- By
- – 01/27/2023
Introduction
Decentralized Finance (DeFi) was one of the first applications of blockchain technology to provide a real use case in the industry. Borrowing financial mechanics from traditional finance, DeFi offers users the ability to earn interest in exchange for their liquidity. DeFi’s “blue chip” protocols – a relative term in this instance given their short history – include Uniswap, Aave, Compound, and MakerDao. Others have emerged to fill gaps in the market, while many more were simply forks (or, copies) of more established DeFi protocols (see appendix for glossary of definitions).
Sufficient liquidity is a key element of DeFi protocols so that, for example, swaps can be executed at fair market value and tokens are available in lending applications to borrow. To achieve this, protocols frequently offer frontloaded incentives to early adopters, typically in the form of their native governance token – a similar approach to a credit card offering bonus miles for signing up and spending some minimum amount. Liquidity providers are rewarded with APYs and the protocol gains liquidity. While providing liquidity is not risk free for users, the risk-reward during periods of high APYs could be worthwhile.
“Yield farming” is a term that describes users moving liquidity from protocol to protocol in search of the highest APYs. It proved lucrative for early adopters during the DeFi peak in mid-2021, but the DeFi conversation has shifted to identify protocols that generate what the industry has now dubbed “real yield”.
Real yield reflects a given DeFi application’s intent to provide users with lower, albeit more stable returns. This is a promising shift in the industry. DeFi was generally designed to distribute earnings to market makers equitably. The current crypto bear market – with the total crypto market cap falling by more than 70% since all-time highs in November 2021 – is a major stress test for DeFi protocols. In this paper, we explore how many of the top DeFi protocols are faring to better understand the direction of the industry.
Covered DeFi Protocols
This analysis focuses on the top DeFi protocols as defined by market cap and total value locked. Protocol categories include the following
- Decentralized exchanges (DEX) – enable token swaps through automated market makers (AMM), which use algorithms to determine exchange rates between tokens
- Lending and borrowing – allows users to earn interest by supplying liquidity to the protocol, and users can borrow against their own collateral
- Collateralized debt position (CDP) – allows users to mint stablecoins (i.e., MakerDAO’s stablecoin, DAI) by providing overcollateralized loans
- Yield aggregator – works across multiple protocols to maximize yield for liquidity providers
- Derivatives:
- Perpetual contracts – liquidity providers can earn yield and traders can speculate on future price action
- Synthetics – token representations of real-world assets on the blockchain
In each category above, protocol users are rewarded in some way by providing liquidity that enables the protocol to execute its core features. DeFi mechanics are modeled after traditional finance but can work without clearing and custody intermediaries through smart contract code – a set of functions performing specific actions. When designed well, guardrails limit a smart contract’s potential vulnerabilities. Table 1 provides an overview of the DeFi protocols (and their tokens) and relevant metrics explored throughout this paper. These protocols were selected based on several factors, including their coverage by Sherlock, market size and popularity, and resilience during market turmoil.
