Where do crypto yields come from & will they last?
Answering the question we all want to ask - is DeFi one big ponzi scheme?
Welcome to the 92 new people who took the red pill since last article! If you haven’t subscribed yet, join our growing community by subscribing here:
I have to admit, crypto yields have for the longest time confused me. When I see APYs of more than 100%, my first thought is - where is the catch? Second thought - this can’t be sustainable. Having grown up in one of the lowest interest rate environments in recent history, I’m accustomed to earning nothing on my bank balances. You have to look back to pre-2000 for the last time we had any significant interest on consumer balances (>5% pa) and we’ve spent the last 20+ years dealing with largely negative real interest rates once you adjust for inflation.
With that backdrop, DeFi yields appear to many like a giant ponzi scheme rather than a sustainable financial ecosystem. This perception has only been exacerbated with the recent turmoil surrounding Luna and the general pullback in DeFi volumes. At the previous peak, we saw over $100bn locked in DeFi contracts which suggests a large number of people are willing to support this ecosystem and believe in its longevity. With arguments on both sides of the table, it prompted me to step back and question - where do crypto yields come from, and are they sustainable?
Understanding the status-quo, where your TradeFi interest comes from
Before unpacking crypto yields, it is important to understand where traditional bank interest rates come from, to get a sense of what a sustainable market structure looks like.
In general, commercial banks are free to set interest rates at whatever they would like. The rate you receive on your balance is a function of competitive forces (to attract customers away from rivals) and the margin commercial banks make on their capital. Central Banks play an important role in this ecosystem as they can impact the cost of capital via their central bank rate - in the US this is the Federal Funds Rate (“FFR”). As we’ve no doubt seen in the news lately, central banks utilise interest rates (ala monetary policy) to manage the dual goals of stable economic growth and healthy base-line inflation. The mechanism here can be quite complex, but to simplify - commercial banks have a minimum required level of capital reserves, set by the Central Bank. Every day their reserves fluctuate (based on how consumers are borrowing or depositing with their banks), which means at the end of the day they need to settle their balances to comply with regulation. To do so, these banks trade overnight amongst themselves (and their Central Bank) using the FFR, which essentially becomes the consumer bank’s cost of capital.
How does this relate to the rate you receive on your balance you ask? That comes back to the first element, consumer rate = cost of capital + bank’s margin. Once you deposit your money into a bank account, the banks invest it in their own ways (generating returns for themselves), and pay you a small amount of interest for using your capital.
The reason to highlight all of this is so we can understand how the existing system works. TradeFi yields are a result of centralised authorities taking advantage of capital arbitrage opportunities. They allocate out capital and take a margin along the way. Those of us with a decentralised mindset are probably thinking the same thing, and to borrow from Chris Dixon - “your take rate is my opportunity”. Simply by removing the middleman and directly matching borrowers to lenders, Web3 should be able to create a structurally higher yield environment for end consumers.
Yield Source 1 - The demand for leverage
Leverage is not a new concept, it is simply an exchange between people who have capital, and those who do not, with one party charging the other interest for the service. Way back in 2000 BC, agricultural communities would borrow seeds against future harvests, and not much has changed since (maybe the soil). We’ve shown above that leverage is the backbone of TradeFi, so it is no surprise that it is one of the 3 pillars of DeFi yields.
Without going into an economics lesson or trying to prove cause vs. effect, asset prices and leverage are correlated. When markets go up, leverage tends to be high. When markets go down, the opposite is true. The Fed has published a paper with the same conclusion - leverage increases asset prices (and increasing asset prices increase leverage). Or more simply we can look at the below chart showing the S&P closing price vs. FINRA reported margin balances (a proxy for equity leverage) to understand how correlated these two elements are.
