Proposal Draft for Lido to buy CVX

Hi, I’m Ouroboros Capital, most commonly known for my Twitter profile. I am a holder of LDO tokens and have briefly tweeted about how I think Lido can benefit from an investment in CVX but would like to elaborate here. I would like to propose for Lido to consider a plan to purchase CVX such that it can solve for the currently unsustainable emissions of LDO tokens into stETH liquidity pools. Disclaimer: I am also a holder of CVX.

Why not AURA?

Before we start, I’d like to preface by saying that given AURA’s still inflationary nature which, as a result, leads to the unpredictability of vlAURA’s future APR, I’ll be focusing on CVX. Additionally, AURA’s lack of liquidity – trades $350K/day on Balancer, means its less pragmatic at this point for it to be considered as a solution for size of LDO’s emission woes. This compares to vs CVX which i) trades $3.5mn/day on Binance, ii) trades $2.8mn/day on Curve and iii) and has existing holders such as Terra (~$10mn CVX holder) which may look to get out via an OTC trade.

The Current Situation – stETH deep liquidity is important but Lido’s current incentives can only last 4 years.

Lido’s key value proposition is the readily available on-chain liquidity on stETH that it provides. However, LDO currently spends ~4.5mn LDO (~$6.3mn) in emissions monthly/54mn LDO ($75.6mn) in emissions annually to sustain these liquidity pools. While it is paramount that LDO maintains its core value proposition of having deep on-chain liquidity venues for stETH, it needs to work towards a sustainable solution for its stETH on-chain liquidity. LDO treasury only has ~$260mn (mostly LDO with some stETH, DAI and ETH) which means at the current run rate, it would last less than 4 years before running out of incentives.

ReWARDS Team is looking this, but we need to accelerate the process as its an opportune time to accumulate CVX.

While the rewards committee has acknowledged “the need to optimize the current program” and has put out a detailed post on its consideration when it comes to incentives optimization, I’m hoping that this post would provide information that would augment as well as accelerate that process given that CVX is currently trading below its intrinsic value – in other words, an opportune window presents itself for Lido to accumulate CVX.

Own CVX as a sustainable solution to the emission issue

The rewards team has already covered in Update on Rewards, a good introduction on how owning CVX would be an investment that would pay off over time (through wielding $CVX/$CRV emissions as as compared to the DAO spending its precious $LDO tokens). The reWARDS committee has also elaborated that the “next steps here would be to discount that back (accounting for opportunity cost) + account for price risk of being effectively long that position as a DAO”. I’m hoping to provide points that would aid that analysis.

2 Key Considerations when owning CVX vs bribing/paying own emissions.

  1. Payback Period – How long for the investment in $CVX to pay off in terms of directable emissions.

  2. $CVX/$CRV price relative to $LDO price and the outlook for that – if $CVX/CRV price is expected to rise vs $LDO then the investment in $CVX vs paying $LDO as reward incentives would be much more effective.

How long does it take for $CVX to payback?

According to Llama Airforce, $ value of emission that $1 of CVX wields (based on the latest round of bribes) is as follows:

$1 of CVX wields 39c of CRV+CVX emissions (annualized) – vlCVX APR of 31% * $1.25 emission per $1 of bribe. In other words, a payback period of ~2.5 years (closer to 3.5 years post adjusting for the declining CVX emissions in the next few years).

How does the outlook of $CRV/$CVX price looks compared to $LDO.

Price = intersection of supply (inflation + sellers) and demand (buyers). The outlook of a token’s price is often an exercise involving pontification. However, I will attempt here to be as methodical as possible, comparing the supply and demand dynamics of the two tokens relative to each other.

On the supply side, I will compare both token’s inflation outlook relative to their liquidity profiles.

On the demand side, both token’s potential demand sources as well as imminent developments (narrative). And finally netting the two against both token’s liquidity.

TOKEN PRICE OUTLOOK - SUPPLY SIDE ANALYSIS

Inflation of CVX – CVX is close to the end of its emission to depositors on Convex. One year forward, there will only be ~9mn CVX ($44mn) emitted to LPs on CVX (mostly to LP token depositors on Convex as well as cvxCRV). Contrasting that to on-chain liquidity venues – selling $1mn of CVX on-chain would incur a 5.9% slippage. Furthermore, it is worth noting that the Frax-3CRV and Frax-BP both in totality ~20% of Curve weekly gauge weights and ~66% of the above-mentioned Frax pools are Frax protocol owned liquidity. Frax has committed in its governance to fully accumulate all CVX going into its protocol owned Frax pool, thereby absorbing a small % of these emissions.

