Summary

Introduce retirement bonds for using the KLIMA token to retire on-chain carbon. Retirement bonds are a form of inverse bonds where the bonder provides KLIMA tokens in exchange for carbon credits that are immediately retired. These bonds are offered with a maximum slippage amount, acting as an extension of current liquidity pools to support larger retirements.

Motivation

The current tokenomics for the KLIMA token are purely inflationary, with no integrated mechanism for dynamically contracting supply. While inverse bonds are useful in regulating supply and performed as expected during our experiment with USDC, they are not an ideal solution as the USDC sitting within the treasury is a finite resource.

While talking to potential partners looking to retire substantial amounts of digital carbon credits, the current limitations of liquidity pool sizes and structure often result in significant slippage. This dictates the need for a more efficient use of both protocol-owned-liquidity and overall treasury resources to empower organizations to acquire and retire carbon at larger volumes.

Proposal

Introduce Retirement Bonds, where users provide KLIMA to the protocol, which is then burned and the reserve carbon asset immediately retired. This allows the treasury to expand the liquidity needed for targeted pools without diluting existing stakeholders. Only KLIMA tokens are accepted to execute these bonds.

  • Pricing: Each bond will be offered at the time-weighted average market rate for the specified carbon credit, with slippage capped at 2% and includes a 30% flat DAO fee.
  • Max Bond Amount: The capacity for each Retirement Bond will be configurable for each carbon credit pool.
  • Duration: The bonds will be available until all allocated reserve assets are retired or at the end of each month. Each month more reserves can be allocated to Retirement Bond capacity by the Policy Team.
  • The KLIMA provided to purchase the carbon would be burned (minus 30% fee sent to the DAO)

Implications

What effect will these bonds have on our key protocol metrics?

Other protocols or projects that want to utilize the rebasing nature of KLIMA for retirement purposes also benefit from this, as their retirements can be directed to the retirement bonds. This reduces the downward pressure on the premium that KLIMA holds on carbon tokens, allowing for better efficiency when bonding more reserves to the treasury.

It is worth noting that creating standard inverse bonds with a carbon reserve asset results in additional pool tokens entering liquidity pools. Retirement bonds behave similarly to that of inverse bonds with the exception of retiring the reserve asset rather than releasing it back into the liquidity pool. Therefore, Retirement bonds naturally increase in effectiveness the closer a reserve asset trades 1:1 with the KLIMA token.

Implementation

  • Develop and introduce retirement bonds.
  • Allow the Policy team to allocate reserve assets to retirement bonds, not exceeding 5% of total reserves.
  • Any reserve token held by the treasury that maintains liquidity paired with KLIMA is eligible for retirement bonds.

Polling Period

The polling process begins now and will end at 12:00 UTC on January 17, 2023. Assuming the forum proposal is approved, it will advance to Snapshot.

Implement Retirement Bonds

This poll has ended.

I am strongly for this proposal as it leverages our unique position to put KlimaDAO carbon assets to more efficient use that ultimately ends in the retirement of more carbon credits and provides a better experience for carbon market participants through better asset price certainty and availability.

As proposed, Retirement Bonds move KlimaDAO from strength to strength.

The 30% fee that goes to the DAO, is the intention to only use it for staking rewards? It would be good to read your reasoning for the 30% DAO fee, aswell as what you intend to do with it if not only for staking rewards.

What's the argument for the pricing model? I would also like to read the reasoning for this.

Additionally, why not have it so that, for example, 1% of Klima could retire 1% of the BCT in our treasury?

    Garmon the 30% is the 30% of the 2% fees as far as I can tell. And my guess is that's gonna be for DAO expenses e.g. strategic partnerships, operational costs etc.

    Great proposal, really appreciate increasing the utility of KLIMA token. Personally I'm not sure about pricing of the retirement bonds. Which one of following examples is valid?
    BCT price would be 1$, KLIMA price would be 2$ ... if I would use 5 KLIMA to retire BCT, then 0.1 KLIMA would be the fee and the rest will retire 9.8 BCT.
    Or is it like 5 KLIMA will retire 4.9 BCT because each KLIMA is backed by 1 BCT?

      I have got 2 questions:

      1. Can you choose which type of carbon credit will be retired?
      2. Will the retirement bonds be in the bonds section? Or will it be an option on the offset page?
      • Cujo replied to this.

