In order to lower the volatility of the sUSD peg against other stables, we need to update the “A” variable on sUSD-V2 pool from 100 to 200.
Motivation:
At the SNX community, we have been discussing the state of the peg extensively in the last few weeks. In order to enable trading on Synthetix, sUSD needs to be close to 1$ to assure traders safe entry and exit passageways.
However, ever since CRV tokens were being distributed and 100 new copy-pasta farmlands have been generated from thin air, sUSD-v2 has been volatile due to farmers exploiting the high A-Factor sUSD-Price.
We have been taking action, by lowering rewards here and here as well as increasing dramatically the supply of sUSD. However our actions are having a limited impact due to the high CRV rewards allocation to that pool and low slippage (due to the A-Factor).
As can be seen here , any time sUSD-V2 allocation of CRV tokens was lowered, the peg improved a lot subsequently so the efficiency of action is on Curve side, not SNX side.
We believe that the change in the A-factor will help ensure a better peg on that pool, allowing snx traders a reasonable volume a passageway towards sUSD.
Fundamentally, the reason why SNX doesn’t increase supply further, is because if you look at the historical trading volume on synthetix.exchange, it is too low as compared to the size of the sUSDv2 pool. So increasing sUSD supply just increases leverage of SNX minters and risk of the system being underwater if SNX price drops significantly. Therefore, the risks on Synthetix are not trivial when deciding to change supply.
Specification:
Commit A 200 on sUSDV2 pool only
For:
All the reasons mentioned above about the peg being vital to SNX, as justwanttoknowathing put it Part of what gives SNX value is that the synths that are generated from it are properly pegged.
The less we have a stable peg on Curve, the more the snx community is searching for alternatives and proposing things that could result in moving away from CRV as the main on-ramp to traders (as you can see here).
Against:
The main reason is more of an imbalanced pool, but honestly it’s not really such a bad thing if you have better pegs. In other words, farmers can dump all the USDC/USDT on this pool and take all the CRV, doesn’t matter as long as sUSD price is aligned with that of USDT/USDC.
Poll:
Not owning enough e-CRV, if someone can help with this.
To me the source of price volatility is a combination of three factors:
Farming game demand (which has nothing to do with CRV or A-factor)
CRV rewards (a lot of people keep voting for sUSD gauge weight), again unrelated to A-factor
The relative scarcity of sUSD (it’s hard to mint more of it since it’s dependent on SNX market cap).
None of the 3 things above ultimately relate to A-factor. If you lower the A-factor, all you will accomplish is probably raising the equilibrium price of sUSD because:
The relatively sensitivity of farming games to sUSD price is low since they are often earning 1000%+ yield, they don’t care if they pay 1.03 or 1.04.
So lowering A factor will punish the very people you want to help, regular users of sUSD pool. In fact, if you RAISED the A factor, that would help the situation by tolerating higher deviations from ideal proportions in the pool. For example, if A-factor were 1000, the pool would only hold 5-7 mm sUSD (ballpark estimate) while providing the same liquidity. That’s because the problem is the overall level of sUSD required in the pool relative to the other asset levels.
Well probably this A factor is new to me, wanted to throw in a proposal to make it more on peg, rather then the reverse. Anyway changed it to doubling the A, rather then halving it. Thanks for clarifying.
Also if you can vote down any future proposals to increase in sUSD v2 gauge, would be great.
I will be reading the white paper over the weekend I promise … I will wait back for confirmation before changing this proposal, as many confirmed to me that a higher A, will lead to a more pegged sUSD (i.e 1 sUSD ~ 1 USDC or USDT)
Increasing A factor will weaken the bonus/slippage modifiers, making it “easier” to withdraw sUSD to exit the pool and deposit USDC/USDT to enter the pool. This will probably cause the pool to grow in size from more USDC/USDT flowing in, the % of pool for sUSD will drop, but the price could very well stay the same. The Curve pools should not be relied upon to magically hold pegs for other tokens while the CRV rewards are so high. Ironically, this also means the SNX incentives for LPs are not helping their peg either.
Feels like there should be a pool with sUSD and DAI, the synthetic dollar stablecoins. They’re both over-collaterized, which limits their ability to scale with demand, unlike the fiat-backed stablecoins.
@iamaloop
Well the thing is, in the old days we were able to control the peg just fine (pre-CRV rewards) by calibrating supply of sUSD (with collateralization ratio changes) and rewards of SNX shot at the pool.
But now, that CRV is used to incentivize providing to this pool, the peg goes off quite frequently, because of the “crop” harvest. So effectively higher A factor is what I am aiming for in this proposal in order to provide more stable sUSD price to traders on snx platform regardless of the imbalances in this pool V2.