Inflation of LDO – One year forward, based on the existing reWARDS budget, Lido would have spent ~54mn LDO ($75.6mn). Contrasting that to existing off-chain liquidity venues – selling $1mn of LDO on-chain would incur a 5.5% slippage.

Contrasting LDO and CVX’s inflation and comparing it to the current liquidity, Lido has inflation close to 2x the multitude of CVX with similar liquidity profiles and as such on the supply side more likely to experience price pressure. This is not yet considering the above-mentioned “Frax effect”.

TOKEN PRICE OUTLOOK - DEMAND SIDE ANALYSIS

  • CVX’s underlying asset CRV is on the cusp of releasing crvUSD which based on teasers so far suggest a new CDP revenue vertical and likely to lift the underlying value accrual of CRV and consequently CVX.

  • LDO as of now is largely a governance token with little value accrual (ie. no real yield) vs CVX which serves as an outsourced emission management mechanism for protocols is likely to see greater demand pockets.

  • Post the Merge narrative, there is a lack of impetus for speculators to buy LDO.

**In conclusion, combining both supply and demand analysis, CVX and CRV appears to have a higher likelihood of outperforming Lido, thereby positioning it as a favorable investment for the DAO vs spending Lido tokens as reward incentives. Worth noting too that the ratio of CVX and CRV token price vs LDO is at the low end of the range.

Flywheel from Investing in CVX and Reduced LDO Emissions

While there is unlikely sufficient liquidity in CVX and AURA for Lido to fully replace LDO emissions, the partial replacement of LDO emissions would still stem downward pressure in token price to some extent and as such boost $ value of each LDO token. As such, over and above creating a sustainable solution to incentivizing stETH liquidity pools via, the number of LDO emitted per month would also be indirectly reduced from an uplift in the token price.

CRV/CVX wars is protocol (3,3)

I’ve tweet about this before as well. The CRV/CVX wars are essentially (3,3) for protocols. More diamond handed protocols (eg. Frax) joining CVX/CRV cartel will essentially absorb circulating supply and lift price of CVX/CRV. In other words, being an early mover would see greater dividends being paid on the investment. Another point for consideration.

CONCLUSION

I am a holder of LDO as I see it as a great business - sitting on a lion share of the staked ETH market as well as being a core DeFi lego/infrastructure. That said, token price is being heavily pressured by rewards into stETH pools which I hope LDO can solve through my suggestion. I believe my suggestions have presented a good case as to why CVX/CRV are good investments presently relative to LDO price and Lido spending LDO tokens as incentives. I’ve presented many points here that I’m more than happy to discuss with the reWARDS team or even be part of the team to help push this through. Feel free to hit me up on Twitter.

3 Likes

Love this! I strongly support this proposal. Great write up @Ouroboros

Thanks for write up. I do agree with your analysis but I am not sure optimizing the LDO being spent is the right approach.

I would like to see LDO incentives reduced dramatically and this investment would essentially front load several years of expenses. As a result we would be unable to reduce further.

Also it’s worth noting the inefficiency of the curve pool and that it is over incentivized currently.

IMO there are better pools such as uni v3 that are more efficient.

4 Likes

The Current Situation – stETH deep liquidity is important but Lido’s current incentives can only last 4 years. Lido’s key value proposition is the readily available on-chain liquidity on stETH that it provides. However, LDO currently spends ~4.5mn LDO (~$6.3mn) in emissions monthly /54mn LDO ($75.6mn) in emissions annually to sustain these liquidity pools. While it is paramount that LDO maintains its core value proposition of having deep on-chain liquidity venues for stETH, it needs to work towards a sustainable solution for its stETH on-chain liquidity. LDO treasury only has ~$260mn (mostly LDO with some stETH, DAI and ETH) which means at the current run rate, it would last less than 4 years before running out of incentives.

Deep liquidity is only a must pre-withdrawals, there’s no need to budget for years of liquidity rewards. Once withdrawals are enabled, LDO emissions should be redirected to other initiatives.

1 Like

Even with withdrawals you’ll need deep stETH liquidity pools for Lido to be a superior staked ETH product. When a validator wants to exit, it needs to enter an exit queue. Its ~6 validators per epoch, ie. 6 * 32* $ETH per 6.4 mins (or $56mn of ETH per day).
image

Right now theres $18bn of ETH staked. I think its fair to say that in a risk-off event, its possible that we see > $1bn of staked ETH outflow. In other words, the queue will not be enough to process all outflow.

