PART-3 Yield Farming: DMG Burn and New Pairs

Hey guys! The feedback we collected from my prior post, “PART-2,” was great and we have been thinking more about how we can fix yield farming to be more economically sustainable for DMG holders long term.

The first thing we looked more closely at is a withdrawal fee - whenever a withdrawal is processed, a fee of 0.5% is taken from the withdrawal to purchase DMG and burn it. For non-native pairs, a fee of 1.0% will be taken (more on what we consider to be non-native pairs are in the next paragraph). This design mechanism was influenced by yearn and also by a prominent member of the community. The need for a withdrawal fee is simple: the yield farming protocol should discourage people from actively withdrawing and depositing funds on a short term basis. So far, the implementation of a burn would have led to about ~2,475 DMG being removed from the supply, and ~3,300 DMG being added. Keep in mind, the current price of ~$0.33 has a 27% APR for stablecoins and a 13.5% APR for ETH-mETH. In most cases, it would take far less than a month to recover the withdrawal fee, not including the interest being accrued in the mTokens. Note, the burn numbers could change depending on a number of factors, like:

  • The price of DMG at the time of the withdrawal - a lower price leads to more DMG being burned
  • The composition of the deposit - more non-native pairs leads to a 1% withdrawal fee.

At the time of writing, there is about $510,000 locked up (not including DMG) in the yield farming contracts. As a team and a community, we think we can grow this number significantly by slightly re-configuring the protocol in two key areas - the DMG reward and the tokens that can be farmed. In the prior paragraph, I mentioned the idea of a non-native pair - an mToken pairing for yield farming whose base asset is not an underlying token. For example, USDC-mUSDC is considered to be native whereas LINK-mUSDC is not. Meaning, the LINK-mUSDC pairing would have a 1.0% withdrawal fee instead of the lower 0.5%.

We think the addition of other assets that users have an appetite to maintain exposure to could be productive for the ecosystem, or adding other farmable, yet safe, assets. However, this is where we would like the feedback from the community. Would something like LINK pairings be worthwhile to add (LINK-mUSDC, LINK-mETH, LINK-mDAI, etc.)? Are there any other pairings? What about something (potentially wacky) like yUSD-mUSDC so users can double-down on farming more than one protocol at a time?


Hi Corey,

Are you against adding a BUSD pair?

Binance’s USD stablecoin has been given the green light by New York’s financial watchdog, meaning licensed institutions can use the asset without any further regulatory approval.

We must have the BUSD pair for all of the obvious reasons and huge potential above. Do you disagree?

1 Like

I think the burn and buy which I believe is a great idea. Pairings I would like to see is: Link, YFI, UNI, LEND, and XTZ/DOT if possible.

I think there should be a clear message about the mechasnism and how it works, including a minimum -> maximum time range until break even point is reached & it’s profitable to exit, many beginners will most likely not understand the burning mechanism.

Charts & Metrics are always important, people need to be able to visualize what’s on the line, and how quickly they’ll ‘turn into profit’ from locking the tokens, I’m also curious on how 1inch / uniswap will treat trades from say USDC -> Link or viceversa ( if it farming other tokens will work the same as current native pairs ), they’ll most likely overlap sometimes with these liquidity pools and trades will happen, this might actually prove profitable for some, or be heavily affected by impermannent loss

Great idea team, thanks for listening to the community. Will my current crop be affected? Or will the withdrawal fee be for new farmers?


There is potential for impermanent loss for non-native pairs, because they’re quoted against assets that are uncorrelated from the DMM smart contract’s exchange rate. This is definitely something that could be hard to quantify since it’s subject to the asset’s volatility (IE LINK, yUSD, etc.).

We are adding a stats panel to the Farm tab on the site to better show these metrics and we’ll showcase the withdrawal fees (plus other important metrics) in there as well.


Depending on when this gets implemented, it would affect withdrawals after that specific date. IE if we roll this out on Thursday (not an actual release date/promise), and you withdraw on Friday, then it’d be subject to the withdrawal fee.

1 Like

I like the withdrawal fee. As you stated, it prevents people from hot swapping.

Could we get more detailed information on the pair rewards (and would this fluctuate)?

Shouldn’t the m token pairs all be the same (since we want to encourage creating a market for all)? I think m token pairings should all be the same.

I think implementing LINK pairs would be a great start for non-native pairings! I would definitely like to participate in LINK-mUSDC yield farming.


I like this idea a lot! I think a lot of the holders of DMG hold link and would be happy to support this farming.

Although I don’t think 13% APY is substantial enough to get wide usage (given the slippage risks). I think the 30% is much more attractive so perhaps that could be raised so stable coin is the same as non native. Also, perhaps rates could change based on price of DMG? ( at $1, the rate would be much more attractive).

1 Like

I agree regarding the APR needing to be higher. Around 27%, ish if it were in-line with other stablecoins, would work and should help with impermanent loss.


Hey, we’re definitely not against BUSD and will be looking to add support for this plus other exchange stablecoins as we are able to review and onboard them.

1 Like

Excellent to hear this Corey! Staking BUSD for DMG will be very attractive for institutions as Binance’s USD stablecoin has been given the green light by New York’s financial watchdog, meaning licensed institutions can use the asset without any further regulatory approval. This will make it very easy for them.

Any thoughts about CRV pairings?

We like the idea of yUSD, but would need to make sure users are aware of its risks.

Adding to this, I know this would be for the yearn wrapper instead of a direct CRV token. However, it could have strong value accrual since it’s a farmable token that already has attractive yield attached to it.

Honestly just a native mLink/Link pairing would draw a lot of money into the system. Current APY for link on Aave is 0.17%. The last time I looked into the yield rates for Chainlink they were very low across the rest of the large Defi protocols. Point being is we could be attracting a large amount of capital at a low cost to DMM.

Edit: The rate of 0.17% does not include the governance token rewards from protocols such as Compound. With the 6.25% base rate DMM provides, the DMG necessary for this pairing to be competitive would be pretty low.


Has long as the risks are fully disclosed and easy to understand said risk it should be a great addition.

As much as I would love to see an mLINK pairing, the high volatility of LINK may pose a problem to the DMM ecosystem regarding a potential overexposure.
Maybe we could try adding mLINK with a relatively low ceiling to see how it goes and then expand the ceiling over time?

Edit: Just to clarify: I am aware that the DMMF hedges against ETH volatility via options contracts, but are those contracts even available for “smaller” tokens like LINK?


Even if there were options contracts for Link the implied volatility of Link is much higher than that of ETH and therefore the premium on the option contract would be through the roof. As much as we all would like to see mLink it is just not going to happen