# The State of our Tokenomics and a Request for your Thoughts

Behind the scenes, we’ve been researching possible upgrades to SSV’s tokenomics, and I'm excited to share our progress thus far. Since our mainnet launch is approaching, and because it might be appropriate to start executing changes, we thought it seemed like a good time to open up the discussion more formally to the community.

There are still many complex decisions and issues to navigate, and these issues are highly simplified here. But together, **let's do our best to confirm our immediate preparedness for mainnet, then take our time to engineer and refine a long-term design and plan** that will create a solid foundation on which we can build an ambitious future.

The following is the result of many discussions with a long list of talented contributors, so the credit belongs to all of them (thank you all 🙏). However, note that **the plan presented here only represents my own perspective, and consensus must still be reached among the SSV team and community**. I don’t claim that this is the optimal solution, and we won't move forward with anything unless there is wide support (and ultimately, a DAO vote). So I encourage everyone to give their thoughts, ideas, and feedback to help guide the project 🙂

## Our Current Tokenomics
For a complete description of the current tokenomics design, see the [documentation](https://ssv.network/tokenomics/). However to summarize, using the figure below (taken from the documentation), we can see that ETH stakers on the protocol pay operators a fee (in SSV tokens) to run their validators. A small additional “network fee” is also charged to the staker, which is taken by the DAO treasury as revenue used to fund further protocol and ecosystem development, as well as other activities approved by the DAO. This network fee is calculated as a percentage of the operator fees.

## Proposed Changes

### Mitigating Funding Uncertainty Within an Aggressive Growth Strategy

SSV is a very young project— we haven’t launched to mainnet, and most ETH users aren’t even aware of distributed validators as a technology. Even though the potential for our growth is massive, it will not be automatic. We still must convince the ETH staking world that DVT is a superior infrastructure, and then we have to attract them to our product among a field of competitors. To reach this potential and secure our success, **we need to continue pushing into the future, which requires heavy investment**.

Thus far, Blox has been paying most of the expenses required to maintain such a strong pace, but this cannot continue. The DAO needs to be independent and pay its own bills. To illustrate our financial needs, [**a preliminary budget has been proposed on the forum**](https://forum.ssv.network/t/ssv-dao-s-expected-budget-for-2023/932) that itemizes our near-term responsibilities and important growth initiatives. Based on this, we should be prepared to spend on the order of $8M over the next year.

Unfortunately, **our near-term revenue is uncertain**. There are many unknown variables that will be revealed in the first year after mainnet launch, and until then, it is difficult to predict the protocol’s usage, needs, and income.

Instead of depending on network fees initially, **I’d like to propose minting tokens to cover our early budget** until this uncertainty is reduced. Minting is a reliable and controllable funding source that will provide the DAO with the flexibility and space to continue its strategy to maximize early growth. It also provides us with the flexibility to adjust our income in the unlikely case that we encounter unforeseen conditions after launch.

However, we also value our token holders and recognize the importance of minimizing the impacts of new token mintings on the market. Minting can be stretched out over the year and performed according to a planned minting schedule based on the DAO’s approved budget. And **we can offset the effects of minting by burning the tokens we collect through the network fee** (similar to Ethereum's tokenomics), creating an upward market pressure that indirectly creates value for token holders and increases in effect as the protocol grows. Since implementing this is as easy as switching the network fee receiving address to the null address (0x000…), this change can be made in time for mainnet with minimal risk and serve as a placeholder while we develop and solidify a longer-term model.

### Configuration of Network Fees

SSV is poised to potentially be one of the largest protocols ever by TVL, but **the network fee, as it is currently defined, does not extract revenue directly from our TVL**. Instead it as calculated as a percentage of operator fees (not taken from operator fees but charged separately).

For our mainnet launch and initial operation, our current configuration seems to be good enough. We have decent predictions of what early operator fee pricings will be, and since the amount of operator fees collected will roughly scale with the number of validators run on SSV, it’s a fair claim to say that our network fee will be roughly proportional to our overall TVL. However, there are disadvantages to this that I think warrant change. Primarily, **our revenue is tied to the pricing of operator fees, which will likely decrease over time and lead to a gradual shrinking of our revenue per validator**.

I expect operator costs to decrease because 1) the game theory of our staker-operator free market incentivizes price undercutting, and 2) the lower bound seems to depend on the operating costs of running a node, which are somewhat fixed and much lower than the current market rate.

Lower operator fees shouldn’t be a bad thing though. In my opinion, low fees should be encouraged, as it would make the protocol more competitive overall by creating higher returns for stakers and giving an advantage to third-party services that use SSV as their infrastructure. And by decoupling the operator and network fees, we could gain these benefits without hurting our income. Additionally, if we calculate our network fee based on our TVL, we would gain in the likely case of a rising ETH price and a growing valuation of ETH staking in general.

