[Poll] Should MakerDAO print unbacked DAI to solve the peg issue?

MKR market cap is the market evaluation of the equity part of Maker. It’s not on Maker balance sheet. Maybe there is another Maker asset somewhere that I’m not aware of.

You are right. We can print as many DAI as we want. It will be on the asset side and the liability side of Maker balance sheet, let’s say 100. Then you can buy 100 USDC so there will be 100 USDC on the asset side, 100 DAI on the liability side and a small profit (let’s say 2 DAI remaining on the asset side therefore expanding the equity of Maker). That’s the PSM idea.

It’s quite different from buying MKR and burning them. This way you end up with 100 DAI as liability and nothing on the asset side which mean a loss that reduce equity.

You can buy MKR or ETH with such DAI, but there would be no buffer to sustain any drop.

Issuing DAI is issuing an awesome debt (no interest, no redemption date except ES). We let our clients issue the debt but don’t use it ourselves. As we don’t have equity (or so few) issuing DAI would be like having a vault with a LR of 101%.

Maybe we need to issue equity (issuing MKR for ETH), leverage it by issuing DAI debt and investing it in Compound, Aave and Uniswap (directly positive for earnings per MKR token). Stop trying to be a central bank and be a bank.

People want Maker debt, they are ready to pay negative interest rates (by buying above par).

As we can see by the peg situation, letting third parties managing Maker balance sheet isn’t working very well regarding the peg. But not having an equity base limits our ability to manage it. I’m on the side to kept the system solvent but some don’t, which is the reason of this poll.

Obviously, wanting to manage the balance sheet would create a lot of technical and managerial issues for MakerDAO. Maybe it’s too early for that.

1 Like

Thanks for the explanation.
3 points

  • I think the foundation still hold 30% of all MKR.
  • Also burning MKR if you can mint them, should be the same as buying and keeping them minus the administration and the fact it is already more or less inside the protocol.
  • regarding the PSM, I thought the idea was to recycle the migration contract with an already verified and proved code. The conversion would be a fix 1 to 1. like dai to sai you will have USDC to dai with an initial reserve which will play a bigger game here.

Also burning MKR if you can mint them, should be the same as buying and keeping them minus the administration and the fact it is already more or less inside the protocol.

You are right, it’s the same. If you buy your own shares (tokens?) you can either put them on the asset side or deduct them from the equity part (both are accepted methods in accounting).

The equity portion of the balance sheet is the insurance current MKR holders are providing to DAI holders, i.e. almost nothing. The expected demand for issued MKR in case of a need is the insurance provided by future MKR holders for DAI holders.

For the PSM, the conversion could have some in and out fees, creating a spread. Not sure who would own the USDC deposited. I think it’s a vault, so the customers.

It’s off topic, but this can’t be true. If you are talking about multisig wallet it’s circa 10.3%, at least according to etherscan.

Post poll results : The poll was closed 7d after the start and 5d after the last comment. It shows that the Maker community in majority feel that the promise of 1 DAI = $1 of collateral at any time is essential. Still one fifth are okay to have it lower than $1 and one fifth don’t have a clear opinion at the time.

On the side note
I will write what I’ve mentioned already in another thread

I believe that two promises MakerDAO gives

  1. Each DAI is backed by at least 1 USD of collateral
  2. DAI is exchangable for 1 USD

contradict each other since they do not acount for both additional utility of holding DAI instead of holding USD and opportunity cost of it.

For instance due to Compound decisions currently You can have around 30% APY (assuming more or less stable comp token value) with depositing dai and borrowing Tether and making leverage on that several times over

That quite some utility of DAI that USD do not have. And it should (and does) drive DAI price up above 1$


We will never have both. I’m not sure why people prefer #1 over #2. Merchants are starting to accept USDC instead of DAI. I think DAI will have more and more peaks above 1.05 but it won’t get much higher due to people selling it for USDC or other tokens for immediate profit. At that point DAI will be only useful as a store of value for those who bough it at a reasonable price (similar to BTC).