I am not so certain that consistently withdrawn sUSD is the main cause of the de-peg, rather the addition of other stables and CRV rewards. This is confirmed by looking at recent history of gauge changes accompanied by steep corrections in the peg of sUSD against other stables.
sUSD doesn’t aim in my opinion to be the defacto decentralized stable coin of the universe like MKR, as just holding sUSD doesn’t pay for the oracle fees our platform need to pay to chainlink. Rather, sUSD supply is for traders to trade with, take positions and exchange synths. The sole purpose of curve was to provide these traders with a way to enter and exit the SNX universe in a stable manner, however the externality of CRV rewards has negatively affected our control over the peg.
So, unlike the case of DAI, sUSD should normally scale with the appreciation of the price of SNX, as more demand for sUSD should be tied to more trading activity, but the your not so “honest” everyday farming has resulted in this volatility. So unlike MKR, normally controlling monetary supply and snx rewards, is sufficient to control the peg, if we didn’t have CRV, as a rise trading volume will push snx price higher and supply of synths higher (this aside from the synth loans available on the platform that can absorb abnormal shocks to synth prices).
Now finally, convincing people to lower CRV rewards shot at a sUSD v2 pool, is asking people who vote for these things to earn less (since CRV owners vote on proposals), which kind of a hard to do . Alright then, what else can we do in this situation, if we want to continue to use and encourage people to use CRV for the on-ramp? Adjusting the A factor seems like one thing we can experiment with.
@charlie_eth I know you meant well by changing my title to “100” to “50” from ‘100’ to ‘200’. But it seems that most of the people on snx as well as here do believe that a INCREASE in A factor will lead to more stable price of sUSD to other tokens, regardless of imbalances in the pool. So I switched it back to 100 to 200, if you please don’t mind telling me if you want to do some changes on the proposal and I’ll amend it gladly, otherwise it’ll be confusing on what I did change and what I didn’t. Other than that can you please setup a vote for me after the change is agreed on, so I can link it in the post. Thank you.
Sorry there was some confusion about the numbers, I was under the impression you had gotten it wrong.
I think discussions internally are suggesting that this would be a bad idea and could hurt sUSD more when it veers off its peg. Maybe getting a SNX dev to come and give us their thoughts would be helpful.
mmm, can you share some of the arguments against doing this? I’ll try to shuffle the arguments between this universe and the snx universe … But from what I understood, raising the A, will lead to more pegged sUSD against other stables and potentially a more imbalanced pool, which is fine…
In the end, sUSD is not being drained from curve pool, because it has little use aside from being used for exchanging on synthetix.exchange… It can be used to farm on Curve, but that would not result in drainage…
I believe with the farming falling out of fashion and having SWRV now that drains excess farming liquidity from CRV… No need to change this factor anymore, peg is just fine .
I disagree on that… Having a pool with $150M while daily trading volume on the synthetix.exchange is less than <10% of that… well why the heck do we need to increase sUSD supply and leverage/risk on the snx system? This is clearly an imbalanced pool CRV problem which can be fixed with A factor adjustment or CRV gauge adjustment…
Anyway, I know where your vote will go, regardless of the arguments I can ever throw here… So let’s leave it at that
And while I pointed out the negatives, I am open to increasing the A factor a bit.
I just didn’t like the argument of using it as a peg management lever. Note that A is an amplification coefficient, it’ll also means, that it’ll be more aggressive at buying sUSD too. It’s dual sided, it also means that the curve pool will try to be more aggressive at propping up the sUSD peg too.
One thing to note is that LPs will have a bit more risk if sUSD depegs downwards.
sUSD will still be above the peg if A is increased. It’s possible that the sUSD price will be lowered, closer to the peg, but it’s also possible that the lower price will lead to increased demand for synths, and you end up still above the peg.
Changing the A coefficient is quite different from the CRV gauge adjustment. There’s still going to be a lot of sUSD bid pressure even if you adjust the A coefficient.
I agree that the CRV gauge adjustment is a very powerful monetary policy tool. Maybe the Synthetix DAO needs to lock up some veCRV to have a bigger influence on decisions.
… I will wait back for confirmation before changing this proposal, as many confirmed to me that a higher A, will lead to a more pegged sUSD (i.e 1 sUSD ~ 1 USDC or USDT)
. Alright then, what else can we do in this situation, if we want to continue to use and encourage people to use CRV for the on-ramp? Adjusting the A factor seems like one thing we can experiment with.
.