Liquidity Mining Response Summary

Very good news. Most FYI liquidity mining is now done, watch the Dai price unwind:


Added the recent debt ceiling increase. It looks like we’re going to be dealing with this for a while, and while it’s important to stay reactive to the changing market conditions, I don’t want us to fall into the trap of seeing this as the new normal, so I’ll keep updating this for the foreseeable future.

Blowing out past 1.035 again. It is time to put aside focus on onboarding marginal new collateral types and work on something to fix the peg directly. Maker and Dai are losing their PMF the longer this peg failure continues. No one wants to borrow Dai for spending when their real debt value increases by 2.5% in a couple days.

1 Like

There will be interest in borrowing DAI. There is quite a bullish sentiment on ETH, and there are now many high yield farming incentives for stablecoins. Seems great to borrow DAI from ETH, sell it when it’s above the peg and also yield farm with it.

IMHO our best shot to solve this is to trust a little more in the only stablecoin Vault we have available now: USDC-A. Even with no big capital, anyone could leverage 10x on it and generate a lot of DAI. But there are 2 things holding anyone from doing this:

  1. 80M cap (not a problem right now, but it’ll be if we want to have really a solution)
  2. 4% risk premium (the most important issue)

Imagine somebody leverages 100k right now to 1M, that could have some impact right now in the price, but with 1M of DAI paying at 4% yearly, and no guarantee that the PEG can be restored in the short term, it could be a very bad business decision to do this (accumulating 100DAI daily of debt can be scary for many)

Lowering that risk premium to something like 1% could help a lot in this regard.

Lowering that risk premium to 1% and increasing slightly the USDC-A debt ceiling to 100M is the quickest response I see right no (one executive vote away). Another one which could have a good impact is raising the WBTC debt ceiling to 34MM (currently in on-chain poll). All of them add some risks of course, but keeping those for 1 - 2 weeks shouldn’t be that big of an issue.

If we manage to solve the issues surrounding TUSD, specially after the good response received from TrustToken in

plus onboarding others stablecoins as collateral, we could reduce the risk of having only USDC for this 10x leverage process (which is IMO, the best process in existence to keep the 1:1 PEG).

1 Like

@smaugho The USDC-A could be close to filling. Was 40/80 yesterday. 50/80 now. 110% CR just got executed yesterday. Who knows how fast that could fill. Raising the DC seems more important.


Well, USDC-A is 80/80 right now, just 3h later. How fast could we vote to add 40M more to the DC?

And yes @Jiecut, it seems that the demand is big even with 4% given this result today, I guess that we don’t have to worry yet for the risk premium as long as the demand is there.

@cyrus, increasing +60M or even +80M will increase too much the risks? It is not like we’re having a USDC crash anytime soon, only big risk I see is the contract being blacklisted, but I don’t think that will happen without MakerDAO knowing in advance (sounds too unreal IMO)


How much usdc locked in vaults are we willing to stomach? 33% of total supply? 50% of total supply? I would be pretty strongly opposed to exceeding 50% of total supply, as this would turn us into essentially a wrapper for usdc. Having a more diverse set of stablecoins would make >50% stablecoin supply a little easier to stomach.

We’re already at ~23% total supply and these are important questions we need to start asking ourselves. There’s no price risk vulnerability in usdc like with eth for the Debt ceiling, so we should decide what percent of total supply in usdc we’re willing to stomach, and just raise the debt ceiling all the way to that level.

We can bundle in raising other debt ceilings on collateral types so that the next time we have to raise usdc, we will have a little more room still.


Total supply of USDC (following Coin Market Cap is 1.1B, I would vote up to 20% in the current conditions (that would be around 220M).

Onboarding USDT and solving issues with TUSD will help us to get a further increase if needed and reduce risks. I know many of us don’t like USDT, but a 10B total supply is not something to ignore.

A proper liquidity mining response IMHO is raising those DC as suggested by @befitsandpiper. The yield farming craze will not end anytime soon, I see nothing wrong in we using what we have to provide the loans needed in DAI, and profit as well from it.


Sorry, I meant as a percent of total dai supply. But you’re right, the total supply of usdc is also something to be considered, although Circle would likely just raise the supply if demand increased dramatically or usdc started breaking peg on the upside.