Protocol | Category | Market Cap ($Million) | Fully Diluted Value ($M) | TVL ($M) | Price/ Sales | BTC Relative Performance, 1Y | Commits, 1Y | Unique Developers, 1Y |
---|---|---|---|---|---|---|---|---|
Uniswap (UNI) | DEX | $3,998 | $5,300 | $4,535 | 31.4 | -13% | 2868 | 355 |
Aave (AAVE) | Lend/ Borrow | $744 | $837 | $3,244 | 14.0 | -44% | 2536 | 70 |
MakerDAO (MKR) | CDP | $517 | $2,385 | $2,159 | NA | -25% | 2740 | 53 |
Pancakeswap (CAKE) | DEX | $ 458 | $ 511 | $ 5,922 | 44.2 | -40% | 5601 | 124 |
Osmosis (OSMO) | DEX | $372 | $731 | $155 | 168.9 | -70% | 6996 | 204 |
Synthetix (SNX) | Derivatives | $362 | $ 466 | $257 | 64.1 | -27% | 4257 | 139 |
Curve (CRV) | DEX | $349 | $550 | $464 | 11.9 | 412% | 2715 | 12 |
GMX (GMX) | Derivatives | $341 | $1,740 | $3,618 | 69.5 | -76% | 1749 | 53 |
dYdX (DYDX) | Derivatives | $213 | $311 | $ 1,376 | 18.2 | -58% | 323 | 19 |
Compound (COMP) | Lend/ Borrow | $204 | $510 | $ 1,509 | 89.9 | -18% | 3297 | 55 |
Sushiswap (SUSHI) | DEX | $184 | $239 | $ 380 | 8.1 | -70% | 5779 | 38 |
Balancer (BAL) | DEX | $ 175 | $ 196 | $ 358 | NA | -58% | 4196 | 133 |
yearn.finance (YFI) | Yield aggregator | $ 162 | $ 1,140 | $ 397 | NA | -61% | 100 | 20 |
Gains Network (GNS) | Derivatives | $ 77 | $ 77 | $ 19 | NA | 171% | NA | NA |
Trader Joe (JOE) | DEX | $45 | $69 | $ 56 | 39.1 | -81% | 1393 | 57 |
Protocol Versus Token
Token economics, commonly known as “tokenomics”, refers to the utility and supply distribution of a given blockchain or protocol token. Note the distinction between the protocol and the token. For example, Aave the protocol is a set of smart contracts that enable lending and borrowing of assets. Aave’s token (AAVE) is used for governance – token holders can vote on proposals that shape the direction of Aave – and can be distributed as a reward to users supplying liquidity.
From the supply perspective, distribution refers to the allocation of tokens. Initial token distribution may include founders, developers, and investors (i.e., VCs). The initial distribution is used to reward early contributors to the project. Subsequent distribution is known as the “token emissions schedule”. Most tokens have some degree of annual inflation, which is often frontloaded to reward public token holders for supplying liquidity. In Table 1, Fully Diluted Value (FDV) is a useful metric to determine outstanding token supply. All but one of the protocols covered in this report (GNS) have a fixed maximum supply that has not yet been reached. Notably, Pancakeswap, dYdX, and Curve have less than 20% of their maximum supply currently in circulation. These tokens will enter the market over a predetermined time period – likely as rewards for providing liquidity – causing inflation.
The distinction between token and protocol is also relevant when considering price action. Even the DeFi blue chip token prices are down over 85% since their all-time highs, with lesser used protocols down upwards of 95%. The value of a protocol is often determined by its token price rather than fundamentals like usage and innovation, which is the focus of the remaining sections in this report.
DeFi Deep Dive
This section explores the value proposition of DeFi protocols mentioned above. We explore several metrics to gain insight into the sustainability of the top DeFI protocols and, by extension, DeFi as an industry. These metrics are distributed into the following: protocol revenue, total value locked, rewards for liquidity providers, token price action, token contract interactions, and GitHub development activity.
Protocol Revenue
Figure 1 displays Total Revenue and Price-to-Sales (P/S) ratio for DeFi protocol tokens available on Sherlock. Revenue is generated through fees1 and accrued interest depending on protocol design. For example, Uniswap charges fees (typically 0.05 - 0.3% of trade value) for token swaps, which are distributed to liquidity providers. Aave charges fees to deposit or swap tokens, and interest for borrowing assets. In addition to liquidity providers, Aave also distributes a percentage of revenue to the Aave treasury.
The P/S ratio can indicate a token’s value similar to how price to earnings (P/E) helps measure traditional assets. Those with lower ratios may be an indication that a stock is undervalued, and vice versa. Price in the P/S ratio is defined as an asset’s fully diluted value (FDV, where available), which is calculated as the current token price multiplied by the maximum number of tokens that will ever be created. For example, there will only ever be 21 million bitcoin so bitcoin’s FDV would be its current price multiplied by 21 million. Many assets have not specified their maximum token supply. In these instances, the current market cap is used as the price numerator. In Figure 1, however, all digital assets have predetermined maximum supplies and therefore we are using FDV in the P/S ratio. Sales is the annualized total revenue of the previous seven days generated through transaction fees, accrued interest, or liquidated positions.
Figure 1 shows that Uniswap had significantly higher annualized revenue than any other protocol included in this analysis at roughly $171 million. Given a FDV of $5.47 billion, Uniswap had a P/S ratio of 31 at the time of writing. GMX, Sushiswap, Aave, and Trader Joe also had P/S ratios below 15.
