[Signal Request] Adjust PSM-USDC DC


Since we set DC on USDC-A to 0 and increased the DC for PSM-USDC to 1 billion DAI in this exec we have seen a pretty stable increase in the usage of PSM-USDC.


At the current rate we will hit the DC in 26 days - once we are there we will risk drifting upwards on the DAI-peg.

Nobody is going to like that, and probably also nobody likes to increase our USDC-exposure, but I guess we don’t have a lot of other options.

Changing the DC for PSM-USDC


  • more DAI in circulation
  • the best weapon we have right now to defend the peg


  • higher USDC-exposure

Please vote for all options you would support in an onchain poll

  • 1000 MM (no change)
  • 1250 MM (+250 MM)
  • 1500 MM (+500 MM)
  • 1750 MM (+750 MM)
  • 2000 MM (+1000 MM)
  • Abstain

0 voters

Next Steps

The Poll will run until 2021-03-25T14:00:00Z; its outcome will either result in a on-chain-poll assuming the outcome of the poll deems it necessary or its intermediate results are going to be taken to another initiative - another signal, onchain-poll or urgency executive.

  1. Given that ETH-C will soon be available and this would effectively be decreasing the SF on ETH vaults with high enough collateralization, perhaps that will be sufficient to defend the peg? 26 days gives us some time to evaluate how successful ETH-C is at increasing DAI supply.

  2. Can we decrease the tout fee on the PSM to encourage removal of USDC?


Dai price has not been shown to be highly responsive to stability fee changes in the past. We can’t rely on “maybe a SF decrease will stop an unmanaged peg”. It has been shown that if we cannot defend the peg, we lose integrations, quickly.

Anyway, at this point it doesn’t matter what we set PSM to, any blacklisting of USDC is already existential for Maker.


Dai price has not been shown to be highly responsive to stability fee changes in the past.

This was true last summer but remember that Maker had just suffered Black Thursday losses. It seems reasonable that people were cautious with opening vaults, even if the SF was zero because the protocol did not work the way many expected. Moreover, it took a while before people were willing to leverage ETH again. Pre-2020, as far as I know, changing the SF has had a reasonable effect on DAI supply.

Anyway, at this point it doesn’t matter what we set PSM to, any blacklisting of USDC is already existential for Maker.

All the more reason to do everything possible to wean DAI off USDC dependency than go further in. Increasing the DC should be an absolute last resort measure because this is dangerous in the long term.



Would you consider adding a signal for changing the fees on the PSM ?
Personally I would vote for lowering this in order to better compete with Curve.

would certainly do that, but i doubt this is going to change anything. as long as we are >= peg nobody is going to unwind DAI for USDC - no matter how low the fees are

1 Like

True. There is no rush - I am thinking more long term when possibly more stablecoin types are added to the PSM. Also, the good times in crypto will at some point end, possibly resulting in peg swings between stablecoins. I want Maker to be in a position to profit from this.

In that case it is the ‘tin’ that needs to be changed but that means more usdc and less fees.

Also we turned the usdc to zero relatively recently and there is plenty of them waiting to be at 0.99 to be sold back. I doubt we will sell any usdc for a while.

So one caveat to this is that holders of ETH-A vaults (and any other non-stablecoin vault) also really want the price of DAI to be as stable as possible. The reason for this is that if they have to de-lever in a crash scenario, they don’t want to have to pay $1.05 for DAI (because everyone wants DAI in a crash scenario this pushes the price up.)

So in this way having the PSM facility actually makes other vaults safer to hold as well. I do agree with you that in an ideal world we would keep the total amount of USDC below ‘existential risk’ level. I think most people agree with that part.

The question is more how do we get there from here. The key solution is to onboard more non-stablecoin collateral, but how to go about this?

1) Reducing ETH (and other) vault fees is an option, but what happens if (like last time) we do that and it has no significant effect on the total amount of stables? We’ve both cut revenue (and system safety, given less buffer less quickly) and in all likelihood not fixed the problem (brought USDC below ‘existential risk’ level.)

2a) The other option (that is currently being pursued) is to onboard new assets for use as collateral. The difficulty here is that ETH and BTC are by far the largest crypto-assets, so there is limited opportunity in the purely crypto space. Liquidity pool tokens are an opportunity here, but it remains to be seen how popular these are.

2b) The other part of this option is to pursue off-chain assets as collateral. This has its own complexities and problems, but can potentially scale much more quickly once it’s up and running.

One of the key arguments against point #1 is that reduced revenue will also reduces our ability to pursue #2a and #2b. Currently this is more theoretical than observed, but in principle more resources gives a greater capacity for action in these directions.

In theory if we pursue #2a and #2b the long-term situation should be much improved. The danger is higher in the short and medium terms (before we can get real world assets and additional valuable cryptoassets onboarded.)


Yes, I do see the difficulties but note that the SF is effectively being reduced due to the launch of ETH-C. Maybe this PSM vote can wait to see what effect that change has. We still have 26 days before there is any danger to the peg and we have the opportunity to see what effect lowering the SF has in the next days.

The other thing I really think the community needs to start talking about is how we view the peg. Our peg is a soft peg and it would be really beneficial to define service standards e.g. DAI will exceed 1.01 only x% of the time. x can be arbitrarily small depending on what the community decides but must be a nonzero value. This will help understand what resources we need to absorb demand and supply shocks - in particular, how big the PSM needs to be relative to total DAI issued.

1 Like

well, even if we don’t hit the ceiling AND eth-c adoption helps a lot - if the demand for DAI from PSM is going down we don’t lose anything by increasing the DC. It is just a way of reducing the risk to depeg in case the ceiling is reached.

I too hope that DAI from ETH-C is helping (a lot), but hope is not a winning strategy.


Fair enough. I guess for USDC a large gap between usage and debt ceiling isn’t really an issue.


I know a lot of people are waiting for increased DCs on the LP assets (particularly uniswap V2 LP tokens).

Frankly what I am seeing in the space is such that demand is high for capital leveraging. Which as this increases could lead to a very nasty de-leveraging event which will bring Maker and the entire DeFI space right back to the same place. Where does one find DAI liquidity when collateral values are compressing hard and fast?

From my seat the ONLY collateral that probably would retain value in such an event is RWA (Real Estate, PMs, etc.) your 2b above. Currently the ability for this collateral to mint DAI is exceedingly limited and during a crypto crash event neither ETH, wBTC or even the LP tokens etc. are going to be able to mint DAI sticking us right back into the stablecoin PSM solution.

My general take on this is that giving the markets free reign without heaping on a risk fee is basically like heaping more explosives on a bomb that you hope will never go off. I also think everyone knows that Maker is probably very close to existential event if Circle or authorities decide to clamp down on stablecoin use leaving Maker not just with a potential collateral deficit, but also wholesale inability to mange the PEG. Probably now the greatest weakness of Maker short of someone seeking a way to ES this system unexpectedly, or manipulating network access at the miner hubs during a high demand network event to exploit the liquidation system.


Yes, if this happens I’m okay with MKR going back down to $200 and minting another couple of hundred thousand MKR via Flop Auctions, so be it. Nothing surprises me in life.

thanks everybody for participating in this Signal Request - the corresponding on-chain poll is live, please vote until 2021-04-01T16:00:00Z


Final Update: the exec containing the raise got scheduled - we are going to have a new Debt Ceiling of 2 B for PSM-USDC on 2021-04-11T00:00:00Z


This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.