Urgent Fix: Gas Refund

Tokenomics, staking, Gas Refund

Simple Summary
Since early April, the ParaSwap DAO started refunding PSP to ParaSwap users who staked on ParaSwapPool or the Safety Module. The goal is to grow our user base, increase the trading volume and encourage more participation in staking.

So far, 9.5M PSP, equivalent to $460k has been distributed. The community is quite happy about the outcome, where PSP staking has given 3 big benefits:

  • Making ParaSwap Protocol more efficient & competitive
  • Earning auto compounding rewards on ParaSwapPool
  • Getting gas refunds on top

We, unfortunately, noticed some increasing abuse. The $30k limit was enough to limit that but we’re seeing few wallets getting around that by farming & market selling PSP the moment they get the gas refund.

Here are the top 3 issues we found:

  1. Trading Bot can add significant sell pressure when claiming the PSP. Example: address with 736k PSP today. Some may still bring some good volume but a good distribution design will align the interests of both parties: The ParaSwap DAO & those traders.
  2. Wallets can get around the $30k limit by recycling wallets. It’s a limited case since it requires moving all assets to the new wallet, but it’s doable given the incentive.
  3. Wallets can stake at the last few hours before the end of the epoch and make random trades to get relatively low-risk rewards.

To address those issues, we propose making emergency adjustments applicable in this current epoch with a two days vote instead of 5 starting Wednesday at 12 pm UTC:

  • Keep the $30k yearly limit but make it a monthly limit: $2.5k. That way, the upside will be limited to $ 2.5k
  • 1-week virtual lockup: Only users who held sPSP/stkPSP for longer than 7 days will qualify.

The proposed actions will limit over-gaming the system as the upside will be capped per month.

Our main goal is to ensure the Gas Refund is healthy and contributes to the growth and sustainability of the ParaSwap DAO.

Please let us know what you know think!

Voting Options

  • For
  • Against
  • Abstain.


  • For
  • Against
  • Abstain

0 voters


i voted for but problem is there is no ask for psp in buy side… rn every issuing psp in either stake or gas refund makes sell pressure…also psp decreasing in value makes lping worthless…i think paraswap need to implement a dynamic psp issuing due to market sentiment …if in an epoch sell pressure getting higher so reducing distribution to make price optimization…

I like the idea but I feel like the parameters should simply be aligned on the epoch duration. Basically 1.25k$ limit per epoch and account for minimal stake hold for whole epoch. It makes it just simpler to reason about.


Something else, If I stake today and start to trade today, my transactions won’t get refunded before the virtual lockup period is due right ?

I am personally in agreement of this proposal. As the original post already mentioned, these edge cases have never been the intended purpose of the gas refund programme. Not only are some instances of these clear abuse (the sPSP transfer to get over the limit, for example) , but they are harming the experience of the average token holder who is using this system with good intentions.

As most token holders (even frequent traders in Ethereum mainnet) will rarely , if ever, reach those amounts organically, people should not notice any significant changes after the vote passes.My only concern with this vote is that due to the nature of this vote, it will be counter intuitive to wait as long as usually expected for it to go for a vote.

The next epoch is less than two weeks away, and these changes have to be implemented as quickly as possible to let stakers get their stakes ready for a virtual lock + ensure that the next reward distribution already has these limits in place (since now that the cat’s out of the bag, bots might run rampant for as long as they can). For this reason, I only suggest two additions to the proposal:

  • Due to the extraordinary circumstances, post the vote on snapshot earlier than expected, and let the vote run for the usual 5 days. This way , users have enough time to have their stake ready for the virtual lockup by the next epoch.

  • For a future vote (probably the next one), introduce a framework for emergency votes that may also require such fast-tracking. Example of this could be to quickly amend potential abuses in the system (like in the current case).

I feel that with PSP 2.0 around the horizon, we have an opportunity to also introduce some new changes to the GRP along with it. Until then however, this fix will be a great start!



I will vote yes to solve this problem.

I think it would be interesting for transparency to do the same analysis with regard to PMM and staking rewards.
It was said that PMM was not the cause of the selling pressure.
A factual analysis of the on-chain data would be super useful to help the community make these types of decisions (especially the PSP 2.0 decision), if anyone in the community is up for it that would be great.

1 Like

Everything is already transparent, you just need to track and follow what’s happening here RewardDistribution | Address 0x8145cDeeD63e2E3c103F885CbB2cD02a00F54873 | Etherscan. The transparency is already on chain :wink:

1 Like

Of course transparency is on-chain, that’s the whole point!

Now I think it needs a data analysis to understand and present in a simple way to the whole community.

Personally I’m not able to do that so I’m looking to see if anyone in the community is able to do it.

But sorry, it’s probably off topic on this proposal.

1 Like

Yes, that will be the simplest approach

GM, thanks all for your feedback!

So here is the plan is:

  • Put a limit per epoch instead of month: $1.25k
  • 1 vLockup of 1 week cooldown period
  • Vote tomorrow, we took extra time to evaluate the tech implementation before going for the final proposal.
  • We keep an eye on every epoch & adjust if needed
  • Improve the Proposal Framework by introducing an emergency vote (This will require a new discussion).

I believe that if we make good progress on PSP 2.0, the GRP issues will be efficiently solved!