[Discussion] Unwinding Stablecoin Exposure

Hey all,

So we managed to lower DAI price by implementing 101% LR on stablecoins and increasing debt ceilings to meet DAI demand. We knew that by implementing this, MakerDAO’s stablecoin exposure is going to shoot up and currently represents 45% of debt exposure. I think it doesn’t hurt if we already start discussing how we can reduce this exposure in the near to long term.

Assuming 4% SF on stablecoin vaults, we should have 3 months time before stability fees on some stablecoin vaults stop effectively accruing fees for the system. We can however lower SF and prolong the time this happens, but essentially it doesn’t change anything for the system as a whole because Maker can collect maximum 1% from 101% LR vaults. Also note that “losses” from potentially undercollateralized vaults due to SF accrual should be offset by surplus that fees on same vaults generated.

The main question is how we can generate enough DAI supply from other assets so these stablecoin vaults unwind by themselves or through manual liquidations when DAI price is low enough.

Community voted in favour of speeding the process of RWA onboarding, but it is less probable that the system generates few hundred million DAI from those assets in the near term. There are also initiatives to speed up onboarding of crypto collateral to generate more supply and put more focus on assets that are able to produce a lot of DAI.

We know that currently the assets with most DAI minting potential are farm related ones. Somehow the most sustainable ones seem to be LPs from Uniswap or Balancer, yVaults and secondary lending tokenized deposits such as c-tokens. In addition to having high potential to generate DAI supply, they also carry high yield which means Maker could set SF proportionally to those yields and receive good compensation for increased risks. Adding these assets would imply that Maker is indirectly swapping one stablecoin exposure for another (instead of higher USDC we may have higher cUSDC exposure), but we can be sure that Maker will definitely be able to collect much more than 1% as in current case.

I am curious about community desires to onboard assets such as:

  • Yearn
    • yyCRV
    • yDAI
    • yUSDC
  • Compound
    • cUSDC
    • cDAI
  • UNI LPs
    • ETH/DAI
    • ETH/USDC
    • ETH/USDT
    • ETH/WBTC
  • BAL LPs
    • ETH/BAL
    • ETH/DAI
    • ETH/USDC
  • Aave (when liquidity mining starts)
    • aUSDC
    • aDAI

SF management with these kinds of assets will need to be a bit more flexible and frequent, assuming the goal is to maximize them based on farm yields generated, which we know can be very volatile. Also keep in mind that adding something like yyCRV in the past was risky due to low DAI liquidity in Curve pool and related issues to withdrawing DAI and unwinding those vaults (case with yETH). But since we were able to fix the peg and improve liquidity, this is less of an issue now.

Note that by swapping USDC exposure to cUSDC or yUSDC we aren’t exactly unwinding stablecoin exposure, so the title for this discussion might be inappropriate. But I see this only as the first step before onboarding RWA and other assets, where we switch stablecoin exposure to farm related stablecoin exposure and achieve better risk compensation. It enables MakerDAO to indirectly participate in yield farming.

Also keep in mind the other regular measures to stimulate more DAI supply from other sources apart from RWA onboarding is regularly adding vanilla tokens such as BAL, YFI, LEND, PAXG (planned in October cycle) and potentially introducing lower LR ETH Vault types.


100% agree on the proposal, this will buy us time for rwa to be implemented and we could accrue additional resources


LP tokens are a great long term solution to the peg problem imo. They can provide much of the stability of stablecoins, plus a yield. They are also versatile in that they can be held directly by the protocol without too much fear of volatility or custodial risk (depending on the token pair), as well as being open to the public. We would also be improving defi liquidity in general, which would help us indirectly.

I don’t have issue with any of the tokens talked about in the post, and think most of them represent excellent opportunities to improve dai stability while also being less complex to implement than RWA.


How come noone thought of that 5 months ago? It comes with a price. 1 USDC might be worth 1 USD in a bank, but does MakerDAO want to open a bank account? Selling USDC for something else than 1 USD in a bank might be worth a different price when MakerDAO is a large seller.

So, that’s how decentralization dies … with thunderous applause (59% including wBTC). I wonder when people will realize that DAI is not the same DAI as 9 months ago: it’s just a wrapper for a weird combination of centralized stablecoins and (soon) yield farming tokens with an unpredictable value mostly in the range of $1.01 to $1.06.

I’m not against those assets but unwinding them when shit happens will incure big losses for the MakerDAO system.

The mentioned protocols are natural candidates for issuing stablecoins like DAI - they have almost everything on their side (capitalization, income, decentralized governance, more developers…) to outcompete MakerDAO so we have to be careful not to give them too much power.

EDIT: I forgot to offer a solution: do anything that is needed to make borrowing from ETH attractive. Limit borrowing from stablecoins and centralized assets and do not subsidize that risk. Raise more capital…


I fully support your suggestion. I hope the team will first consider products that can produce large-scale DAI. Please don’t put time on the research of small products, our time is very tight.


For the uniswap tokens, you need to stake them to earn the Uni rewards, So I don’t think there will be demand to deposit them into a CDP as you will not be able to receive rewards on the deposited tokens. I’m not sure if this will be an issue for the other tokens but should research as people will likely only deposit tokens into a CDP that they cant earn farming rewards on.