Total Value Locked
Total Value Locked (TVL) is a crypto-specific term for how much money is deposited, or “locked”, in a protocol. Protocols with higher TVLs can indicate their “stickiness” – essentially user loyalty and trust to deposit significant sums. Popular reasons to deposit funds in these DeFi protocols are to earn yield by providing liquidity and borrowing against collateral. Many of the first mover DeFi protocols, like MakerDAO and Uniswap have the highest TVLs shown in Figure 2. These protocols have survived several stress tests during periods of significant market growth and shrinkage.
A downside to TVL is that the metric is typically denominated in USD; thus, TVL volatility is highly sensitive to token prices. The ETH-USD exchange rate is down 75% since its all-time high in November 2021, but the volume of ETH locked in DeFi protocols is roughly unchanged over that time.
Protocol Rewards
Despite a significant decline in USD-measured TVL for the industry overall, DeFi APYs have remained steady, on average. This is in part because APYs are typically annualized forward-looking estimates based on past 24-hour or 7-day trading volume. During periods of volatility, increased transaction volume results in higher protocol fees and, by extension, APY estimates. Figure 3 shows daily TVL-weighted average APY for a subset of DeFi protocols. We set a TVL minimum of $1 million for pool data in Figure 3 to remove noise since many smaller pools with newer, unproven tokens may offer unsustainably high APYs. Inclusive of rewards, the TVL-weighted averaged APY across the protocols and pools in our subset was 5.4% over the past six months. The average APY without TVL weighting is higher because of incentives to pools with lower liquidity.
There are two APY spikes evident in Figure 3, which are reactions to significant industry events. The first being a result of Terra’s UST depegging and related LUNA token crash; and the second is the recent fallout stemming from FTX insolvency. Exploring the data more closely, DEXs account for nearly all the APY increases. Protocol users are likely swapping assets or withdrawing liquidity at much higher volumes than ordinary. Liquidity pools with stablecoins (i.e., WETH-USDC pair) also experienced higher increases in their APYs (driven by volume and fees) during volatility relative to those pools without a stablecoin.
While Figure 3 displays the daily average APYs, Figure 4 details the APY distribution by pool within each protocol. Each pool’s APY was averaged over the past 90 days to smooth daily volatility, and the data in each protocol’s Box and Whisker plot shows the range of pool APYs. Protocols can host any number of pools, and protocol category largely determines the quantity of pools. For example, lending and borrowing protocols like Aave and Compound have fewer pools and lower APYs because demand for borrowing top tier crypto assets like wrapped versions of BTC and ETH far outweighs the demand for lesser-known assets. DEXs, on the other hand, like Uniswap and Osmosis have many more pools to support token swaps; they also feature a much wider range of APYs.
Market Cap Relative to BTC
The market cap for the entire crypto industry is down roughly 70% from its all-time high, as measured in USD. However, token prices relative to BTC – the industry leader by a number of measures – can indicate how a protocol and its token fare on a level playing field. Since the beginning of 2022, most DeFi tokens have underperformed relative to BTC (Figure 5). This is typical of altcoins during bear markets, albeit based on the short history of the industry. Uniswap’s UNI token performed the best out of those presented in Figure 5, losing 12% of its value against BTC. Trader Joe (JOE), Curve (CRV), and Sushiswap (SUSHI) lost 65-80% relative to BTC during this period.
A subcategory of DeFi that has outperformed BTC (and ETH) is perpetual trading, which allow users to trade futures with leverage and no fixed time-period. Although relative performance varies widely between the specific protocols in this subcategory, the top three perpetual trading tokens by market cap (DYDX, GMX, and GNS) outperformed BTC by over 200% year-to-date, on average. These tokens were excluded from Figure 5 because they would greatly skew the y-axis scale.
Token contract interactions
Figure 7 displays the number of unique wallet addresses interacting with token contracts in a 24-hour period, with a 7-day moving average applied to smooth the data. Interactions with token contracts include activities such as sending tokens to a different wallet address, swapping on DEXs, and depositing tokens in liquidity pools.