**So I propose that our network fee should be formulated as a percentage of the average ETH staking rewards of the validators running on SSV.** The total cost to stakers could then be, for example, 1% of the average ETH rewards + operator fees. Note that this network fee would not be taken directly from a validator’s ETH rewards, but instead would be calculated and billed to the user separately as a fee for continued operation of their validators, in the same way the fee is charged currently.

### SSV’s Use as a Utility Token

In SSV’s current tokenomic model, the SSV token is used to pay fees, but this is not our only option, as the fees could instead be paid in other currencies, like ETH or DAI, leaving the SSV token’s function to be primarily for protocol ownership and governance. This is a complex issue with many conflicting viewpoints being debated, but I think it can be reduced to 1) What would create the best product? and 2) What would bring the most value to the token?

In regards to 1), **my opinion is that using ETH for fee payments would simply create a better product**, because by using SSV, we are worsening the user experience and creating an unnecessary cost inefficiency that will reduce the profitability for both stakers and operators. The need to purchase SSV adds a non-trivial barrier of entry, it adds the requirement for stakers to manage a balance of SSV (instead of paying fees from their ETH income), it adds an unnecessary cost for stakers (fees to swap ETH->SSV), and it adds an unnecessary cost to operators (fees to swap SSV->ETH/USD). In contrast, with the use of ETH, users could deposit ETH, pay fees in ETH, and earn in ETH… seems much simpler, right?

In regards to 2), **my opinion is that using SSV tokens for fee payments probably won’t add much value to the token**. I think this because by having an inferior product with less utility (see 1), we’re going to attract fewer customers (lower TVL). Also, because the flow of fees into the protocol will equal the flow out (the amount of SSV sold by operators and the DAO will roughly equal the amount bought by stakers to pay fees), the impact on the market is neutral… although, I should state here that this is a highly simplified claim with a great deal of analysis and contrary opinions omitted.

In other words, **by focusing on creating a product that provides the most utility and best experience for users, not only will we attract more customers and make them happier; we will increase the “value” of the protocol, which in turn will bring value to the token, because token holders will own a piece of that increased protocol “value”**.

I know this might seem quite counter-intuitive. If SSV isn’t used for some functional purpose on the protocol, why would it even have value? How does it benefit token holders? The short answer is that **there’s great deal of value in the ownership and control of a protocol that has clear and significant utility, a large number of users, a lot of TVL, and a big treasury**. By holding the SSV token, you indirectly own part of the DAO’s treasury and can influence how it will be used, you have some power to set the protocol fees and who they go to, and you get to help decide what updates are made and have the ability to protect aspects of the protocol that you find valuable. For more on this topic, I recommend [this article that articulates this concept better than I ever could](https://jacob.energy/hyperstructures.html).

Beyond simply speaking theoretically, **there are many successful projects that follow this design**, with Lido being the closest example to our own project (all transactions and fees are in ETH, and LDO is only for governance and protocol ownership).

However if we choose to move in this direction, the transition could be tricky. Because the token is currently used for fees and the code has already been written and must be audited soon, **it is too late to make this conversion without delaying mainnet launch**. And since many users might have bought large amounts of SSV with the expectation of using it to pay their fees, we could potentially unfairly cause trouble for these users.

Nonetheless, I think we can make the conversion after mainnet and help to mitigate the largest obstacles and headaches from the transition as follows:

- Phase 0 (current) – Only SSV is used for fees.
- Phase 1 – SSV and ETH can be used to pay fees. Fees go to a hot wallet controlled by DAO. Operators receive payments in only SSV from this wallet. DAO manually refills wallet with SSV as needed and takes ETH as revenue.
- Phase 2 – Only ETH can be used to pay fees. Operator fees go directly to operators. Network fees go directly to DAO.

I am suggesting ETH as the new currency for fees primarily because it minimizes complexity and cost for the user. However, to mitigate volatility, **operator fees should be denominated in USD (but still paid in ETH)**, which can be handled by the use of an oracle.

### Improving Decentralization via Dynamic Network Fees

Since I've already written an [**article on the topic**](https://hackmd.io/@fod/ByIuRx1A5), I won't linger on the details here. But in summary, our network fee offers a great leverage point that can be used to encourage good behavior from users and strengthen the protocol overall.

There are several ways this can be utilized, but **the most popular, and in my opinion the most valuable, variation of this mechanism involves using a dynamic network fee to incentivize decentralization**. Specifically, this could be done by discounting the fee for users that choose small operators and penalizing the choice of large operators. We could also reward a high diversity of clients, geolocations, and other properties that would improve the protocol's health and altruistic value.

In my opinion, these discounts and penalties should be fairly small— large enough to meaningfully incentivize good choices, but not so large that apathetic users are chased away to our competitors. For example, this might mean a change of about 10% on the network fee, or about a 0.1% change of a user's overall staking APR. But that specific value should be discussed and optimized.

