I create open-source tutorials for Web3 product designers.
I create open-source tutorials for Web3 product designers.
Subscribe to Travis Kassab
Subscribe to Travis Kassab
Share Dialog
Share Dialog
<100 subscribers
<100 subscribers
Swap tokens with Uniswap
Mint stablecoins with MakerDAO
Lend crypto with Compound Finance
Liquidity mining to boost yield with Compound Finance
Yield farming with Curve Finance
Automated investing with Yearn Finance
DeFi dashboard with Zerion & Zapper
DeFi 2.0 & Stablecoin Wars
Deep liquidity with OlympusDAO
Disclaimer
The previous three design course modules (1, 2, 3) cover important background information on Web3. Now, we enter the Web3 ecosystem, starting with Decentralized Finance (DeFi). This article will quickly give UX Designers a quick understanding of what DeFi is, DeFi’s foundational products, and where DeFi is heading in the future. By the way – there’s also a video version of this article just below.
DeFi, or decentralized finance, refers to a complex ecosystem of dapps that offer financial services without the centralized intermediaries of traditional finance like banks, stock/bond brokers, and fintech companies. It offers foundational services – somewhat infrastructural – that the greater Web3 ecosystem can build on. These services include:
token swaps (Uniswap and Curve)
stablecoins for payments (MakerDAO and DAI)
lending/borrowing crypto (Compound)
automated fund managers (Yearn Finance)
dashboards (Zerion)
Just like in traditional finance, DeFi users want to earn yield on their crypto to grow their assets. DeFi is the cutting-edge of fintech. In this course module we review the flagship protocols that have laid the foundations for a thriving DeFi ecosystem. These are some of the most mature and battle-tested dapps in the entire Web3 ecosystem. After discussing them, we move on and talk about the next wave of innovative protocols, called DeFi 2.0.
Decentralized exchanges (DEXs) allow users to swap between different tokens. It’s a simple use-case, but one of the most foundational dapps in the Web3 ecosystem. Before, users had to temporarily give up custody of their tokens to a centralized exchange to make swaps. DEXs, on the other hand, facilitate token swaps with open-source smart contracts and no intermediaries. Uniswap is one of the most well-known DEXs – let’s see how it works.
After connecting their wallets, users select which token they want to swap from (Token A), and which token they want to swap to (Token B). Uniswap then calculates the exchange rate between this token pair. This tells users how many of Token B they should expect to receive. In the case above, I’m swapping .02 ETH for about 40 DAI, with an estimated network fee of $7.32. I will need to have extra ETH in my wallet to pay this network fee.
Now let’s talk about what is happening under the hood. For example, where did the DAI come from that I swapped into? Decentralized exchanges use a different method than centralized exchanges for making these trades. Centralized exchanges (CEXs) use something called a central limit order book, or CLOB, to match buyers and sellers. Basically, buyers place bids for how much they are willing to pay for an asset, and sellers place bids for how much they are willing to sell an asset. The transaction is executed when there is overlap between the buy and sell bids. An order book for Bitcoin is shown above, where the green area represents buy orders, and the red sell orders.
CLOBs are costly to run on-chain because they are transaction heavy and benefit from fast transaction processing. Thus, DEXs implement a different approach to token swaps. Every token pair you can swap on a DEX has a liquidity pool. So, in the example above, I made my swap using Uniswap’s ETH-DAI liquidity pool. We see $54.82M of tokens are locked in the ETH-DAI liquidity pool, made up of 23.81M DAI and 15.52k ETH. As users make swaps on these liquidity pools by depositing Token A and withdrawing Token B, an algorithm automatically recalculates the exchange rate for the token pair. These algorithms are known as “automated market makers”, or AMMs, and are responsible for updating token prices on DEXs based on supply and demand dynamics.
Now, where do the tokens in the liquidity pools initially come from? Liquidity providers (LPs) are other Web3 users who want to earn yield on their tokens. Uniswap charges a trading fee (0.3%) on each swap, and this trading fee is distributed to the liquidity providers for that pool. LPs deposit Token A and Token B into a liquidity pool, and receive LP tokens in return. At any point, users can burn their LP tokens for their initial liquidity plus whatever trading fees have accrued to them.
So, those are the basics of DEXs. Liquidity providers deposit tokens into liquidity pools to earn trading fees. Traders use these liquidity pools to make token swaps. And, AMM algorithms reprice tokens within liquidity pools based on market conditions. Uniswap is not the only DEX out there. Other DEXs have been created that excel for different use-cases. For example, Balancer allows liquidity pools containing up to 8 different tokens, whereas Uniswap liquidity pools always contain only 2. And, Curve is a DEX focused on stablecoin swaps, resulting in less slippage and reduced risk of impermanent loss.
Speaking of stablecoins… The vast majority of Web3 tokens have volatile prices, but a subset of them, called stablecoins, are designed to maintain a stable price over time. This makes them more attractive as a medium of exchange compared to volatile tokens. As Ryan Selkis puts it, “no one wants to spend currency they believe will be worth 10% more tomorrow, and no one wants to accept currency they think could be worth 10% less tomorrow.” This leads into another foundational DeFi protocol, MakerDAO, which allows users to deposit their volatile tokens (e.g. ETH), and mint stablecoins for tax-free spending.
In the previous section, we swapped ETH for DAI stablecoins on Uniswap, but this is a secondary market for DAI. DAI is quite unique in how it is minted. DAI comes from a DeFi protocol called MakerDAO, where users deposit volatile assets into a Maker vault and then mint DAI stablecoins.
This is known as a collateralized-debt position (CDP), because the user borrows DAI against their collateral, and pays interest (i.e. “stability fee”) on the loan. In the picture above, users open a vault, deposit ETH into the vault, and then take out a DAI loan. The ETH is locked in the vault until users pay back the DAI loan, which is the principal amount plus whatever interest is owed.
MakerDAO requires that the collateral deposit is worth more than the stablecoin loan. This is known as “over-collateralization”, and is necessary when using volatile assets as collateral. For example, let’s say you deposit 1 ETH (worth $3000), and take out $2000-worth of DAI – this makes for a collateralization ratio of 150% (i.e. over-collateralized). If the price of ETH drops by 20% the collateral is now worth $2400, changing the collateralization ratio to 120%. The vault would still be over-collateralized, but could soon be under-collateralized if the price of ETH continues to drop. In fact, in the ETH-B vault pictured above, MakerDAO automatically liquidates the collateral to pay back the DAI loan, and closes out the vault, when the collateralization ratio drops below 130% (i.e. “min collateralization ratio”).
You may wonder what the point of this is. If these vaults need to be over-collateralized, then you’re essentially borrowing with money you already have. Aren’t loans about spending money you don’t have? We may eventually see under-collateralized crypto loans as Web3 matures. At a minimum we would need verified Web3 identities in order to track who has defaulted on their loans and assign credit scores accordingly. But still, for now, there are beneficial use-cases for crypto-collateralized stablecoins.
These stablecoin loans allow users to go long their collateral asset, while unlocking a portion of its value, in the form of DAI stablecoins, which can be used for payment throughout the rest of the Web3 ecosystem.
For example, if I’m bullish on ETH, and believe it will 2x within a year, then I want to continue holding ETH. In fact, I want to accumulate as much ETH as possible, and certainly do not want to swap my ETH for a stablecoin. Swapping ETH for a stablecoin hurts me in two ways. First, it triggers a tax event, and I will have to pay capital gains on my ETH. Second, I lose exposure to ETH’s upside potential during a bull run. Instead, borrowing DAI against my ETH with a Maker vault does not trigger a tax event, allows me to keep ownership of my ETH, while also freeing up liquidity so I can make payments in the meantime.
