at once, the RToken would not depeg (as the RSR insurance would not run out).

If we look at the future of RTokens, and if certain RTokens become popular, the situation may as well change. We might end up in a scenario like on the image below, where the RToken market cap is much larger than the amount of staked RSR on that RToken. If we’d still have the same RToken configuration as initially (3 collateral tokens), then we’d find ourselves in a pretty risky situation indeed.

However, there’s a solution to this scenario, and it’s one the Reserve team has been planning for. The more diversified the collateral basket of an RToken becomes, the lower the risk of a large portion of RSR being needed at once.

The long-term vision of Reserve has always been to create a stablecoin that is backed by 50+ assets ranging from precious metals, government debt, commodities, and perhaps even equities. To make sure the risks on these assets themselves are diversified, they would also need to be diversified over issuers & jurisdictions.

If you apply the same kind of analysis as above on this token, and if the diversification I’m speaking of is applied correctly, you’ll find that - at each point in time - the RToken would only need (e.g.) 3/50th or 6% of the RToken’s market cap in RSR insurance (just in case 3 of the collateral assets would all at once come under regulatory pressure or other issues).

Your points are definitely mostly correct, and it is a fact that it is up to the RToken governor(s) to keep the risks of their RToken as low as possible. What these examples show, however, is that the right configuration of an RToken can make these risks negligible.

Feel free to reply with any comments you might have (but keep in mind that this forum is being rebuilt towards the governance forum, so you might start to see some changes happening).

Unchained · December 28, 2021, 12:33am

Thank you for the analysis.
The issue is that the Reserve platform will support multiple RTokens and RSR will be split among them. Since the platform is open to anyone to issue any stablecoin (which I am in favour of), RSR will be staked on a bunch of RTokens that will be offering some kind of incentives.
In my opinion, this will hinder the possibility of having a flagship RToken that would be the GoTo for everyone using the Reserve and Defi App(s).

I don’t agree with you regarding the low risk of Defi protocols : if we take Aave, the protocol that Reserve will be using for its next USD Stablecoin, it had the same problem as Cream which resulted in 130M USD hack.

A quick look at https://rekt.news/ and you’ll see that the risk is not really as low as you expect especially if you are depositing other people’s money not yours.

Let’s not forget that these RTokens will be using centralised stablecoins that could be blacklisted very easily by the issuer especially if the app is being used in countries that go againt the interests of the US (for US centralised stablecoins).

I agree that adding more collateral types would reduce the risk significantly but we are not there yet. It would take years to tokenise real world assets and choosing the right mix to put as collateral.

Having some RToken(s) backed fully by RSR with RSR DAO managing the risk of the protocol as a whole (via insurance funds, buying insurance on other protocols, RToken marketCap ceilings …) would give more confidence on these stablecoins and most importantly protect other people’s money.

The other RTokens launched by the community members could be backed by any ERC20 asset freeing up RSR to be stacked on the flagship stablecoins issued and insured by the Reserve DAO.

If I am taking the time to write about this, it’s because I adhere to the Reserve team’s vision of having stable currency is a human right and I don’t want the risks to be underestimated.

I trust that the Reserve team have already thought deeply about these risks, however I don’t really agree with the assessment of their probability & severity.

Good luck with the mainnet and thank you for what you are building