How much DAI is needed to get the peg back to 1:1

As a follow-on to the general theme a discussion surrounding the peg and to get an idea of why intense action may be justified, please the below informal poll (not a signal request)

How much DAI needs to be minted in order to get DAI back to $1.00 as measured with a 14d trailing average?

  • 1MM - 5MM
  • 25MM -50MM
  • 100MM -150MM
  • 200MM - 500MM
  • 750MM - 1Bn
  • more

0 voters

similar to the above

once the above 1:1 is achieved,

How much further should we internationally overshoot before we start to use the DSR to cause the demand (instead of tapping the breaks on DAI issuance)?

Starting at

  • $1.00000
  • $0.99999
  • $0.99990
  • $0.99900
  • $0.99000
  • $0.98000
  • $0.95000

0 voters


I said 200-500m. But I’d say in the upper range. Much closer to 500M+ DAI needed. At least 350m is needed to reach 1.0 And the lower price might cause more demand.

2nd question is a bit weird since I don’t really think it’s the DSR to stimulate demand, the main lever is raising the SF to curb supply.

I said $0.99 because nowadays with all the AMMs market making DAI, $0.99 is quite significant. It takes a lot to go from $1->$0.99. Not that I’m expecting us to reach $0.99 anytime soon. A bit of wishful thinking that we’ll break 1.0.


We have start using the risk premiums based on the risk alone… It really should be divorced from monetary policy.

It is specifically the goal of the DSR to reduce the traded supply of DAI by removing it from circulation. It has a demand impact when a return is generated from it.

We will. As RWA are brought on, it is just a question of when, what scale, and what pace.

Please add multiple choice next time. I’d like to choose more than one option as I’m torn between 2.


Will do! Probably a good idea to start doing this informal poll quarterly (or even monthly).

If only there was someone who could give us a weekly report on the “state of the peg” :thinking:

There’s still the base rate. Unless you’re saying all stability fees will be fixed.

IMO the DSR is quite a weak lever considering all the yield farming going on.

I’m speculating that the SF/(base rate) going from 0% to 5% has a much bigger impact then DSR going from 0% to 5%.

1 Like

Fair points.

The acronyms RP and SF and then base rate get used a lot. From my sense, I am referencing the aggregate cost of capital to the borrower. On a per collateral basis, the cost of capital should be known (and only really change when the risks associated with that collateral shift) and then the DSR crosses all collateral. At least this was the last cycle of what I understood. There are discussions about removing the base rate. @Primoz no doubt has a better handle on that.

I am no expert on yield farming. However with any inefficient market that generates a yield (e.g. algo trading)… the incremental returns decline over time. The DSR however, really is not trying to compete in that space. As a synthetic US Dollar, it can be marketed and acquired by anyone wishing to do commerce and earn a core return (which is probably better than many other currencies vs. their local inflation).

I think this should be more specific when it comes to timeline. If I dump 100M Dai onto the market tomorrow, I’d bet we get under the peg on most venues. If you want Dai to remain sustainably under the peg, that’s a different story.

1 Like

Fair point.

Lets amend with the notion that the measurement would be the trailing 14d avg.

1 Like

I don’t see the point of those questions. Whatever you do, you have to do it continuously. The design of MakerDAO must be changed to be able to control the DAI price in all circumstances.

If you want numbers - look at the USDC+USDT market cap from about 6 months ago and now.

The second question makes even less sense to me. We don’t overshoot intentionally.

1 Like

AMM’s can absorb 300M+ of DAI and we won’t go under the peg.



@Jiecut , do you have a sense on the breakdown between AMM that produce DAI vs those that consume it (and then trade for other tokens)?

I am trying to gauge how unique yETH was in the big picture of things.

I believe it is fair to say that we need more DAI supply in the market to rebalance the supply and demand to push the DAI price to 1:1 …

As for the question on when to start the DSR, if we start increasing it right when we hit 1.00000 it will all come down to the “inertia” of the price coming down. We may have an excess of supply such that we may need to start anticipating at 1.0005 … likewise, if the inertia is minimal, we should probably overshoot a bit such that after we increase the DSR we have room to have the price go back up to 1:1 …

I was hoping this was a multiple choice. I picked 500-800 when I really wanted 200-500.

On the overshoot we wouldn’t use DSR I think we’d want to use SF and raise some funds.

I don’t want to see a DSR return until the SF is like 2-4% and sustained so I abstain from the second vote just because of what what used to sop up liquidity. In general however we do it I’d like to overshoot to .99 if we could just to prove we can provide enough liquidity. With USDC selling DAI we won’t get to .99 - maybe .999 after perhaps 100M DAI-USDC changed hands - at least.


I think it should be around 500 million.

The current market SF should be maintained above 2%, there is no reason to continue to maintain SF =0. The idea that SF=0 can help restore PEG has proven to be a mistake.

1 Like

Agreed- we need the SF at 2-4% for all major assets before we re-institute the DSR.

1 Like

The stability fees had to be reduced to incentivize DAI minting. Had the SFs been increased, the peg would have been much higher.

1 Like

not when we have PSM and lower LR on stablecoins to create direct arbitrage opportunities. Then people will mint dai to fix peg despite the SF because its still profitable. Look at whats happening with USDC right now and we are finally getting good income for MKR from it.

The market is flooded with farmers. Just look at what happened to USDC-B and you will know how trivial it is to charge them 2% SF. I insist that SF=0 does not help solve PEG. The fact is also true.

1 Like