Figure 3 shows an average overall decline of roughly 25% since our anchor date (January 1, 2022). Token contract interactions show noticeable spikes during significant negative events – Terra and FTX crashes in May and November, respectively. Some likely reasons could be users moving their assets off centralized exchanges and into self-custody and increased DeFi activity given market volatility.
Development activity
GitHub code contributions can provide insight into project innovation and interest among developers. We created scatterplots (Figure 7) that chart development activity relative to FDV (or market cap where FDV is not available) and added a linear trendline to approximate patterns. With a limited subset of assets, the trendline suggests a slight positive correlation between development activity (for both unique developer count or commits) and token value.
FDV (or market cap) is used on the vertical axis on both charts. The horizontal axes differ, however, with the left chart being unique developers in the past year (from January 2, 2022 to January 2, 2023) and the right chart unique commits in the past year. Notably, Uniswap is excluded from this chart because its coordinates skew chart scaling due to its significantly higher market cap of roughly $4 billion at the time of writing. Uniswap also had 355 unique developers over the prior year; its commits in the past year were in the mid-range at 2868.
Figure 8 also explores monthly commits by project. There has been an overall decline from early 2022 based on the subset of protocols we explored, though it is not evenly distributed across all protocols. Since the summer, overall commit activity has been flat.
Table 2 also breaks down the data in Figure 8 into protocol-level descriptive statistics. Osmosis (OSMO) had the highest commit activity with 529 average monthly commits; Osmosis also observed growth from January to December – 1 of just 5 protocols to do so. We want to note that, compared to every other protocol included in this code contribution analysis, Osmosis is the only DeFi application that is also a Layer 1 blockchain.2 Therefore, developers must maintain the blockchain infrastructure beyond the typical features and upgrades of a DeFi protocol’s development.
Some protocols, namely Aave (AAVE) and Trader Joe (JOE) had high variance in commit activity month-to-month, as is evident by their standard deviations being nearly equal to the mean monthly commit values. Such a high variance can skew the Jan-Dec change column in either direction.
Protocol/Token | Mean | Std Dev | Min | Median | Max | Jan-Oct change |
---|---|---|---|---|---|---|
OSMO | 592 | 281 | 339 | 530 | 1383 | 0.49 |
MKR | 485 | 271 | 149 | 400 | 1147 | -0.87 |
SUSHI | 483 | 340 | 237 | 407 | 1444 | -0.80 |
SNX | 361 | 115 | 170 | 350 | 533 | 0.10 |
YFI | 357 | 189 | 169 | 284 | 702 | -0.43 |
BAL | 284 | 88 | 191 | 265 | 520 | 0.15 |
UNI | 244 | 108 | 86 | 256 | 458 | 0.16 |
CAKE | 231 | 58 | 138 | 230 | 338 | 0.69 |
GMX | 231 | 110 | 35 | 256 | 409 | -0.89 |
AAVE | 216 | 243 | 57 | 82 | 639 | -0.86 |
CRV | 149 | 61 | 82 | 118 | 256 | -0.66 |
JOE | 123 | 125 | 31 | 94 | 493 | -0.94 |
COMP | 27 | 24 | 0 | 23 | 80 | -0.83 |
The Case for DeFi and Potential Risks
If the events in this industry over the past six months have taught us anything, it’s that the core ethos of crypto (decentralization and transparency) is more important than ever.
DeFi matters because it minimizes the human variable and the need for extensive regulation. The term “code is law” refers to the deterministic actions that smart contracts enforce, effectively providing a self-governing framework. Other elements of blockchain technology promote decentralization and transparency. In each of the instances mentioned above, a centralized authority acted in self-interest; decentralization can reduce this risk, and transparency of code and on-chain transactions promotes trust in the system. There is no need for a protocol to provide “proof of reserves” – as many exchanges have been pressured to provide recently – because the smart contract holdings can be queried.
The flipside to this argument is that DeFi has experienced its own share of problems, namely in the form of scams and hacks on code vulnerabilities. According to DefiLlama, these accounted for over $3.2 billion in lost funds in 2022.i Rekt, a site that tracks DeFi and crypto hacks and significant losses, estimates $2.78 billion in losses in 2022.ii Some of the most prevalent reasons for lost funds include compromised private keys, bridge vulnerabilities, and rug pulls.3, iii Although developers are addressing common vulnerabilities and users are becoming savvier with respect to potential risk factors, billions in lost funds from crypto-native applications could serve as fuel for regulatory scrutiny. It is not our place to speculate on regulatory guidance in this space, but we do acknowledge the potential impacts.