Crypto leverage is primarily used for speculation - people expect prices to go up, so they borrow and buy more tokens to amplify gains. FTX were at one stage offering 100x leveraged positions, before reducing it to 20x to curb reckless activity. This is in contrast to more traditional uses of leverage such as collateral for home loans, or general household consumption (e.g. to buy goods & services). The rising crypto price environment lifted all ships, and demand for crypto liquidity skyrocketed in the past 12 months. The best known Web3 lending protocols, Aave, MakerDAO and Compound, have seen explosive growth in TVL (“Total Value Locked”) albeit with limited impact of this leverage in the real world. Protocols like Terra burst onto the scene thanks largely to the demand for stablecoins to be used for leverage (you need a stable underlying token to lend against) although Luna’s implosion has only raised more questions about the validity of algorithmic backed stable coins (vs. asset backed like USDC).
Unlike the TradeFi model explained above, yields for these assets are set based on pure demand vs. supply dynamics, with the majority trading on Decentralised Exchanges (DEXs) via Automated Market Makers. It is likely as more liquidity floods into the ecosystem we will see some compression of yields,, however I believe we will continue to see attractive rates relative to TradeFi for many years to come.
One parting thought on leverage driven yields - DeFi borrowing is heavily overcollateralised at the moment, typically requiring 150% of the value to be locked up (i.e. you'll need $150 worth of ETH to borrow $100 of DAI). As collateral shifts to more stable, low-risk assets (rather than those as volatile as ETH), we should see collateralisation rates closer to 100%, increasing capital efficiency and in turn bringing higher yields for lending assets. The likes of Maple and Rose/Oasis also enable uncollaterialised lending with on-chain credit scoring, further broadening the market and providing greater liquidity.
Yield Source 2 - Native token issuance
Hundreds of new crypto projects are made every day, however we have a frighteningly small number of daily active users. This creates a challenge for new projects, as they need user data to iterate & improve the product. Proof of Stake networks need critical user mass for security/stability, and in the case of lending protocols, more users = more liquidity = better for everyone.
One method for short-circuiting this problem is the issuance of native tokens to encourage adoption in new projects and incentivise participation. Similar to how Web2 businesses subsidised their products via freemium of free-trial models, Web3 has created its own go-to-market mechanism. The two main forms of token issuance are 1) Staking rewards and 2) Participation rewards, outlined as follows:
Staking is simply the process of allocating tokens to a pool to achieve a collective outcome. The type of pool, and form of outcome, can vary massively, yet the principal is the same.
At the mature end of the spectrum we have Proof-of-Stake (PoS) rewards which blockchains such as Solana use to incentivise validators and improve network security. Native SOL tokens are issued on a periodic basis (approx. every 2 days) and the overall token supply is subject to an 8% annual inflation rate to accommodate the new issuance. Staking is typically subject to a time lock-up, however liquid staking providers such as Lido are helping remove that burden. If you believe in the fidelity of the underlying blockchain, staking rewards are an easy win for consumers. The yields will diminish overtime due to inbuilt ratchets where issuance slows as adoption increases (shown below for Solana), however staking is still a highly attractive & sustainable yield generation mechanism.
I need to insert a word of caution here - not all staking is created equal. For every Solana, there are hundreds of sub-scale projects hoping to attract users, offering 300%+ APYs. Some of these will be legitimate projects where you benefit from being an early adopter, some of them will be failures. These projects also typically use staking to reduce circulating supply (and hopefully increase prices), rather than as a security mechanism. Do your research (and perhaps stick to large cap projects with realistic yields).
Staking can introduce consumer friction. As mentioned above, this is typically a minimum time commitment, and the processes can be complex. As a result, some projects are incentivising use through participation rewards instead, which is in turn a source of yield. Some of these require active participation (e.g. trading) and some are more passive (such as providing liquidity to lending pools). LooksRare is an example of the former, rewarding traders with their share of the daily $LOOKS issuance. Rewards are subject to a diminishing schedule (shown below) and will expire completely after 361 days of launch, obviously not a sustainable yield generator.
Compare this to Compound, who distribute 2,312 COMP (>$150k) tokens daily to users who contribute liquidity to the protocol. The distribution schedule was made in 2020 with ~5m COMP tokens initially set aside, enough for about 6 years of daily issuance, a more sustainable mechanism, but still finite.
