yIndex development?

[Niffler]

Sorry if I missed this but has there been much discussion on yIndexes? In a world where passive investment is highly sought after I believe index strategies could be super interesting to develop.

Further, a strategy that provides long exposure to a sector of cryptoassets (e.g. DeFi, large caps, NFTs, etc) and provides automated rebalancing following a strict methodology could be a very attractive product.

Another cool thought for indexes is that the yearn ecosystem is uniquely positioned to provide the highest benefits and returns to passive index strategies. Let’s think about a DeFi index for example, where YFI holdings go into the yYFI vault, LINK holdings into the yLINK vault, SNX holdings into the upcoming ySNX vault, etc. By doing so, not only can indexes track the performance of the underlying, but provide a yield ontop of it!

[yEtienne]

Hey,

I’m super bullish on Indexes as well. What DeFi Pulse and TokenSet have made with $INDEX is awesome, and I believe we can leverage Yearn success to launch a similar product. I will support this initiative .

Thanks

[DeFiChad]

Since there are serious projects in the space already with working products we need a twist in this to make it take off. I really like your idea with getting yield on the assets as additional return because that is yearn´s core expertise. If we construct an index made solely out of yield producing assets that would probably crush it so hard.

Combined with automatic rebalancing according to market cap every so often (like once daily) it would imitate industry standard ETFs like SPY. In this context I suggest that the index vault actually pays out the yields like dividends to the investors on a regular basis (e.g. at rebalancing). That would be the perfect analogy to giants like SPY and especially appealing to rent seekers.

Let´s build the yYIELD index.

Edit: I know I would invest myself. What a simple yet briliant idea, thanks @Niffler

[ejbaraza]

This is something to develop for sure, but I think we are still waiting on StableCredit, synthetic rebased dollars, and single-sided AMM. I think that an index that tracked the performance of yETH, yWBTC, ySNX, yLINK, etc. with a yUSD dividend would be truly innovative. It would have a superior yield and plenty of diversification. It is hard to ideate about such things when there is little in the way of progress reports or timelines so everything is kind of on a wait and see approach. Ultimately, this is the direction DeFi is going.

[Ceazor]

Both INDEX and PIEDAO have appealing products. I have begun to farm one of those token myself. But I think both protocol s have finite farming so this is not really a good idea for yearn type vaults. They are also using uniswap and balancers LPs which exposes one to ImpLoss.

However, I think a sort of mixed bag vault where people deposit an array of assets like the DefiPulse 9 (the ones that are collaterable) which are then used to borrow DAI or what have you to run the usual yCRV (or 3pool) strategy might be appealing the masses and get yearn in on the indices hype.

[Ceazor]

I like your thinking DefiChad. One vault to rule them all, and in the glory, bind them.

[Niffler]

I think daily rebalancing might result in too much noise and costs associated with index turnover and withdrawing from vaults, but here is basically what my high level thinking is as of now. Design a methodology that:

• Design an set of eligibility criteria for the index to select assets that meet minimum criteria such as liquidity, size, availability, etc
• Select index constituents that are ERC-20 (including synthetic tokens (partner with an exchange like Synthetix on this) if they meet other eligibility)
• Weight the index at the start of every month based on their circulating supply
• Deposit the assets into yVaults at the start of every month (if a yVault for that asset exists), otherwise just hold the asset passively (note if an asset like MKR has enough weight in the index maybe this would incentivize a MKR strategy that lends MKR on Aave to withdraw stables and put in a stable vault?)
• Harvest the yield attributed to the index allocation in the yVault monthly and re-invest in the yIndex in proportion to a users holdings (or pay a dividend in yUSD)
• Rebalance index assets at the end of each month

[mattdw]

Why am I just seeing this for the first time?

I’ve been working on coding up a vault strategy that involves depositing into other vaults to use their yields as profit. This is pretty similar idea.

IMO the yIndex should only take in a single deposit token, say ETH or stablecoin or whatever, and it should be extremely well defined as having capital at risk.

