A huge thank you to @puniaviision for helping me to flesh this out and his analysis on minting vs exchange rates.
Background
A core problem for the Index Coop is the necessity to target and maintain a liquidity threshold for its Indices on AMM(s) such that sufficiently large trades can be made with low slippage.
There is competition for the capital of liquidity suppliers, and thus, many attractive (risk-adjusted) yield opportunities. The yield from supplying liquidity in order to facilitate volume for the Coop’s DPI product on Uniswap is not attractive enough, alone, to acquire sufficient liquidity.
In order to remedy this, the Coop has been subsidizing liquidity providers for the last 60+ days to the tune of a cumulative 10%+ of the $INDEX total supply. This type of program is fundamentally unsustainable, as the treasury would quickly burn through its reserve. Additionally, this program exerts downward pricing pressure on the $INDEX token, and thus the purchasing power of the treasury.
Solution
One patchwork solution is to funnel larger purchasers of DPI into an issue/mint flow. Fundamentally, the buyer’s experience would be to take some amount of ETH (or USDC, wBTC, etc, ideally) and purchase newly minted DPI shares.
On the backend, an exchange on Uni for the DPI underlying tokens would take place, followed by a mint via the Set protocol. The user would receive DPI in their wallet and be none the wiser as to the backend process. Ideally, we could rebate the cost of gas.
The benefit of this flow is it doesn’t run into the same liquidity issues as DPI, because it is able to leverage the (typically) liquid pools of the underlying tokens.
Running an analysis on the effective slippage differences between each method, we’d really only need to incentivize at most a $10M liquidity pool in Uniswap, which is a third of what we are incentivizing presently. Check out the analysis here: Issue Vs. Gas Cost - Google Sheets
A third of the $INDEX token costs affords us the ability to liquidity mine every new “basket of tokens” product fairly and thus gives an even playing field for any Methodologist reward program we may wish to run. It also allows other products to compete fairly with the DPI in terms of the demands for their methodologies.
Risks and Mitigations
Decreased Governance Power
One of the reasons why we want to keep a fairly large liquidity pool on Uniswap is that a large AUM, incentivized or not, gives the Index Cooperative governance power over other protocols. However, as it stands right now, the benefit of governance power is not nearly high enough to warrant distribution of 1% of the total Index supply every month.
A lot of the meta-governance is a very powerful marketing tool which exists even with a $10M incentivized pool. Furthermore, meta-governance is a long-term moat that we as the community hope to have, and so we should have a long term plan for developing it as opposed to relying on liquidity rewards.
Decreased Liquidity for Integration Partners
Another risk of decreasing liquidity is decreasing the probability of being on-boarded onto Aave, Maker, or Compound in the event that the DPI needs to be liquidated. Currently, the liquidity of the token is a major factor used in determining the suitability of a piece of asset as being added as collateral onto each money market protocol.
As we had outlined in the original Aave proposal, while selling the asset directly on Uniswap is an easy way to liquidate it, fundamentally it is a basket of other tokens that can be redeemed and sold off in one block. We see this happen every day with the arb transactions that keep the price at NAV. Decreasing the liquidity of the DPI:ETH Uniswap pool should not heavily impact the true liquidity analysis of the DPI.
Awareness
Right now a lot of users are simply super familiar with interacting with Uniswap for buying and selling assets. Some large buyers may be discouraged by seeing the low liquidity numbers on Uniswap and not be aware that they can buy directly for lower slippage on the Index Coop website.
This is a valid concern. However, looking at the transaction data on Uniswap, most are very much below the $50K transaction size amount that is being optimized for in a $10M pool. Whales that do larger purchases are likely to be more informed about the product and ecosystem and would want to use the lower gas cost issue option. Fundamentally, the cost of marketing and awareness is much lower than what we are paying out in liquidity mining programs currently.
Next Steps
While this gives us a virtuous effect of being able to launch more products without spending too much, which would lead to a rise in token price, which would lead to another decrease in the amount of emissions, this still isn’t a long term enough plan.
We need to fundamentally figure out a program by which we incrementally decrease and retire the liquidity mining incentives and let liquidity exist on its own merit and incentives.
Finally, we need the developer team to comment on whether or not this is technically feasible and how much effort it would take.
Something else we could do that is significantly less promising, but worth considering, is a Balancer pool that we weight 80:20 DPI:ETH which lets people enter easily, but it won’t be as easy to exit.