Having covered the main types of native token issuance, we need to discuss the sustainability of these yields. Many of these projects operate with a fixed token supply which will naturally cause the yields to decline over time, or stop completely. I do not expect this to be an issue for high quality projects over the next few years as we are still so early and distributions are often planned for 5-10 years, however it is something to be aware of. It also sounds obvious, but native token yields are predicated on the value of the underlying token. Earning additional yield in $COMP sounds great, but that yield could evaporate if $COMP crashes.
Yield Source 3 - Protocol Fees
The third and final pillar of DeFi yield generation comes from protocol fees. Similar to a dividend or normal equity ownership, some tokens pay out cash flows. These tokens are in essence digitally native assets with visibility on potential earnings. Unlike native token issuance, these fees are distributed in third-party tokens and result from other users’ activity in the protocol / platform.
Taking Uniswap as a notable example, they charge on average 0.30% for every swap on the platform (ignoring V3 fees for simplicity). This fee is paid in the swapped token (instead of a native token) and the protocol can take 0.05% (1/6th) of the fees for itself, the value of which goes back to the $UNI holders.
Of the three sources of DeFi yield, protocol fees are arguably the most sustainable, but rarest in the wild. We are accustomed to businesses charging fees and distributing profits to shareholders, and this is no different. A handful of DeFi projects have experimented with protocol fees (e.g Bancor and 0x) yet there is still a reticence to move to this model as it can be seen as extracting value from the users. Indeed, Uniswap was one of the first to implement the framework however it is still not operational for many of their liquidity pools.
Opportunity and risk goes hand-in-hand
So, where does this leave us?
DeFi is an incredibly fast growing space, with numerous opportunities to make money - It all comes down to risk appetite. Some people will be chasing high APYs and willing to accept the risks (ala yield farmers). Others will prefer low volatility, low yield scenarios. The market has evolved to offer something for everyone, with yield being generated from an array of sources. It is impossible to paint everything with the same brush.
Coming back to the main question around sustainability, this also really depends on the source of yield in question. At a fundamental level I believe aggregate DeFi yields will be higher than TradeFi for years to come, and sustainably so. Demand for leverage is perennial and will be a constant yield generator for both TradeFi and DeFi. Staking rewards in large-cap blockchains will consistently deliver 3%+ APY in native tokens, which in good likelihood will appreciate in value in the long-term. Protocol fees are structurally sound, albeit less widespread, and native token issuance has a degree of predictability despite largely finite time horizons.
As with any investment these strategies bear risks, some of which have not been covered here (e.g. Impermanent loss, regulatory question marks, rug pulls), however hopefully anyone who has made it this far, now has a better sense of what they might get themselves into and can identify the sources of yield. Sadly, it is likely Terra’s demise will throw a shadow over this industry for a long time to come, and cause even more confusion/fear with the general public. I hope that doesn’t detract from the promise of DeFi and I personally take comfort from the fact that many of the brightest minds are working towards a better future. DeFi has the potential to change the financial services landscape, for the banked and unbanked alike, and being incredibly early into that journey there are bound to be more bumps along the way. It is the price we pay as early adopters.
Written by Joseph Pizzolato (JPizzolato.eth)
Special thanks to Luca Prosperi, Joyce Chin and Akash Bajwa for their contribution and help on all things DeFi.
For anyone looking for further reading I recommend The True Sources of DeFi Yield by Yuga.eth, or this thread by by Ming Zhao (@FabiusMercurius)
😎 About the Author
I’m an Investor at Felix Capital, a London based early-stage and growth fund specialising in the intersection between consumers and technology. We have been fortunate to support the likes of Ledger, Sorare, Lightspark, Rally, Flooz and others, and actively looking for great crypto founders to invest in. For entrepreneurs looking for funding (or wanting to chat), you can reach me via email - Joseph@felixcap.com