Trade that deposit (eg ETH) into the basket of chosen tokens, and then send those tokens to do yield things. I would recommend keeping the list of tokens smallish at first, say 10-15 maximum. The more tokens involved, the more complicated the code would get. 10 is a good number to start with.

Weight via square root market cap method (my personal favorite), rebalance the fund once a month in terms of weighting.

Seems doable!

[byolo]

I support this. The twist I want to see is that the index funds can then be placed into its component vaults. Say the index contains YFI. Some of the YFI should be staked for more yield.

[Niffler]

@mattdw

IMO the yIndex should only take in a single deposit token

I agree on this point. I was going to even suggest stables. Much in the same sense that one would deposit USD into equity/commodity markets, so too could they here in crypto markets. Further, people might have a tendency to not want to sell ETH into other crypto, where-as USD should not have this issue.

smallish at first, say 10-15 maximum.

Agree on this too, even less perhaps. IMO given the audience of vaults at the minute, it would likely need to be an ‘exciting’ index, like DeFi or ERC-20. Ideally at least 50% of assets ideally have vaults so the underlying can be deployed and put to work too.

Weight via square root market cap method

Another fun and unique way to weight (I have not backtested yet) could be to do so using TVL. That way you are capturing and reflecting the utility of the protocol (and potentially its long term viability). The only issue with this of course is that underlying asset price does not necessarily correlate to the TVL

[mattdw]

@Niffler

Ideally at least 50% of assets ideally have vaults so the underlying can be deployed and put to work too.

Honestly, I think this is the rate limiting step, here. Before creating the yIndex (as I’m imagining it), we should probably create vaults for each of the underlying assets. That way the funds in the yIndex can be placed into those Vaults, and then the strategies of those v2 vaults take over in regards to their deployment.

Alternatively, we can just create an Index comprised of all the vaults we currently have, though in my opinion that almost sounds like a different project, a yyVault or something.

[Niffler]

Completely agree on this. The way I am imagining it a yIndex would be most impactful once there are at least (a few) the following vaults: SNX, ETH, Aave, CRV, wBTC, YFI, COMP

Thankfully with v2 it sounds like we might be getting there though!

[mattdw]

After thinking about this whole system some more, I think this can be an entirely separate product offering under the YFI umbrella. I started thinking about it as we’re discussing copying mutual fund fee structures… so why not offer similar service?

For example, some basic starting products could be:

• WBTC/ETH
• DeFi basket
• equal weight of every vault
• etc

Creating an equal-weight share would be the simplest way to get started. With the V2 vault structure coming out, this will be a natural product offering.

I think there’s a lot of promise here, and I really want to see this deployed. I’ll keep working on it in the background while finishing up my current projects.

[Niffler]

I’m very down for this. Eager to see what vaults are released once v2 is up and running and I guess that will help with designing what cool products we could consider.

• wBTC/ETH definitely makes sense as many are very happy to passively hold these assets long term.
• Defi vault is sexy exotic and high yielding and we are all largely bullish defi if we are here
• Even a stable vault could be cool (deploy into DAI strats/USDT strats/USDC strats evenly)

I think even weight is the easiest to do for sure, but might get tricky depending on the liquidity of assets within the vaults (e.g. if a defi basket has majors like SNX/YFI but also allocations in small cap vaults like Hegic)

[mattdw]

Another thing I was thinking about is that stacking vaults may be double-dipping on fees. We may need to come up with a separate fee structure for these style of “vaults.” May be tough to iron out while the yVault fee structure is still in flux, but I’d imagine a “meta-vault” style strategy do be cheaper than the core vault strategy. Maybe a 2% maintenance, 5-10% performance.

[Niffler]

Good point re fees, I think there would need to be a very clean structure here as well. Maybe just a 2% management fee plus maybe something on assets not allocated to vaults. As for assets allocated to vaults (ideally the bulk of them), yIndex strategists might be considered ‘contributors’ to those vaults and share their vault fees rather than create more fees?