SIP-0066: Curtailing Zero Borrowing

Bitocracy SIP | Sovryn Improvement Proposals
1

As we expected, Zero has proved to be an extremely desirable product. Zero exists as part of a two-sided market. On the one side are those who wish to borrow against their BTC, and in doing so, issue new DLLR into existence. On the other hand are Sovryn Dollar holders, who create the demand for Sovryn Dollars that are issued by Zero.

As is frequently the case in two-sided markets, one side is much easier to grow early on. In the case of Zero and the Sovryn Dollar, it has proven to be much easier to grow Zero than the Sovryn Dollar. This is due to the fact that Zero is immediately useful to a single individual, whereas the Sovryn Dollar becomes more useful at scale. The Sovryn Dollar is a currency, and like all currencies, it requires and benefits from a network effect.

Growing a two-sided market is a common challenge in the digital realm. There are well worn strategies to achieve this. I will write more on this in a future post, to be added to this discussion thread.

Currently the development of the Sovryn Dollar ecosystem lags behind. While we are progressing in developing the Sovryn Dollar ecosystem, we expect it will be a few months until efforts begin to bear fruit. In the interim, this poses a challenge to the Sovryn Dollar peg and causes cannibalization of Zero LoCs through redemptions.

Our highest priorities are:

  1. Maintaining the DLLR peg
  2. Minimizing redemptions.

We have seen some success on both these fronts. However, until we begin to see consistent growth in DLLR demand, this progress remains fragile.

For this reason it is proposed that we should curtail new Zero borrowing until DLLR demand shows significant and consistent growth. Given that Zero is permissionless, we cannot shut borrowing down with the current smart contract. Instead it is proposed that the Origination fee be increased to 1000%.

Concerns and Considerations

How will this impact protocol revenue?
In the short term, we can expect to see a significant reduction in protocol revenue, as originations are currently the most significant revenue driver. This short term negative impact is prudent, if it positions us for further revenue growth in the future.

Protecting the two priorities described above, while growing the demand base for Sovryn Dollar is exactly the set up required for this kind of future growth.

This will also allow us to more easily direct borrowing demand to the fixed-interest borrowing product, which will help drive DLLR lending demand.

How will this impact users currently holding LoCs?
Existing LoCs will not see any change to their fees. What we expect they will see is a lower chance of redemption.

How will this impact new users?
New users will be unable to open LoCs at attractive rates. Adjustments to the frontend can help make sure they do not create very expensive LoCs in error.

Resolved:

  1. Increase the Origination fee - Base: 99%; Max 100%

Code changes:

Updating BORROWING_FEE_FLOOR from 5% to 99%
Updating BORROWING_FEE_MAX to 100%

2

Very interesting your analysis. Still, I have this doubt:

I have always thought that redemptions increased or decreased based on the redemption fee. I can’t understand how redemptions will be reduced by changing the origination fee instead of the redemption fee. Could you help me understand?

Perhaps the idea is that there are no new LOCS issued, and redemptions gradually reduce the existing LOCS (currently decreasing at 10% per month) until reaching a point of balance between the demand and supply of DLLR?

3

There is noticeable causation between borrowing activity and redemptions. Either the borrowers directly redeem the DLLR/ZUSD they have borrowed, or they sell the DLLR/ZUSD on the AMM, which depresses the price of DLLR and incentivizes redemptions. Curtailing borrowing is the “lowest hanging fruit” thing we can do to reduce inflation pressure and slow down redemptions.

That said, this doesn’t get us completely out of the woods: there are still millions of ZUSD in Mynt and BabelFish and it’s only a matter of time before DLLR and XUSD holders will want liquidity. We have already seen them use redemptions for liquidity, given the lack of other options. And so the next thing we need to do is to increase liquidity in other stablecoins in Mynt and BabelFish, to provide an alternative path for DLLR and XUSD holders to get liquidity if needed. This is outside the scope of this proposal but nonetheless an important initiative that the community will need to focus efforts on.

3 Likes
4

I see 2 problems.

  1. The DLLR lending pool is not working. Even if a liquidity provider deposits DLLR there and almost 100% of the available liquidity is lent out, liquidity providers get no yield. Literally Zero yield. Although the tooltip says 6-7%. We should fix this problem. A non-functional DAPP is the biggest obstacle to the adoption of DLLR. Here we lose a real use case for dllr. I won’t even start on margin trading again.