MakerDAO’s stablecoin, DAI, is decentralized and maintains its peg to the US dollar without reliance on reserves held by centralized entities, like USDT’s Tether and USDC’s Circle. Stable, decentralized assets are an important DeFi building block, and an open area of experimentation, which we’ll see later on.
You can think of MakerDAO as a protocol for self-loans; however, another class of lending protocols exist, like AAVE and Compound, that support peer-to-peer token lending. Lending protocols like these are foundational to DeFi because they establish lending/borrowing interest rates for a variety of tokens. Now, Web3 tokens have a time-value aspect to them, just like risk-free interest rates for fiat currencies in traditional money markets.
Compound creates a peer-to-peer lending marketplace for a variety of Web3 tokens. Suppliers are paid interest for depositing their tokens into lending pools. This has become a popular passive income strategy for Web3 users in DeFi.
Similar to how Uniswap liquidity providers receive LP tokens, Compound suppliers receive cTokens for depositing tokens into lending pools. At any point, users can burn their cTokens for their initial deposit, plus the interest that has accrued. For example, cETH represents ETH that is actively earning interest in Compound.
On the other end of things, borrowers take loans out on these lending pools. Borrowers first must supply tokens to a lending pool, and select to use the token as collateral for a loan.
Just like with MakerDAO, all Compound loans must be over-collateralized based on a collateral factor (similar to collateralization ratio). The collateral factor varies by token, but let’s say it’s 75%. It means I can only borrow up to 75% of the collateral I have deposited. So with $100-worth of Token A collateral, I can borrow at most $75-worth of Token B. This borrow limit is indicated to the user at the bottom of the modal (“Borrow Limit Used”) in the picture above. If the value of a borrower’s collateral drops then he is at risk of liquidation.
To close the position, borrowers repay the token loan back into the lending pool plus the interest payment they owe. Note that interest, for suppliers and borrowers, is paid in the token that is being lent/borrowed. Since I borrowed DAI in the example above, I made a small interest payment of DAI.
Compound is powered by smart contracts that automatically set interest rates based on the liquidity of tokens in the lending pools. Tokens with lower liquidity will have higher interest rates, as supply and demand dictate. These are “floating” interest rates meaning they change over time. Compound calculates an interest rate for each lending pool at every Ethereum block.
You can see why liquidity is important here. The lower the token liquidity, the more it will cost to borrow the token. This issue of liquidity is a challenge for all DeFi protocols – token liquidity is required to get the protocol off the ground, and to continue offering a usable service. Generally, the more liquidity the better, so DeFi protocols compete to attract liquidity to their platform.
So how do DeFi protocols attract liquidity in the first place? Compound was the first to try out liquidity mining. Compound distributes COMP – its native governance token – to its current suppliers and borrowers. This strategy worked well for Compound, and other DeFi protocols followed suit, leading to an explosion in liquidity in summer of 2020 (i.e. DeFi Summer).
Notice that suppliers on Compound are actually paid two interest rates. In the example above, where I’ve supplied ETH, one interest rate pays me back in ETH. This is from the interest paid by borrowers on their ETH loan. The other interest rate is paid in COMP. This is Compound’s liquidity mining reward, which is meant to incentivize me to keep my ETH liquidity locked in the protocol.
Back in 2020, Compound minted its COMP governance token with a fixed supply of 10M COMP. 4.2M COMP were allocated to its liquidity mining program, which distributes 1139 COMP tokens across its suppliers and borrowers on a daily basis. This will run for around four years until the allocated COMP runs out.
Compound was the first to implement liquidity mining, but many other DeFi protocols followed suit, using their own governance tokens as reward. These tokens give holders the ability to vote on governance proposals that could be anything from technical improvements to adjusting protocol parameters like interest rates and the list of assets approved for collateral (e.g. MakerDAO). Discussing protocol governance leads to the concept of Decentralized Autonomous Organizations (DAOs). This is an entire course module in itself that will be covered later. For now, understand that DeFi governance tokens have real-world value as they are traded on secondary markets, seen in the picture above.
Compound launched its liquidity mining program in June 2020. Other liquidity mining programs cascaded from there, leading to a boom known as “DeFi Summer”. Liquidity flooded into DeFi protocols as yields increased across the entire ecosystem. Look at the increase in Total Value Locked (TVL) in the chart above. The TVL chart shows all the token liquidity “locked” in all of the DeFi protocols on Ethereum. This could be tokens in DEX liquidity pools, Maker vaults, lending pools, and more. TVL is a great metric for comparing the health of DeFi protocols (e.g. AAVE versus Compound), as well as entire DeFi ecosystems (e.g. Ethereum DeFi versus Solana DeFi).
So far we’ve talked about several ways of earning yield on Web3 tokens, such as becoming a liquidity provider on a DEX, supplying tokens to lending pools, and participating in liquidity mining programs. In the next section we discuss another DeFi yield activity – staking – and show how DeFi users stack multiple yield activities on top of one another to maximize yield. This is known as yield farming.
Staking is the idea of locking tokens to receive other additional token rewards, paid out over time. This is an odd idea at first, but locking tokens in stake pools can be beneficial to DeFi protocols – let’s look at why this might be.
Remember back to the first section how users receive LP tokens for depositing their tokens in a DEX liquidity pool. Not only is it important for DEXs to initially attract liquidity, but they also must retain liquidity to maintain good UX for its users. Some DEXs encourage users to stake their LP tokens in order to receive additional incentives.
Curve is a DEX for stablecoin swaps that offers this kind of LP token staking. Liquidity providers deposit tokens into a Curve liquidity pool, and receive Curve LP tokens. These LP tokens can be staked in Curve Gauge Pools where the user receives regular payouts of CRV tokens. Users are unable to remove their liquidity so long as their LP tokens are staked, and they are receiving CRV rewards. Do you see how this incentivizes users to keep their liquidity on Curve?
Staking increases yield for Curve liquidity providers. For example, in the first Curve liquidity pool above (“tricrypto2”), the yield is 1.89% from the pool’s trading fees; however, users receive additional yield – anywhere from 4.23% to 10.59% – paid in CRV tokens. This means, if I deposit $100-worth of liquidity in the pool, I’d expect to earn $1.89-worth of trading fees, plus $4.23-worth of CRV, annually.
Things do not end here on Curve. It has additional staking opportunities. Users can stake CRV in the Curve DAO for veCRV in return. veCRV is the governance token, so it allows users to make Curve governance decisions; however, it also boosts CRV rewards by up to 2.5x, which is veCRVs main draw. That’s why the CRV APYs – shown two images above – are displayed as a range (e.g. +4.23% – 10.59% CRV). Looking at the image above, CRV is time-locked anywhere from one week to four years. This takes CRV off the market, making it a deflationary asset, which some believe has led to its favorable price performance compared to other DeFi tokens.
Are you starting to see how yields can be stacked on top of each other? On Curve, users can deposit tokens into a liquidity pool to earn trading fees. They can then stake their LP tokens to earn CRV rewards. And then stake their CRV tokens to boost their CRV rewards. This is what “yield farming” is all about – stacking yield to maximize return on investment.
