DeFi protocols live on liquidity and have therefore built economic incentives for anyone willing to add liquidity to their platform. In our low-interest-rate macroeconomic environment, where idle cash earns next to nothing, DeFi yields become highly enticing. However, aspiring liquidity providers may find understanding and predicting the profitability of their strategies complex and confusing. In this article, we break down the factors affecting profitability, explore the trade-offs, and learn about the diverse range of risk-return profiles available with different liquidity provision strategies. The above inequality represents the fundamental bet liquidity providers make: they bet that gains (left) would outweigh losses (right). This article would be developed around the above inequality and the various sub-factors that influence each variable in it. Note that variables and sub-factors may vary depending on the specific liquidity pool used. Even within each liquidity pool type, protocols may have variations that introduce new factors into the mix. It is, therefore, best to read the documentation. Examples of liquidity pools include: Liquidity pools in decentralized exchanges (DEXs): Pancakeswap, Uniswap, Curve, etc. Lending pools in two-sided lending protocols: Aave, Compound, Cream Finance, etc. Other liquidity pool types (loose definition): Liquity’s stability pool (Lending), Unslashed Finance’s Capital Buckets (Insurance) LP token value When adding liquidity to a pool, you are locking up tokens in a protocol’s smart contract. In exchange, protocols often provide you with liquidity provider (LP) tokens that are representative of partial ownership of that pool. These LP tokens can then be exchanged back at a later date when you wish to unlock your liquidity. The value of LP tokens is dependent on 3 main variables: price gain of tokens in the pool, impermanent loss, and fees earned and distributed by the pool to LP token holders. The diagram below summarizes the content of the next few sections and highlights sub-factors that influence each variable. Price gain of tokens Since LP token holders are owners of liquidity pools, they fundamentally have a long position in tokens held in the pool. The number of token types in pools varies. Usually, lending pools have one token type, while most DEXs have two token types per liquidity pool at specific ratios. When prices of the underlying tokens appreciate, so does the value of LP tokens. Impermanent loss In a single-token pool, an x% gain/loss in the token’s value would result in an x% gain/loss in the LP token’s value. Meanwhile, multi-token pools experience an effect called impermanent loss that amplifies percentage losses while dampening percentage gains in token price. In other words, an x% gain/loss in any of the underlying tokens’ values would result in an ≤x% gain or ≥x% loss in LP token value. “Impermanent loss (IL) describes the percentage by which a pool is worth less than what one would have if they had instead just held the tokens outside of the pool”
~ Fernando Martinelli, Balancer co-founder [ ] source The equation above shows that IL is a function of the change in exchange rate of a liquidity pool’s token pair: the greater the price ratio change, the greater the IL. Hence, pools with highly correlated tokens experience less IL, while pools with asymmetric token price movements experience greater IL. As IL is only realized upon exiting the liquidity pool, what matters is the price ratio difference between the entry and exit point. Fluctuations between the entry and exit points can be disregarded. It may be easy to become averse toward liquidity provision upon learning about IL. Hence, it is important to put its scale into context: in standard XYK pools (eg. Uniswap V2, Sushiswap, Pancakeswap), a doubling of price ratios would only result in a 5.72% IL. To better understand IL, you can to explore numbers. Note that the equation and calculation for impermanent loss shown are different for pools that are not based on the XYK model (eg. Balancer), do not have a 50/50 composition, or apply concentrated liquidity (eg. Uniswap v3). use this calculator To illustrate how impermanent loss calculations are different in other pool types, we look at Balancer. Balancer pools are multi-token, not necessarily 50/50, and their pools are governed by the value function shown above. Nevertheless, the general relationship between IL and asymmetric changes in token price holds. Fee earnings/ Revenue Share Liquidity pools enable dApps to provide and charge users for services, generating revenue. A portion of this revenue is then distributed to liquidity providers. As protocols often also distribute revenue to other stakeholders, it is important to understand the revenue sharing structure and how much goes to you as an LP. Revenue is often shared with: Holders of their governance token (often by burning token supply, the conceptual equivalent of share buybacks in traditional finance) Protocol reserve (reserves may have a pre-determined specific use or may be overseen by governance token holders) Protocol development team An LP’s expected fee earnings can be seen as a function of supply and demand in the liquidity pool. Higher demand increases revenue generation, while a larger supply (pool size) means revenue has to be shared among more people. The table above summarizes common metrics that can be used to determine current supply and demand. However, what is arguably more interesting is expected future demand and supply. Hedging against IL Hedging strategies can be optionally adapted to reduce IL. From the IL graphs earlier in this article, it is clear that IL is bidirectional and non-linear. Hence, linear tools like perpetual and futures do not work. To mimic the non-linear nature of IL, you use options. Options are non-linear as losses are capped at the price of purchasing the option (option premium) while gains are uncapped. Meanwhile, the bidirectional nature of IL and its shape can be mimicked and roughly hedged against by purchasing a basket of call and put options with different strike prices. However, a lack of liquidity in options could hamper this strategy. You can learn more . in this article There are other emerging methods of IL hedging, such as the creation of two-sided insurance markets for IL hedging. You can . Regardless of your hedging method, hedge only when your expected savings from hedging exceed the costs of hedging instruments used. learn