ForthDAO SPOT-AMPL pool on ElasticSwap


This is a proposal to discuss having the Forth DAO seed liquidity for the initial SPOT-AMPL pool on ElasticSwap.


Having received a lot of support from the Ampleforth community, I want to start a discussion about ElasticSwap being the launch AMM for a SPOT-AMPL pool, with the Forth DAO minting SPOT and seeding the pool with the minted SPOT and a commensurate amount of AMPL. The amount to be seeded is subject to any limitations on the initial supply of SPOT, but could be capped at 10-20% of the DAO’s AMPL supply (this is a cap not the amount needed). The deeper the pool the more efficient the pair will trade.

This proposal allows for the Forth DAO to have that additional SPOT liquidity while maintaining its AMPL exposure, in essence providing a zero-cost way for the treasury to diversify their treasury and grow the ecosystem, while also increasing the total capital the DAO treasury controls. Furthermore, increased liquidity for both SPOT and AMPL is a win-win for the Ampleforth ecosystem as a whole.


ElasticSwap is the first AMM that specializes in dealing with elastic-supply tokens, getting rid of impermanent loss as a result of the rebase. ElasticSwap is a partner of the Ampleforth ecosystem and has been operating pools on both Avalanche and Mainnet since April/May 2022, and is therefore a logical starting place for the initial pool.

The Forth DAO currently has FORTH, AMPL, and LP positions from various pools in the treasury. It is logical that the DAO mint SPOT as a way to diversify the treasury and deepen the amount of available SPOT but rather than just holding that SPOT in the treasury, the DAO could use that SPOT and some additional AMPL to bring liquidity to SPOT, provide a deep pool of initial liquidity, and put the assets to work via an LP position.

ElasticSwap is committed to supporting the Ampleforth ecosystem and will integrate SPOT into the frontend from day 1.

LFG frens!


LPing SPOT/AMPL is not zero-cost. There can be impermanent loss. There can also be additional losses when the leveraged long Z tranche earns less than the forced buying/selling of AMPL (as an LP). The DAO should consider how AMPL market cap changes affect the DAO’s net AMPL holdings.

Another thing to consider is what happens at tranche maturity.

When a tranche matures, the DAO’s SPOT is rolled over to the tranche of the next maturity, maintaining reduced exposure to AMPL. However, the leveraged long Z tranche (that the DAO holds when it minted SPOT) has also matured, so the DAO no longer has leveraged exposure to AMPL. The DAO has a net decrease in exposure to AMPL. If AMPL market cap goes up, the DAO will have lower % of market cap.

To be “zero-cost” at tranche maturity (neutral exposure to AMPL), the DAO needs to offset their reduced exposure to AMPL by finding some way to get leveraged exposure to AMPL, but Z tranches cannot be rolled over (yet). Another way is to mint more SPOT at the next maturity to get new Z tranches, then sell SPOT for AMPL.


Thanks for the proposal @Loki_VT !:pray: I can add some context as well.

Someone from ElasticSwap can correct me, but ES pools should be free from almost all IL you’d see in an AMM like Uniswap. Elastic Swap accounts for AMPL’s rebase and automatically rebalances accordingly. @Loki_VT do you have any docs that describe the IL exposure?

Yes, this is true-- we would end up trading some AMPL (which has upside and downside) for SPOT which is mostly stable. This could diversity the treasury in the short term… but I don’t view this as a diversity play because there is still correlated risk between AMPL and SPOT.

Here’s where I see some upside, though:
Providing liquidity for SPOT creates direct value for the SPOT ecosystem. The early days of SPOT will be primarily about seeding deep liquidity to increase availability and allow the protocol to set a track record of performance. AMPL / SPOT liquidity is essentially “free” to create for an AMPL holder, because the hardest part of seeding liquidity is finding the pair token (USDC, ETH, etc.) SPOT and AMPL both come from the treasury and the treasury would still own all the assets in the pools.

The LP position could earn trading fees for the treasury, depending on how the pool is configured. However, I can see a compelling argument for a no-fee pool for the most efficient market possible at launch, when we care most about growth.

Assume also that this isn’t the only SPOT pool. I also think it’s prudent to encourage a deep pool of SPOT/${stablecoin} on Uniswap v3 through other means where we could leverage concentrated liquidity. So SPOT/AMPL is just one piece of a puzzle.

A higher level perspective about using AMPL in the DAO treasury for liquidity is that this represents a slight shift away from liquidity mining and towards DAO-owned liquidity. The AMPL owned by the DAO was originally earmarked for liquidity purposes. If nothing else changes and we continue with the current approach, then all of the AMPL will be given away through geysers over the remainder of the 10 year program.

