Proposal to Increase the D3M DC from 50M to 100M

Proposal to Increase the D3M DC from 50M to 100M

The Direct Deposit Module has hit its current debt ceiling of 50M. Protocol Engineering and Risk are proposing to increase the D3M debt ceiling line to 100M. We propose keeping the gap at 25M and the APR Target Borrow Rate bar at 3.9% as they are today.

D3M Background

The D3M acts as a stabilizer of interest rates on Aave. It has the capability to mint and supply DAI to reduce interest rates, and can likewise remove liquidity if demand is low in order to increase interest rates. This is done with the intention of providing DAI liquidity at a stable and predictable rate. The predictable rate helps to retain DAI’s competitive advantage as a favourable stablecoin for secondary market borrowing.

Recent changes as seen on Makerburn:

Current Market

The effective borrow rate on Aave is currently approx 2.07%. This is the equivalent to a user opening a 200% CR ETH-DAI vault on Aave, where a user would receive 0.66% rewards on ETH and 1.3% rewards on DAI. Subtracting this from 3.9% means the user has an effective borrow rate of 2%. As a result we do not suggest reducing the target rate from 3.9%.

The proposal to increase liquidity at Aave will not change utilization or decrease rates, but will give us more supply capacity if demand increases at Aave, all while keeping us well below 20% of the Aave v2 DAI supply market.

Analysis & Discussion

Simulation of the 100M debt ceiling parameters and the existing target borrow rate of 3.9% can be explored on the Risk Team’s D3M Dashboard.

There has also been a good discussion on Risk Team’s forum post D3M - DC Increase & Target Borrow Rate Decrease Proposal.

Following community discussion, the intention is for GovAlpha to post an onchain poll for increasing the D3M debt ceiling to 100M next week, followed by an executive to make protocol changes.

21 Likes

One step after one step - thanks @Derek for pushing :slight_smile: can’t wait for the moment we have a DC higher than necessary (at least most of the time)

3 Likes

Is @Primoz with risk supportive here?

Yes, I believe 100m DC is logical next step now that the audit was officially received. 100m is still safe from liquidity risk perspective and credit risk of DAI borrow market is low (see our D3M risk assessment or other posts on D3M). We are currently working on separate dashboard that will enable everyone to monitor these types of risks on continuous basis.

That said, I am planning to run a poll on potential further increase of DC and/or lower target rate. As @Derek pointed out, it is hard to expect increase in D3M utilization even with a higher DC - we’d need to either decrease target borrow rate or if DAI borrow demand at Aave v2 increases organically. In any case, this becomes a decision made by community - how much more competitive does community want Aave to be? @hexonaut made a similar signal already in July where community voted in favour of 1-2% spread to ETH-A (excl. Aave rewards). If this still holds, we are able to decrease D3M target rate to 3.5%, which would according to this simulation increase D3M utilization to about 250m. But note that in such case borrowing DAI from ETH (assuming 200% CR) would equal effective rate of only about 1.5% after Aave rewards.

8 Likes

I was looking around a bit and it looks like DAI deposit and borrow rates are normalizing somewhat (at least on L1). There is some variation to be sure between centralized, decentralized, various LP farming situations.

Keeping an eye on the PSM gives me a better feeling for what markets are thinking/doing.

After the market ETH other stuff 10-20%ish dump and GUNI DC increasing I see after 300M USDC leaving over the past 10days or so another ~150M moving back in, in the past 2… GUNI vault up over 100M in about the same time. So 100M @ 3.5% might be a reasonable target.

Where the larger rate and price disparity exists is pretty much on sidechains at this point where liquidity is somewhat thin and DAI is there just because it is looking for some returns LP staking etc.

I guess one thing we will have to be careful of is the total costs (oracle and manpower) on operational management of these facilities to make sure they stay profitable over longer time periods and to keep a real eye on DC usage/target rates fees related to operational costs.

Thanks for chiming in.

1 Like

lowering the target rate to 3.5% doesn’t only encourage more borrowing but also will force some DAI holders to withdraw from Aave looking for better yield somewhere else which means more space for D3M… I think that with 3.5% we can increase D3M DC to even 300m

2 Likes

Yes please and yes please for going to 3.5% target rate and eventually functionally unlimited DC.

I would say eventually we should down to something like 2.5%, so that Aave/compounds borrowing rates become equal to our own borrowing rates +1%. (how much leverage do they allow for ETH again? I’m assuming we can compare them to ETH-A which might be wrong, if ETH-B is the better comparison then it’s another calculation entirely because we don’t want to be subsidizing something that captures ETH-B users)

Needs to be done step by step and with careful security monitoring and maybe even audits or something. But this is such a powerful tool for growth and can easily pay for all the increased security costs it will require.

5 Likes

Aave leverage is even higher than ETH-B
I think their collateral ratio is like 125%
but it has been like this for more than 1 year