1 Like

Some info about UMA’s new incentive program for yUSD:

Tl;dr - UMA offers fixed rate/term synthetic USD loans against ETH collateral, which currently bear negative interest rates due to liquidity mining demand for yUSD.

I am trying to think about this situation a little more holistically.

Obviously there is massive demand in the Ethereum ecosystem for stable coins. There is near $13B in stable coins issued on blockchains, and the vast majority of that is on Ethereum. Even though DAI is a relatively small part of that, it is arguably the most in demand. Money markets have significantly higher interest rates for DAI compared to other stable coin and it is considered safer than fiat backed coins due to its decentralized nature. People seem willing to pay a 2-3% premium for this and no other widespread stable coin fetches a premium (in fact the largest one is consistently traded at a discount). DeFi systems are incentivizing DAI as collateral over other stable coins and are consuming more than 90% of the supply just as collateral, even though there is a premium. Despite the amount of DAI that is being minted recently, very little of this is going into normal circulation, so it is doing little to help stabilize the peg.

Interesting to me is that of course these DeFi systems have full access to ETH as collateral as well. And while they do use this to some small degrees, there is apparently not much borrow demand for it, so it is not incentivized as collateral. There is obviously work required to create a system to convert ETH value -> USD and maintain that system. Then there is ongoing risk due to volatility, etc. All this means that these systems are willing to pay a premium for Maker to supply this conversion for them so that they can focus on the finance aspect of the business. This is great for Maker long term.

In an ideal environment, we would be charging a stability fee for the risks Maker is taking. And we are for some assets, but not for our primary asset ETH because we are trying to scale the supply side faster. But the simple fact is that we are not meeting the demand for DAI, and likely are not even close to doing so since it is probably a number somewhere in the billions. This is a cautious approach since we are more focused on mitigating risk than supplying product, which I think is fair given the nature of the system in its immature state. If we mis-judge the risk, the system can quickly become insolvent and worthless.

I would encourage people to think about this from a different direction though. Hypothetically, lets say we were to completely remove all the debt ceilings tomorrow. What risk parameters would we want in place in order to do this? Debt ceilings are the training wheels in our system - we are using them to ensure that any mistakes we make are limited. It is a smart move for a new system. However, we need to move in a direction where the risk parameters and the demand from the market are the determining factors in how much DAI is in circulation. IMO these parameters are clearly not set correctly for most of our assets and we are instead relying on the debt ceilings to manage risk. However, until we can remove the debt ceilings and allow the market to work, the peg will never fully stabilize.

1 Like

I agree. We can spend a whole year adding new collateral and get exactly the same results as so far. I’m strongly against that failed tactics and I vote against new collateral every time.

On the surface this seems to make sense, but looking at the DeFi services, this does not seem to be the case. It seems like there is lots of demand to borrow DAI even at steep rates. Am I reading this wrong?

1 Like

Farmers borrow DAI to lend DAI to borrow DAI …

The price of DAI is almost irrelevant to them.

You are right that we need a mental framework to set parameters.

@smaugho- Agreed. We’re working with the community to address concerns around TUSD and have already committed to provide $1mm TUSD to borrow DAI and sell into the market to help stabilize the peg.

Assuming it becomes possible to borrow against TGBP, TAUD, TCAD, and THKD in vaults as well, there’d be a basket of un-correlated (or at least not fully correlated) stablecoins to borrow against, which would make the system more robust. Additionally, this would attract traders looking to lever-up on these non-USD currencies by creating vaults with them and then borrowing DAI and selling it, further stabilizing DAI price.


A Chinese YFII is on their way, already locked into their system more than 50M DAI:

It is a Fork from the YFI project. I hope that measures taken today be effective over all the demand we’ll experiment on next days. Coinbase also started their DAI Reward Program

IMO, we should be prepared for the following:

  1. USDC Vault growing steadily and having effect on DAI price reduction: Keep monitoring
  2. USDC Vault growing very slowly and DAI price continue to climb: Change Risk Premium there to 1% - 2%

We should keep monitoring the stats.

1 Like

Added a section on the most recent debt ceiling raises.

1 Like

Compound is potentially planning to update the USDT interest rate model to match DAI:

This could take some pressure off of DAI by rebalancing COMP farming towards USDT. Caveat: USDT isn’t approved as collateral so it can’t be recycled several times like DAI.