Sources of Yield with Digital Assets
In the traditional capital markets, return on investments comes from two sources – price appreciation and/or yield. Price appreciation (or loss) comes from the realization (or disappointment) of growth prospects compared to discounted growth expectations. Yield comes from the income generating capacity of the entity into which the capital was invested – be it an equity stake in an enterprise, a piece of real estate, or a debenture. Presumably, that entity was doing something productive with the capital it was given to have the capacity to generate income.
We should expect no less from the digital asset protocols and projects that we invest in. Many protocols, such as Bitcoin, don’t provide a yield so their valuation is simply based on the social consensus of the value of its network effect, perhaps proxied by Metcalfe’s Law. Others do generate a yield. Let’s dive into what productive ends capital is used for and where these yields come from.
TVL and the State of DeFi
One of the characterizations of a healthy financial market is the free flow of capital across jurisdictional borders, the ease by which assets may be lent and borrowed, and the free interchange of assets among market participants to promote price discovery. In TradFi (traditional finance), prime brokers are important agents who provide access for assets to be borrowed for leveraging and shorting. Customers (lenders and borrowers) engage with the prime brokers through legal agreements and take their assets to the exchanges to construct their preferred portfolios.
In DeFi (decentralized finance), protocols play the role of prime brokers and exchanges, while smart contracts replace the legal agreements. In a short span of just over 5 years, DeFi activities have been bootstrapped from nothing to around $50 B, as measured by the total value locked (TVL) of the various assets on the DeFi protocols.
While one may be tempted to look at the chart and write off the 70% drop in TVL as lack of interest for DeFi activities, one needs to recall the TVL is measured here in USD and the dollar value of cryptoassets has fallen by a similar 65% over the same period. Measured in quantities of the cryptoassets, DeFi activities remain vibrant.
Lending and Borrowing
The top 22 protocols capture about 75% of the DeFi TVL and of these, over half of the TVL are placed on lending platforms (i.e., the prime brokers of DeFi) while the rest of the protocols are mostly DEXs (decentralized exchanges).
Top lending protocols include Aave and Compound on Ethereum and Justlend on Tron. A peek onto the Aave platform indicates the top assets being borrowed are eth (and its wrapped variations), stablecoins, and wrapped btc. Yields received on supplying assets are sensibly lower than those paid to borrow assets.
Generally speaking, hedging and arbitraging price discrepancies are the main drivers for investors to borrow assets. We described an example of an arbitrage opportunity that allows an investor to capture about 2.6% over a window of about a month by buying underpriced cbeth (liquid staking derivative of eth issued by Coinbase) using borrowed eth. At a borrow rate of 3.61% per annum or 30 bps per month, that is well worth the trade, especially if the position is levered up.
Liquidity as a Service
Most TradFi and centralized digital asset exchanges use central limit order books (CLOBs) to trade assets. Limit orders of what quantities to buy or sell and at what price are aggregated and ordered by price. A market order to buy (sell) an asset is matched against the lowest (highest) prices and quantities of the sell (buy) limit orders.
In contrast, many digital asset DEXs, such as Uniswap or Sushiswap, use automated market making (AMM) that relies on a pricing or a bonding curve, obviating the need for CLOBs. While bonding curves can take on different forms, a typical one is X * Y = k, where X and Y represent the quantities of assets X and Y, and k is simply a constant. Liquidity providers provide quantities of assets X and Y, in equal monetary value, into the AMM. Traders can withdraw quantities of X (or Y) but pay for it in quantities of Y (or X). Withdrawals of larger quantities of X require paying exponentially larger quantities of Y, implying the marginal purchase of X is becoming increasingly more expensive.
For example, to add liquidity to an eth/usdt (tether stablecoin) pool, and assuming the value of eth is 1,600 usdt, the liquidity provider can choose to deposit 1 eth and 1,600 usdt tokens. This provider is awarded liquidity pool (LP) tokens to represent her share of the liquidity pool, earn a proportionate share of the trading fees, and can be used to withdraw her funds from the pool at any time.
Liquidity providers are subjected to the risk of an impermanent loss due to prices moving away from the prices at which the liquidity is provided. The loss is temporary and disappears when pricing reverses but is realized when liquidity is withdrawn at adverse prices.
Staking for Consensus and Security
Proof-of-Stake (PoS) is one of the two major consensus mechanisms used to secure a blockchain, with the other being Proof-of-Work (PoW). Ethereum, after the Merge, is the largest chain that uses PoS.
