Navigating the waters between now (SCD) and MCD with the DSR

Today we have Single Collateral DAI… and we are using the Stability Fee for monetary policy. In the coming months we will start the migration to launch Multi-collateral DAI… and thereafter to launch the DSR.

The objective: Arrive at the end with MCD and DSR with DAI not changing the price off the peg more than it would have if we stayed on SCD.

The challenge: introducing new market forces & changing fundamentally from a monetary policy to a risk & monetary policy

Suggestions: Forum rough consensus as to the big picture steps to arrive…

Recommended Step #1: Launch MCD and use ether as the sole collateral with the identical risk parameters as in SCD including a materially similar debt ceiling… further use the same stability fee. {Logic says that if there are no market force changes, the price, nor the supply nor the motivations of the participants should change}

Recommended Step #2: Add an additional collateral package of ETH but with a stability fee that is approved by the interim-risk team. {note this is purely a risk question… nothing to do with monetary policy}… with one material change of a debt ceiling that is extremely low. Around $100 {by setting the amount very low, we will not cause the minting of any material amount of DAI thus not disrupting the price equilibrium}

Recommended Step #3: Launch the DSR with an interest rate that is extremely low {by setting the amount very low we will not cause the usage to be high thus not disrupting the price equilibrium}

Recommended Step #4: Increase the debt ceiling on the new collateral package by $10,000 {something but enough to cause the price of DAI to be impacted

Recommended Step #5: Increase the DSR by 10bps {something but not enough to move the price of DAI}

Recommended Step #6: Iterate of steps 4 and 5 over and over until we get the debt ceiling to not be the limiting factor on the amount of DAI issued. {the core objective here is to control the increase of issuance of DAI withOUT moving the DAI price}

Recommended Step #7: Start to reduce the debt ceiling on the original collateral package until down to zero.

the above are purely recommendations. as a community we need to discuss / debate the merits of the above or come up with something different all together.

most importantly, we need to form rough consensus on the concept of bifurcating the existing monetary policy only via SF to one where there is distinct and segregated risk policy (via the SF(MCDx)… and monetary policy… one where the monetary policy is controlled via the DSR…

1 Like

After Step #7: we can start to introduce other collateral packages of ETH or other underlying collateral (on-chain or off).

I approve a launch that’s basically a no-op. The specific steps are a good first draft but we’ll probably get more information as we’re coming around step #4 and will reevaluate then (just trying to signal that we should have low attachment to multi-step plans).

I’d also like to reduce the risk of having the training wheels for longer than necessary, so if we manage to:

form rough consensus on (…) monetary policy (…) where there is distinct and segregated risk (…) controlled via the DSR

before MCD launches, I’d love launching with all collaterals (instead of just ETH) at super low debt ceiling, then increasing the ceiling as we increase the new-eth-collateral-type ceiling.

fair point. primary objective must always be to keep the peg stable. if we launch with multiple collaterals that is fine as long as we don’t cause a surge of DAI to be minted. We need the DSR to control the monetary policy.

once we find the point where participants will absorb excess DAI via the DSR (by increasing the risk free rate), then each time we add new collateral… and increase that specific collateral package’s debt ceiling, we will have a known tool to absorb the excess… but we have to find that point before we start bringing on excess DAI… otherwise we risk the price being eroded…

My understanding is that during the transition period, both MCD and SCD will exist simultaneously, if there is a period when both are active, we may have to maintain the peg of both SCD Dai and MCD Dai at the same time.

As a SCD CDP owner you can move your CDP to the MCD CDP through the redeemer portal that you’ll find at makerdao.com at launch. Or close your CDP by paying back your debt and redeeming your ether back.

If you hold your Dai in a wallet where you control your private keys, then head to a page on makerdao.com which will be announced at launch. Follow the instructions to upgrade to MCD and optionally activate Savings Dai, which allows you to earn savings.

Both taken from https://github.com/makerdao/developerguides/blob/master/mcd/upgrading-to-multi-collateral-dai/upgrading-to-multi-collateral-dai.md, last updated June 8th.

Some thoughts:

  1. There will be the ability to like for like migrate CDPs from one system to another. This means that there will need to be ‘room’ for migration within the new debt ceiling for Eth, we would want to ensure this because we want migration to be as easy as possible, forcing CDP holders to wait, or to migrate a bit by bit is a really crap user experience.
  2. How can this even work regarding migrating old Dai to new Dai? Dai has to be backed by a CDP. The only migration possible can be by selling old and buying new, in which case holding the peg for both systems will be both essential and nightmarishly difficult.

Would be good to know how this is going to work behind the scenes.

if we have to maintain both, all the more reason to have a collateral package in MCD that superimposes SCD with identical parameters…

1 Like

I agree, broadly speaking I’m on board with your proposed steps, my only worry is the amount of time/governance the process will take. The iteration especially could take a while.

do we know or have an approx time between the MCD launch and the DSR launch?

If I understand correctly they’ll be launched at the same time. Unless of course the DSR is just kept at zero at launch, which seems highly unlikely to me.

Could you share why you feel this is unlikely?

From the blog post:

“At launch, Multi-Collateral Dai will include a feature called the Dai Savings Rate (DSR).”

“At launch”

they were pretty clear about it

it is a smart contract… it will be a governance question on when we use it… and how much of a delay between launch and DSR increase the community will approve…

Seems reasonable to include a feature but not initially have it enabled. It’s still ‘included in the launch’. As mrabino says, it is up to governance to adjust.

If it isn’t launched at 0%, what do you launch it at? The only way to decide is to ask governance, and without wanting to generalise or assume too much, so far it appears that there is a view that MCD should come online in a piecemeal fashion rather than all at once.