CIP-Draft: Should CowDAO consider revenue-share tokens to reduce COW sell pressure?

The purpose of this CIP is to gather feedback and support within the community to consider a modified version of the “ve” token structure that will protect the COW token.

Simple Summary

  • Sell pressure on the COW token destroys treasury and project value
  • Forced selling at all-time lows is a sure fire way of destroying value
  • Similar to the “ve” token model, CowDAO can implement revenue-share tokens
  • Revenue-share tokens can be expressed as transferable ERC-20 tokens
  • ERC-20 revenue-share tokens give CowDAO a way to reward contributors, customers, and liquidity providers without putting sell pressure on the native token during a bear market

CowDAO has 1 tool (the COW token) but 2 types of holders: short-term (looking for immediate yield (namely liquidity providers and some customers)) and long-term (core team and core community). By using the COW token to incentivize long-term and short-term stakeholders, the COW token is exposed to unwanted sell pressure by the short-term holders, especially in a bear market. Using revenue-share tokens gives CowDAO a tool to reward short-term stakeholders that are in search of yield while protecting long-term holders from value destruction as a result of COW token sell pressure.

Motivation

The CowDAO treasury is made up primarily of the COW token. The value of CowDAO as a project, and the project’s runway, depends on the value of the COW token. Unfortunately, this is what has happened to the token since inception:

The COW token is currently the only tool CowDAO has to manage the project. This applies, among other things, to customer incentives and liquidity provider incentives. Unfortunately, not all stakeholders who receive COW tokens are long-term believers in the project. For example, it is well documented that liquidity providers sell their native token rewards back into the pool they provide liquidity for (Pool 2 rewards). While having COW listed on DEXs is useful for price discovery, rewarding liquidity providers in COW can be overly costly in dilution and sell pressure.

A proxy for expected price reduction is slippage, which as we know is the “loss” incurred by liquidity providers as a result of the price of the assets moving relative to one another in liquidity pools.

1inch is a DEX aggregator that can be used to estimate the slippage from native token emissions. In his post (link) on Affiliate Program Incentives, @middleway.eth suggested 600k COW retroactive rewards to customers that participated in the referral program. Based on the illustrative transaction in 1inch below, these rewards could push the price of COW down another 30% if they are all sold (note that I strongly support @middleway.eth and his proposed incentive program, I am just using this as a current example to illustrate my point):

When stakeholders that do not believe in the long-term potential of CowDAO receive COW tokens, excessive sell pressure becomes costly for the project in terms of value destroyed and lost runway.

Specification

This problem has become so acute that we’ve seen DAOs across DeFi experiment with various strategies to mitigate token selling. The “ve” model popularized by Curve incentivizes native token holders to lock their token for a portion of protocol revenue (and locking means they cannot sell) [link]. “POL” or protocol owned liquidity, a model popularized by OlympusDAO, was designed in response to mercenary liquidity providers in Pool 2 liquidity pools with the goal of reducing the sell pressure on native tokens [link]. DAOs recognize they need to diversify their treasury but they cannot afford the market slippage, so they are opting for asset swaps instead [link]. Finally, DAOs are experimenting with vesting schedules for customer rewards [link].

All of these potential solutions are designed to improve the long-term outlook of the treasury by diversifying assets or by redirecting native token towards long-term holders. However, these solutions remain anchored in the idea that the native token is the only tool DAOs and protocols have at their disposal.

It seems to me like there is an alternative solution that would be easy to implement.

Taking example from Curve’s “ve” token revenue-share model, CowDAO could implement a revenue-forwarding mechanism to holders of a specific revenue-share token (this could take the form of a transferable ERC-20 token).

For example, 600k COW is currently worth $81,600 ($0.136/COW). Rather than issuing 600k COW tokens to customers that will likely sell them into the market (again, just an example for now), CowDAO could mint revenue-share tokens. The revenue-share tokens could have the following structure:

  • 10,000 ERC-20 tokens (1 basis point per token)
  • CowDAO forwards 20% of revenue to wallet addresses that hold revenue-share tokens
  • Revenue-share tokens are distributed the same way the 600k COW would have been distributed
  • Tokens are burned and revenue-forwarding ends after ~$85,000 is received

Rationale

Structuring a temporary revenue-share agreement this way allows CowDAO to pay for things without exposing the COW token to unwanted sell pressure.

