KTX Education: Liquidity Mechanisms 🌊
GMX V1 GLP model vs GMX V2 GM Pools
In this week’s edition of our Educational Series we’ve decided to go a slightly different route than normal. This time we’re going to discuss the benefits of a GMX V1 GLP model (like we have with KLP) over GMX’s V2 GM model.
Now of course, there are things that the GM model does better than the GLP mode, but there are trade-offs between them that we think are worth highlighting for those who may not be in the know.
Let’s start off with a look at both the GLP model and the GM model so you have an understanding of where the differences lie and where the GLP model stands out.
A Look at GLP 🔎
So what is GLP?
GLP is the liquidity provider token for V1, and these LPs earn fees from leverage trading, borrowing fees, and swaps.
Think of GLP like an index of assets that gets used for swaps and leverage trading. This token can be minted using an index asset and can be burnt to redeem any index asset.
Essentially GLP acts as a basket exposure to many different assets, generally being 50% exposure to stablecoins and 50% exposure to Crypto assets like BTC, ETH, and various blue chip alts.
For example:
- User wants GLP and has ETH ready to use to mint GLP.
- ETH gets put into the pool of assets and the user receives GLP in return.
- User can then stake GLP for yield in the form of fees from leverage trading, borrow fees, and swaps.
How does the GLP model impact traders? Well for one, there is no price impact but you do have a 10 basis points fee for opening and closing trades.
This basis point fee is essentially 0.1% of the position size with another 0.1% fee when closing the position. If you had a position size of $100K, then 0.1% of that would be a $100 opening and closing fee.
For the V1 GLP model, the collateral of long positions is the token being longed.
- If you long ETH, your collateral is ETH. If you long BTC, your collateral is wBTC, etc.
The collateral of short positions is different as it is any of the supported stablecoins like USDC, USDT, etc.
When you take profits on longs your profits come back in the form of the token being longed, and vice versa with shorts.
A Look at GM Pools 🌅
What is GM you say? Besides being the preferred greeting used by Crypto Twitter, it’s also the liquidity provider token for GMX V2 markets.
The V2 GM model changed up things pretty significantly as the pool is no longer a multi-asset version. Instead, each GM pool consists of 3 things:
- Index Price Feed — Long and short tokens get opened and closed based on this price feed.
- Long Token — This is the token that will back long positions.
- Short Token — This is the token that will back short positions.
A good example of this is you have an ETH/USD (ETH-USDC) market which would look like:
- Index Price Feed — ETH/USD
- Long Token — ETH tokens backing the long position
- Short Token — USDC tokens backing the short positions.
One fundamental difference between the GM V2 and the GLP V1 is that the GM pools have the ability to have multiple kinds of collateral available for a selected market.
For our prior ETH-USDC example, you can use ETH or USDC as collateral for either longs or shorts. This can be useful for various reasons allowing you to create delta neutral strategies or giving you the flexibility to swap between longs and shorts rapidly.
GM pools also have the ability to be a synthetic market, like a DOGE perpetual backed by ETH-USDC. This is different from the GLP model as GLP is limited by what assets are allowed in the indexing pool.
The trading fee structures for GM pools are different as well as price impacts are now introduced. Price impact can either go positive or negative for increasing/decreasing positions depending on the balance of long and short positions.
There is also a .05% fee for swapping assets. This is important to know for understanding why there are reasons why you would use V1 over V2 and vice versa.
Since we have a fundamental understanding of both V1 and V2, let’s dive into the differences and see where our KLP can give better options for users over a V2 GM user.
KLP > GM Pools?? 🎉
As we know, KLP is a operates on a GMX V1 GLP model so any of the aforementioned attributes apply to KLP as well. There is an exception, however, as the assets in a KLP pool differ from those in the GLP pools.
With that disclaimer out of the way, let’s get into the benefits of KLP over the GM pools!