But all this makes my head spin. The yield farming I’ve described here is complex, manual, and costly. For example, CRV rewards accumulate over time and users must manually claim CRV on Curve’s UI (image above), which requires a transaction and costs a network fee. Yield aggregators – another class of DeFi dapps – solve these problems by automating yield farming for users. Users’ funds are pooled together in vaults that follow pre-programmed yield farming strategies designed for efficiency and maximal return.
Yearn Finance was one of the first yield aggregators. On Yearn, DeFi experts propose yield farming strategies and implement them as Yearn vaults that Web3 users can deposit tokens into. Essentially, these vaults aggregate a lot of tokens, and deploy them to various DeFi protocols, sharing the yield with its depositors (i.e. yToken holders).
Yearn vaults generally accept Curve LP tokens, and automate the yield farming process from here, such as moving tokens into different lending protocols in pursuit of the highest interest rate. Or, stake the LP tokens in other protocols, and automatically recycle the rewards back into the vault for a compounding effect.
This brings us to a concept called “token composability”, which is at the heart of yield farming. Essentially, Web3 tokens are interoperable with other protocols. DeFi tokens can be used throughout the DeFi ecosystem in order to stack rewards.
Let’s look at Yearn’s “Curve USDN” vault strategy. Users deposit crvUSDN tokens into this Yearn vault. Yearn then stakes these LP tokens in another DeFi protocol, Convex Finance, to earn CRV and CVX. As the vault earns tokens, Yearn automatically harvests them and converts them into more crvUSDN, which it cycles back into Convex Finance. This automation simplifies yield farming for users, and reduces their network fees. They only need to send two transactions: one to deposit tokens into a vault, and the other to withdraw tokens from the vault.
At this point in the article, we’ve discussed all the major DeFi building blocks: tokens swaps, minting stablecoins, lending/borrowing, staking, and yield farming. This all takes place across different protocols, often requiring the user to visit multiple web apps. This can make the DeFi ecosystem feel scattered. DeFi dashboards bring everything together for a more seamless UX. Let’s look at how dashboards take advantage of the open-source nature of Web3 to stitch DeFi protocols together into one dApp.
DeFi dashboards provide the user access to multiple DeFi protocols all from one UI. The first way to think of these dashboards is that they extend wallet functionality. Popular browser extension wallets like MetaMask have small UIs with limited feature sets like viewing token balances and sending tokens to other wallets.
However, users can connect their wallet to a dashboard like Zerion to view their wallet as an asset portfolio, not just simple token balances. These portfolio views help users see how their wallet’s value changes over time. Also, Zapper allows users to bundle multiple wallets together so they can track the combined value of multiple wallets.
Dashboards also directly integrate multiple DeFi protocols into one UI. This is an example of open-source software composability. Web3 protocols are open-source so other projects can use them as building blocks in their own dApp. This is juxtaposed to the proprietary APIs of Web2 applications. Users can search for the best yields across the entire DeFi ecosystem, and directly deposit tokens into lending protocols (AAVE and Compound), liquidity pools (Curve, Sushi Swap, Uniswap), staking pools (Curve and Convex), and yield aggregator vaults (Yearn and Harvest). All accessed from Zapper’s “invest” tab.
These dashboards are packed full of other useful features. For example, the Zerion dashboard has “swap” functionality that searches across multiple DEXs to find the best exchange rate to execute a token swap. This is called a DEX aggregator, and is a convenience utility for Web3 users.
The thesis that we will live in a multi-chain world continues to be proven correct. More and more, people will use multiple blockchain networks, and will increasingly want to transact tokens between these cryptonetworks. Blockchain interoperability has a long way to go and Web3 ecosystems feel siloed; however, dashboards are well positioned to improve the UX here.
Zapper tracks the value of a user’s portfolio across 11 different cryptonetworks (all EMV-compatible). And, Zerion aggregates blockchain bridges so that users can transfer tokens to other Layer 1 cryptonetworks (BSC, Avalanche, Fantom), Ethereum Layer 2s (Optimism, Arbitrum), and sidechains (Polygon), all from one UI.
So that wraps up the first wave of DeFi from token swaps to dashboards. There are other protocols I consider slightly less foundational than what we’ve covered, but the curious reader may be interested to look into some of them. DeFi Derivatives platforms allow users to trade advanced financial instruments like options (Opyn), perpetuals (DyDx & Perp), and synthetic assets (Synthetix). Also, there are token index funds that diversify a single investment across multiple tokens (see Index CoOp). This is the same idea as ETF’s and mutual funds in traditional markets. Finally, there are insurance protocols that protect users from things like impermanent loss, and DeFi smart contract exploits (see Nexus Mutual).
Interest rates have been at all time lows in traditional markets over the past several years as the U.S. 10 Year Treasury Bond hit 0.60% APY in 2020. At the same time, DeFi was on the rise, giving investors alternative markets to earn much higher (albeit riskier) yields. All the DeFi products we’ve talked about were some of the first to find product-market fit in Web3, evidenced by the fact that they onboarded millions of users onto Ethereum; however, like any first iteration of technology, there exist some problems in the current DeFi ecosystem. Let’s discuss some of the problems facing DeFi 1.0, and peer into the emerging DeFi 2.0 landscape, which seeks to address them – starting with stablecoins.
Demand for stablecoins is rapidly increasing as the Web3 ecosystem is more and more in need of non-volatile assets for things like payment. There are many types of stablecoins each with their pros and cons. And it’s also an active area of experimentation as entrepreneurs search for optimal stablecoin designs. Let’s take a look at the different stablecoin categories.
Source
The first category is fiat-collateralized stablecoins. These are, by far, the most popular right now based on market cap. Here, regulated companies, like Tether and Circle, issue stablecoins that are fully-backed by their fiat reserves. USDC and USDT are pegged to the US-dollar, but some stablecoins track other currencies (e.g. EURS). Historically, fiat-collateralized stablecoins have maintained their peg better than others; however, their fiat reserves are custodied within the traditional financial system. Governments have the ability to seize these reserves and undermine fiat-collateralized stablecoins, which many consider unacceptable for a means of payment in the world of Web3.
Enter crypto-collateralized stablecoins. This brings us back to MakerDAO where users can deposit crypto into Maker vaults and mint DAI – a stablecoin also pegged to the US-dollar. This is a big deal, because the crypto collateral backing the stablecoins cannot be seized by centralized entities. Thus, crypto-collateralized stablecoins are truly decentralized; however, the problem is that crypto is volatile, and using it as collateral is risky because the value of collateral can quickly fall below the loan’s value. Therefore, crypto-collateralized stablecoins must be over-collateralized, which isn’t an efficient use of capital.
MakerDAO was the first crypto-collateralized stablecoin protocol. Since then, other DeFi protocols have iterated on, and improved the original MakerDAO concept. For example, the crypto locked in Maker vaults is just sitting there, not earning yield. On Alchemix, users deposit crypto collateral to generate alUSD – Alchemix’s stablecoin. But here, Alchemix deposits collateral into Yearn Vaults that earn interest, which is used to automatically repay the user’s alUSD loan. It’s a self-repaying loan. Similarly, Abracadabra.Money supports yield-bearing tokens as collateral, like Yearn (e.g. yvDAI) and Convex (e.g. cvx3pool) tokens. But I digress – let’s move onto the third stablecoin category – algorithmic stablecoins.