Remember that DAI is backed by USD, but USD is backed by nothing as it is the only way to be useful as a currency. I think we need to be somewhere in between those 2 extremes.


I want to chime in agreeing with @bit and @Adam_Skrodzki. I don’t see how DAI can both be backed by $1 of collateral and trade at $1, given that there is extra utility from being part of a widely-trusted and widely-adopted network.

From my perspective as a user “DAI is exchangeable for 1 USD” is the core value prop of DAI, and thus an end-in-itself. “DAI is backed by 1 USD of collateral” is just a means to that end. So if they are in conflict, it seems clear that the DAI backing should be sacrificed, rather than the stable peg.

For me, this is not just hypothetical–I run the SourceCred treasury and currently we pay all of our contributors in DAI, but Grain is pegged to USD, so I’ve started needing to do spot currency price conversions every time someone redeems their Grain. It’s a pain. For us to keep using DAI long term, it’s important that Maker finds a way to keep a stable peg.


So your argument is that because DAI is superior to USD in almost every way it will be hard to keep it at $1? I think you should think that its such a good product with high demand that we will be able to suck in almost all assets in the world with super low rates due to high DAI demand. Your argument sounds like a short term problem until A) We scale to billions and B) We make efficiencies in collateral/liquidations so that LR can be lower. Perhaps one day we can get to a point when the trust of the system alone allows the whole system to be under-collateralized but I hope not as that just returns to the current economic problems we face today.

The point is not that it’s bad to be better than the usd in almost every way, but it’s then impossible to be worth only what a usd is worth. The whole problem is that dai is extremely attractive. That’s what it means that it’s more valuable than the usd. If we want to break peg we can, but otherwise we have to find ways to be a LESS robust/attractive currency.

I think it is true that we either have to blaze our own trail and break usd peg though, or break some of our rigid ideals around collateralization.

An interesting compromise here is simply lowering the vault collateralization requirements across the board. Atm we are not just 100% collateralized by usd, but significantly overcollateralized. I think the average right now in between 200-300%. It’s actually incredible that we’ve kept the peg so well with such low capital utilization. We could either print unbacked dai, or lower minimum CR and still maintain overcollateralization on a system wide level.

1 Like

instead of printing unpacked DAI we should accept inferior collaterals (centralized stablecoins)

the idea that was almost done with PSM before the foundation stopped it

there are 2 scenarios with PSM:
1- we will have significant RWA available in the future so we can easily remove PSM without hurting DAI peg and adoption

2- significant RWA will take much longer time than expected (more than 5 years) then we should adapt with PSM

what are we doing right now is hurting and limiting DAI adoption and growth for the wrong reasons

1 Like

It is a paradox,

part of intended DAI utility is Unit of Account, that is why it should be on 1$ all the time. We are not building store of value here.

DAI being on value other than 1$ is not what it was intended to be.

I believe important difference between DAI and USD is in DAI full transparency
not in it being fully backed

redeemability of DAI for collateral is very limited anyway (only to Global Settlement i believe) so it is more important for user that it will be able to redeem it for 1$ in much more practical way - that is selling it on a market and that is the place where dai should have value of 1$ to keep promise of redeemability

You can actually do both and earn high APY in between. For instance Maker could mint x amount of unbacked DAI and deposit it into yCRV until the point DAI is $1 or whatever price ceiling you want (same idea as PSM). The difference with PSM is that Maker starts earning 60%+ on farming fees, the downside is that Maker gets directly exposed to other protocols and collateralization is closer to 1:1.

Also, if community was in favour of PSM and potentially increased stablecoin exposure with lower collateralization, similar mechanics could be reached quickly by having very low liquidation ratio for USDC or other potential stablecoins. For instance having 102% LR for USDC puts a price ceiling on DAI at $1.02. The difference with PSM is that protocol takes more risks when unwinding those vaults. Only certain users can unwind it when DAI price falls enough or governance would need to manually liquidate such vaults when price stabilizes.


Minting “unbacked” dai to be put into AMMs or other defi protocols where the protocol controls the defi tokens seems like a great option too. This is something I was pushing for with uni v2, but it could also potentially be done with balancer, curve, compound, etc.