  2. The new swap function of the new dapp will take the cheapest route for the user, including redemptions. With the split/low liquidity in the amm pools and with the other costs involved, the cheapest way may very well be via zero redemptions. This means that first of all the big honeypot (BTC in Zero) is used for any swaps, if this is the cheapest way. This mechanism will not work.
    We should increase the redemption fee so that this way is no longer cheaper than the regular AMM swaps.

Before voting on the steps described in the SIP, the old dapp should be put in order and a smooth user experience should be presentable.
Have we really done all we can to create demand for DLLR? Certainly not with the old dapp. Let’s start there.

A clear no from me.

4 Likes
5

As far as I know, the lending protocol and UI is working as expected. Keep in mind that in order for lenders to earn interest, there have to be borrowers paying interest! If a lender has supplied liquidity to the pool, and no borrowers have paid interest since then, then that lender will not see any positive value in the “Profit” column. However, if a borrower does come along and pay some interest, then the lender will see it reflected like this:

Screenshot 2023-07-26 140631

This would contradict priority number 1 that Yago identifies, which is maintaining the peg.

Since the beginning of April, about 2.1m ZUSD has been borrowed. During this same time period, the cumulative redemption volume has almost tripled from 1.3m ZUSD to 3.4m ZUSD, an increase of about 2.1m ZUSD. Looking at onchain activity we can see that the similarity of these amounts is no coincidence: there is direct causality between borrowing and redemptions.

Bitocracy cannot control demand. Bitocracy cannot force anyone to hold ZUSD/DLLR. And though holding can be incentivized such efforts so far have still been outstripped by new supply hitting the market.

This leaves us with one short-term option: Bitocracy does have a lever that can be used to slow down or stop inflation by dramatically increasing the cost of issuing new ZUSD into circulation. For the sake of line of credit owners, to maintain credibility as governors of the protocol, Bitocracy should use it. The origination fee rate must be increased to curtail borrowing and reduce redemption pressure until liquidity recovers and DLLR is at or above peg more often than not.

6

Thanks Sacro. Very valuable feedback.

To point 1 - The lending pool should be paying out yield - I am looking into the reason it is not showing up.

To point 2 - I agree. I have previously voiced objection to the bundling of redemptions with swaps in this way. I’d like to see an analysis of what % of swaps would see a price below redemption?; what % of swaps would result in a redemption? Without such an analysis, I don’t think we should have this bundling.

However, I think both the above points are orthogonal and unrelated to this SIP. Even if we grant both as true, that does not change the calculus where originations drive redemptions.

As @light points out above there is almost perfect symmetry between originated and redeemed value. This is make perfect sense, as the demand function of DLLr has not much changed and so the only place for excess DLLR to go is to redemptions.

1 Like
8

I don’t understand all that technical stuff about maintaining the peg etc.
But for me the main questions are:

  1. If the only noticeable source of income for stakers is cut, what do you think will happen?
    Stakers won’t extend their stakes, no new stakes will be added (what for if no worthy yield is generated)
    For the same reason (lack of yield on staked SOV) paid out SOV will be sold/swapped immediately

  2. Price of SOV will drop like a stone once again - why knowingly cripple the protocol and its loyal users?

  3. How long is “short term”? half a year? more?

  4. Who is left then? Give me a reason why anyone should recommend Sovryn to anyone?
    There is no use for an “army” as you were asking for recently.

I am a bit shocked here.

2 Likes
9

If Zero isn’t an attractive protocol to borrow from because there’s a high risk of redemptions, that will also lead to a reduction in staking revenues. As I see it, the decision SOV stakers have to make here is not between “less revenue from Zero if this proposal is approved” or “more revenue from Zero if this proposal is rejected”, but rather the choice is between either “if this proposal is approved, less revenue from Zero in the short term (because of the effective pause on Zero borrowing) in exchange for more revenue from Zero in the long term (due to increased confidence among borrowers that their collateral will be safe)” or “if this proposal is rejected, maybe more revenue from Zero in the short term if people continue borrowing, but less revenue in the medium and long term because Zero’s reputation will be ruined due to chronic redemptions”. I’m hoping that stakers see the big picture here and think about the long term health of the protocol, and not overly fixate on lost revenue in the short term.

3 Likes
10

This is a declaration of surrender and nothing short of a catastrophe.
The exact opposite of what should be done.
Wrong on too many levels to even count.
I not only will vote no, like any sane person should, but I never thought a SIP would be proposed to essentially kill the protocol…it is pretty much the only thing that ever could have been proposed to kill off Sovryn for good. I would have to unstake and eat my losses, since there would be no coming back from this. Ever.