The first two stablecoin categories are asset-backed. Either fiat reserves, or crypto collateral in vaults, support the price of the aforementioned stablecoins. A lot of recent experimentation has been done on algorithmic stablecoins, which have no assets backing them. An algorithm automatically maintains the peg of a stablecoin based on real-time market conditions for that stablecoin. This is usually done through mint and burn mechanics. Let’s pause and talk about the most popular algorithmic stablecoin – UST.
UST is built on the Terra blockchain, and is pegged to the US-dollar. UST is paired with Terra’s native token – LUNA, which is a volatile crypto asset. UST and LUNA can be exchanged in a 1:1 ratio with each other. At any point, 1 UST can be minted by burning $1-worth of LUNA. Alternatively, 1 UST can be burned in exchange for $1-worth of LUNA. Arbitrageurs utilize this mint-burn mechanism when UST is trading on centralized exchanges for a premium, or at a discount. For example, let’s say UST breaks its peg, and trades at $1.02 on a CEX. Arbitrageurs can burn $1-worth of LUNA, and mint UST in order to realize a 2% gain. On the other hand, if UST is trading at $.98, then it can be exchanged for $1-worth of LUNA. This is how UST maintains its peg.
However, it’s ironic that I’m writing this now. Just last week UST failed massively. There was a major sell-off in UST as investors lost faith, reflexivity kicked in, and caused an even greater sell-off. Billions of dollars of UST were liquidated over several days, and UST is sitting at around $.09 at the time of writing. This points to the immaturity of algorithmic stablecoins, and puts into question the feasibility of algorithmic stablecoins in general.
There is another category of stablecoin – fractional reserve stablecoins – that is a middle ground between algorithmic and crypto-collateralized. These stablecoins are partially crypto-collateralized hence the term “fractional reserve”, and partially rely on an algorithm to maintain their peg. FRAX is the most popular example of fractional reserve stablecoins.
All the stablecoins we’ve discussed so far are pegged to fiat currencies. The issue with this is that pegged stablecoins are beholden to central banks that control the monetary policies of fiat currencies. Some argue it’s unacceptable to have any reliance on central institutions in the decentralized Web3 ecosystem. The final category of stablecoins we talk about is non-pegged stablecoins. OlympusDAO is the most popular project in this space with its non-pegged OHM stablecoin.
“Stablecoin” is a bit of a misnomer for OHM. Currently, its price is around $17 – and it’s been as high as $1,300+. So OHM can be highly volatile, but one of the main features of OHM is its floor price, or risk-free value. Users exchange crypto with OlympusDAO for discounted OHM in a process called bonding. All bonded crypto is held in OlympusDAO’s treasury, which ensures a floor price of at least $1 OHM. Thus, the value of OHM is free-floating, but users can always have faith in the floor price, backed by Olmypus’ war chest.
This leads into the next problem that DeFi 2.0 is attempting to solve – that of deep, permanent token liquidity. OlympusDAO controls a massive treasury, which is a great example of an upcoming concept – “protocol controlled value” (PCV). The protocol can deploy liquidity as it sees fit to ensure liquidity for its native token, or to earn yield and grow the treasury larger. Also, their protocols can rent out their liquidity to help other protocols bootstrap token trading, or bolster their liquidity pools.
We know that liquidity is important in all aspects of DeFi. Without tokens in liquidity pools you can’t make swaps on DEXs, borrow tokens in lending protocols, and enact automated yield farming strategies. Without liquidity DeFi grinds to a halt. And the more liquidity the better. For example, the more liquidity in a DEXs liquidity pool, the less slippage users are subjected to when making token swaps. That’s why DeFi protocols compete with one another to attract liquidity, and increase the TVL of their protocol.
Liquidity mining, first seen with Compound in 2020, attempted to overcome the liquidity bootstrapping challenge. Protocols mint their governance token, and distribute it incrementally as a reward to its liquidity providers. Unfortunately, this only solves the liquidity problem in the short-term. Liquidity mining programs attract mercenary capital. Without loyalty to any single protocol, users switch their liquidity between protocols whenever higher interest rates popup elsewhere. It’s difficult for protocols to maintain this liquidity, and they have to continue inflating their governance token in order to do so. But more and more, liquidity is managed at the protocol-level, called “protocol controlled liquidity” or “protocol owned liquidity”.
DeFi 1.0 protocols do not own user liquidity. Users own their LP tokens, and can redeem them for the underlying liquidity at any time. This is changing in DeFi 2.0. Let’s go back to OlympusDAO, and users bonding their crypto in the treasury. Here, user crypto is exchange for discounted OHM. In this case, the users no longer own their liquidity – the liquidity is now permanently held in OlympusDAO treasury. In fact, OlympusDAO controls over 99% of its liquidity in DEXs (e.g. OHM-FRAX LP & OHM-DAI LP).
This ensures that the liquidity remains in place so that users can trade in and out of OHM, with low slippage. Also, the protocol earns almost all of the liquidity pool trading fees, and is becoming a large holder of SUSHI, which all work to further grow the protocol.
Frax Protocol is another good example of PCV. Users mint FRAX stablecoin by depositing their crypto assets into Frax protocol’s reserves. Frax Protocol applies automated market operations (AOMs) on its reserves for a variety of reasons. For example, its collateral investor AMO moves USDC reserves into lending protocols like Compound and Yearn, which earns interest. Another AMO provides FRAX liquidity on exchanges like Curve and Uniswap to ensure deep liquidity pools for FRAX-stablecoin pairs. Fei Protocol is similar – the protocol controls reserves for yield farming strategies to grow its reserves, as well as to provide FEI liquidity on Uniswap.
These protocols can also rent out their liquidity to other DeFi protocols, known as liquidity-as-a-service (LaaS). Tokemak is a great example of this. Essentially, individual liquidity providers deposit tokens into reactors, and liquidity directors (LDs) rent out this liquidity to various DEXs. LD’s create liquidity on-demand – they vote on which token reactors get paired together, and which DEXs to send these token pairs to. Of course, Tokemak earns DEX trading fees, which its building up a treasury with. This treasury, controlled by TOKE governance, will be utilized for additional LaaS operations in the future. Tokemak is like a liquidity aggregator, supplying liquidity to other specific areas in the DeFi ecosystem.
Finally, LaaS is being used to help other up-and-coming protocols launch their native token. Before, to start sellings a token on a DEX you either needed to seed the liquidity pool with a lot of up front capital, or attract users to bring their own liquidity. This brings us back to liquidity mining, which we’ve already established is ineffective in the long run. However, now, OlympusDAO launched a product called Olympus Pro, which generalizes its token bonding mechanism for other protocols. For example, users can bond their LP tokens in exchange for the protocol’s discounted governance token. Now that protocol owns the LP tokens, so the liquidity will remain in place without continuous liquidity mining. This is known as “protocol pwned liquidity” (POL).
Also, FEI and FRAX have partnered with Ondo Finance, which provides liquidity pools to protocols looking to list their governance token in an IDO – initial DEX offering. The protocol deposits its governance token in a pool, which is matched by either FRAX or FEI stablecoins. The Frax and Fei protocols earn 5% APR on the liquidity it provides. This is like renting out stablecoin liquidity so that other protocols can quickly get up and running selling their governance tokens.
I, Travis Kassab, am not a financial advisor, tax professional, broker, or legal advisor. None of my podcasts, videos, or articles constitute financial advice. This content is purely for educational purposes, and is not intended to endorse any cryptocurrency or other investment vehicle. At the time of writing this, my cryptocurrency holdings include: BTC, ETH, CEL, SNX, BANK, SOL, DOT, KSM, USDC, GERO.
