[AIP-32] TOKE WAR (Bonds)

scoopy

maaaaybe it would be cool if TOKE bonds rewarded you with tALCX?????

mastermojo83

scoopy I like exploring this idea!

0xFelix

We already discussed it earlier, but I just wanted to also publicly voice my support for this proposal. In the end, it's a bit of a risk since the reactors are not live and we don't know if the product will work as intended. However, this is - in my opinion - a risk worth taking since if Tokemak succeeds, Alchemix will be in a very comfortable position to direct liquidity to all the necessary pools.

I do think we should be a bit more conservative with proposals A) and B) and avoid doing another AIP-20 (as listed under other options).

My suggestion would be to tweak the numbers as follows:

TOKE bonds @ 5% of emissions (taken from the SLP)
Stake max 20% of the treasury in the ALCX reactor

barree

Sounds really great, if Tokemak will be successful, then heavily investing in the project is by far the best decision as it really does wonders in terms of liquidity cost for protocols.

Things that I would add:

You mentioned how much the toke rewards for the LP side of the tALCX reactor would decrease if the treasury ALCX was deployed there. This is true, however by deploying this amount of tokens there, it pushes the rewards up on the LD side as well, which means people will allocate their votes more to the tALCX reactor, taking the LP side of rewards back to previous levels (minus the total system-wide dilution of toke emissions that this action would cause). This makes this proposal even better IMO, as it has relatively minimal downside for the tALCX stakers (just the system-wide dilution of toke rewards, which I suppose would be no more than 1-2%).
As for the other proposal, it might be prudent not to reduce the ALCX/ETH SLP pool any further, before other liquidity boosts activate (including mainly tokemak), to keep the ALCX/ETH liquidity at least level in the meantime and not have it drop. This assumes that the ALCX/ETH pool is deemed to have the optimal size right now. If that is not the case, and the thinking is that it makes little difference if the ALCX/ETH pool gets a bit smaller, then by all means, go for the full 7.5% toke bond allocation as soon as possible.
Alternatively, the DAI bonds could be partially reallocated to toke (?), as there is a lot in the transmuter as it is, which means whatever DAI is bonded is a relatively miniscule part of the total DAI supply in the transmuter+treasury. (Or the treasury wants DAI and does not want to take any out of the transmuter?)

mastermojo83

scoopy Adjustments look good to me.

IMIMIM

dixie

What are the risks of NOT implementing said proposal, and how do the risk/reward profiles of doing the thing and not doing the thing compare?

ov3rkoalafied

IMIMIM The risk is potential future scaling of the alchemix platform. If TOKE works it gives us access to a lot of future liquidity indefinitely, which is imperative in having large alETH,alUSD,alBTC debt caps.

ov3rkoalafied

A few thoughts:

Bonding for tALCX sounded cool, but the numbers are kinda meh. 7.5% of emissions x 15k tokens/week at 27% APR would only net the DAO about 35 toke/week at current prices. Not sure that's worth the added friction.
We are discussing allocating 7.5% of emissions to TOKE bonding - but what if we just gave those to tALCX stakers instead? With current prices, bonding would net about 6.6k TOKE/week. Taking those 7.5% emissions to incentivize stakers and up the TOKE pool TVL would net about 6.7k TOKE/week. Ie, bonds tALCX could be better short and long term!

We'lll need to run the numbers as that DOES NOT account for decreased APR from having a higher TVL on the LP side. But I think the fundamentals all favor just doing more tALCX rather than setting up bonds:

As barree mentioned market forces should help offset that imbalance by attracting more LDs
Bonding tALCX creates more ALCX dumping. We still have plenty of emissions going to mercenary liquidity, and they all have to make the decision to stake or dump. A stronger tALCX pool will mean more folks will stake rather than dump, which creates a stronger ALCX token (which creates more efficient emissions, etc, etc).
While I didn't account for dampened APR on the bond side, I also didn't account for ALCX being sold at a discount and giving a 3.3% fee to OlympusDAO on the bond side.

So it would take a pretty big dilution effect (ie, cut in APR) to convince me that we should do bonding isntead of just more tALCX.

So in my opinion: 20% DAO ALCX to tALCX, no toke bonds, take 2.5% emissions from CVX bonds and 5% from SLP and put those into tALCX.

lazyguac

I dont think it really makes sense to cut 2.5% from cvx bonds and allocate it to accumulating toke regardless of the method. CVX has proven to be effective while TOKE hasnt yet.

Besides that, I like the proposal and discussions. Personally, I like the idea of parking 20% of the treasury and also moving 5% from slp to the tALCX pool.

ethacct

TOKE and ALCX are literally my 2 biggest bags outside of ETH, and for this very reason. I disagree with the 'wait and see' approach because by the time anything has proven itself, you've lost a ton of upside. Makes sense to get in early while it's cheap, in order to capture the advantage going forward. I'd default to Scoopy's knowledge for the specific numbers-tinkering, but support both proposals tremendously.

dixie

OK, so there are a few points to respond to here. Thanks to everyone for the comments, feedback and suggestions. I'll do my best to keep this response focused 🙂

There seems to be general consensus (amongst the small group here, at least) on the following:

Allocate 20% of ALCX Treasury to the ALCX Token Reactor (36k ALCX => $10.5M @ $292/ALCX).
I estimate this will deliver an additional 124 TOKE in the next cycle, corresponding to a boost of 18% on the cumulative harvest for Alchemix.
Reallocate a maximum of 5 points from the ALCX/ETH SLP emissions (i.e. go no lower than 20%).
Consider reallocating some points from the DAI bonds (1 - 2 points?).

