Temperature check - Swapmining : Should we incentivize swapping?

[allo]

I want to propose something new – I think it hasn’t been brought up here.

So after having watched the second Community call, where it was discussed that we needed new models of distribution of the UNI supply to the public that wasn’t good old LM (Liq. mining, which favors large funds), I had this idea of incentivize the “client side” of Uniswap.

It may sound crazy at first… but… here’s the way I see it. At the current stage, I think it’s food for thought and discussion and I’m not sure of the technical feasibility of this idea.

Rationale :

• At first glance, the metrics of success for us should be seen as volume. But volume is to a certain extent dependant on liquidity or what is for us TVL. No liquidity no volume. Volume also attracts liquidity, and liquidity attracts volume. The two are intertwined. Therefore both Liquidity and volume are the ingredients of success.

• The current, standard and fashionable way to compete for the 1st rank in DEX is with liquidity mining campaigns. We can define LM as a direct incentive for liquidity (or what amounts to TVL). Liquidity providers get attracted by bonus APY in the form of tokens they can either sell or hold.

• What would then be a direct incentive of the other cause of success : volume ? Instead of liquidity mining, we would have client-side mining, or what we could maybe call Swapmining.

Pros:

• Users are encouraged to use Uniswap rather than the competition because of this, even if the reward you get for a swap aren’t a lot, there is a psychological aspect to it and people love it.
• UNI gets distributed to actual swappers. These users are probably more diverse than LPs. LM can also happen simultaneously.
• As a consequence, it could drive volume (we don’t have data for this, yet, so Swapmining would be an experiment) mainly by “luring” customers from other competing DEXes or CEXes
• Technically, it offsets a portion of the gas fees
• Cost to distribute UNI fairly is lower because we can’t give too much UNI per swap
• Last but not least, It’s another publicity stunt for us.

Cons:

• The system could be gameable under many aspects : volume between valueless/fake pairs, bots executing useless trades and introducing unforeseen effects in the market.
• Claiming UNI could be hard/not worth it for a lot of people, because of gas prices.
• It could mean we need to decide of a preset list of incentivized pairs (whitelisted), which is always arbitrary to some extent.

I am fairly certain that we could factor parameters so as to reduce those risks.
The essential parameter being : amount of UNI mined per trade and USD value of this amount of UNI. A good way to mitigate gameability is to calculate the rewards and make them available only something like a month later. We would need to look at the data of average swaps per days and months, etc, VS what budget we want to allocate. Then, make sure it doesn’t become a way to buy cheaper UNI (which it should not be, because of gas mainly) or a way to subsidize useless trades. It would be important that a potential bad actor doesn’t exactly know what he’s gaining from a trade in terms of bonus UNI until after the rewards are allocated and made available to users.

Also, I think excluding contract addresses would be somewhat essential (exclude a lot of bots and non-end-user addresses). (This would also have the benefit of disqualifying 1inch, etc., but alas wallets like Argent would probably not be able to participate).

Thanks for reading this ! Curious to hear your thoughts !

PS : I don’t really like the term “swap mining”,…
Liquidity mining is supply-side mining
demand-side mining is… Volume mining ?

[Shawman]

Love this idea. I think this is much more effective way to distribute uni to the community of users that are adding value to the protocol. Due to gas fees, I think it will be important to allow the rewards to accrue/compound and be able to be claimed at any time.

[Unicorn33]

I agree i think its very important to incentivies the community who actually swap their tokens on Uniswap !
I think this has the potential of been interlinked with the rewarding of current UNI token holders.
For example " Join UNISWAP and become a UNIMINER"
Which i think has more atraction than the name swap mining.
Current holders of UNI can stake UNI to get interest on their UNI holdings but also are rewarded UNI for swapping their coins on the platform. The UNI they earn can be automatically restaked on top of their current stake.
Instant reward updates are a nice touch as its important to remind the users as often as possible how important they are to the company!

If the customer feels valued they will spread that value for us to other people encouraging new UNIMINERS to join by word of mouth.

[damiwack]

i like this idea… incentive users of swap is best for volume and earn more fees for pools … Compound have a similar mechanism.

[allo]

It’s true that Compound distributes COMP to those who take out loans/repay/etc. Thanks for reminding us ! The idea suddenly becomes less weird.

It’s a simple idea after all : rewarding users, not just LPs.

I think one of the only problem here is that bot addresses could end up raking up a lot of the rewards if we don’t have good parameters/rules for the usage mining campaign. Not all (arbitrage, etc.) bots use contracts : some of them are just regular addresses I reckon (not an expert, but I know one…), so they are hard to differentiate. We could have a UNI reward cap per address as some kind of mitigation. Sometimes bots will front-run traders by buying and selling right after (one block difference ?). Well, we could probably exclude that type of behavior from the UNI distribution.

[Shawman]

I really like how Compound has integrated the COMP rewards dashboard into the governance voting page. It makes the community that go to track rewards accrual, claim, and participate in governance voting all in one place. Really smart. I’d like to see Uniswap do something similar.