Swap tokens with Uniswap
Mint stablecoins with MakerDAO
Lend crypto with Compound Finance
Liquidity mining to boost yield with Compound Finance
Yield farming with Curve Finance
Automated investing with Yearn Finance
DeFi dashboard with Zerion & Zapper
DeFi 2.0 & Stablecoin Wars
Deep liquidity with OlympusDAO
Disclaimer
The previous three design course modules (1, 2, 3) cover important background information on Web3. Now, we enter the Web3 ecosystem, starting with Decentralized Finance (DeFi). This article will quickly give UX Designers a quick understanding of what DeFi is, DeFi’s foundational products, and where DeFi is heading in the future. By the way – there’s also a video version of this article just below.
DeFi, or decentralized finance, refers to a complex ecosystem of dapps that offer financial services without the centralized intermediaries of traditional finance like banks, stock/bond brokers, and fintech companies. It offers foundational services – somewhat infrastructural – that the greater Web3 ecosystem can build on. These services include:
token swaps (Uniswap and Curve)
stablecoins for payments (MakerDAO and DAI)
lending/borrowing crypto (Compound)
automated fund managers (Yearn Finance)
dashboards (Zerion)
Just like in traditional finance, DeFi users want to earn yield on their crypto to grow their assets. DeFi is the cutting-edge of fintech. In this course module we review the flagship protocols that have laid the foundations for a thriving DeFi ecosystem. These are some of the most mature and battle-tested dapps in the entire Web3 ecosystem. After discussing them, we move on and talk about the next wave of innovative protocols, called DeFi 2.0.
Decentralized exchanges (DEXs) allow users to swap between different tokens. It’s a simple use-case, but one of the most foundational dapps in the Web3 ecosystem. Before, users had to temporarily give up custody of their tokens to a centralized exchange to make swaps. DEXs, on the other hand, facilitate token swaps with open-source smart contracts and no intermediaries. Uniswap is one of the most well-known DEXs – let’s see how it works.
After connecting their wallets, users select which token they want to swap from (Token A), and which token they want to swap to (Token B). Uniswap then calculates the exchange rate between this token pair. This tells users how many of Token B they should expect to receive. In the case above, I’m swapping .02 ETH for about 40 DAI, with an estimated network fee of $7.32. I will need to have extra ETH in my wallet to pay this network fee.
Now let’s talk about what is happening under the hood. For example, where did the DAI come from that I swapped into? Decentralized exchanges use a different method than centralized exchanges for making these trades. Centralized exchanges (CEXs) use something called a central limit order book, or CLOB, to match buyers and sellers. Basically, buyers place bids for how much they are willing to pay for an asset, and sellers place bids for how much they are willing to sell an asset. The transaction is executed when there is overlap between the buy and sell bids. An order book for Bitcoin is shown above, where the green area represents buy orders, and the red sell orders.
CLOBs are costly to run on-chain because they are transaction heavy and benefit from fast transaction processing. Thus, DEXs implement a different approach to token swaps. Every token pair you can swap on a DEX has a liquidity pool. So, in the example above, I made my swap using Uniswap’s ETH-DAI liquidity pool. We see $54.82M of tokens are locked in the ETH-DAI liquidity pool, made up of 23.81M DAI and 15.52k ETH. As users make swaps on these liquidity pools by depositing Token A and withdrawing Token B, an algorithm automatically recalculates the exchange rate for the token pair. These algorithms are known as “automated market makers”, or AMMs, and are responsible for updating token prices on DEXs based on supply and demand dynamics.
Now, where do the tokens in the liquidity pools initially come from? Liquidity providers (LPs) are other Web3 users who want to earn yield on their tokens. Uniswap charges a trading fee (0.3%) on each swap, and this trading fee is distributed to the liquidity providers for that pool. LPs deposit Token A and Token B into a liquidity pool, and receive LP tokens in return. At any point, users can burn their LP tokens for their initial liquidity plus whatever trading fees have accrued to them.
So, those are the basics of DEXs. Liquidity providers deposit tokens into liquidity pools to earn trading fees. Traders use these liquidity pools to make token swaps. And, AMM algorithms reprice tokens within liquidity pools based on market conditions. Uniswap is not the only DEX out there. Other DEXs have been created that excel for different use-cases. For example, Balancer allows liquidity pools containing up to 8 different tokens, whereas Uniswap liquidity pools always contain only 2. And, Curve is a DEX focused on stablecoin swaps, resulting in less slippage and reduced risk of impermanent loss.
Speaking of stablecoins… The vast majority of Web3 tokens have volatile prices, but a subset of them, called stablecoins, are designed to maintain a stable price over time. This makes them more attractive as a medium of exchange compared to volatile tokens. As Ryan Selkis puts it, “no one wants to spend currency they believe will be worth 10% more tomorrow, and no one wants to accept currency they think could be worth 10% less tomorrow.” This leads into another foundational DeFi protocol, MakerDAO, which allows users to deposit their volatile tokens (e.g. ETH), and mint stablecoins for tax-free spending.
In the previous section, we swapped ETH for DAI stablecoins on Uniswap, but this is a secondary market for DAI. DAI is quite unique in how it is minted. DAI comes from a DeFi protocol called MakerDAO, where users deposit volatile assets into a Maker vault and then mint DAI stablecoins.
This is known as a collateralized-debt position (CDP), because the user borrows DAI against their collateral, and pays interest (i.e. “stability fee”) on the loan. In the picture above, users open a vault, deposit ETH into the vault, and then take out a DAI loan. The ETH is locked in the vault until users pay back the DAI loan, which is the principal amount plus whatever interest is owed.
MakerDAO requires that the collateral deposit is worth more than the stablecoin loan. This is known as “over-collateralization”, and is necessary when using volatile assets as collateral. For example, let’s say you deposit 1 ETH (worth $3000), and take out $2000-worth of DAI – this makes for a collateralization ratio of 150% (i.e. over-collateralized). If the price of ETH drops by 20% the collateral is now worth $2400, changing the collateralization ratio to 120%. The vault would still be over-collateralized, but could soon be under-collateralized if the price of ETH continues to drop. In fact, in the ETH-B vault pictured above, MakerDAO automatically liquidates the collateral to pay back the DAI loan, and closes out the vault, when the collateralization ratio drops below 130% (i.e. “min collateralization ratio”).
You may wonder what the point of this is. If these vaults need to be over-collateralized, then you’re essentially borrowing with money you already have. Aren’t loans about spending money you don’t have? We may eventually see under-collateralized crypto loans as Web3 matures. At a minimum we would need verified Web3 identities in order to track who has defaulted on their loans and assign credit scores accordingly. But still, for now, there are beneficial use-cases for crypto-collateralized stablecoins.
These stablecoin loans allow users to go long their collateral asset, while unlocking a portion of its value, in the form of DAI stablecoins, which can be used for payment throughout the rest of the Web3 ecosystem.
For example, if I’m bullish on ETH, and believe it will 2x within a year, then I want to continue holding ETH. In fact, I want to accumulate as much ETH as possible, and certainly do not want to swap my ETH for a stablecoin. Swapping ETH for a stablecoin hurts me in two ways. First, it triggers a tax event, and I will have to pay capital gains on my ETH. Second, I lose exposure to ETH’s upside potential during a bull run. Instead, borrowing DAI against my ETH with a Maker vault does not trigger a tax event, allows me to keep ownership of my ETH, while also freeing up liquidity so I can make payments in the meantime.