Holders can of course hedge via ETH perps but the staked ETH product that differentiates will still be the one with the most liquid venues for holders to get out. I agree that liquidity pools may not have to be as large post withdrawals as arbitrageurs will help the peg. Right now, stETH market cap is $5.7bn and existing stETH pools on Ethereum can only handle ~$0.75bn of stETH on a 5.95% depeg. Essentially only ~15% of stETH can rush for the door in a risk-off environment while some will take a 5% depeg. I am too of the view that stETH is not a product that guarantees no depeg and staking needs to come with the understanding that holders have to weather volatility. But from a product perspective the lion share will still sit with the staked ETH product with the most available liquidity for exit.

At 6% depeg, arbitrageurs can come in and defend the peg for a ~72% per annum arb opportunity (long stETH, short ETH perp) assuming they have to wait 1 month to exit. That said, as you can imagine, funding rates can be quite negative (< -100%) in a risk-off environment which means arbitrageurs would then not be a reliable defence for the peg.

As mentioned in the original proposal draft, Lido currently spends ~$72mn of $LDO tokens incentivizing the stETH pools. The call to action here is not for Lido to completely replace LDO incentives with CVX or CRV and embark on a nuclear purchase of CVX/CRV. There isn’t even enough CVX or CRV for that (~$150mn of CVX required to completely replace $LDO emissions, $CVX market cap is $350mn. The call to action instead is for Lido to start buying SOME CVX and CRV at predetermined limit price given CVX is trading below intrinsic value and CRV has crvUSD on the horizen. In other words, the relative price outlook of those two tokens looks better vs LDO, making them good investments for the DAO as an emission management tool.

image
Worth mentioning too that Frax has expressed an interest to do a tripool of frxETH-stETH-ETH with co-incentives. A tripool would mean larger TVL and naturally a more stable peg.

Also, Alunara the founder of Llama Airforce, confirming the CVX payback math.

I like it and strongly support this proposal !

Such a deal benefits only frxETH because stETH is already very liquid against ETH. In fact, it would directly boost a competitor’s liquidity while also exposing stETH to LP risk.

9 Likes

You mean pairing with frxETH right? I don’t disagree with that. And naturally any stETH-frxETH-ETH tripool will not be the main stETH liquidity. It would likely entail Frax contributing an equal amount of votes or bribes with Lido and the main stETH liquidity pool would be too huge for that. That said, stETH-frxETH-ETH tripool as a side pool is still a good idea since it would involve 2/3 of the pool being emissions/bribe contributors (frxETH and stETH) vs just stETH vs ETH (50% of the pool).

All that said, believe the MAIN discussion here is the decision to purchase CVX.

reWARDS member here, wanted to chime in a bit. We’ve discussed CVX/CRV purchase couple times through the year, and figured that estimated payback period of year+ (your analysis at shows 2.5y+ at this time) doesn’t sound like a good deal for our timelines / focus. There’s a decent chance we’re be exploring new pools with different mechanics (i.e. UniV3 & other “concentrated liquidity” AMMs), where CVX voting power won’t yield us benefits.

3 Likes

Hey! I’m a reWARDS member as well and wanted to chime in. Thanks for the thorough write-up and thought you put into this. Thanks also for the conversation we’ve had to discuss the topic.

These are personal thoughts here:

On Acquiring Governance Power for Liquidity Incentives

The main concerns for acquiring other governance tokens for directing liquidity incentives (under the ve tokenomics model) have been laid out in a couple places but imo they’re mostly:

  1. There’s loss in optionality by doing such a purchase. It commits at least some of the liquidity strategy to one DEX and ecosystem longer term.

  2. Payoff periods are not that great (at least in current market conditions).

    • A couple examples here and more detailed for CVX on the message below.
  3. Like you also pointed out, withdrawals will reduce (though not eliminate) Lido’s reliance on secondary market liquidity as there will be a primary market for stETH:ETH. I would expect this to happen much earlier than the EV for payoffs.

  4. Above all, in my opinion again, purchasing one of these governance tokens is above all a directional long bet on their price. More so than the pure rent vs. own discussion or the LDO incentives savings (altough those considerations matter a lot too but that goes without saying).
    CRV or BAL going up or down in price over the months/years would me more impactful to this being a profitable or unprofitable decision than all the other considerations we make.

    I personally am very wary of taking these directional views.

  5. I think this last one is not as important as the others, because I’m sure we could find a way, but this initiative probably couldn’t the carried out with just open market buys and sells (just due to size required and current depths available).

(I do think there are pros and reasons for a ‘Liquidity Seeker’ to do such a purchase but you laid out those well. These are just the cons and risks I see. Purposefully not a balanced take)

1 Like

On CVX Specifically

Again, here speaking personally, based on data. I don’t have many strong directional views so this isn’t a bear case on CVX — more so the risks I’m looking at that would make me very wary of supporting the proposal.

