Two proposals to introduce ve tokenomics and trading incentives to Serum

Hi all, posting a couple of proposals for the community’s consideration.


As the first fully on-chain order book, Serum represents a great innovation in the world of decentralized exchanges. Serum is currently the largest spot exchange on Solana, and provides liquidity for many projects in the ecosystem. But Serum has the potential to grow much more, and become a dominant protocol among all decentralized and centralized exchanges.

The motivation for these proposals is to address two areas that can be improved to achieve this vision:

  1. Low liquidity ⇒ suboptimal execution ⇒ low volume

    One of the advantages of an order book exchange over an automated market maker is its ability to provide efficient and fair price discovery and minimal price impact. However, these characteristics are only achieved when market makers provide sufficient liquidity and narrow spreads. Serum could benefit by bringing more market makers to the exchange to improve its execution and draw more volume to the exchange.

  2. Aggressive emissions + limited token utility ⇒ low demand for SRM

    Community members have expressed concern (see forum topic #205) about SRM tokenomics, specifically that the aggressive emissions schedule deters potential investors from buying the token. Additionally, SRM provides limited utility. Currently, the main benefit of holding SRM is reduced trading fees, which mainly incentivizes GUI providers to hold SRM so that they can pass low fees on to their users. SRM also allows holders to vote on DAO proposals, but it is unclear how much this utility encourages community engagement.


This post presents an approach to tackle both problems by applying liquidity incentives systems used by AMMs to Serum DEX, and introducing a voting-escrow token (veToken) model (pioneered by Curve finance) to Serum’s governance.

We’ve split the design into two proposals, which are related but not necessarily interdependent. These are each explained in separate forum posts, so that they can be independently discussed and voted on.

The two parts to the proposal are:

  1. Voting escrow tokenomics (veSRM)

  2. Volume-based trading incentives

A DAO vote will be created for each of the above proposals. The purpose of these votes will be to solicit feedback and gauge community sentiment toward the proposal, in order to determine if work should begin on implementation. New DAO proposals will be opened to introduce concrete implementation changes, should the above proposals pass.


I support these proposals. However, can you elaborate the Preventing wash trading section. This is a critical part of this proposal and it is barely addressed.

@bonfida agree that this is a critical part of the design, and an area we’ve given a lot of thought to (we didn’t include too many details here to avoid distracting from the high-level goals of the rewards program).

As mentioned in the proposal, the most basic way to address this is to screen for blatant wash trading activity (e.g. two accounts always trading against each other). The trading rewards implementation should include this check. AFAIK, this is all dYdX currently does to address wash trading.

The more complicated way to address the issue is by tuning the parameters of the reward program. The main parameter would be the amount of SRM rewards allocated to the program, which will determine how easy it would be to profitably wash trade. If the quantity of SRM is too high relative to trading volume, wash trading could basically guarantee a profit. We’ve done some analysis on this to determine what reasonable reward quantities might be, but more work should be done here as part of the implementation effort. The trade-off is obviously that if you cap rewards to discourage bad actors, everyone earns fewer rewards (when most actors are probably good).

What’s interesting about this model is that it’s impossible to actually predict your trading rewards, since they depend on the total market volume. In other words, there would always be a risk that wash trading would result in a loss, if the market volume exceeded expectations and therefore the trader received fewer rewards. This kind of creates a self-regulating dynamic–e.g., if everyone thinks wash trading is profitable, everyone will wash trade, which will should increase volume and result in wash trading no longer being profitable (wash traders lose).

A potential approach could be to begin with conservative rewards, where the expected volume threshold beyond which wash traders lose is within the range of historical volumes. If the rewards program is effective, volume on Serum should increase, at which point the reward allocation could be increased as well. More thought is required on how that would work.

As a final note, there would always be edge cases in this system, since the pool weights and market volumes are inherently unpredictable. If somehow a market with incredibly low volume is allocated a disproportionately high percentage of rewards, wash trading would likely be profitable there. But due to the dynamics mentioned above, there is at least some amount of self-correcting force at play.

Interested in if others have thoughts on this problem–it’s a bit tricky.


DYDX token price has been a total unmitigated disaster, their emissions related to incentivization have only benefitted: VCs with a purchase price <1, big MMs who can wash trade all day.

We should do better here. I am not saying I have the answer as to how tho!

also this has come out of the locker after being dormant for a month and it seems it has passed. unsure if this affects the above

I am a big fan of this proposal and I’m excited to see SRM move away from the buy and burn. I believe value accrual occurs more optimally in the form of distributing fees.

Think this proposal has good merit, and would be excited to see further discussion / implementation of this within the Serum community.

DAO votes for both proposals are now live:

1 Like

so they both passed right?