MakerDAO’s stablecoin, DAI, is decentralized and maintains its peg to the US dollar without reliance on reserves held by centralized entities, like USDT’s Tether and USDC’s Circle. Stable, decentralized assets are an important DeFi building block, and an open area of experimentation, which we’ll see later on.
You can think of MakerDAO as a protocol for self-loans; however, another class of lending protocols exist, like AAVE and Compound, that support peer-to-peer token lending. Lending protocols like these are foundational to DeFi because they establish lending/borrowing interest rates for a variety of tokens. Now, Web3 tokens have a time-value aspect to them, just like risk-free interest rates for fiat currencies in traditional money markets.
Compound creates a peer-to-peer lending marketplace for a variety of Web3 tokens. Suppliers are paid interest for depositing their tokens into lending pools. This has become a popular passive income strategy for Web3 users in DeFi.
Similar to how Uniswap liquidity providers receive LP tokens, Compound suppliers receive cTokens for depositing tokens into lending pools. At any point, users can burn their cTokens for their initial deposit, plus the interest that has accrued. For example, cETH represents ETH that is actively earning interest in Compound.
On the other end of things, borrowers take loans out on these lending pools. Borrowers first must supply tokens to a lending pool, and select to use the token as collateral for a loan.
Just like with MakerDAO, all Compound loans must be over-collateralized based on a collateral factor (similar to collateralization ratio). The collateral factor varies by token, but let’s say it’s 75%. It means I can only borrow up to 75% of the collateral I have deposited. So with $100-worth of Token A collateral, I can borrow at most $75-worth of Token B. This borrow limit is indicated to the user at the bottom of the modal (“Borrow Limit Used”) in the picture above. If the value of a borrower’s collateral drops then he is at risk of liquidation.
To close the position, borrowers repay the token loan back into the lending pool plus the interest payment they owe. Note that interest, for suppliers and borrowers, is paid in the token that is being lent/borrowed. Since I borrowed DAI in the example above, I made a small interest payment of DAI.
Compound is powered by smart contracts that automatically set interest rates based on the liquidity of tokens in the lending pools. Tokens with lower liquidity will have higher interest rates, as supply and demand dictate. These are “floating” interest rates meaning they change over time. Compound calculates an interest rate for each lending pool at every Ethereum block.
You can see why liquidity is important here. The lower the token liquidity, the more it will cost to borrow the token. This issue of liquidity is a challenge for all DeFi protocols – token liquidity is required to get the protocol off the ground, and to continue offering a usable service. Generally, the more liquidity the better, so DeFi protocols compete to attract liquidity to their platform.
So how do DeFi protocols attract liquidity in the first place? Compound was the first to try out liquidity mining. Compound distributes COMP – its native governance token – to its current suppliers and borrowers. This strategy worked well for Compound, and other DeFi protocols followed suit, leading to an explosion in liquidity in summer of 2020 (i.e. DeFi Summer).
Notice that suppliers on Compound are actually paid two interest rates. In the example above, where I’ve supplied ETH, one interest rate pays me back in ETH. This is from the interest paid by borrowers on their ETH loan. The other interest rate is paid in COMP. This is Compound’s liquidity mining reward, which is meant to incentivize me to keep my ETH liquidity locked in the protocol.
Back in 2020, Compound minted its COMP governance token with a fixed supply of 10M COMP. 4.2M COMP were allocated to its liquidity mining program, which distributes 1139 COMP tokens across its suppliers and borrowers on a daily basis. This will run for around four years until the allocated COMP runs out.
Compound was the first to implement liquidity mining, but many other DeFi protocols followed suit, using their own governance tokens as reward. These tokens give holders the ability to vote on governance proposals that could be anything from technical improvements to adjusting protocol parameters like interest rates and the list of assets approved for collateral (e.g. MakerDAO). Discussing protocol governance leads to the concept of Decentralized Autonomous Organizations (DAOs). This is an entire course module in itself that will be covered later. For now, understand that DeFi governance tokens have real-world value as they are traded on secondary markets, seen in the picture above.
Compound launched its liquidity mining program in June 2020. Other liquidity mining programs cascaded from there, leading to a boom known as “DeFi Summer”. Liquidity flooded into DeFi protocols as yields increased across the entire ecosystem. Look at the increase in Total Value Locked (TVL) in the chart above. The TVL chart shows all the token liquidity “locked” in all of the DeFi protocols on Ethereum. This could be tokens in DEX liquidity pools, Maker vaults, lending pools, and more. TVL is a great metric for comparing the health of DeFi protocols (e.g. AAVE versus Compound), as well as entire DeFi ecosystems (e.g. Ethereum DeFi versus Solana DeFi).
So far we’ve talked about several ways of earning yield on Web3 tokens, such as becoming a liquidity provider on a DEX, supplying tokens to lending pools, and participating in liquidity mining programs. In the next section we discuss another DeFi yield activity – staking – and show how DeFi users stack multiple yield activities on top of one another to maximize yield. This is known as yield farming.
Staking is the idea of locking tokens to receive other additional token rewards, paid out over time. This is an odd idea at first, but locking tokens in stake pools can be beneficial to DeFi protocols – let’s look at why this might be.
Remember back to the first section how users receive LP tokens for depositing their tokens in a DEX liquidity pool. Not only is it important for DEXs to initially attract liquidity, but they also must retain liquidity to maintain good UX for its users. Some DEXs encourage users to stake their LP tokens in order to receive additional incentives.
Curve is a DEX for stablecoin swaps that offers this kind of LP token staking. Liquidity providers deposit tokens into a Curve liquidity pool, and receive Curve LP tokens. These LP tokens can be staked in Curve Gauge Pools where the user receives regular payouts of CRV tokens. Users are unable to remove their liquidity so long as their LP tokens are staked, and they are receiving CRV rewards. Do you see how this incentivizes users to keep their liquidity on Curve?
Staking increases yield for Curve liquidity providers. For example, in the first Curve liquidity pool above (“tricrypto2”), the yield is 1.89% from the pool’s trading fees; however, users receive additional yield – anywhere from 4.23% to 10.59% – paid in CRV tokens. This means, if I deposit $100-worth of liquidity in the pool, I’d expect to earn $1.89-worth of trading fees, plus $4.23-worth of CRV, annually.
Things do not end here on Curve. It has additional staking opportunities. Users can stake CRV in the Curve DAO for veCRV in return. veCRV is the governance token, so it allows users to make Curve governance decisions; however, it also boosts CRV rewards by up to 2.5x, which is veCRVs main draw. That’s why the CRV APYs – shown two images above – are displayed as a range (e.g. +4.23% – 10.59% CRV). Looking at the image above, CRV is time-locked anywhere from one week to four years. This takes CRV off the market, making it a deflationary asset, which some believe has led to its favorable price performance compared to other DeFi tokens.
Are you starting to see how yields can be stacked on top of each other? On Curve, users can deposit tokens into a liquidity pool to earn trading fees. They can then stake their LP tokens to earn CRV rewards. And then stake their CRV tokens to boost their CRV rewards. This is what “yield farming” is all about – stacking yield to maximize return on investment.