1 Like

Tokensets has a WETH/DAI UNI farming strategy. Tokenized form of the staked WETH/DAI UNI pair.



Hey folks, I’d like to understand the factors for wanting to reduce stablecoin exposure- there are some we might be able to help with and some we can’t. Potential factors:

  • Earning a higher yield (can be solved with tokens such as yTUSD, yUSDC, etc.)
  • Diversifying collateral beyond USD stablecoins (can be solved with tokens such as TGBP, TAUD, TCAD)
  • Avoiding or minimizing centralized counterparty risk (can help here e.g. by working with Chainlink oracles to post collateralization info on-chain, but certainly can’t solve this entirely)
1 Like

Could you point me to that vote?

Hey Hasu,

Signal Request here and on-chain Poll here.


I think the main downsides to high stablecoin exposure are:

  • counterparty/custodial risk
  • no organic demand for usd-denominated stablecoins except for arbitrage.

Having non-USD stablecoins as collateral can help reduce the 2nd issue by turning MKR into a kind of forex exchange. But, it doesn’t eliminate the first issue.

There are very few assets though that have low volatility and no custodial risk. LP tokens are probably the closest thing we have rn, and any other non-custodial assets like this will have to have their value derived from decentralized currencies.

We are not trying to remove custodial risk altogether, but it is worrisome when a single counterparty like circle makes up a large percentage of the total dai debt. The best way to deal with the custodial asset types we do want to hold is to diversify across multiple counterparties and also multiple legal jurisdictions.


@befitsandpiper- makes sense. I’d be interested to know if there are things that would make Maker more comfortable with the counterparty risk from an issuer like TrustToken. For example, if we had a third-party like BitGo hold the keys to the multisig that owns the smart contracts and refills the mint pools.

Well, we do have some Bitgo exposure with WBTC already. Might not be the best to lump you two together.

Though not technical advice, Bitgo could be more secure than your current set up.

I’m not sure how much this is actually the issue, at least for me personally. A larger concern is that you guys (and all centralized stablecoin issuers) are required to comply with directives from at least one government. Unfortunately that’s not really something that can be addressed.


I agree, we should reduce our exposure to centralized stablecoins. There’s nothing wrong with stablecoins in general (although it should be noted that too much exposure to any one stablecoin, centralized or no, may be unhealthy).

While we wait for RWA to become a thing, and since new crypto collateral seems to be a little slow to onboard at the moment, we should look at what we can start doing right now to start influencing our exposure to these assets.

I’ve been saying that we need to try to make other assets more attractive to borrow against. Lower LR, lower RP, higher DC. I can’t guarantee that this would work; I can say that if I had a pile of wealth, and I was considering putting some of it in a vault and borrowing against it (probably to do some yield farming), I wouldn’t necessarily care exactly which asset I was borrowing against, so long as it gave me the best of those three things.

What I’m saying is that I think yield farmers don’t necessarily care what they’re putting into their vault; I think they care about how much Dai they can get out of it, and how much it costs them; exactly how much leverage they can get. Should we form a strategy around yield farming? I don’t know. It seems to be a booming sector right now, so I think it’s not a bad idea to move in that direction.

1 Like

While this can’t be solved in general, listing stablecoins from different fiat currencies may help hedge this risk. If US regulators, for example, passed strict regulation around USD stablecoins such as TUSD, this might not affect non-USD stablecoins such as TAUD, TGBP, TCAD, and THKD.


The problem with that is–once the U.S. bans stablecoins made from a basket of USD – the other Authorities will follow, and in due time they will also restrict non-USD stablecoins, like TGBP.

I highly doubt the U.S. wants to abolish innovative Tech that pegs to the USD and has potential to score world exposure. Hence, you have some big time players pushing USDC to overtake USDT. The Washington DC guys would prefer USDC, versus USDT, or other CBDCs.

However, this was an interesting interview:

Laura Shin: “Valerie Stefanik of the SEC said that Certain type of stablecoins have raised issues under security laws. Which type of stablecoins do you believe veer more in that direction?”

SEC Commissioner, Hester Pierce: “I think Sometime when you see a stablecoins that’s build on a basket of underlying assets, it could implicate certain rules”

“We don’t think every stablecoin implicate security laws, but you should talk to us ”

What to do…

1 Like

Definitely a hard nut to crack. At the moment I think the only practical solution is to find a way to onboard as many different kinds of assets as soon as possible to minimize concentration risk which is rising day by day. This is easier said than done of course. I raised my hand to help with RWA, but most people seem to want Maker DAO to work wonders as they sit on their hands. If we want things to change, we have to be part of the solution in whatever way we can contribute. Complaining alone does not move the needle.

1 Like

I think it is healthy that stablecoins account for about 50% of the system. In fact, it has been proven that it can make the system more stable and balanced, so we are on the right path. As long as the stable currency does not exceed 60%, we do not need to change it. The only thing to do now is to concentrate human resources and accelerate the introduction of assets that can generate a lot of DAI. Those assets that cannot generate large amounts of DAI in the short to medium term should be postponed.

1 Like

100% this…