### SSV Token Staking and Holder Incentives

I should state immediately that I am personally fairly neutral on this topic, and I believe that we should be very careful if we choose to implement any such mechanisms. With that said, holder incentives are common in the space, and **while SSV doesn't seem to have any problems that would create an explicit *need* for such incentives, a well-designed program has the potential to add significant value to the protocol nonetheless**.

To illustrate what I mean, let's look at the example of Rocket Pool’s RPL staking, where node operators have the opportunity to stake RPL. In my opinion, RPL staking solves a specific problem: operators must provide a large amount of collateral to participate, and by earning income on that collateral, their risk is mitigated over time. However, a secondary effect is that those who wish to accept that risk (or even voluntarily take on more of it by staking more than the required minimum) have the potential to significantly increase their overall return, which creates additional utility within the protocol.

I think it’s clear that a similar mechanism could be designed for SSV with the same potential to add utility. But we should ask ourselves if the costs and complexities of such a mechanism would be outweighed by that additional utility. **With those caveats, I would like to propose the following SSV staking mechanism:**

- Allow only node operators and ETH stakers using SSV to stake their tokens (up to a maximum per validator).
- Rewards are minted, effectively paying users holding SSV at the expense of non-user holders.
- This increases overall profitability of the protocol for users that are willing to accept additional risk (by holding SSV). By choosing to take on that risk, stakers will effectively earn extra returns on their staked ETH, in the form of SSV tokens. In the same way, operators could earn extra income.
- This also creates an incentive for more SSV to be held by users of the protocol, increasing their ownership and governance power relative to non-users.
- Optional feature: Allow users to lock up their SSV to increase their returns, where the return increases with the duration of locking.

Alternatively, we could allow anyone to stake their SSV, not just users of the protocol. Although this would probably have the benefit of having more SSV staked than the option above, I think it loses the primary benefits, like the potential of increased profitability for users and healthier DAO governance. And with too much SSV staked, the mechanism would likely act more as a shelter from self-imposed inflation than an additional incentive to hold the token. Therefore, **I think keeping an SSV staking mechanism limited to protocol users is a better option than having it completely open.**

Beyond staking, there are alternative mechanisms we could consider, such as the use of dynamic network fees to give discounts to holders, but in my opinion, staking is a superior approach. For example, staking allows the magnitude of rewards to be greater than the network fee and has a more desirable flow of value (from non-users to users, instead of from users to other users).

However, in regards to the necessity of having some holder incentives at all, if we look at projects that have operated without them, we can find several examples of successes. For example, Lido has no extra incentives, and investors don’t seem to care because their expected growth is high. Chainlink is another example (prior to their recent upgrade). I think SSV also has a similar high growth expectation, and you could use these cases to argue that **there is not a *need* for SSV staking or extra incentives to be available, at least not immediately at launch**, especially when there are many other high-value tasks to focus on (e.g. getting to mainnet as soon as possible).

### Proposed Roadmap

The following is a tenative roadmap meant to give a concrete initial starting point with the intention of further iteration. Please don't view it as a finalized or polished plan.

**Phase 1** (before mainnet) – Slightly modify existing tokenomics to secure our near-term finances.

- Agree on a DAO budget for the first year post-mainnet
- Mint tokens to fund that budget
- Burn all collected network fees

**Phase 2** (after mainnet) – Start transition away from SSV’s use as a utility token

- Change network fee to be defined as a percentage of ETH staking rewards, not a percentage of operator fees.
- Add ETH as payment method for fees (denominated in USD to mitigate volatility). Operators will still receive payments in only SSV.
- If needed, implement a DAO-controlled “hot wallet” that receives all fees and pays out SSV to operators. It can be manually refilled with SSV as needed from the DAO treasury.
- Stop burning fees and collect the new ETH fees as DAO revenue.

**Phase 3** – Implement highly desired mechanisms

- Implement dynamic network fee to reward decentralization — discount small operators and penalize the choice of large operators; reward diversity of clients, geolocations, etc.
- Optional: Allow node operators and ETH stakers to stake their SSV (up to some maximum, on a per validator basis). Pay for rewards by minting tokens. Maybe include supplementary features, like a token locking mechanism.

**Phase 4** – Finish transition away from SSV’s use as a utility token

- Remove SSV as a fee payment method.
- Change operator earnings to be paid in ETH.

**Phase 5** – TBD

### A Request for your Thoughts

How do you feel about the DAO minting its early budget? What rate of inflation would you see as excessive?

Would you like to see early network fees burned? Or should the DAO just keep them in addition to what it mints for its budget?

How do you feel about SSV’s use as a utility token? Would you prefer to pay/receive your fees in SSV? ETH? DAI?

Would you stake your SSV tokens if the option was available? Would it cause you to buy more SSV or change how you use the protocol? Would it affect your business as an operator or a third-party using SSV as your infrastructure?

How do you feel about operator pricing? Should we work to encourage lower prices to improve our competitiveness? Should the market be left alone?

Please don't hesitate to give your thoughts on the above. And please feel free to reach out to me or any other member of the team. Thanks for reading and contributing to this great community! 🙏🙂