more here LP token farming Since protocols need liquidity to function, many incentivize the provision of liquidity by creating farming opportunities for LP token holders. However, not all LP tokens have yield farming opportunities. In practice, this usually involves locking LP tokens in a vault or staking them and receiving other tokens overtime at a variable APR. The LP tokens and generated yield can then be withdrawn at a later date. Some dApps don’t make this a separate step. For example, Liquidity Stability Pool providers earn LQTY incentives automatically. Farming opportunities can usually be found on: The same dApp that provided you with the LP token (eg. DEXs like Pancake Swap have a farm feature where you can lock LP tokens for yield) Farming dApps that employ a range of different strategies to enhance yield (eg. Yearn Finance, Beefy Finance, Autofarm) dApps trying to grow their native token’s liquidity pool on a major DEX (eg. Alchemix provides rewards when you stake alUSD3CRV LP tokens from Curve DEX) Yields from LP token farming depend on many factors: APR/APY offered LP token staking pools and vaults usually cite an APY (annual percentage yield) and/or APR (annual percentage rate) figure. While both express yields earned on an annual basis, APR does not account for the effect of compounding yields while APY does. Hence, APR < APY. Less established dApps often offer higher APR/ APYs to attract liquidity in their early days. When considering APY, one should also understand the assumptions made in calculating it. APYs vary depending on the compounding frequency, with higher frequencies producing higher APYs. To produce attractive advertised APY figures, some vaults assume high compounding frequency in calculations while not actually compounding yields at that frequency (a bad market practice). The bonus is, therefore, on the user to manually compound at that given frequency to achieve the advertised APY. Manual compounding is cumbersome and costly, as transaction fees are incurred when withdrawing yields and adding them into the vault. 2. Farming reward type Yields generated from LP token farming often come in other and sometimes multiple token types. Hence, the value of your yields may change over time. Also, this can cause APR/ APY to fluctuate in line with the value of the reward token types. It is thus important to know what reward tokens you would obtain and if there are multiple token types to understand how yield is split between them. Common reward token types include: Native dApp tokens (eg. $CAKE in Pancakeswap farms, $AUTO in Autofarm vaults) L2 tokens (eg. sidechain Polygon provided $MATIC to LPs in Aave hosted on Polygon) Base LP token (farms like Beefy finance may automatically convert reward tokens to LP tokens regularly as part of their farming strategy) In short, preferably find farming opportunities that provide reward tokens that you are bullish on. If you are not bullish on the reward tokens, you may convert them to another token regularly. To make this economically sensible, your trade size should be large enough such that transaction costs do not eat into yields excessively, and you may want less volatile reward tokens that do not lose significant value between conversions. 3. Liquidity pool size and rewards supply The supply of reward tokens distributed is often fixed and follows a rewards distribution schedule. To attract early liquidity and incentivize early adopters, rewards are usually high initially and decrease over time. Moreover, as the pool size grows, rewards are distributed among more holders. This can lead to a decrease in APR/ APY and yields over time. Token Pair Profiles (Trade Offs) While we all want opportunities with high farming yields, high fees accrued, low risk, and low impermanent loss, they rarely exist. Even if they do, they only exist for short periods of time. The table below makes some broad generalizations, but I hope it illustrates some possible trade-offs liquidity providers have to consider: From the table, you can understand why small cap token-small cap token pairs rarely exist: naturally, there is little demand for such a pool from users in the first place and little incentives for liquidity providers. This means pools are small with higher slippage, further deterring any would-be users. When thinking about farming yields, it is useful to consider who is offering them and why. Token creators and dApps want their native token to be liquid and thus incentivize liquidity pairs in major DEXs between their token and major large cap tokens or stablecoins. DEXs compete with each other for liquidity and want to provide low slippage. Hence, they incentivize liquidity provision for pools that often have the highest natural demand: those between large cap tokens and stablecoins. Final thoughts Processing the many factors and trade-offs can be overwhelming for liquidity providers. Nevertheless, mapping it out brings clarity and structure to the thought process. While significant qualitative judgment is still needed, quantitative historical data could serve as guide rails for decision-making. With on-chain and public data, you could easily check out metrics related to the discussed factors. Have fun and all the best as you provide liquidity! I try to share thoughtful content and maximize signal:noise ratio. Catch me on the following platforms: . linktr.ee/gideon.tay

Polygon

Uniswap

PancakeSwap

Aave

Chain

Dapp

fees

Ethereum Tokenomics 2021: Impact of Eth2, EIP 1559, and L2 Scaling Solutions on Demand/Supply

Read My Stories

Check out linktr.ee/gideon.tay

Too Long; Didn't Read

DeFi Liquidity Providers: Factors Affecting Profitability, Trade-Offs, and Risk-Return Profiles

DeFi Liquidity Providers: Factors Affecting Profitability, Trade-Offs, and Risk-Return Profiles

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

Untitled Story

Binance Smart Chain's CeDeFi Is Not Likely To Rain On Ethereum's DeFi Parade

$1B Capitalization: Meme Token Launch Strategy Under the Microscope

(1/100) Crypto Countdown: Golem

06/09/2018: Biggest Stories in the Cryptosphere

07/09/2018: Biggest Stories in the Cryptosphere

Binance Smart Chain's CeDeFi Is Not Likely To Rain On Ethereum's DeFi Parade

$1B Capitalization: Meme Token Launch Strategy Under the Microscope

(1/100) Crypto Countdown: Golem

06/09/2018: Biggest Stories in the Cryptosphere

07/09/2018: Biggest Stories in the Cryptosphere

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps