Summary
- We demystify some of the sources of yield in DeFi and look at why stablecoin yields differ across DeFi protocols.
- Frax (FRAX) and Magic Internet Money (MIM) registered the largest decrease in market cap again. USD Coin (USDC) and Binance USD (BUSD) are still the only coins to register increases in market cap.
- FEI is still the most volatile stablecoin across a one-month to one-year horizon of those we track.
- Yields on stablecoins are up on the week, except Dai (DAI). Fiat yields are also up.
Where Does DeFi Yield Come From?
Traditional savings accounts typically offer interest rates under 1% now. Why, then, are yields in DeFi higher – double-digit interest rates are common, and sometimes we see much higher. But what is the source of this yield, and why do yields differ across different DeFi platforms?
The Different Types of Yield
There are several different sources of yield in the DeFi world. Some of the more robust sources include:
- Earning yield on staked coins – this is one of the most fundamental sources of yield in modern crypto frameworks. Validators (analogous to miners) earn a passive yield on their staked coins as compensation for validating and securing the blockchain. Ethereum is set to move to a proof-of-stake consensus protocol after the merge, which will replace the mining block rewards with yield on staked ETH.
- Lending rates – when you deposit your funds on a lending/borrowing platform like AAVE, borrowers will pay an interest rate to borrow those funds – you receive (a part of) the interest charged by the DeFi platform.
- Exchange fees – when you deposit funds to a DeFi platform/liquidity pool, you are providing liquidity to that pool. This enables trades to settle efficiently. The protocol charges a fee for trades made between pairs, and if you are a liquidity provider for that pair you receive a portion of that fee.
- Platform incentives – if you buy into the long-term vision of a DeFi platform, you may be inclined to stake a portion of the native coin with the platform. You may be rewarded for doing this. That is, a portion of the fees collected by the platform are redistributed to stakers to compensate them for their commitment to the platform.
There are more unstable sources of yield, too. You may have seen yields in the order of thousands of percent before – why is that? One explanation is that when launching a new platform/token, sometimes the makers will incentivise users to engage by rewarding liquidity providers with native tokens. The problem is when the issuance of incentives occurs too aggressively. This can be unstable and cause a collapse of the native token if too much selling occurs.
So we now know the sources of yield in DeFi. But why do yields differ across platforms if the sources are similar across them? To answer, we must first understand how interest rates are determined on DeFi platforms like Compound and Aave.
What Is the Utilisation Rate?
Most DeFi platforms rely on something called utilisation to determine borrow and lend rates on stablecoins (and other crypto). Utilisation is a metric that is defined for each stablecoin or cryptocurrency – it is the share of total reserves currently being borrowed for each currency. It can be defined as the total amount borrowed divided by the total liquidity for a currency.
Interest rates continually change based on fluctuations in the utilisation rate. When utilisation is low (little borrowing), interest rates are low to encourage more borrowing. When it is high (lots of borrowing), interest rates are high to encourage repayment of loans, thus injecting more capital deposits. When the utilisation rate is high, liquidity risk materialises. One way to look at the liquidity risk profile of an asset is its historical utilisation rate over time (Chart 1).
Why Do Yields Differ Across DeFi Platforms?
Take Compound and Aave as examples. They both have a similar interest rate model. There is an optimal value of the utilisation rate that is designed to manage liquidity risk by being calibrated to market conditions. Most markets follow a jump rate model in which interest rates increase with a shallow slope as the utilisation rate increases toward the optimal utilisation. When it exceeds this optimal level, the slope of the interest rate curve rises sharply (Chart 2). So, interest rates are changed relative to supply and demand. Therefore, differences in utilisation and perhaps the methods used to calculate/optimise it across DeFi platforms are one reason yields vary.
For example, on Compound, the optimal utilisation for USD Coin (USDC) is 81%. Currently, around $412mn is being borrowed (a 37% utilisation rate), which translates to a borrow rate of around 2%. Meanwhile, on Aave, the optimal utilisation for USD Coin (USDC) is 90%. Currently, around $1.2bn is being borrowed (a 50% utilisation rate), which translates to a borrow rate of around 2.3%. So, the two are at different locations (and proximity to the optimal utilisation rate) along their respective interest rate curves, which translates to a difference in yields.
Other reasons for the differences in yields across platforms include:
- Changes to the interest rate model parameters. When market conditions change, risks change, so the utilisation rate is monitored by each of the platforms to mitigate liquidation risks. One way of doing this is to adapt the model parameters which will of course cause changes to yields.
- Changes linked to the platform incentives. Some platforms, like Curve, will not only provide fee distributions and yield to users that stake the native token. They also incentivise further by giving stakers higher interest rates on all other liquidity pools too. The longer you lock up your CRV, the more yield you earn from staking, and the more platform fees you receive.
To summarise, the main differences in yield across DeFi platforms arise from differences in the utilisation rate, liquidity risks, and interest rate models, as well as changes in interest rate model parameters and more exotic platform incentives.
Latest Developments
Market Cap and Peg Risk
Frax (FRAX) and Magic Internet Money (MIM) have again seen the largest drop in market cap over the past 30 days (Chart 3). Between the two, around $2.4bn has been wiped from their market cap. USDD market cap is up the most at 174%, but this is because it was only released on 2 May.
All other stablecoins we track have continued to see their market cap drop over the past 30 days, except USD Coin (USDC) and Binance USD (BUSD). They registered a 11% and 5% increase in their market cap, respectively.
Volatility
The one-month annualised volatility of Fei USD (FEI) and Magic Internet Money (MIM) are now jointly the highest at 5% (Chart 4). USDD comes in at a close third with a one-month annualised volatility of 4%. The common thread between the three is that they are algorithmic or crypto collateralised stablecoins. Fei USD (FEI) remains far above the rest in terms of its three-month annualised volatility. Over the past year, Fei USD (FEI), Magic Internet Money (MIM), and Frax (FRAX) lead in terms of annualised volatility with values between 6% and 10% across the three.
Yields
Turning to yields, on Compound, average lending rates are below 2% for all stablecoins except Tether (USDT) (Chart 3). But they are all up on the week except for Dai (DAI). Similarly, average borrowing rates on Compound are all up on the week except Dai (DAI) (Chart 4). Tether (USDT) has the highest lending (2.2%) and borrowing rates (3.7%) on Compound. Looking across other decentralized finance (DeFi) protocols, we find that lending and borrowing rates are mixed across the stablecoins (Tables 2 and 3).
Appendix
USDT: Tether is a fiat-collateralised stablecoin primarily issued on the ethereum and bitcoin blockchains. It aims to be pegged 1:1 against the US dollar. Tether’s reserves are not backed 100% by US dollar deposits. Instead, they are backed by reserves that include cash, cash equivalents, short-term deposits, commercial paper, corporate bonds, funds, precious metals, secured loans, and other investments including digital tokens.
USDC: USD Coin is a fiat-collateralised stablecoin issued as ERC-20 tokens on the ethereum blockchain. It is 100% backed by cash and short-dated US treasuries. USDC publishes a monthly public attestation of 100% reserves.
BUSD: Binance USD is a fiat-collateralised stablecoin issued as ERC-20 tokens on the ethereum blockchain. It is backed 100% by USD held in Paxos-owned US bank accounts and US treasury bills (including through repurchase agreements and/or money-market funds invested in US treasury bills). Paxos is a New-York-regulated financial institution and publishes a monthly public attestation of 100% reserves.
TUSD: TrueUSD is a fiat-collateralised stablecoin issued by the TrustToken platform that is issued as ERC-20 tokens on the ethereum blockchain. It aims to maintain its 1:1 peg against the US dollar by being fully collateralised by US dollars using multiple escrow accounts to reduce counterparty risk.
USDP: Pax Dollar is a fiat-collateralised stablecoin issued as ERC-20 tokens on the ethereum blockchain. It aims to be pegged 1:1 against the US dollar by holding USD reserves in Paxos owned US bank accounts.
DAI: Dai is a crypto-collateralised stablecoin that attempts to maintain a 1:1 peg against the US dollar by depositing other crypto assets into smart contracts on the ethereum blockchain every time a new DAI token is issued. DAI is maintained by a decentralised autonomous organisation (DAO) called MakerDAO. And since the mechanism is maintained by a system of smart contracts, it has higher decentralisation than the centralised entities controlling USDT, USDC, or BUSD.
MIM: Magic Internet Money is a crypto-collateralised stablecoin launched by the DeFi platform Abracadabra. MIM is backed by interest-bearing tokens (ibTKN).
UST: TerraUSD is a crypto-collateralised hybrid stablecoin native to the Terra blockchain. To mint 1 UST, $1 worth of UST’s reserve asset, LUNA, must be burned. The idea was to try and ensure LUNA’s long-term growth. More people buying into UST means more LUNA gets burned, which should make the remaining LUNA supply more valuable. However, the system collapsed recently when UST de-pegged from the US dollar.
FRAX: Frax Finance is a fractional-algorithmic stablecoin that uses both collateralisation and an algorithmic process to create its decentralised stablecoin that is pegged 1:1 to the US dollar. Only stablecoins (currently, USDC) are accepted as collateral by the protocol.
FEI: FEI is an algorithmic stablecoin that aims to be pegged 1:1 against the U.S dollar that is backed mostly by ETH.
Dalvir Mandara is a Quantitative Researcher at Macro Hive. Dalvir has a BSc Mathematics and Computer Science and an MSc Mathematical Finance both from the University of Birmingham. His areas of interest are in the applications of machine learning, deep learning and alternative data for predictive modelling of financial markets.
Bilal Hafeez is the CEO and Editor of Macro Hive. He spent over twenty years doing research at big banks – JPMorgan, Deutsche Bank, and Nomura, where he had various “Global Head” roles and did FX, rates and cross-markets research.