[Danpi314]

This might work rather well, perhaps some sort of gas rebate system (70% of gas paid claimable back) that way users that do many trades should get something back.

This way people won’t be incentivised to spam the network (as they’ll still be spending Ethereum) but at the same time it gives a unique advantage to Uniswap, and especially to smaller users where the gas fees is often more than the 0.3% swap fee.

[Danpi314]

According to ethgasstation $12.3M has been spent on gas fees on Uniswap (over the last 30days).
This would make a plan like I’ve outlined not only cheaper (Assuming a 70% rebate) than rewarding another 5M Uni a month but in my opinion much more of a benefit to the average user that doesn’t do 100ETH+ trades.

[Shawman]

We might want to consider a flat UNI reward per trade that would be less than the lowest transaction fee on the network. Currently at 20 gwei, a swap costs 1 dollar. Considering gas costs probably won’t go below 5 gwei (swap cost of 25 cents), perhaps we start at a 0.05 UNI (17 cents) reward per trade. We have roughly 4 million trades per month, so this would be a community reward of ~$680,000 per month.

[Danpi314]

That’s actually much better, as it’s massively easier to implement a flat rate I’d a thought.

The only risk of setting it too high is that some might try and game the system if the rebate exceeds the fee from a transaction that is not being included in a block by a miner. This would essentially set a floor on the gas fees.

At a level of 5 or 10 GWEI it will help those doing Uniswap trades (particularly non-whales), as it’ll also get people using Uniswap to stay on Uniswap as we’ll be cheaper than any other onchain platform. This should create more volume and organically increase TVL

[Jacob]

You’d probably want to not set it as a set amount of UNI per trade but rather set it proportional to the # of swaps that block/epoch.

This would:

• put an upper bound on how much value Miners could extract by using these tx’s to fill empty blockspace
• limit exposure to the UNI block/epoch reward in the event of an uncaught economic exploit
• With the added benefit of rewarding those who trade in periods of low volatility/volume more than those who trade during high traffic times

@allo why do you see excluding aggregators like 1inch as a benefit? Excluding contracts doesn’t make sense to me because presumedly this rebate program could incentivize developers to integrate Uniswap which means more volume.

[Danpi314]

I’ve spoken with a user on Reddit and they’ve suggested a better idea might be to try an pay for the gas fees directly in a similar way 1inch already does with the CHI gas token.

Advantages
Direct, less gas cost to user, can reduce gas costs by up to 50% or so. This would give Uniswap USP over other DEX’s as Uniswap would be a market leader due to these reduced gas costs. The gas fee often exceeds the swap fee so this would incentivise users to trade on Uniswap
Would be effective at any gas price.
Miners wouldn’t be incentivised to include Uniswap blocks in needlessly.

Disadvantages
Would need a rewrite of the transactions being sent to uniswap but not the code that holds the funds.
E.g. 1inch already uses this so it’s definitely possible
May result in slightly higher gas costs for users not using Uniswap
CHI would somehow need to be acquired, whether that be minted on bought on Uniswap.

[allo]

Thanks for this precious input ! A list of agregators could be compiled and whitelisted. The idea here is to exclude bots to make more place for “retail” or “normal” end users. But contracts could also be accepted in general to foster integrations, yes. The way I see it, though, is that 1Inch is pretty indifferent here. They don’t care that much about Uniswap.

So, now that ETH is going up in price, swaps are going to get more expensive and the network could go back to the congestion levels we’ve seen this summer (200+ gwei gas).

At these rates, I wouldn’t be surprised to see 100$ swaps soon.

Suddenly, incentivizing swaps makes more sense, if it didn’t already.

Imagine if we had an ETA for Uniswap on Layer 2 (where gas costs way less), we would be able to bridge the gap of really high gas prices by subsidizing swaps until L2 lands. I know I’m dreaming, but still…

High ETH cost + Network congestion (high gas cost) could prove to be a worse enemy than the japanese dish competitor.

@Danpi314 CHI is basically an ugly hack that’s frowned upon by many ETH devs. It will be deprecated soon, I hear… The real solution here is Layer 2 support.

[Danpi314]

I agree wholeheartedly with your idea of subsidizing swaps until L2, this would be a much better use of Uni as opposed to just providing LP rewards, as it benefits a larger number of our users. Not only that it has a sunset clause rather than perceptually handing out rewards.

Fair enough, I’m just throwing up out ideas that aim to help the average person on Uniswap. Your right though exploiting a refund by creating and destroying dummy contracts would be bad for the ecosystem as a whole.

[Shawman]

@Jacob, so if I understand this correctly, we would set a unit value of UNI rewarded per block, and then the UNI reward would be distributed proportional to the number of trades in the block? So for example, we could start the program with 1 UNI rewarded per block, and then number of trades per ~13 seconds would get their proportionate share? For the last few months we have been averaging ~4,000,000 trades per month, or about 20 trades per block. So, at 1 UNI per block of reward, each trade would receive 1/20 UNI, or 0.05 UNI (18 cents). This would cost us ~$750,000 per month which I believe is a good value for all stakeholders.