1 Like
11

Appreciate your quick response, thank you.
I understand what you are saying and Travis shares your opinion in the dojo.
I am certainly not “fixating” on short term revenue since I am not a top staker
so I was never overly excited about the numbers anyway.
I am trying to see the big picture from a different angle here:

Who will be left after whatever time period you decide to lift the new hurdle?

How many are still there to listen?

Do they fit in a pub or in a phone booth?

How do you intend to win them over again?

“This short term negative impact is prudent, if it positions us for further revenue growth in the future.”

Sounds great but how is this going to be achieved?

For the time being there will be no incentive and that is fine if there is a ‘real’ concept behind it.
I am just concerned and I try to understand your process but for now it looks like Zero has failed its first
baby steps and fell on its face. Hope the it can get up on its feet again.

12

It is not killing the protocol. It’s governing the protocol, as SOV stakers are supposed to, in this case to protect line of credit owners from excessive redemptions. If you have any ideas for alternatives to achieve this, I’m open to suggestions!

A good comparison here is Maker DAO, which early on in the life of the protocol had a debt ceiling that restricted the amount of DAI that could be issued into circulation. Zero does not have a pre-programmed debt ceiling, but based on what we are able to observe in the market I think it is prudent for Bitocracy to impose its own limit by effectively pausing borrowing, since liquidity has been completely exhausted and new borrowing is only leading to more redemptions. If you are unfamiliar with what this means, you can read my post about redemptions from earlier this year here. (Yes, it is a problem Zero has been struggling with for a while, and hasn’t found a sustainable solution for yet.)

I didn’t mean to imply you were! I was moreso referring to stakers who would bail due to the immediate revenue implications of this SIP.

If the choices really are as I put them in my previous post, then, again, the choice is not “lose revenue → lose stakers” or “don’t lose revenue → don’t lose stakers”. If Zero redemptions continue then we are going to lose revenue and stakers either way. The question is which path is more likely to lead to recovery and growth? Allowing borrowing and redemptions to continue unabated, or pausing borrowing and allowing liquidity to recover and restarting once demand has caught up to the supply?

So I would turn the question around to people who oppose this SIP: what do you propose should be done to increase DLLR demand in the immediate/short term and slow or halt redemptions? How are you going to offset the millions of DLLR worth of borrowing we’ve seen in the last few months and would likely continue to see over the next few months? I can speak from the perspective of the contributors working full time on this and say that a lot of ideas have been tossed around, now a few of the most promising are being focused on, but since DLLR is new it takes time to build trust in this currency and so these initiatives will not have an effect overnight. The purpose of pausing borrowing is to buy time for these initiatives to yield results, for DLLR demand to grow and for liquidity and the peg to recover. If we don’t pause borrowing the redemption situation will only get worse, and by the time our DLLR demand initiatives begin yielding results it may be too late to repair the reputation of Zero (and by extension DLLR).

Yago spoke about a few initiatives in progress on the governance call yesterday that was organized to discuss this SIP. The recording should be uploaded soon. At a high level these include new distribution channels via third party integrations and finding more companies such as Exodus to hodl DLLR in their corporate treasury. So, more applications and use cases for DLLR, and more hodlers who recognize its value as a censorship resistant digital dollar.

3 Likes
13
14

The discussion around this SIP has just begun and the vote is already being started.

The increase in the origination fee is arbitrarily high. We could have discussed raising it further without having to come up with such drastic values.

The announcement of a zero governance call was also very short notice, just 24 hours before the call took place.

24 hours later, the SIP is live.

This makes me very concerned and sad. I would have expected more foresight and rationality when it comes to making decisions in the bitocracy.

6 Likes
15

It’s possible this was unnecessary and a mistake. However, as I explained in the governance call, Bitocracy is slow and can be front run - and sometimes we need to be able to enact votes fast. In this particular case, there has been many, many months of discussion - but this SIP is at risk of being frontrun and I think there is a strong argument for seeking to avoid that.

1 Like
16

It is also worth pointing out - this is a temporary measure and we can, at any time, have a vote to reduce the origination fee.

17

Yago, to pass this we need >70% support. Likewise, to overturn it would require >70% in favor of overturning. So it’s not easy to overturn once it’s in place.

I understand the concern about front-running. But we’re talking about people rushing to borrow with a 5% fee. I really doubt there are lots of people ready to do that on a moment’s notice. I think it would’ve been good to have more discussion. (I realize I could’ve spoken up sooner, but I needed this much time to think about it, which is part of the problem.)