This seems like an easy opportunity for the DAO to provide long term liquidity for AMPL/SPOT while maintaining complete ownership. So I think this proposal has significant upside, and the DAO could always unwind the LP positions, if it ever decided it needed to.


Thanks for replying. I agree with all your main points. Here are my thoughts.

It’s interesting to consider 2 ways to use X amount of DAO-owned AMPL:

  1. Geyser rewards for SPOT/AMPL

  2. Conversion into SPOT/AMPL

Once rewards stop there is no guarantee the liquidity will stay deep. So I think it’s wise to convert DAO-owned AMPL into SPOT/AMPL.

I think the DAO should not run a geyser yet. It would be nice to know if AMPL holders provide SPOT/AMPL willingly without incentives. Maybe run a small geyser to overcome friction. If the pool is deep enough, the DAO can provide less SPOT/AMPL.

Over medium time horizons (not long enough for inflation to push up the target price), AMPL looks like a stablecoin. So other than SPOT/stable, the DAO should consider also providing concentrated SPOT/AMPL in the medium term.


If we take this approach, then we should include a trading fee for LPs in the ES pools. This seems like a good option to me as well.

1 Like

A fee would be great. What fee % would we be thinking? Something comparable to Uniswaps DAI/USD at 0.01% - 0.03%! Or would be shooting for something closer to 0.3% with the Uniswap DAI/ETH pool.

I did not do the math, but I believe the downside of Z tranche + SPOT/AMPL is greater than the downside of just holding AMPL. What happens when AMPL market cap decreases?

  • In the absence of other trading activity, arbitrageurs force SPOT/AMPL LP to sell SPOT for AMPL, increasing exposure to AMPL. As AMPL market cap decreases 1%, 2%, 3%, the LP increases exposure to downside along the way. This is the usual IL curve
  • Remember Z tranche? At a SPOT:Z ratio of 50:50 the underlying value of Z decreases 2x. So the IL curve of the entire Z + SPOT/AMPL position should be steeper than usual

The combination of geyser rewards (if any) + fees should at least be high enough to offset a steeper IL curve.

I think you might have a misunderstanding of Buttonwood Tranches(docs) or Elastic Swap(docs).

For ES, when AMPL rebases downwards ES doesn’t force sell SPOT the way Uni would because it accounts for the rebase and rebalances the pools. The price before and after rebase stays the same.

For Z tranches, imagine tranching AMPL into As+Zs. If you hold them both through the lifetime of that bond, it has the same AMPL exposure as just holding the original AMPL. Now imagine you swap those As for SPOT. SPOT is not exactly the same as As, but it’s meant to be very similar–like a fungible version of As. The only change to AMPL exposure happens at the end of that particular bond’s lifetime after the tranches mature, because SPOT would not rebase the way the AMPL from the matured As would.

So I believe these are true:

  • ES removes almost all significant sources of IL from being an LP on AMPL/SPOT
  • Keeping SPOT longterm (in the LP pool or otherwise) does decreases the treasury’s exposure to AMPL slightly

Great question. @Loki_VT does the ES team have any recommendations?

I think AMPL/SPOT (on ES) is much closer to a stable/stable pool than a floating-price-token pool.

Thanks, I had another look through the docs and your writeup on the original AMPL/USDC Balancer smart pool where you explained how rebases cause IL. Yes, ElasticSwap does not have IL from rebases.

And a long term LP like the DAO should not care about IL from the price ratio, SPOT’s long term price should be stable at 1 AMPL.

Since the pool is not deep (the proposed 10% - 20% of treasury is just 400K - 800K) and not concentrated, big traders will incur slippage losses. I don’t see a problem with charging a relatively small 0.5% or even 1% fee to incentivize market participants to deepen the LP. If the pool becomes much deeper (through more LPers or through positive rebases), we can consider decreasing to the stable 0.05% fee.

I once again ask the DAO to consider providing concentrated SPOT/AMPL liquidity, so that traders incur less slippage and we can deploy with a 0.05% fee.

Currently 25bps go to the LP. It gets complicated having different fees for different pools, but this is a lever we can pull if the DAO felt strongly about a different fee.

I support this proposal.

ElasticSwap works seamlessly on Avalanche, so great job by the team. I suggest preparations for porting ES to Arbitrum (and ZkSync main-net when live) as well, to help circumvent MEV and centralization vectors on Avalanche. AMPL-SPOT should remain chain agnostic, but it is favorable to prioritize decentralization in these early stages.

1 Like

Hey guys, I have put up a Snapshot vote related to this discussion thread! Appreciate the support!