Emitting tALCX instead of ALCX
There have been additional proposals regarding the use of tALCX as an Alchemix emissions token. For example, for farming emissions, Alchemix would stake the ALCX to be emitted in the ALCX Token Reactor, and would instead issue the derivative tALCX tokens. This would then allow Alchemix to harvest additional TOKE, although it's difficult to model because of the Tokemak cycle-based rewards system.

Relating specifically to bonds, I can see that the Tokemak staking model would be a disincentive for cyclical bonders - both due to the additional gas cost to withdraw (request to withdraw, withdraw), and also because of having to wait for the cycle to complete after requesting to withdraw. This makes it much more difficult to be efficient as a "professional" bonder, and my take on these types of participants is that they should help to eat away the economic profits in bond discounts (i.e. so it settles around the 2 - 4% mark).

I can however see that tALCX may be useful in non-bonding scenarios, so would really like to explore that a little bit more...

Questions

What is the appetite for reducing the DAI allocation (5% -> 3%) in order to redirect emissions to some form of TOKE collection (Bonds or other) ?
Are there any other creative options for Alchemix to collect TOKE? (i.e. not mentioned here)

Summary
Based on all of the above, I will clarify in the initial post that there is an amendment to Proposal B), such that 20% of treasury is proposed instead of 40%. It seems like we still have some distance to go on whether to bond for TOKE or not.

ButlerAndTheThirdStringers

dixie I personally would rather take from ETH bonds than DAI bonds for a few reasons:

(1) As Scoopy has pointed out before, our liquidity for alETH is actually very strong right now. We don’t really have any difficulties on that front atm
(2) Users tend to care much less if alETH is trading at a slight discount to the value of ETH compared to if alUSD is trading at even a discount of a few cents against DAI or USDC.
(3) There are more USD derivatives that will likely be added as collateral in V2 than are ETH derivatives, which means at that time there will be a larger chance of low liquidity against any one stable, since that the total liquidity against all stables will likely be split across all stable pools. Best to continue to focus on accumulating DAI bonds in order to front run this possible pain point

dixie

OK, so after some additional discussion in the Discord, here is an adjusted proposal:

Alchemix to stake ₂₂% of current treasury in the ALCX Token Reactor (69.420k ALCX).
New TOKE Bond @ 2.5% of weekly emissions.
Boost tALCX emissions from 10% to 15%.
wETH/DAI bonds to remain as-is @ 5% each.

Emissions are to be drawn from:
5 points ALCX/ETH SLP emissions (25% -> 20%)
2 points CVX bonds (20% -> 17.5%)

All other emissions to remain unchanged.

Notes:

ALCX LP on Tokemak
When I prepared this proposal, the ALCX in the treasury was 180k ALCX. It's now over 300k ALCX due to a delay in harvesting ALCX for supply of the bond contracts, so 20% => 60k ALCX. It's so close to 69.42k that I've bumped it up to that, so we can sail on the power of memes. We've had some back and forth discussions about whether it's responsible to allocate this much to Tokemak. I'm in favour of this allocation and I would argue that it's easy to reduce this amount via an AIP if the situation changes.

tALCX Emissions
tALCX emission adjustments have not been discussed yet, but a deep dive with @ov3rkoalafied and @barree in the Discord has indicated that it probably makes sense anyway. The justification for increasing tALCX emissions is that we can see there is a decent chance of capturing some low hanging fruit (tALCX holders who do not stake in Alchemix). These holders attract over 50% of the TOKE emissions to the LP in the ALCX reactor, and it delivers the most benefit to Alchemix to attract them across. Additionally, boosting the tALCX emissions by 50% (to 15% total) rewards ALCX/tALCX holders, and does not incur the same dumping effect that cyclical bonding would.

TOKE Bond
Also a comment on the TOKE bond @ 2.5% - this might seem low, but it delivers some immediate benefit in TOKE collections. I see it as a positive that the DAO can choose to readjust the emissions allocation afterwards in response to feedback/observations about the performance.

hekmat

Saw this late but to leave my thoughts for posterity:

I'm in favor of the small allocation in bonding to TOKE. Once they prove themselves as a viable product we can look into increasing it. The addition of increased tACLX emissions is a nice touch also.

I would personally have liked a reduction in the DAI bonds to fund it but I see the logic of using the ALCX/ETH SLP instead. In general I don't love DAI bonds having equal weight of their ETH equivalent since...fiat. Literally everyone here expects ETH to be more useful than DAI in the long-term so the weighting should reflect that IMO.

Still that's another discussion entirely and looking forward to seeing green text Dix. Nice job man 😃