But all this makes my head spin. The yield farming I’ve described here is complex, manual, and costly. For example, CRV rewards accumulate over time and users must manually claim CRV on Curve’s UI (image above), which requires a transaction and costs a network fee. Yield aggregators – another class of DeFi dapps – solve these problems by automating yield farming for users. Users’ funds are pooled together in vaults that follow pre-programmed yield farming strategies designed for efficiency and maximal return.
Yearn Finance was one of the first yield aggregators. On Yearn, DeFi experts propose yield farming strategies and implement them as Yearn vaults that Web3 users can deposit tokens into. Essentially, these vaults aggregate a lot of tokens, and deploy them to various DeFi protocols, sharing the yield with its depositors (i.e. yToken holders).
Yearn vaults generally accept Curve LP tokens, and automate the yield farming process from here, such as moving tokens into different lending protocols in pursuit of the highest interest rate. Or, stake the LP tokens in other protocols, and automatically recycle the rewards back into the vault for a compounding effect.
This brings us to a concept called “token composability”, which is at the heart of yield farming. Essentially, Web3 tokens are interoperable with other protocols. DeFi tokens can be used throughout the DeFi ecosystem in order to stack rewards.
Let’s look at Yearn’s “Curve USDN” vault strategy. Users deposit crvUSDN tokens into this Yearn vault. Yearn then stakes these LP tokens in another DeFi protocol, Convex Finance, to earn CRV and CVX. As the vault earns tokens, Yearn automatically harvests them and converts them into more crvUSDN, which it cycles back into Convex Finance. This automation simplifies yield farming for users, and reduces their network fees. They only need to send two transactions: one to deposit tokens into a vault, and the other to withdraw tokens from the vault.
At this point in the article, we’ve discussed all the major DeFi building blocks: tokens swaps, minting stablecoins, lending/borrowing, staking, and yield farming. This all takes place across different protocols, often requiring the user to visit multiple web apps. This can make the DeFi ecosystem feel scattered. DeFi dashboards bring everything together for a more seamless UX. Let’s look at how dashboards take advantage of the open-source nature of Web3 to stitch DeFi protocols together into one dApp.
DeFi dashboards provide the user access to multiple DeFi protocols all from one UI. The first way to think of these dashboards is that they extend wallet functionality. Popular browser extension wallets like MetaMask have small UIs with limited feature sets like viewing token balances and sending tokens to other wallets.
However, users can connect their wallet to a dashboard like Zerion to view their wallet as an asset portfolio, not just simple token balances. These portfolio views help users see how their wallet’s value changes over time. Also, Zapper allows users to bundle multiple wallets together so they can track the combined value of multiple wallets.
Dashboards also directly integrate multiple DeFi protocols into one UI. This is an example of open-source software composability. Web3 protocols are open-source so other projects can use them as building blocks in their own dApp. This is juxtaposed to the proprietary APIs of Web2 applications. Users can search for the best yields across the entire DeFi ecosystem, and directly deposit tokens into lending protocols (AAVE and Compound), liquidity pools (Curve, Sushi Swap, Uniswap), staking pools (Curve and Convex), and yield aggregator vaults (Yearn and Harvest). All accessed from Zapper’s “invest” tab.
These dashboards are packed full of other useful features. For example, the Zerion dashboard has “swap” functionality that searches across multiple DEXs to find the best exchange rate to execute a token swap. This is called a DEX aggregator, and is a convenience utility for Web3 users.
The thesis that we will live in a multi-chain world continues to be proven correct. More and more, people will use multiple blockchain networks, and will increasingly want to transact tokens between these cryptonetworks. Blockchain interoperability has a long way to go and Web3 ecosystems feel siloed; however, dashboards are well positioned to improve the UX here.
Zapper tracks the value of a user’s portfolio across 11 different cryptonetworks (all EMV-compatible). And, Zerion aggregates blockchain bridges so that users can transfer tokens to other Layer 1 cryptonetworks (BSC, Avalanche, Fantom), Ethereum Layer 2s (Optimism, Arbitrum), and sidechains (Polygon), all from one UI.
So that wraps up the first wave of DeFi from token swaps to dashboards. There are other protocols I consider slightly less foundational than what we’ve covered, but the curious reader may be interested to look into some of them. DeFi Derivatives platforms allow users to trade advanced financial instruments like options (Opyn), perpetuals (DyDx & Perp), and synthetic assets (Synthetix). Also, there are token index funds that diversify a single investment across multiple tokens (see Index CoOp). This is the same idea as ETF’s and mutual funds in traditional markets. Finally, there are insurance protocols that protect users from things like impermanent loss, and DeFi smart contract exploits (see Nexus Mutual).
Interest rates have been at all time lows in traditional markets over the past several years as the U.S. 10 Year Treasury Bond hit 0.60% APY in 2020. At the same time, DeFi was on the rise, giving investors alternative markets to earn much higher (albeit riskier) yields. All the DeFi products we’ve talked about were some of the first to find product-market fit in Web3, evidenced by the fact that they onboarded millions of users onto Ethereum; however, like any first iteration of technology, there exist some problems in the current DeFi ecosystem. Let’s discuss some of the problems facing DeFi 1.0, and peer into the emerging DeFi 2.0 landscape, which seeks to address them – starting with stablecoins.
Demand for stablecoins is rapidly increasing as the Web3 ecosystem is more and more in need of non-volatile assets for things like payment. There are many types of stablecoins each with their pros and cons. And it’s also an active area of experimentation as entrepreneurs search for optimal stablecoin designs. Let’s take a look at the different stablecoin categories.
Source
The first category is fiat-collateralized stablecoins. These are, by far, the most popular right now based on market cap. Here, regulated companies, like Tether and Circle, issue stablecoins that are fully-backed by their fiat reserves. USDC and USDT are pegged to the US-dollar, but some stablecoins track other currencies (e.g. EURS). Historically, fiat-collateralized stablecoins have maintained their peg better than others; however, their fiat reserves are custodied within the traditional financial system. Governments have the ability to seize these reserves and undermine fiat-collateralized stablecoins, which many consider unacceptable for a means of payment in the world of Web3.
Enter crypto-collateralized stablecoins. This brings us back to MakerDAO where users can deposit crypto into Maker vaults and mint DAI – a stablecoin also pegged to the US-dollar. This is a big deal, because the crypto collateral backing the stablecoins cannot be seized by centralized entities. Thus, crypto-collateralized stablecoins are truly decentralized; however, the problem is that crypto is volatile, and using it as collateral is risky because the value of collateral can quickly fall below the loan’s value. Therefore, crypto-collateralized stablecoins must be over-collateralized, which isn’t an efficient use of capital.
MakerDAO was the first crypto-collateralized stablecoin protocol. Since then, other DeFi protocols have iterated on, and improved the original MakerDAO concept. For example, the crypto locked in Maker vaults is just sitting there, not earning yield. On Alchemix, users deposit crypto collateral to generate alUSD – Alchemix’s stablecoin. But here, Alchemix deposits collateral into Yearn Vaults that earn interest, which is used to automatically repay the user’s alUSD loan. It’s a self-repaying loan. Similarly, Abracadabra.Money supports yield-bearing tokens as collateral, like Yearn (e.g. yvDAI) and Convex (e.g. cvx3pool) tokens. But I digress – let’s move onto the third stablecoin category – algorithmic stablecoins.
The first two stablecoin categories are asset-backed. Either fiat reserves, or crypto collateral in vaults, support the price of the aforementioned stablecoins. A lot of recent experimentation has been done on algorithmic stablecoins, which have no assets backing them. An algorithm automatically maintains the peg of a stablecoin based on real-time market conditions for that stablecoin. This is usually done through mint and burn mechanics. Let’s pause and talk about the most popular algorithmic stablecoin – UST.
