In the previous blog post, we show that DEX adoption is an increasingly prevalent phenomenon and that Uniswap is the market leader based on volume. However, a significant majority of the order flow on Uniswap is informed (also known as toxic flow), which begs the question: why are LPs providing liquidity to toxic order flow, and are they profitable doing so?
For many years, market making has been a test of speed and information — participants aim to minimize their adverse selection and manage a carefully selected inventory, requiring them to operate in the order of milliseconds for volatile markets. While blockchain technology lends itself to open participation, it falls far short to HFT-grade market making technology when assessed on latency. Correspondingly, the automated market maker (AMM) was designed to meet the limitations of a 10-TPS blockchain, as AMM’s are computationally efficient and have minimal storage needs.
While intuition indicates that the complex matching engine of an order book should theoretically create the most efficient market, we observe that the deep liquidity in AMMs can provide an attractive trading venue for rational traders. As shown in the figure below, with the advent of concentrated liquidity, Uniswap has had higher market depth and thus more competitive pricing compared to its centralized counterparts at every price level.
Since spreads are often under 6bps for stable pools — which makes it difficult for market makers to compete against due to the low margins of profitability — we observe that DEXs have much deeper liquidity compared to centralized exchanges with more sophisticated LPs. For instance, Uniswap v3 has about ~5.5x more liquidity than Binance on its USDC/USDT pool.
AMMs were initially designed to thrive in pricing the long tail of cryptocurrencies, where the slippage of the trade corresponds to the bid-ask spread on a traditional order book. This idea has been quite successful, as new projects are able to bootstrap liquidity for their tokens in minutes without having to work with a centralized market maker. In this way, projects can incentivize their own community to provide liquidity through various incentive mechanisms. In many ways, these incentives correspond to the fees that would be taken by a traditional market maker, but instead distributed to the community. Some DEXs also use call options and time-based incentives as a means to bootstrap volume for LPs, while reducing the amount of sell pressure on the project’s token from profit-seeking yield farmers.
Since Uniswap attracts traders due to its competitive pricing and strong brand, we remain confident that there will be volume to attract LP’s. However, given the scale of Uniswap’s deep liquidity, and the pillar of quantitative finance outlining the inverse relationship of a strategy’s scale and Sharpe Ratio, we now delve into the profitability of the LP.
As mentioned in our Beginner’s Guide to LPing, a sizable portion of TVL on DEXs comes from LPs who have parked their funds into stable-stable pools. While the portfolio risk is low due to the stable payoff, the risk-adjusted return may be severely underperforming the current risk-free cost of money (i.e. treasury bills). While many stableswap protocols have handled hundreds of billions in volume, there is inherent risk due to smart contract security on a permissionless and non-custodial protocol. This risk is particularly enhanced on side-chains and L2s that do not have battle-tested code. It is also important to note that the risk-free cost of money of crypto may not be comparable to that of traditional finance, and instead of comparing to treasury bills, a more suitable juxtaposition would be with a crypto-native yield source such as liquid staking.
Moving on from stable pools, which offer a low risk and low yield solution, we now analyze the traditional LP into blue-chip pools (ETH-USDC, WBTC-USDC, SOL-USDC, etc.). The profitability of LPs on DEXs is a complex topic due to several factors, including impermanent loss and adverse selection. Impermanent loss arises when the prices of the two tokens in a liquidity pool fluctuate relative to each other. This can result in losses for LPs when they are forced to rebalance their portfolio by buying more of one token and selling their existing token holdings in order to remain compliant with the automated market maker’s pricing function. Furthermore, adverse selection happens when traders place orders that are designed to exploit LPs’ stale quotes, attempting to profit through arbitrage strategies. As such, these two issues can reduce the profitability of LPs due to the additional costs incurred when mitigating these risks. Many protocols often airdrop incentives to LPs to offset these risks, but this has proven to be a temporary patch until the underlying mechanisms are improved — either the LP or the underlying AMM need to guard against the forces of IL and MEV in order to create a robust and scalable market. However, despite these drawbacks, LPs on DEXs still tend to be profitable if they are able to properly manage their positions.
We believe products that hedge market exposure, rebalance LP positions, and offer the ability to sell LP positions as options, can significantly improve LP profitability. The final portion of this blog post gives further insight on each of these issues and identifies a few of the products that are attempting to solve them.
While depositing both assets of a pool is the default route, LPs may choose to craft a more nuanced position. In this section, we explore the PnL of a delta-neutral position, which is crafted by using the LP position as collateral to borrow the same amount of alt token that was deposited into the position. Before fees, the payoff is represented by the dark blue curve, while traditional LP’ing is the green curve:
Delta-neutral LPing is an innovative trading strategy that has been gaining in popularity as a way to help LPs maximize their profitability. It is based on the concept of hedging risk by maintaining a neutral Delta position — meaning that the LP’s position is equal in both long and short positions, and they are not taking a directional bet on markets. By maintaining a delta-neutral position, LPs can effectively hedge against the volatility inherent in markets, allowing them to take less risk while still maintaining the potential for profitable trades. As part of this strategy, the LP will use market data to establish the correct levels of long and short positions in order to stay in a neutral delta state. While derivative markets to achieve this payoff can be the most capital efficient mechanism, most DeFi users often use spot markets due to the higher liquidity and durability of the protocols. The LP will also regularly review their position and make necessary adjustments to take advantage of changing markets. This allows them to capitalize on potential profit opportunities and minimize losses should the markets turn against them. By using this strategy, LPs can increase their overall profits without having to take on more risk.
Below is a backtest of historical performance for the delta-neutral strategy on Uniswap’s ETH-USDC pool from June 1st, 2022 to November 18th, 2022. The simulation shows a 7.8% return, which corresponds to an annualized return of 16.75%. It should be noted that this backtest simply shows how a naive delta-neutral payoff would have performed, and does not incorporate any advanced rebalancing strategies that can be offered to further minimize drawdowns.
As shown above, traditional LP’ing is primarily profitable when the underlying asset has netted a positive return, while the delta-neutral strategy can be profitable in market drawdowns of 20–40% across a certain time horizon.
OpenBlock is a platform that empowers Web3 organizations with the data intelligence needed to take action and drive growth. From managing grants to optimizing incentive spend, leading DAOs and blockchain networks utilize this technology to maintain billions of dollars in value within their ecosystems.
If you are interested in working with OpenBlock Labs, please visit us at openblocklabs.com or reach out via Twitter at @openblocklabs.