By using the ERC-20 standard, revenue-share tokens would be easy to understand, easy to transfer across all existing infrastructure, and easy to manage across all existing tools. CowDAO could reward customers and incentivize liquidity providers with non-dilutive revenue-share tokens instead of continuously issuing COW tokens.

As mentioned in the Summary, the crux of the issue is that CowDAO has 1 tool (the COW token) but 2 types of holders: short-term (looking for immediate yield, namely liquidity providers and some customers) and long-term (core team and core community). By using the COW token to incentivize long-term and short-term stakeholders, the COW token is exposed to unwanted sell pressure by the short-term holders. Using revenue-share tokens gives CowDAO a tool to reward stakeholders in search of yield. The COW token can be used to incentivize believers in the long-term potential of the project.

Further, being a forced seller of any asset in any market represents a sure way of destroying value and leaving money on the table. Using the native token to pay for things in a bear market is extremely damaging to the runway and value of the project. CowDAO would benefit greatly from having another tokenized asset at its disposal.

Finally, with something like revenue-share tokens, CowDAO could develop more sophisticated treasury management solutions. It is easy to imagine a scenario where CowDAO purchases or sells revenue streams for other project assets with the dual purpose of (i) generating attractive returns and (ii) adding a diversified currency to the treasury.

I look forward to discussing this proposal with everyone.

Cheers,
Habs4Lyfe

4 Likes

Thank you for the well written proposal! You are specifically talking about revenue sharing (not profit sharing), right? It’s true that the protocol is generating between $50k-$100k in revenue at the moment which could potentially be passed on to token holders.

At the same time the protocol has significant operating cost (beyond paying for its development) which mainly stem from solver rewards and are in itself still slightly exceeding the current revenue.

Without any change in the fee structure, a revenue sharing model would therefore accelerate the rate at which the DAO’s treasury would be used. This would likely eventually also lead to sell pressure (as you are correct COW is the largest asset in that treasury).

Therefore I have two clarifying question:

  1. Does this proposal assume the DAO will actually have excess cash flow from operations (ie. by increasing the trading fee) as in would it be fair to change it to be a “profit-sharing” proposal?
  2. How would this proposal fundamentally differ from a buy-back program where the DAO agrees to purchase a fixed $ amount of COW per week?

An interesting opinion was shared by @Hasu on bankless recently where he argues protocols should not jeopardise their treasury to pay premature dividends.

One side note:

I disagree with this statement, as this assumes the entire buy demand for CoW is currently expressed by LP providers.
I think what your screenshot (I personally prefer to use CowSwap over 1Inch for these type of queries :wink: ) shows instead is that the market is very illiquid - you cannot buy/sell larger amounts of tokens in one trade without incurring price impact - and therefore the current price discovery of CoW is innefficient/incomplete. By no means would I equate price impact with “loss” for liquidity providers as some parties which aren’t actively providing liquidity would likely step in correct a price drop that steep.

4 Likes

@fleupold

Thanks for reading the whole proposal and providing a clear reply! Your feedback is much appreciated.

My suggestion is not to pay dividends prematurely. On the contrary, I am suggesting raising non-dilutive capital using future revenue instead of using COW tokens, thus protecting the COW token from further selling and increasing the treasury value as a result.

Revenue-share tokens are just a way of accessing non-dilutive capital based on future revenue, without collateral.

Revenue-share tokens would never lead to any treasury depletion because the payment amount is variable; it is calculated as a proportion of incoming revenue, not as fixed obligation. For example, payment to revenue-share tokens would always be 50%*[revenue]. The lower bound of the payment is $0 if the DAO generates $0 of revenue.

In certain instances this does accomplish the same goal. However, constantly issuing COW tokens and constantly buying back COW tokens isn’t a cohesive capital allocation strategy. Revenue-share tokens provide an alternative reward mechanism (or fundraising mechanism) for periods when minimizing COW token issuance is a priority (I would argue that this is the case in the current environment).

You’re right that slippage on 1inch is a far from perfect proxy for sell pressure, and you could be right that there will demand at lower COW prices. However, given the recent performance of the COW token (and the market broadly), it is far from certain that demand would step in at any level.

I completely agree with you and @Hasu’s Bankless post that DAOs should be run more like businesses. If we believe the “equity” in the DAO (in the form of COW tokens) is valuable, it’s not clear to me why we would choose to use it as a currency (i.e., to to pay for things) if there is a viable alternative (future revenue). Conversely, if the DAO constantly uses the token as a currency, it’s not clear to me why we should expect anyone to hold it like equity.

The goal is to treat the COW token more like equity: minimize COW issuance to short-term holders, minimize dilution, and minimize unnecessary sell pressure so as to maximize treasury value. Using revenue-share tokens can help us do that.

1 Like

Interesting, I indeed understood your initial proposal differently. My understanding now is that you are suggesting to introduce a new type of equity (a claim on future excess cash flows) - veCoW - which is more senior to CoW (assuming [v]CoW holders voted to share future profits amongst governance token holders).

That is, as soon as the protocol makes a profit veCoW holders would get a fixed share of that cash flow before the rest gets redistributed to the treasury or any other token holder.

This new token would then be used for rewards instead of the more junior claim.

I deliberately replaced “revenue”-sharing with “profit”-sharing because of your statement:

Clearly this is only true for profit sharing tokens. If a company has $100 in revenue and $100 in cost, assuming it gives away 50% of revenue, it would start making a loss and deplete its treasury.

Please let me know if I’m still mis-understanding this proposal.

1 Like

You are right, @fleupold, that the revenue-share token, taken together with other expenses and capital outlays, would not absolve the DAO from requiring treasury funds if revenues were insufficient to meet the sum of its obligations. My statement pertained to the revenue-share instrument by itself: because revenue-share payments are calculated as a % of revenue, they can never be greater than revenue. My apologies for this not being more clear.

Like any form of financing that is not equity, the revenue-share tokens protect equity from being diluted (or treasury from being depleted) because they bring revenue from future years to the present, which is critically important considering this price environment and considering the potential sell pressure on the token and the impact on the DAO’s runway from excessive issuance.

See below for an illustrative example of how moving money through time via revenue-share can reduce dilution or treasury depletion.

Revenue growth via partnerships

The key in the illustration above is the revenue growth, and I think we are all in agreement that volume is the critical requirement towards driving revenue increases the DAO.

In this discussion thread from the spring, @fleupold and @defiparasite you guys talked about veTokens and increasing fee revenue. Another thing that came up was network effects. While I don’t have an opinion on the fee increase, I tend to agree with what @0xSami_ posted over the weekend that distributing revenue in the form of dividends and share buy-backs should be reserved as later stage capital allocation decisions.

That being said, issuing short duration revenue-share tokens to some of the institutional DeFi funds that seek sustainable yields could be an interesting way to increase the notoriety for CowDAO and kick-start network effects. Thinking about it further, there are 3 ways finding an institutional DeFi investor for short duration revenue-share tokens could help CowDAO drive trading volume (and therefore revenue growth):

  1. CowDAO can promote the new partnership (earned media)

  2. Institutional investor meets the team, tries using the product, and (hopefully :crossed_fingers:) becomes a long-term repeat customer for CowDAO

  3. Institutional investor becomes a brand ambassador, incentivized to promote CowDAO among other institutional players because they benefit from revenue increase via faster repayment

Ultimately, revenue-share tokens can be used to bring on partners the same way native tokens are used. Issuing native tokens to partners is a well-trodden go-to-market path in DeFi and I think CowDAO could benefit from the same kind of publicity. As a shorter duration, less volatile, revenue-generating asset, revenue-share tokens would attract a different kind of investor than the COW token, thus broadening CowDAO’s partner base.

What do you think?

2 Likes

Thanks for clarifying and thanks for the added calculations to help understand the idea behind the revenue token better.

My personal impression is that the reason COW is trading at some positive value today mainly comes from the fact that people expect to vote on sharing part of protocol’s future profits with token holders. Creating a revenue token should reduce those future profits and consequently lower the value of COW by roughly the same amount the token is valued at.

However, I must admit that this kind of financial engineering is beyond my expertise, so I will defer to more financially literate people to comment on the effects and drive the discussion forward.

@fleupold appreciate your continued feedback! :grinning: It’s a very valid consideration though this kind of structure would not affect the long-term profit potential of the project:

1 Like

From my (naieve) perspective, paying COW to solvers is an expensive endeavor over the long term that contributes to excess COW to be sold when solvers cash out.

Is it possible to switch the solver reward from COW to a non-COW token?

@0xEvan yes, in my opinion it is possible to issue rewards in a different token!

Having an “IOU” token that is repaid using DAO revenue creates a non-dilutive alternative to issuing the COW token. This token would (i) also be an ERC-20 and (ii) have it’s own liquidity pool. As a result, it’s easy for anyone to sell it for yield, which is what most entities do already. As long as they get their yield, they don’t care what kind of token they receive!

What do you think?

I agree with most of what you said except that it be the native token itself instead of an IOU token. The major downside to an IOU token is that you need to provide liquidity for it. It’s not clear to me what benefit you get with an IOU token as opposed to the underlying tokens itself (e.g. ETH, USDC, DAI)

@0xEvan @fleupold

The problem is CowDAO can’t use the ETH/USDC/DAI in it’s treasury because treasury funds are way too important. Treasury health dictates project runway AND is used to derive valuation. Projects live and die by their treasury, so the treasury should remain untouched as much as possible (i.e., hopefully it only ever increases).

If CowDAO wants to protect its treasury, and wants to use a non-COW token to reward Solvers, then IOU tokens are the ideal solution.

And to answer your question directly: an IOU token would not require complicated liquidity management like the COW token because the goal would not be to make the IOU token accessible to the mass market. IOUs would be issued to Solvers and short-term stakeholders (think: DEX liquidity providers). Plus, IOUs would only be issued in proportion to the amount of revenue earmarked for repurchases. As a result, IOUs wouldn’t need full liquidity on all exchanges, they would just need a single Uniswap liquidity pool:

  • Each $1 IOU token is redeemable for $1.03 after 90 days
  • Uni v3 liquidity pool is funded by the DAO with a maximum price of $1.03 in case some holders want liquidity before 90 days
  • The DAO commits to setting aside x% of fee revenue to repurchase IOU tokens every week
  • Revenue goes to repurchasing IOUs from the Uni v3 LP OR begins accumulating in the “Redemption Wallet”
  • Holders that did not sell into the LP can sell their IOU tokens into the “Redemption Wallet” for $1.03 after 90 days

Result: think kind of structure would GREATLY reduce the dilution experience by COW holders.

→ Less COW dilution
→ Less COW sell pressure
Higher COW price

What do you think?

In this proposal the DAO would basically issue 90 day bonds to solvers, right (which apart from equity, is definitely a standard way to raise money)? I think for it to protect the treasury, the term would have to be much longer (and thus create additional liability), but that’s possible.

I guess one question to ask is as what do we see solvers? As a service provider that should get some fixed income for their service (via a bond, cash, etc), or as a more integral part of the system (a co-owner of the protocol) that should be rewarded with a claim on future cash flows (of course in practice they are free to trade off one for the other).

I wonder if at the end of the day most of this is financial engineering around how to best utilitze market inefficiencies. To me, at least in theory, the value of CoW should be that same if a solver receives $1 in cash, bonds or in tokens. But I might be wrong, since this is definitely not my expertise.

@fleupold

This is a great point Felix. It’s something all protocols in DeFi grapple with when issuing tokens to their many stakeholders. Allow me to suggest that the key metric should be whether Solvers are selling the COW tokens immediately upon receipt.

Yes, $1 of COW is equivalent to $1 of IOU, but if Solvers sell their COW, it pushes the price of COW down. Next time, more COW will be needed to pay out $1 of COW.

Btw I totally understand your skepticism here. I think the sheer size of the debt markets globally, as well as the amount of debt companies issue (including various forms of non-dilutive financing for startups) speaks volumes to the importance of capital allocation.

What do you think?