        So how does the pricing exactly work? Assuming, user buys 1 KLIMA from Sushiswap at $2, then bond-retires it for BCT at a price of $1 (twa). Does this mean, the $1 contains 30% DAO fee, meaning that this 1 KLIMA amounts to 0.7 tons of CO2 retired via BCT? (P.s. I would be fine with it.)

        Also: allocation limit 5% --> would 2% work, too?

        • Cujo replied to this.

          Can you elaborate a bit more all the calculations of the 50k example? Cfr comments above, different interpretations are possible. Thank you.

          • Cujo replied to this.

            Garmon
            The current pricing model is more about providing an extension of the current liquidity held by the token being retired through the bond. For instance right now with the current liquidity that is deployed for KLIMA/BCT, you can get around 2% slippage for roughly 16,000 BCT. If you were to retire anything less than that through a retirement bond then you would get same cost as if you went and did the swap on Sushi yourself.

            Some key differences for utilizing the retirement bond to make this "swap" however:

            • All of the KLIMA used to acquire the token would be removed from circulation.
            • Premium of carbon per KLIMA remains the same, rather than decreasing if you had directly swapped into the liquidity pool.

            The 30% fee would be 30% of the KLIMA that is used to retire. So say you provided 1,000 KLIMA to a retirement bond. 700 would be immediately burned and 300 sent to the DAO. That percentage is the same percentage fee we have for normal bonding circumstances.

            hodlmao
            In your example (assuming no slippage), you would retire 10 BCT. 3.5 KLIMA would be burned and 1.5 sent to the DAO.

            Kimbal
            We would have the ability to make different markets for each token supported under the aggregator! Not sure exactly where they would show up under the UI at this time.

            rrrmmmmm
            See my earlier examples, I think it helps clear that up. If not let's discuss further 🙂

            SVNatK-Lo
            I just went through and updated our sheet with some new numbers, here's a screenshot of the result, and a link off to the google doc if you want to dive in a bit more on the effects of a bond being executed.

            https://docs.google.com/spreadsheets/d/1s6g7eCobg34TyHG2z93ttpZ2W5Z2y1FlFknsK1Y4Lz4/

              One other important aspect here regarding the pricing of these bonds. We propose to continue the practice of utilizing the market to determine the current (and by extension time weighted) price of a token through the use of the existing liquidity pools. This is also why there is a max bond amount to ensure that any demand in excess of what is available within a current retirement bond flows back to liquidity pools.

              Cujo

              Thank you for sharing!
              I have a follow up question: I see the value accrual for klima here is linked to the backing increase. This increase happens (partly) as you don't count anymore the klima sent to the DAO as outstanding klima

              This implies however that the klima send to the DAO are not backed anymore (otherwise you'd have to further reduce the bct reserves with 15k and not count them as backing for the outstanding klima). That doesn't seem correct, does it? Maybe I'm missing something?

                As of now if I want to make offset it happens like this: I choose a pool, let's say BCT. The offset event takes BCT from the pool paired with klima and retairs it meanwhile klima token is not burned. Meaning that Klima is backed less with the BCT. -> this puts growing pressure on BCT bonds (once active) and more klima is minted by new bonding -> inflation for klima token.

                KIP-31 makes it so that you retaire BCT and equivalent amount of Klima is burned (minus the dao fee) so that the weights of the pool BCT/Klima stays the same.

                As of now, If there is small pool (let's say NBO) and I need to make huge retairment on that specific credit, I would need first sell my klima tokens and then swap them to NBO. This would lower the price of Klima token and rise the price of NBO on AMM (slippage) and that would be inefficient. With the new bond I'm covered from that slippage.

                Did I get it right?

                  Nikodemos
                  The first part of your deduction is correct, retirement bonds would be a new piece in the toolkit that makes retirement and regular bonds (when they reopen) more synergistic in the long run.

                  As for your last question, instead of slippage in the open market, you would still face a slippage cap of 2% (per the original proposal) when you retire NBO but without selling pressure on the KLIMA token.