If I had a substantial Zero loan, I would be running to repurchase ZUSD to pay off my loan, for two reasons: 1) the exposure to redemption would make me want to pay back the loan and possibly move to a standard loan with interest, and 2) anyone can buy ZUSD at a discount right now. I realize that not everyone has the funds to buy back ZUSD, but some should. Others could use a small portion of their collateral to buy ZUSD and pay off their loan in stages. So most people could pay down their loan and reduce ZUSD that way. Buying ZUSD to pay off loans would drive up the price of ZUSD/DLLR relative to RBTC and therefore relative to other stablecoins, removing the incentive to redeem.

Redemption really isn’t the only way supply can be reduced in Zero. If the user base were acting rationally, paying back loans would close that gap quicker than redemptions. ZUSD is available at around a 2-3% discount, but redemptions can only profit 1-2% due to the 1% fee.

I believe we haven’t done enough to educate the user base (and the new user supply) about the risk of redemption and the opportunity to pay back the loan. That still seems more likely to address this situation in a healthy way than de facto turning off the protocol. At least I think it should be tried first.

For that reason I don’t support the SIP.

1 Like
18

Thank you for your message, always bringing a lot of common sense to your statements and analysis.

I agree with your observation about users’ lack of understanding of how ZERO works. Most of us, myself included, are unaware of the purpose of ZERO as it has evolved over time. Initially, for me, it was what Yago explained in his videos, but currently, it serves a different purpose for expert users. The defense of the PEG is the red line, and everything should support that objective.

Reading the hundreds of messages in the Dojo, we discover that we haven’t been educated to understand ZERO properly. Of course, we have the fantastic control panel you’ve built, but it might have been very helpful to have a couple of posts or videos explaining what happens from the birth of an LOC until it’s liquidated or redeemed. And why it happens, what elements are involved, what market agents participate, and how and why they do what they do. I have the feeling that many of us want to help or give ideas, but we don’t fully grasp the big picture to analyze it. That’s why it often happens that no one launches proposals or changes them on the fly, sometimes saying absurd things with the best intentions.

This SIP will be approved. The debate that you mention as nonexistent would have been futile even with an extra week, as everything seems to indicate that the decision has been made, and thus, 24 hours are enough for the idea, the proposal, and the voting.

I have some questions for you, and I hope you can answer them, regarding what would happen after the SIP approval:

  1. Is there any reason to think that redemptions will decrease?
  2. Does this SIP guarantee a more perfect PEG?
  3. Is it possible to defend the PEG and minimize redemptions, or are redemptions not the mechanism to defend the PEG?
  4. Can the equilibrium value between the supply and demand of DLLR be calculated?
1 Like
19

What is missing is a pathway to create fiat demand for DLLR/ZUSD to earn yield in the stability pool. Since DLLR is not listed on any exchange, no direct ZUSD demand is present except via the bridge which is a bottleneck and cumbersome.

What would it take to get DLLR or ZUSD listed at e.g. Bitfinex who already support RBTC? I’ve heard it might take some $$$ to get a coin listed: can the bitocracy use some protocol revenue to get that done?

20

But what does temporary mean? Unknown, right? Origination fee raise was supposed to be temporary :man_shrugging:

All decisions are “temporary”, in some respect, no?

This decision is to shield current LOC’s from redemption, right? Why suddenly are we worried about being front run? Redemptions have been running rampant by someone “Interacting with the smart contract” for months. Suddenly its an immediate concern that could be “front run”?

21

I am still in the Free Zero camp. The “we need to defend the peg with our life” has not worked in the past months to make this exosystem really grow. I think the concerns above that there never was any really good edducational videos about how it all plays together is very valid. Id say lower origination fee back to 0,5% educate people how it all works in easy to understand terms. Make redemptions possible for everyone lower rdemption fee to 0.5% and then market the hell out of both DLLR and ZERO with lots of explaining along the way why the peg may be loose at the beginning why there are so many redemptions etc pp. People will start trading DLLR that means they might hold it cause it may go up in price that in turn makes demand generally there will be more eyes on the whole thing - which is what we are lacking most - more interest from the crypto community at large. That only happens when people actually use the protocol. Shutting down half of it will onl mean even less interest even less discussion about it. We need all publicity we can get - good and bad. You can see this even here. - even a shitty zip that everyone seems to hate is actually waking up the community and all of the sudden we have deeper discussions more engagement. That will work the same way out in the field. Every “Zero doesn´t work” headline will also generate people thinking “Hey the idea of DLLR is actually really interesting I may want to hold it a bit - yeah peg is bad but every other stablecoin had bad pegs in the beginning”.
We need to gain users and fast as project has not enough runway to “wait for big things to happen” another two month. I agree we need drastic changes but shutting down half the protocol is not the way to increase interest. Its illogical even.

1 Like