Be Composable like Water 🌊
There is something to be said about the ease in which protocols are able to integrate KLP over something like the GM pools. The reasoning there is that it’s much easier to integrate because there is just one pool to integrate with as opposed to GM pools which have multiple independent pools.
What also may not be widely known is dynamic fees like price impact and adaptive funding fees make building on top of the GM pools that much more difficult for protocols. KLP fees however, are fixed and predictable which makes it much easier to build strategies on!
The amount of hype generated by the GLP launch as opposed to the GM pools launch is a good barometer to use as well!
If you remember, various protocols like Jones DAO, Plutus DAO, etc created their own products built on top of the GLP technology. Whether it was creating delta neutral vaults or auto-compounding features, there was a lot of innovation in a short amount of time.
When the GM pools launched, it was much more difficult for ecosystem projects to integrate because it meant picking and choosing certain pools, while with GLP/KLP it’s considerably less of a headache.
This all creates an easier environment for partnerships between various projects and KLP for building more and more lego blocks on top of KTX.
Going Under the Hood 🚗
We discussed the fee distribution for both the V1 and V2 models above so make sure you go back and review those as they will come back into play in this section!
Now let’s talk about traders and how they impact each model.
The Trader’s Impact 💥
For V1, losses get socialized to all KLP holders which is a fancy way of saying when traders profit, LPs lose. The caveat here is that traders lose more often than they win so KLP holders tend to earn solid yields more often than not!
But in the GM model, trader profits (and LP losses) are pushed onto the liquidity providers in each specific GM pool, which exposes them to unnecessary (and higher) risks.
Because KLP is a multi-asset pool, when traders do end up winning, the profit is better spread out through the pool and is a more sustainable model for all LPs. In a way, buying KLP/GLP is akin to buying the whole casino instead of specific slot machines (GM pools).
There is also a unique advantage that KLP pools have over the GM model that some people might not realize! And that would be the fees!
Watch for Hidden Fees 🥷
KLP/GLP have that 10 basis point spread on opening and closing positions but have no price impact while GM pools have price impact along with the .05% swap fees.
While one may seem cheaper than the other, the V2 fee model is variable and can change based on the pool while the V1 stays the same. Which leads to an interesting option for users…
- If the basis point spread + price impact of V2 > basis point spread of V1 → trade on V1 pools.
- If the basis point spread + price impact of V2 < basis point spread of V1 → trade on V2 pools.
This is why you might see V1 pools still receiving volume because there are plenty of opportunities for it to be cheaper than the V2 pools.
KISS: Keep It Simple Stupid 💋
Let’s see if we can make an even simpler way of breaking down the differences between these two models:
KLP Pool
- Traders get to function on autopilot and get predictable fees, making it easier to trade if you’re trying to get the most bang for your buck on fees.
- LPs get one unified liquidity pool that makes it easier to LP in and out of, essentially getting to set and forget it.
GM Pools
- Traders have to operate everything manually needing to choose which pools to use, having extremely variable fees and skew along with potential decision fatigue from having to operate multiple open positions at once.
- Managing these positions can be more of a pain considering that the max leverage for each individual GM pool can vary as well.
- LPs are more like market makers with this model, having to consider long and short sides of each pool. Very much not a set and forget model with the user needing to have more of a hand in it.
Too Long Didn’t Read 🛑 📖
Maybe that was all too much for you and you want to understand why you should use KLP over the GM pools, in even simpler terms.
Don’t worry, we got you!
Imagine that you’re betting on a horse race. You go up to the ticket counter and place a bet on a single horse to win against 12 other horses. Not the greatest odds but you’re confident in that one horse. That’s the GM pools.
Now instead, you go up to the counter and just like the Peaky Blinders, you get a little benefit from the worker. He allows you to bet on the top 3 horses at the same time. Much better odds! That’s the KLP pools.
To really sum it up, there are benefits to both models but we think the utility of KLP is underexplored and underappreciated.
We look forward to KTX being a base for those looking to build financial products on and to utilize KLP in new, efficient ways!