[rezwile]

@Jacob

limit exposure to the UNI block/epoch reward in the event of an uncaught economic exploit

I like the general idea to incentivize swaps a lot. And your limit is a very thoughtful security aspect. In any case there should be upper limits per block or time interval.

[rezwile]

@allo

PS : I don’t really like the term “swap mining”,…
Liquidity mining is supply-side mining
demand-side mining is… Volume mining ?

Usage reward.
Transaction reward.

I see no real reason in using the word mining, as there is no mining involved.

[Figu3]

I’m not sure I see the point of rewarding users.

While LPs are essentiel to the well being of Uniswap, its users are the ones who benefit the most from it.
Subsidizing usage is dangerous for a number of reasons :

• It doesn’t really contribute to the protocol’s long term usage (see Aave vs Compound TVL even thought Comp rewards its users)
• Creating a precedent is dangerous for the stability of governance (we can’t just start handing money over, there has to be a clear strategy)
• By distributing tokens this way you create a downwards pressure on prices
• The small amounts disitrbuted would be so insignificant it would have no utility (when the amounts are too small decentralization becomes pointless.

All in all, while it allows new category of actors of Uniswap to be rewarded, I believe it’s an idea that would create too many complications.

Cheers,
Figue

[Shawman]

Thanks for your take Figue. I respectfully disagree on some of your points. Interested to get your and others feedback on my counterpoints.

I see the initial early success of Uniswap to a number of factors, but most obvious is that the trading experience is simple, convenient, and many times faster than interacting with a centralized exchange. What Uniswap is not is cheap due to three main factors: Network fee, Swap fee, and Slippage.

Network fee solve: V3 is rumored to utilize rollups greatly reducing network fees, but until then, we are currently not serving customers the best we can because of high network fees. Giving a temporary micro reward will help here.

Swap fee solve: There is no current solve for the high swap rate of 0.3%, and it is a reality that centralized exchanges beat us here. For example, Binance is 0.1% and offers reducing trading fees with BNB.

Slippage solve: With the liquidity mining program and a general organic increase in transaction volume, slippage has progressively improved, and hopefully will continue to improve in the coming months. It remains to be seen if/when liquidity mining will return, but it appears that a scaled down version will happen at some point to accelerate the rate of slippage reduction on certain high value pairs. There definitely appears to be a law of diminishing returns on incentivizing liquidity, and this is why there is a lively debate of the forum on further liquidity mining, including size of program and targeted pairs.

On your point that this type of reward does not contribute to long-term usage (Compound vs. AAVE TVL), I would say that TVL does not tell the whole story on how successful a protocol is. I believe key performance metrics we need to also look at is active users (daily, weekly, monthly) and transaction volume. On these metrics there is a stark difference:
Active users (monthly): Compound (105,000) AAVE (7,000)
Transaction volume(monthly usd): Compound (7.2 billion) AAVE (2.5 billion)

On the hazard of setting a precedent, to mitigate this risk, we need to be clear about our intentions of this program up front to our customers. The goal of this is to earn loyalty to our existing and incentivize potential customers to try/choose us over other decentralized and centralized exchanges during a time when the network and protocol do not have the infrastructure complete to scale.

As far as this program creating downward pressure on price, I would say that micro rewards to our hundreds of thousands of users will create less sell pressure than large concentrated rewards to liquidity miners. If our goal is broad ownership of this community asset (and I think it is), this is the way to go.

I also believe no matter how small a user position is in UNI, they are still an owner and stakeholder in the success of the protocol, and we should encourage participation in governance from people that have even just 1 UNI. This program will create more small UNI holders and hopefully more distributed governance. We should track Gini coefficient to ensure continued decentralization of UNI token.

Finally, I think this is a great promotional marketing tool. Imagine the headline something along the lines of “In appreciation of our users, for the next 3 months we will be rewarding 1,000,000 dollars per month in UNI to our community of traders.”

[Figu3]

Hey Shawman, love yout insights, especially those tables you put up comparing UNI/SUSHI TVL per LPs !

I like the way you highlighted all the issues and how they’ve been fixed, will or won’t.
Yes DEXes are more expensive, but they offer a different kind of service. I believe decentralization can’t be cheaper because it needs more incentives (to fix coordination issues created by lack of central authority).

I think your point about TVL not being the most important metrics is fair, even it helps solve some of the key problems you highlighted above.
Asa large and ancient Aave holder I still used comp as a lending platform until last week, which makes no sense. In the end it’s a lot about branding (and incentives).
That’s why I believe we don’t really need to incentivize swapping : Uniswap is the most popular DEX, hands down. Normies won’t go to sushi right now, we have all new user traction.

So UNI “swapmining” is great for advanced DeFi users but useless for normies. They have no idea how governance works, probably don’t care, and would use Uniswap anyways.

I’m a firm believer we should invest the treasurey wisely. Marketing (user distribution) can be smart, but it has to be targeted correctly. Massive distribution would be a waste of bullets we might need otherwise.

A Gini coefficient would however be really useful, would you be open to chat about it privately ? I’m working on something similar !