                  Will these bonds be available for TCAs which price is way smaller than the price of KLIMA?
                  eg. BCT => 1$, KLIMA 2$ ... so to retire 20 tonnes I need to use 10 KLIMA ... counting with fee to DAO we'll burn 7 KLIMA tokens.
                  The supply will change in a way that breaks the backing 1 tonne == 1 KLIMA. With our current situation this would decrease runway, but in later stages of the project it could break the backing condition where there would be more circulating KLIMA than carbon tonnes in treasury.

                  • Cujo replied to this.

                    Clever mechanism. I like it.

                    Interested in a response to the question by SVNatK-Lo as I'm also unclear as to why the 15,000 KLIMA fee to the DAO in this example would be removed from total KLIMA supply (assuming it will be used to pay contributors now or in the future).

                    • Cujo replied to this.

                      SVNatK-Lo WAGMIcapital

                      The KLIMA tokens held by the DAO are considered out of circulation. The treasury still enforces the requirement that there must be at least one carbon token for each KLIMA minted. We also exclude the KLIMA held by the DAO from any market cap figures since they are not in circulation.

                      Taking the assumption that the KLIMA sent to the DAO would eventually be paid out, with this scenario if someone comes to retire carbon on chain, they had to acquire the KLIMA some way initially. Assuming a direct USDC swap to KLIMA as part of the retirement, there was still more demand on KLIMA for the retirement than if the DAO immediately distributed and sold its fee back for USDC.

                      hodlmao

                      So that case should be outlined in a prior reply of mine. When looking at backing per circulating KLIMA, the increase/decrease amounts have been using purely the direct reserve balance of the treasury. In this proposed implementation only excess reserves would be utilized for funding a retirement bond. So any retirements made through this contract would not affect the backing of currently issued KLIMA tokens.

                        Cujo

                        I still don't agree, sorry.
                        In your example you take relative to respectively the bct reserves and the klima outstanding circulation an equal % amount of bct and klima out. So this doesn't generate value. The fact that 30% of the klima is not burned, while the full equivalent amount of bct is burned makes every klima weaker. That you don't count the 15k klima anymore in your backing calculation is an accounting trick, because you state that they are still backed by the same treasury reserves that are still, in full, part of the backing calculation.

                        The real effect here seems to be shifting excess reserves from the klima holders to the DAO. There is no real fee charged to the user for retiring.

                        I don't mean to be negative so forgive me challenging this hard. I want to understand this to be 100% sustainable and positive for klima.

                        An alternative design could be that if you want to retire for $10000 it wil cost you $10500 (in klima) and of this $10000 in BTC is burned, $10300 in klima is burned to accrue value on the remaining klima and $200 in klima is sent to the DAO for funding further operations and initiatives. Only if users pay real fees to use your product you'll obtain a sustainable model.

                          Cujo In the screenshot of the spreadsheet, why is BCT burned 83k? Since only 35k of the supplied KLIMA is burned and used for BCT retirement, shouldn't it be 35,000 x 1.66 = 58,100 BCT

                          An alternative to the 30% DAO fee issue would be to price BCT (or any other digital carbon) at 30% premium and utilize this for DAO expenses.

                          Would the retirement bonds only be utilized when the order size is expected to cause a slippage > 2% ?

                          • Cujo replied to this.

                            Many have asked for a way to burn KLIMA and clean up older credits in the treasury since the beginning. As long as the math adds up, I support this KIP.

                            @Cujo
                            I agree with the objections by @SVNatK-Lo and @andybash12 . The numbers in the table do not seem to make sense. Providing 50k KLIMA to retirement bonds should amount to 58k = 50k x 0.7 x 1.66 BCT being burned. At the same time, 50k x 0.7 = 35k KLIMA would be burned and 15k KLIMA kept by the DAO as the fee.

                            The numbers in the table suggest that the 50k KLIMA lead to 83k BCT being burned. Hence, the bond-retirer (or "customer") gets BCT offsets/retirements with no fee at all. At the same time, not burning the whole 50k KLIMA, but keeping 15k alive seems to lead to undue inflation. If this is indeed the intended model, then a different word than "fee" should be used.

                            Is it possible to receive a better insight into how the policy team performed the modeling, testing and parameter evaluation for this concept?

                            Also, on a different note, is it planned to undergo an audit with the new functionality or what is the opinion by the policy team here?

                            • Cujo replied to this.