UST is built on the Terra blockchain, and is pegged to the US-dollar. UST is paired with Terra’s native token – LUNA, which is a volatile crypto asset. UST and LUNA can be exchanged in a 1:1 ratio with each other. At any point, 1 UST can be minted by burning $1-worth of LUNA. Alternatively, 1 UST can be burned in exchange for $1-worth of LUNA. Arbitrageurs utilize this mint-burn mechanism when UST is trading on centralized exchanges for a premium, or at a discount. For example, let’s say UST breaks its peg, and trades at $1.02 on a CEX. Arbitrageurs can burn $1-worth of LUNA, and mint UST in order to realize a 2% gain. On the other hand, if UST is trading at $.98, then it can be exchanged for $1-worth of LUNA. This is how UST maintains its peg.
However, it’s ironic that I’m writing this now. Just last week UST failed massively. There was a major sell-off in UST as investors lost faith, reflexivity kicked in, and caused an even greater sell-off. Billions of dollars of UST were liquidated over several days, and UST is sitting at around $.09 at the time of writing. This points to the immaturity of algorithmic stablecoins, and puts into question the feasibility of algorithmic stablecoins in general.
There is another category of stablecoin – fractional reserve stablecoins – that is a middle ground between algorithmic and crypto-collateralized. These stablecoins are partially crypto-collateralized hence the term “fractional reserve”, and partially rely on an algorithm to maintain their peg. FRAX is the most popular example of fractional reserve stablecoins.
All the stablecoins we’ve discussed so far are pegged to fiat currencies. The issue with this is that pegged stablecoins are beholden to central banks that control the monetary policies of fiat currencies. Some argue it’s unacceptable to have any reliance on central institutions in the decentralized Web3 ecosystem. The final category of stablecoins we talk about is non-pegged stablecoins. OlympusDAO is the most popular project in this space with its non-pegged OHM stablecoin.
“Stablecoin” is a bit of a misnomer for OHM. Currently, its price is around $17 – and it’s been as high as $1,300+. So OHM can be highly volatile, but one of the main features of OHM is its floor price, or risk-free value. Users exchange crypto with OlympusDAO for discounted OHM in a process called bonding. All bonded crypto is held in OlympusDAO’s treasury, which ensures a floor price of at least $1 OHM. Thus, the value of OHM is free-floating, but users can always have faith in the floor price, backed by Olmypus’ war chest.
This leads into the next problem that DeFi 2.0 is attempting to solve – that of deep, permanent token liquidity. OlympusDAO controls a massive treasury, which is a great example of an upcoming concept – “protocol controlled value” (PCV). The protocol can deploy liquidity as it sees fit to ensure liquidity for its native token, or to earn yield and grow the treasury larger. Also, their protocols can rent out their liquidity to help other protocols bootstrap token trading, or bolster their liquidity pools.
We know that liquidity is important in all aspects of DeFi. Without tokens in liquidity pools you can’t make swaps on DEXs, borrow tokens in lending protocols, and enact automated yield farming strategies. Without liquidity DeFi grinds to a halt. And the more liquidity the better. For example, the more liquidity in a DEXs liquidity pool, the less slippage users are subjected to when making token swaps. That’s why DeFi protocols compete with one another to attract liquidity, and increase the TVL of their protocol.
Liquidity mining, first seen with Compound in 2020, attempted to overcome the liquidity bootstrapping challenge. Protocols mint their governance token, and distribute it incrementally as a reward to its liquidity providers. Unfortunately, this only solves the liquidity problem in the short-term. Liquidity mining programs attract mercenary capital. Without loyalty to any single protocol, users switch their liquidity between protocols whenever higher interest rates popup elsewhere. It’s difficult for protocols to maintain this liquidity, and they have to continue inflating their governance token in order to do so. But more and more, liquidity is managed at the protocol-level, called “protocol controlled liquidity” or “protocol owned liquidity”.
DeFi 1.0 protocols do not own user liquidity. Users own their LP tokens, and can redeem them for the underlying liquidity at any time. This is changing in DeFi 2.0. Let’s go back to OlympusDAO, and users bonding their crypto in the treasury. Here, user crypto is exchange for discounted OHM. In this case, the users no longer own their liquidity – the liquidity is now permanently held in OlympusDAO treasury. In fact, OlympusDAO controls over 99% of its liquidity in DEXs (e.g. OHM-FRAX LP & OHM-DAI LP).
This ensures that the liquidity remains in place so that users can trade in and out of OHM, with low slippage. Also, the protocol earns almost all of the liquidity pool trading fees, and is becoming a large holder of SUSHI, which all work to further grow the protocol.
Frax Protocol is another good example of PCV. Users mint FRAX stablecoin by depositing their crypto assets into Frax protocol’s reserves. Frax Protocol applies automated market operations (AOMs) on its reserves for a variety of reasons. For example, its collateral investor AMO moves USDC reserves into lending protocols like Compound and Yearn, which earns interest. Another AMO provides FRAX liquidity on exchanges like Curve and Uniswap to ensure deep liquidity pools for FRAX-stablecoin pairs. Fei Protocol is similar – the protocol controls reserves for yield farming strategies to grow its reserves, as well as to provide FEI liquidity on Uniswap.
These protocols can also rent out their liquidity to other DeFi protocols, known as liquidity-as-a-service (LaaS). Tokemak is a great example of this. Essentially, individual liquidity providers deposit tokens into reactors, and liquidity directors (LDs) rent out this liquidity to various DEXs. LD’s create liquidity on-demand – they vote on which token reactors get paired together, and which DEXs to send these token pairs to. Of course, Tokemak earns DEX trading fees, which its building up a treasury with. This treasury, controlled by TOKE governance, will be utilized for additional LaaS operations in the future. Tokemak is like a liquidity aggregator, supplying liquidity to other specific areas in the DeFi ecosystem.
Finally, LaaS is being used to help other up-and-coming protocols launch their native token. Before, to start sellings a token on a DEX you either needed to seed the liquidity pool with a lot of up front capital, or attract users to bring their own liquidity. This brings us back to liquidity mining, which we’ve already established is ineffective in the long run. However, now, OlympusDAO launched a product called Olympus Pro, which generalizes its token bonding mechanism for other protocols. For example, users can bond their LP tokens in exchange for the protocol’s discounted governance token. Now that protocol owns the LP tokens, so the liquidity will remain in place without continuous liquidity mining. This is known as “protocol pwned liquidity” (POL).
Also, FEI and FRAX have partnered with Ondo Finance, which provides liquidity pools to protocols looking to list their governance token in an IDO – initial DEX offering. The protocol deposits its governance token in a pool, which is matched by either FRAX or FEI stablecoins. The Frax and Fei protocols earn 5% APR on the liquidity it provides. This is like renting out stablecoin liquidity so that other protocols can quickly get up and running selling their governance tokens.
I, Travis Kassab, am not a financial advisor, tax professional, broker, or legal advisor. None of my podcasts, videos, or articles constitute financial advice. This content is purely for educational purposes, and is not intended to endorse any cryptocurrency or other investment vehicle. At the time of writing this, my cryptocurrency holdings include: BTC, ETH, CEL, SNX, BANK, SOL, DOT, KSM, USDC, GERO.
No activity yet