I want to push back a bit on the premise that using COW for incentives is bad.
We’re aiming to build a decentralized, self sustaining protocol which needs well-aligned stakeholders (IMO still need much more than what we currently have…)

Ideally, every COW that is given in incentives is held long term and pushes the recipient to contribute to the protocol and participate in governance. Overall creating more value to the protocol than the dilution caused by the COW incentive.
Needless to say, this is never 100% the case.

Given that anyway the DAO would like to utilize some capital for user acquisition, solver acquisition etc, I think it makes sense to use COW in the mix.

Of course I’m fully aware that using COW at the current low prices is extra dilutive.
So if we want to use non-dilutive form of payment, I think that it is easier and simpler to use the proposed “bonds” to outright raise capital for the DAO that will be used to make those payments.

The problem is, that I assume this will be quite expensive with the current environment of rising rates and the risk premium investors will assign to a bond issued by CoW DAO.

Try to make a simple thought experiment.
How much would you pay today on a claim for 1$ in two years from the CoW DAO.
Unfortunately my guess is that it’ll sell at significant discount. Would bet on 15-25% a year at least

1 Like

Thanks @middleway.eth for the thoughtful response!

I 100% agree that COW is the primary incentive driver of CowDAO. Anyone interested in contributing to the DAO in any, shape, or form should want to own and hold COW tokens. Anything suggested here is meant to complement the COW token and support it.

The question is whether there is a more sustainable way to reward stakeholders that put sell pressure on the COW token (DEX liquidity providers, some solvers, etc.)

I like your thought experiment. Two years is a long time in crypto - maybe we think about 6 months? Do you think members of the community would pay ~$0.92 to receive $1 from the DAO in 6 months?

This would create a sustainable, non-dilutive source of capital to the DAO and provide attractive yield/return to community members.

TBH I really don’t know…
And still this is 16% annually. There’s no good reason to assume CoW DAO will have significantly different cash flows in six months. So either we dilute then to pay the debt, or we look for more funding.
So, I’m not against it. But I also don’t think it is a very attractive source of funding.

@middleway.eth

I see. You’re right that 16% is not cheap, though it’s not bad on a relative basis (top rated market makers with ~$1bn balance sheets get charged ~8% by TrueFi/Clearpool for on-chain loans)

What has been CoW Protocol’s go-to-market strategy ? Is there a customer acquisition strategy beyond the affiliate program?

If CoW Protocol is looking for growth strategies, I think we may be able to help.

Lending protocols (Idle, Centrifuge, Ribbon, mStable, etc.) are working with us to implement a customer-specific revenue royalty that attract institutional lender liquidity. The yield from the revenue royalty gives the institution added incentive (via better risk/return) to stake into the lending product, and using Cinch gives the lending protocol a whale customer without the hassle of having to chase down and pitch multiple institutions themselves.

It sounds like raising capital via a revenue royalty could be beneficial for CoW protocol in two ways: access to non-dilutive capital + protocol volume growth. In fact, having potential customers invest in your company is a well trodden path in FinTech (example: Maven invested into Maple’s seed round and is one of their delegates/customers). Here is an example:

  • CoW Protocol could raise non-dilutive capital by selling a [20]% revenue royalty to a DeFi hedge fund. Estimated price is $0.9 per dollar.

  • CoW Protocol onboards an institutional partner that can be promoted on Twitter, blogs, case studies, etc.

  • DeFi hedge fund is incentivized to use CoW Protocol so that their revenue royalty is worth more (higher volume for CoW Protocol)

  • DeFi hedge fund is incentivized to tell other institutions to use CoW Protocol so that their revenue royalty is worth more (x2 higher volume for CoW Protocol)

  • This would require $0 from the treasury and 0hrs of dev time. Treasury actually increases by the amount received from the sale of the royalty (and currency can be chosen by CoW).

It’s not a sure thing but we have DeFi hedge funds in our network at Cinch that could potentially be interested in something like this.

What do you think?

Definitely interesting. Happy to arrange a call to learn more.
If you know funds and defi players who drives decent dex volume, I think it is worth collaborating / partnering.

Because the DEX market is quite competitive and because CoW Protocol is in its growth phase, at this stage the pricing should be as competitive as possible IMO.
This means the protocol will not generate profits that could be used to pay royalties / repay loans etc.

Did something came out of this?

Has there been progress regarding this?