Dynamic Automated Market Makers (dAMMs)

From CryptoWiki

Revision as of 04:39, 1 February 2021 by wiki_crypto>Zeb.dyor (Created page with "== Basics == * A new take on AMMs. * Introduced by Kyber in their [https://blog.kyber.network/kyber-3-0-architecture-revamp-dynamic-mm-and-knc-migration-proposal-aca...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Basics

  • Introduced by Kyber in their blog (21-1-2021):

"Kyber will be launching DeFi’s first automated Dynamic Market Maker (DMM), which is also the first new liquidity protocol added to the network and is accessible to anyone, including retail LPs and token teams. The new Kyber DMM addresses two of the most critical problems in AMMs today — capital inefficiency and impermanent loss.

Unlike the static nature of typical AMMs and other liquidity platforms in the space, the Kyber DMM protocol is designed to react to token pairs and market conditions to optimise fees for liquidity providers and rates for takers. This is achieved via two simple yet novel mechanisms: Programmable Pricing Curve based on the nature of the token pairs and Dynamic Fees based on market conditions.

1. Programmable Pricing Curve To Improve Capital Efficiency

The current generation of AMMs adopts a ‘one size fits all’ approach to the pricing curve, with the most common curve used in Uniswap’s model, which aims to cater to all possible price variations.

While this type of pricing curve is useful in ensuring that the pool can always be traded on both sides, it comes at a serious downside to LPs, who need to contribute a huge amount of liquidity to make low price slippage possible. In addition, takers will suffer high slippages for pools without a very high amount of liquidity.

For example, a stablecoin pair (eg: USDC/USDT) works poorly on regular AMMs (e.g. Uniswap), since the fixed price curve is built to cater towards all types of pairs, which means that the amount of slippage is the same for stablecoin pairs and a new Token/ETH pair. This results in an extremely inefficient capital allocation for LPs, since they will need to provide a huge amount of capital (and hence capture much less fees in proportion to funds provided).

Kyber DMM’s Programmable Pricing Curves allows liquidity pool creators to customize the pricing curve and set the capital amplification factor of the pool in advance:

Stable pairs with low variability in price range (like USDC/USDT, ETH/SETH) will be able to support pools with a very high amplification factor, which means given the same liquidity pool and trade size, slippage can be up to 100X better. For other pairs, (like WBTC/ETH), capital efficiency can be up to 5–10X better as well. This means that LPs can have the opportunity to earn much more fees relative to their contribution size.

Liquidity providers can then determine which pool to put capital in and we believe that ultimately the market will determine the best parameters for different pairs to achieve higher capital efficiency.

2. Dynamic Fees To Mitigate IL and Increase LP Profit

The risk of impermanent loss (IL) is a very well known issue in current AMMs, with the main losses happening when the price of one token moves dramatically relative to the other token in the pair. During these volatile periods of high price movement between the pair, we can expect a high level of volume.

Kyber DMM hence watches the volume happening on-chain and adjusts fees accordingly. In a normal market, the DMM will behave like other AMMs. In a market where volume is higher than usual, the DMM will increase the fees, and in periods of low volume, the DMM will reduce the fees. This is similar to the strategy undertaken by professional market makers to maximise returns.

By increasing the fees for trades during these periods, the DMM recoups what would usually be ‘impermanent losses’ for LPs during periods of sharp token movements and absorbs more fees for LPs, thereby helping to reduce the impact of IL. Conversely, the DMM dynamically optimizes for lower fees to encourage trading and higher volume during periods of low volatility."