In this situation the dai is only unbacked instantaneously, since it is immediately collateralized by the value stored in the redemption tokens for these protocols. This looks remarkably similar to a central bank purchasing debt in exchange for the liquid currency.

1 Like

Rather than try to build new technical solutions, the community should just use the tools that already available.

Reduce USDC and other stablecoins LR to 101%, putting a hard ceiling on the Dai price of 1.01 USD

Speed up onboarding of new collateral assets, especially DeFi tokens such as Uniswap LP tokens, compound tokens (with yield farming), Curve tokens, etc. These DeFi tokens would be able to easily drive 100s of millions of Dai supply, we just need to actually onboard them quickly.

The community should create bounties with signal request or MIPs that allows developers to build the smart contracts and oracle code for new collateral onboarding and get paid for each successful collateral onboarding.

This then allows the protocol to raise rates and earn even more money, setting the stage for community driven exponential growth.


I agree with you, there are plenty of ways to scale DAI.

waiting for RWA is not a solution, sure it will great when it is ready but in the mean time we should act faster… we have lost the peg for almost 6 months now

Maker perception has transformed from being the most successful Defi project to almost a failed project because of the peg


Agreed but we would have to reduce the risk premium, correct?

Definitely agree on this but DD must be done and I believe domain teams are 150% on workload, we should scale the :muscle:t2: but aside from the foundation which is doing all of the spending we simply don’t have the :bank:.

The rest I also agree but there is no cash from the community stand point. This brings me to the minting of mkr idea proposed previously, I totally agree with it as long as there is a clear destination of resources & accountability.

1 Like

I fail to see why we can’t have DAI backed by 1 USD of collateral and DAI exchangeable for 1 USD. USDC do just that. Your bank deposit do just that. Free banking notes were just that.

The issue is that the demand, instead of pushing the supply, is impacting the price. The USDC-A vault with 110% LR is providing a cap at 1 DAI < 1.1 USDC but it’s not enough. Having USDC-A vault at 101% LR will make liquidation difficult.

The intent on strategic reserves is a way to provide MakerDAO the ability to generate the supply leveraging a USDC-A vault. One step further would be to have a USDC-M vault (M for Marker or Market Operation) only available to the Strategic Reserves with a 100% LR and a negative risk premium (so SF is always 0%). It shouldn’t be too hard to do (let me know if I’m mistaken).

This hasn’t any liquidation issue has MakerDAO can close those vaults any time (MakerDAO control the strategic reserves). At worst (USDC goes to 0$) this has the same effect than printing unbacked DAI. Aside of that, each DAI still has 1$ of collateral (as each DAI is backed by 1 USDC = 1$). The SF is useless as any SF would be both an expense and an income for Maker (netting 0), just making things complicated.

We can print almost unlimited DAI and still keep 1 DAI being backed by 1$ of collateral with almost no initial capital. By construction, we can push 1 DAI = 1$ and make a profit.

When available, same can go with others stablecoins (to diversify custody risk) and with cDAI (no custody risk, but a protocol and liquidity risk). But just with USDC, we can mint almost 130M DAI and not exceed current risk appetite.

Does that seems acceptable to both those wanting to keep 1 DAI = 1$ of collateral and those who want to print unbacked DAI to solve the peg issue? It seems to me a kind of middle ground.

If there is traction and nothing I missed, I can push the idea (as is or improved by the discussion) forward.


This is basically the same as the PSM. Anything like this would be good - so many ways to benefit MKR holders when the peg is >$1 - we should be taking advantage of it.

PSM was a good and relatively simple idea to start and experiment with

I hope the foundation reconsider it

we can later build further strategies on PSM or we can remove it if we were able to scale DAI supply

right now for the DAI to keep the peg that means that the demand for DAI = the demand for leverage

the problem is that the demand for DAI is much larger than the demand for using Maker as a leverage platform… and I see this problem will only get worst

so we are hurting DAI adoption (lowering the demand) to keep the peg

PSM can change this equation and let us meet the huge demand for DAI