Proposal: Pay off impermanent loss with datestamped IOUs

I am operating under the assumption that if the deficit was standing at $1, the Bancor Foundation or some Bancor-affiliated entity would have paid that $1, recovering the lost reputation of the project.
So there does exist a number somewhere between 1 and 32,000,000, the current deficit, which if the deficit is lowered to, will create the perfect win-win opportunity for depositors to be made whole, and for Bancor to completely restore their reputation. I can’t know where that number stands, but what we can know for sure is that lowering the deficit brings us closer to it. The question is, how can the deficit be lowered…

The way I see it there are three major mechanisms:

  1. The desperation of depositors who give up all hope and take the cut (bad for reputation long-term).
  2. An increase in BNT value.
  3. The entire crypto market crashing down at the same pace, thus lowering the USD value of the deficit for all pools that are not stablecoin-based.

In my opinion, relying on #1 until the very end would cause irrecoverable reputational damage, and #3 is already on the way, and in fact we have a limited time to take advantage of it. So we need to focus on #2.

The issue with #2 is the absence of trust in the project, preventing any potential buyer from showing interest in BNT.
Meanwhile, at first glance, in order for the project and the BNT token to gain trust, all depositors need to be made whole first - a classic catch 22…
Unless we think a bit outside the box - perhaps we can break the catch 22 by improving confidence even before the funds to make all depositors whole are present.

How? By only partially recovering some of the lost funds for the users. Advertised as a “first step to making all depositors whole and proof that the Bancor team is on the way to a full recovery”.

Of course, there are a number of “horribly wrong” ways to go about it. Paying off 1/3 of the deficit, for example, would cost Bancor upwards of $10m and will not do much to recover reputation. So obviously this is not how it should work.

I believe that instead, there should be a limited number of wallets that get their funds fully recovered, while the rest are promised to be next in line. And it should be done in a way, in which the depositors who are left behind for the time being are not disappointed with the decision. In my opinion, there is one clear way to make this happen:

Return impermanent loss protection

I propose returning impermanent loss protection, except, instead of paying out in BNT, the team is covering the impermanent loss gap with date-stamped IOUs. For example, if a ENJ holder decides to withdraw their 10000 ENJ from the pool, in the month of December, they get 6000 ENJ and 4000 BNT-ENJ-Dec2022 tokens.

This way, depositors are offered a tradeoff - they can get more than half (at least for most pools) of their funds to use elsewhere as they wish, while also retaining a chance of getting the rest at a future date. In addition, the IOU tokens will be tradable on the open market, meaning that people who withdraw will usually get more in liquid funds than just their outstanding balance minus impermanent loss.
Many will be happy to oblige - thus withdrawing and reducing the outstanding deficit. Perhaps me included.

Sure, the sum of the deficit and the IOUs will remain the same as before, as those IOUs will need to be paid at a future date.

However, it does lower the amount of funds needed to make the first batch of liquidity providers whole. And if it wasn’t obvious, the first batch would be those who left their funds in the contracts. The second batch would be the holders of the most recent IOUs (which is why they had to be datestamped in the first place).
This creates a clear, understandable process that I believe nobody would disagree with.
As for the datestamps, I think creating new IOU tokens every month would provide a fine balance between:
a/ a low amount of overall tokens so that their appearance on the open market doesn’t confuse potential speculators
b/ a high enough separation that would make individual “batches” of holders to be repaid smaller, allowing for more frequent positive news around the project.

This will begin the process of restoring confidence in Bancor and the BNT token might become interesting to traders once more. I am convinced that both selling pressure will be lowered because of it, and buying pressure will be increased - even if subtly.

The combination of withdrawals reducing the first batch deficit, and restoring hope in BNT as a token, would tackle the deficit problem from two different angles. Until at one point, the deficit drops low enough that a full recovery for the most patient liquidity providers is possible (aka we have reached that magic number between 1 and 32m).

I believe that once that happens, it will trigger a cascade of positive events, leading to a full recovery much sooner than it would have happened if the team was to wait until all funds are recovered from v4 profits.

This not only means a quicker recovery of both user funds and project reputation, it also mitigates the risk of a particular coin outperforming BNT massively. I do believe that we are on a timer before something like ENJ, or LINK or some other coin starts gaining traction, and unless the team has positive news to bring to the community, such an event might spell Bancor’s demise as a whole. I strongly believe that we should act quickly in anticipation of such an event.

I am not sure if I managed to elaborate the idea in the most concise manner, so feel free to ask questions. It is not complicated, but I believe that it is a massive improvement over doing nothing, while simultaneously not hurting the project in any way. In addition, it does require minimal coding. I’m curious to hear what the community thinks about it, whether something similar has been discussed before, etc.


The problem with this solution - as with many others that have not made it to governance - is that this provides incentive for LPs to withdraw their capital. Those who LP in Bancor need the liquidity of every other user in order to maximize depth and generate fees on trades.

By allowing users to generate IOUs at the time of withdraw new problems are created

  1. Withdraws → less liquidity
  2. Less liquidity → Greater IL
  3. Greater IL → Withdraws
  4. Withdraws → Back to step 1

The IOUs will be harder to pay back when there is less liquidity, and the lower the liquidity means there will be less volume, and the lower volume means less money is going towards reduction of the deficit.

The best thing that could happen to Bancor right now is a 1 year crab market - totally sideways motion for 365 days - because all of the fees generated on volume go towards deficit reduction. It is a fact that with - enough - sideways price action the deficit would be totally fixed.

And as the vortex is reducing the deficit every day I see no reason at this time to start incentivizing withdraws.


Thanks for posting this proposal in governance. Just wanted to mention that since this thread was originally created, that a large portion of the surplus from v2.1 was migrated to v3 which did lower the deficit across many pools. Most of that is tracked in the following post:

and I think this was quite positive for many token LPs in v3. Overall, I think the deficit is trending in the right direction and is at similar levels from back in June (see

When it comes to regaining trust in the project, the recent announcement of Carbon, a quote driven orderbook on chain, in my opinion was was well received by our community but also by those outside. This shows that despite everything that has happened with Bancor v2.1/v3 and the paused of liquidity protection, that this protocol is still innovating in the DeFi space and building new financial products. I personally think that Carbon is a revolutionary AMM (one of its kind) that will find a niche in the market with those looking to perform automated flexible trading strategies. Here are some resources on Carbon if you haven’t had a chance to dig in:

⁃ Whitepaper:
⁃ Litepaper:
⁃ Intro blog post: Introducing Carbon | by Carbon | CarbonDeFi | Nov, 2022 | Medium

I think we are on the right path to regain trust in the project. If Carbon is successful and starts generating fees for the protocol, I don’t see many reasons for the DAO to not vote to direct fees towards deficit reduction. The precedent of using the vortex is there with v2.1 and v3 and is something that can also be used for Carbon.

1 Like