YIP-52: Make Strategist Skin in Game Partner for Make Benefit of Glorious Brain of Yearn

This is a very good point. Wall Street has learned this the hard way. Traders and portfolio managers without lockups or clawbacks are incentivized to take higher risks and blow up as they are playing with other people’s money. If they win, they win big, if they lose, it’s someone else’s money and they can just walk away from the losing strategy and create a new one. I think strategists should be very well rewarded but only when long term results are aligned and not reward high risk taking for potential short term reward.
Also keep in mind that the largest group of investors long term will not be developers or YFI holders, it will be retail, and retail will deposit their money where they get the highest risk adjusted returns. When high fees are subtracted from the strategy return, Yearn might not be able to compete with other returns offered by other protocols in the space.

3 Likes

Happy to support this proposal, the current setup by no means gives adequate reward to the strategists. However, I’d like to add three caveats:

  1. It does feel like we’re being pushed into an early vote on this as per @xgambitox’s comment. There are some legitimate differences of opinion here around exactly how to achieve the aim of increasting strategist reward. The original mega-discussion was covering too many points; an interim discussion thread on each topic prior to a proposal would, IMO, have been a better option in hindsight.

  2. Related, I’m going to bang the drum once more for including a % of the management fee (effectively a reward for attracting AUM) as part of the strategist reward. I don’t believe the commets in the proposal adequately address this. Regarding predictable income for the protocol, sharing a small % of the management fee does not impact this at all (it’s just a slightly smaller predictable income). Regarding “survival of the fittest”, the proproal assumes that the only thing that matters to vaults is performance and not, for example, risk management. We’re over-rewarding risk with this proposal.

  3. Lastly, I’d strongly support some form of lock up and/or partial reward in YFI as mentioned by various other posters. This is a better alignment of incentives for the long-term.

None of the above negates the fact that this is a good proposal that I would like to see pass. But I do think this is a good, not a great, solution to the problem.

7 Likes

Great breakdown @banteg @lehnburg @milkyklim

Code is highly fungible, strategies are highly fungible, talent isn’t!!!

Code is easy to copy and paste, strategies are trivial to replicate, but there is a finite supply of Talent, Courage, and Awareness at any given moment in time. Yearn should try to suck as much as that finite supply as possible, like Bitcoin sucking in hash rate. This would put it on the right track. A decentralized Citadel or Bridgewater only survives by attracting The BEST.

Reward the talent early and often, and the early yearn team and early adopters will reap more rewards than they could ever imagine just by being slightly less greedy and slightly more giving. What a wonderful thing. Peace and love, y’all. Great work everyone.

4 Likes

This is a fantastic point, and something worth diving into here.


To suss this out a bit more, the question here is “Who is in charge of Risk Management? (of the Vault)”

I personally do not think Strategists should be in charge of risk management, at least not the most responsible party for maintaining Vaults at an adequate risk level. Of course, they should be able to communicate risks well to the responsible party, and also be considered the “domain expert” in their own strategies that they’ve developed, but to put them even partially in charge of the overall Vault seems to be an expansion of their duties and responsibilities. Hence, it doesn’t really make sense for Strategists to be compensated for properly managing risk (of the Vault), or have their incentives aligned in a way that negatively impacts them if the Vault’s risk management were screwed up.

I think this responsibility of proper risk management should be clearly and squarely on the shoulders of Governance instead.

The incentives are aligned in this direction: Half of the earnings for proper Governance is based on the management fee, which rewards larger “assets under management”. What are the reasons why this number would be high? I think the biggest reason is being an attractive and secure place for user’s funds. This is a balance of proper risk management and earning high-enough returns that will attract large amounts of capital where this fee is rewarding to Governance participants. Hence the incentives are balanced between management and performance through the split of the corresponding fees.

The mechanisms of control are also aligned in this direction: Governance (and only Governance) can control: 1) adding new strategies 2) changing parameters on current strategies (debt limits, etc.) and 3) changing Vault parameters (deposit limit, etc.). There is a subset of responsibilities delegated to the Guardian role, but that role is only able to do actions relating to Vault safety e.g. revoking strategies and shutting down the Vault. “Performance” and “Risk” are not a part of that role’s responsibilities. There is no such thing as a free lunch!

Streamlining the responsibility split between Governance and Strategists just makes it easier to know who has to do the work to improve things, and where the design of the overall system is deficient. The incentives are aligned with these behaviors, hopefully encouraging responsible action. Also, keep in mind that these roles are fluid. Strategists can be Governance participants, and vice versa. Reputation is totally at play here too: Strategists who participate in Governance and add to the discussion are going to be more likely to have their strategies approved. Many Strategists will stake YFI and earn rewards. These roles are going to naturally blend together, but for the sake of the protocol we should keep it clear what the individual responsibilities of these roles are.

The negative space in all of these proposals (which are still be worked on and further defined) is that Yearn’s Governance has responsibilities to Yearn, and are paid handsomely for it. I think it’s important to keep this in mind in these discussions.


NOTE: As a reminder, “Governance” is currently automated by the multi-sig, which had a temporary authorization to manage the system until we were able to transition to a more effective and scalable Governance system that streamlines these actions and responsibilities.

7 Likes

Thanks for the comprehensive reply – I hadn’t considered the risk responsibility quesiton from this perspective previously. I think this is another symptom of the trickle of proposals that are all interrelated.

I would agree that, as and when Governance takes on full risk management, there is less of an argument for splitting some of the management fee with the Strategists – as you say, Governance is responsible for risk management, they therefore get the benefit of executing on this reponsibility effectively.

I still think, however, that this doesn’t deal with the issue completely.

Governance is analogous to the executive board in a real-world company. From stragey and planning throguh to product (Vault) offering, resource allocation, setting the right incentives and risk mitigation. They are rewarded for doing this job successfully.

So, let’s say that one day Governance “decide” that the stragey for the entity over the coming months is to attract AUM and, looking at the competitive landscape, the belief is that the best plan for achiving this is to offer a series of wBTC and ETH vaults that are low yielding but have broad appeal due to, say, their built-in contract risk cover. Over to the Stragegists… except now there’s a problem. Strategists could work on designing and implemening such complex bu low yeilding vaults for Yearn, but the reward is, by request, going to be low. Maybe they go with it anyway – small performance fee on a large vault is, after all, still a descent number. But maybe they don’t…

Now, I know I’m taking this to the extreme and that in reality a the fluid roles of various individuals would probably mean these vaults would be built. I also know that what we’re voting on here is really just a stopgap solution – more flexibility (e.g. variable fees for vaults) will definitely be needed in the future. I just don’t want to set a prescedent that performance fees are the only way that Governance will be able to attract Strategist talent for ever more – I think that would be tying Governance’s hands in the long run.

1 Like

Voting has passed with 83.76% For.

1 Like