With PoS, token holders can participate in the network’s consensus mechanism by staking their tokens to act as validators to approve and add transactions into new blocks. In doing so they earn a reward for securing the network. The larger number of diverse validators there are, the more decentralized and secure the blockchain becomes. Bad actors, who try to post invalid transactions or irresponsible ones who stake but don’t participate in validation, risk having their tokens taken away from them, i.e., slashed.
Staking can be done either by running a node on the network or by delegating the task to a trusted staking service provider like Lido Finance (decentralized) or Kraken Exchange (centralized). Rewards may include newly minted tokens and a portion of transaction fees in the new block that the validator creates. The table below shows the current staking yield across various PoS blockchains but these yields can be negated by the tokenomics of the blockchain. Inflationary token supplies lower the effective yield while deflationary supplies have the opposite effect.
Which came first, the token or the protocol?
Designing a DeFi protocol is half the problem. The other half is attracting enough market participants to create an active market. Asset owners are needed to supply assets on DeFi lending platforms for borrowers and liquidity providers are needed to provide pools of asset pairs for traders on DEXs. To bootstrap their platforms, protocol designers have turned to issuing governance tokens that reward lenders and liquidity providers. Compound Finance, an asset lending platform, issues the COMP token to asset lenders in addition to the cTokens which are certificates representing claims on the loaned assets. Similarly, Curve, a stablecoin DEX, issues the CRV governance tokens to stablecoin liquidity providers. The CRV token allows the holder to participate in determining the amount of fees to charge and to receive a portion of those fees.
Double-, Triple- Digit APYs?!
When sovereign interest rates were still close to 0% in the middle of 2021, the hunt for yield was intense. That has since abated with high quality debt such as the US 3-month Treasury Bill yielding over 4.6%. Digital asset yields from staking, lending, or providing liquidity seem less attractive in comparison. Nonetheless, yield aggregators such as Convex or Yearn can generate healthy yields in the double-digits and sometimes triple-digits by taking advantage of the composability of DeFi to harness economies of scale. These strategies range from pooling disparate assets to lock in multi-year liquidity pools to earn larger rewards in governance tokens, to using leverage to take advantage of mispriced lending and borrowing rates across different platforms. The exact mechanisms of how these eye-popping yields are achieved are beyond the scope of this discussion.
Wait, Yields in the Same Token? Risks Abound
Not all yields are created equally and those from digital assets carry with them their fair share of risks. A fundamental law in finance is that there is no free lunch so extra yield, if found, needs to be more closely scrutinized.
In DeFi, smart contracts replace legal contracts. These are self -executing programs that govern the rules and logic of a DeFi protocol. Smart contract risk arises when there are vulnerabilities in the code that could be exploited by attackers, resulting in the loss of funds, theft, or other damages to DeFi users and platforms.
Staked positions on PoS based chains incur the risk of being slashed. Slashing is intended to incentivize validators to follow the rules of the consensus mechanism while penalizing dishonest or malicious ones. Some common activities that slashing tries to avoid are double-signing (validating two different blocks at the same height leading to a fork in the blockchain, thereby causing confusion and security concerns), creation of invalid blocks, and network downtime due to validators going offline.
Perhaps two of the biggest risks in trying to earn yield with digital assets is the token economics (tokenomics) of the reward and the leverage employed to generate the above “normal” yields. Yields are often paid with the same token as the asset being lent, staked, or pooled, or with the protocol’s governance token. It is not paid in the investor numeraire currency, e.g., USD for US investors. Investors will need to exchange the reward tokens for their local currency to be able to purchase real world goods and services, thereby incurring price risk. A close analogy is owning shares of Apple Computer and getting paid dividends, not in USD, but in more Apple stock. This share dividend payment is effectively a stock split, not changing the intrinsic value of the shares owned. If anything, those who do stake and earn token rewards are effectively getting a larger share of the same pie at the expense of those token holders who don’t stake their tokens.
Finally, juiced up yields typically employ some form of leverage to reach those double-digits. Rehypothecation, the act of using the same collateral for different loans, is typically not possible as collateral is locked in a smart contract without any access until the loan is liquidated. Nonetheless, the build-up of leverage in the system is real and excessive levels can lead to a cascade of rising prices during the good times, and a vicious cycle of falling prices in a bear market. A large mitigating factor is that DeFi protocols are generally safer than centralized protocols as these leveraged positions are typically over-collateralized unlike many of the centralized lending platforms that were under-collateralized with loans being extended based simply on a handshake.
In a short span of 5 years, vibrant markets for lending, staking, and liquidity provisioning have been developed in DeFi that have allowed digital asset owners to earn market-based yields on their portfolio. Yields are important not just for earning income but also to properly price the assets and assess their associated risks. Yields form a key component of the asset valuation process, a topic for another time.