It is true that CVX emissions and supply growth has slowed/is slowing down a lot as you rightly pointed out. But we need to also look at the other side of that: with the reduced emissions, growth for key Convex metrics has also stalled.

CVX control of veCRV:

Some of what we care most about here: # veCRV held by Convex and % of total veCRV ownership:

cvxCRV Supply:

Convex veCRV Ownership (% of total):

Bribe revenue:

Also, for multiple reasons (some market, some structural), total bribe revenue as well as bribe per vlCVX have both steadily declined.

Bribe revenue can be a proxy here for valuing the asset and for how productive it is in terms of directing emissions/incentives to LPs (what we care about).

vlCVX Bribe Revenue Over time (in total & per vlCVX):

Of course if CRV’s price increases in the future this would increase too (as each vlCVX vote would be directing more emissions value in USD terms), but again that is just a long token price bet.

Other draggers on growth

With other competing options like st-yCRV currently yielding substantially more than cvxCRV (31.5% vs. 14.8% today) we should expect new CRV locking to flock to the former (and to others whose design leads to higher current yields).

Further, cvxCRV:CRV exchange rate is another risk for growth as if it’s somewhat below 1:1 people who want cvxCRV can be pushed to purchase on secondary markets instead of staking net new CRV with Convex.

CVX Payoff period

At this current bribes and rates it would be >3 years. This seems like too long given that we would expect withdrawals to be enabled well before that and that in general 3 years seems too long-term of a committment in our space. (Lido didn’t exist 3 years ago)

Of course we would expect there to be terminal value of the held CVX at the end of the payoff period which needs to be accounted for in a decision like this. But the reasoning and metrics above would, I think, make a case against that being a very attractive part of this decision.


From this, it is difficult for me to support the proposal, though I see exploring all the different options that Lido may have for liquidity extremely fruitful and thanks again.

5 Likes

Great post and data.

The question in this thread seems to be comparing the strategy of buying CVX as compared to paying out LDO emissions to LPs directly. So wouldn’t it be more appropriate to look at payback period for CVX purchases using the projected LP rewards instead of bribes? i.e. how much value in future CRV emissions can be generated per CVX owned by Lido, and how does that compare to current projected LDO spend.
You addressed this with your point on veCRV per CVX, but the payback math uses bribe savings as the source of payback, I think.

As far as making an assumption on CRV price, I think you could say the same about an assumption on LDO price in terms of the impact that LP rewards have to the DAO - if LDO price goes down, more LDO needed for LP rewards. I think relying on Lido to spend more out of treasury as the price of LDO goes down is quite risky too given it is “wrong way risk”.

As mentioned there’s also the question of how integral the Curve pool is to the long term strategy for stETH liquidity, and I don’t know the answer to that. But if the math was clearly in favor of buying CVX vs paying out LDO rewards in the short term (positive carry), there are in theory options to hedge the CVX price exposure so that Lido doesn’t need to be fully repaid on the purchase price. Lido just needs to break even on the hedging/borrow costs for the hold period for there to be positive NPV. Still requires an NPV analysis, but those hedging costs would at least give you a discount rate to work with and allow for a more concrete assumption on terminal value.

Hey,
Very interesting and well written proposal!
A few thoughts:
1/ CVX is really cool as it will suffer limited inflation in the future, which should be partially offset by future CRV locks. However, it is generally trading above its CRV backing, which is normal as it also has an FXS backing.
image
Therefore, to gain an hedge specifically on Curve’s governance, CVX might not be the most efficient tool compared to directly locking CRV (but need a blacklist), or using other CRV wrappers such as sdCRV.

2/ Given the fact that we don’t precisely know how deep will be the need for secondary liquidity in the long run, it might not be the most efficient to invest in governance, compared to renting votes using so called bribe solutions. Acquiring governance tokens takes around 3 years to breakeven, and will need higher cash investment if we want to have rapid impact. The good point of investing in CVX or CRV, or sdCRV, compared to incentivising voting for Lido gauges, is that you keep a residual value (it’s an investment, not an expense), and it will also bring some yield to the treasury. The downside is that it is basically a trade with a downside risk, on which we would need to take a view.

1 Like

I’m in agreement with Carvas here.

There is risk of downside and losing value post acquisition of tokens.

This might be viable in the future but with as many moving pieces as we have now (global economics, war, massive layoffs in tech, FTX ripple effect, etc.) Lido should protect the treasury, not expose it to downtrend.

Probably should be put up for revision in a more stable and reasonable enviroment at a later date.

1 Like