Conclusion
Data presented in this paper indicate that, collectively, DeFi is weathering the market downturn. Protocols have generated significant amounts of revenue and share it with liquidity providers; users have been able to earn attractive returns through protocol participation; and development activity remains strong. The pace of innovation is also impressive. Protocol logic and efficiency, front-end design, and user experience are expected to continue improving, ultimately providing a more robust user experience. Individual protocols will come and go and token price action is still largely driven by speculation, but the technology has shown its worth and resilience during a period of market volatility. Pending regulation aside, the promise of blockchain, smart contracts, and DeFi should not be lost on account of the current market downturn or human error, deliberate or otherwise.
Glossary of DeFi Terms
DeFi – Decentralized Finance is a set of applications built on blockchain technology that operate without the need for intermediaries or traditional financial institutions. DeFi has taken many of the financial mechanics of traditional finance (lending, borrowing, trading, etc.) and enabled their functionality with greater speed, scale, and autonomy using smart contracts.
DeFi Protocol – a set of smart contracts performing financial functions without an intermediary.
Total Value Locked (TVL) – the sum of assets deposited into a given protocol.
Fully Diluted Value (FDV) – a forward-looking market cap for a given token once all the tokens are minted. Calculated as the current price multiplied by the maximum token supply. Ass ets without a hard cap token supply cap do not have a FDV.
Liquidity pools (LP) – a collection of tokens locked in a smart contract that provide essential liquidity to perform financial functions.iv
Important Information
The information herein was prepared by Sherlock Analytics. It is for informational purposes only and is not intended to constitute a recommendation, investment, tax, accounting or legal advice of any kind, or an offer or the solicitation of an offer to buy or sell securities or other assets. Please perform your own research and consult a qualified advisor to see if digital assets are an appropriate investment option.
This information is not intended for distribution to, or use by, any person or entity in any jurisdiction or country where such distribution or use would be contrary to local law or regulation. Persons accessing this information are required to inform themselves about and observe such restrictions.
Digital assets are speculative and highly volatile, can become illiquid at any time, and are for investors with a high-risk tolerance. Investors in digital assets could lose the entire value of their investment. Some of this information is forward-looking and is subject to change. Past performance is no guarantee of future results. Investment results cannot be predicted or projected. Sherlock, the Sherlock Logo are service marks of FMR LLC. © 2021 FMR LLC. All rights reserved.
1. Fees can be either gas fee or protocol fee. A gas fee is the amount of ETH (or another token, depending on the blockchain used) charged by the network to process a given transaction. It is a payment for computational services and an incentive to those individuals running the network. Gas fees are determined algorithmically based on network congestion (i.e., how many other users are transacting). A protocol fee may be charged by a given DeFi protocol, such as Uniswap, for using their application. Depending on the transaction type, users may pay a gas and a protocol fee in order to execute their transaction.
2. Osmosis is built with the Cosmos SDK, a modular blockchain framework developed that is part of the Cosmos ecosystem. Cosmos supports an “app chain” thesis whereby blockchains are application-specific yet can seamlessly interoperate with one another. Similarly, DyDx announced in the summer that they are migrating from an Ethereum-based dApp to their own blockchain using the Cosmos SDK.
3. A rug pull occurs when a team holds a disproportionate share of their own token, pumps the price, sells, and disappears with the funds
Sources
i DefiLlama. Accessed November 15, 2022. https://defillama.com/hacks
ii Rekt. Accessed November 15, 2022. https://rekt.news/leaderboard/
iii Rosie Perper, Coindesk. “What Is a Rug Pull? How to Protect Yourself From Getting ‘Rugged’”. August 30, 2022. https://www.coindesk.com/learn/what-is-a-rug-pull-how-to-protect-yourself-from-getting-rugged/
iv Mike Antolin. “What Are Liquidity Pools?”. November 16, 2022. https://www.coindesk.com/learn/what-are-liquidity-pools/