Our team’s experiences and candid thoughts using the compound.finance interface.
Last year we decided that we needed to be more proactive about using the exciting protocols and dapps around the Ethereum ecosystem. We noticed that we were talking about a lot of these protocols and dapps, but rarely got around to actually using them.
We have a feeling a lot of you are in the same boat.
For January, Compound was our protocol of choice. We gave every MyCrypto team member 20 DAI and told them to lend (supply) that DAI using Compound. We then collectively shared our experiences — around what was easily-understood and/or confusing about the experience and the concept as a whole. The below shares a bit about what we learned.
Please keep in mind that our backgrounds range from technical to non-technical, but none of us are financial wizards, traders, or from a finance background. If you see wrong information, especially if you have a deep knowledge of financial systems, please politely enlighten us. We would love to learn more.
Some of our team member’s answers:
- To earn interest on your crypto
- Gain interest, probably some other fancy financial shit
- To earn interest on otherwise static funds
- It’s a way to gain interest on your ETH. You supply ETH for others to take loans out (if anyone needs loans…) and you get interest back on that ETH.
So, that’s pretty clear cut: you want to make a bit of money off the assets you have. Cool beans. The more interesting question though….
The answers we came up with fell into two categories:
- There is still risk in lending out your assets (e.g., the Compound smart contracts are hacked).
- You don’t have as much access to your funds (e.g., if you lent out ALL your ETH, every time you wanted to send or trade ETH, you would have to withdraw from Compound first).
This helps distill our previous answer a bit more: you want to make a bit of money off the assets you aren’t planning on doing anything else with…with minimum risk. Just like you don’t put 100% of your money into crypto, you don’t lend out 100% of your crypto for similar reasons.
Because you’re doing some fancy financial trading stuff? Hedging your bets? Shorting something? Moving piles of money around?
“Going long” is basically investing. It’s what you did to buy ETH or BTC or invest in an ICO. You use one asset to buy another asset because you think the price of that second asset will go up relative to the first asset.
- Example: You buy 1 ETH for $100 USD because you think you can sell that 1 ETH for $200 USD in the future and make a profit of $100 USD.
- Example: You buy 7 REP for 1 ETH because you think you can sell those 7 REP for 2 ETH in the future and make a profit of 1 ETH.
However, basic “investing” isn’t usually referred to as “going long” in the crypto ecosystem. That’s called HODLing.
Going long usually refers to the practice of using someone else’s money to buy more of an asset than you could otherwise afford because you think the asset you are borrowing will go up more than the asset you put as collateral. Compound explained it in a blog post:
- Alice wants increased long exposure to ZRX.
- She supplies 10,000 ZRX to Compound, and uses it as collateral to borrow 30 WETH.
- Alice unwraps the WETH, and sends Ether to her favorite exchange to purchase 6,000 more ZRX.
- Alice now has long exposure to 16,000 ZRX, and owes Compound 30 WETH.
- She is long ZRX/ETH.
- If ZRX/ETH increases in value, Alice will be able to repurchase the 30 ETH she owes Compound for less than 6,000 ZRX.
- She can then repay her debt, and keep the excess ZRX as profit.
Okay, still with me? Good, because it’s about to get even more weird.
This is probably a common reason to borrow via Compound (we actually have no idea), even if it seems like an exceedingly large number of steps. First, let’s try to understand what shorting actually is.
This is what you do when you think one asset will go down relative to another asset. What most people don’t realize is the steps that are actually occuring when you open a short position.
- Let’s say you think the price of ETH will go down relative to USD.
- You have $100 USD in your pocket.
- You borrow 1 ETH, using that $100 USD as collateral. (+1 ETH, -$100 USD.)
- You agree to pay a fee to borrow that 1 ETH — say $1/month. Meaning, you will get that $100 USD back as long as you give them 1 ETH + $1/month.
- Balance: 1 ETH and $0 USD.
- You immediately sell that 1 ETH on the market for $100 USD. (-1 ETH, +$100 USD.)
- Balance: 0 ETH and $100 USD. Somewhat back to where you started, except you sold on the market — you didn’t settle with Compound yet so you’re still engaged in borrowing.
- A month later you buy 1 ETH on the market for $90 USD. (+1 ETH, -$90 USD)
- Balance: 1 ETH and $10 USD.
- You repay the 1 ETH that you borrowed from Compound, pay the $1/month fee, and get your initial $100 USD back. (-1 ETH, -1 USD, +100 USD)
- Balance: 0 ETH and $109 USD.
- You made a profit of $9 USD!
- Bob wants short exposure to ZRX.
- He supplies 100 WETH (WETH = wrapped ETH. For the sake of learning just think of it as ETH) to Compound, and uses it as collateral to borrow 12,000 ZRX.
- Bob sends the ZRX to his favorite exchange, and sells it for 60 ETH.
- Bob now has long exposure to 160 ETH, and owes Compound 12,000 ZRX.
- He is short ZRX/ETH.
- If ZRX/ETH decreases in value, Bob will be able to repurchase the 12,000 ZRX he owes Compound for less than 60 ETH.
- He can then repay his debt, and keep the excess ETH as profit.
Most exchanges have this all automated. They also may allow you to “leverage” your position, meaning you would take your $100 USD, borrow 2 ETH, sell 2 ETH for $200, buy 2 ETH for $180, repay your original loan of 2 ETH, get back your $100, pay a fee of $1/month, and have $19 in profit at the end of the month, even though you still only had $100 to begin with. (Warning: while this may seem attractive, don’t try this at home. The reason shorting is risky is because if the market goes up instead of down, your loan gets called in and you lose your money.)
Anyway, so this is all automatic on an exchange and you don’t realize the steps that are occurring. However, with the new decentralized finance tools like decentralized exchanges (DEXs), you can really only trade an asset you have for another asset of equivalent value. There is no “short” option on DEXs if you think an asset will go down.
However, there is the “borrow” function on Compound. So you could borrow REP using your ETH (WETH) as collateral. Then you could take that REP to a decentralized exchange and do the above steps manually on that exchange before returning to Compound to repay the loan and make a profit.
- Decentralization > Centralization: Because you don’t like or trust your data/money with a centralized exchange.
- “Regulatory Arbitrage” aka fuck KYC.
- “Regulatory Arbitrage” aka avoiding taxes: Governments have made it clear that you must pay taxes on any gains made by trading crypto. While some people previously assumed that they didn’t have to calculate their gains on every single BTC/ETH, ETH/BTC, BTC/ETH, ETH/REP, REP/BTC trade they made over the entire year, many governments have decided differently. Centralized exchanges are providing the IRS with records of all your trades, etc. By using decentralized systems, some people believe they can hide from the tax man (not advisable, by the way). It’s also possible some are trying to avoid a “trade” by “borrowing” instead. Some believe this “borrowing” would not trigger the realization of your capital gains, which would be especially beneficial for assets that have gained immense value over time. We’ll touch on this below a bit. Again, avoiding the government or tax man is not advisable!!
We looked at this tweet thread to try to gain insight: https://twitter.com/econoar/status/1070174096002973696?s=19
Our understanding of the above: Instead of trading REP into ETH or DAI, they used their REP as collateral and then borrowed all the DAI…for some reason.
- They took their pile of REP.
- They lent it out.
- Now they have a pile of REP they’re gaining interest on.
- Then they borrowed DAI.
- Now they have a pile of DAI they are paying interest on.
- They take that pile of DAI and sold it for ETH.
- They added that ETH to their existing Maker CDP.
- They now have a pile of DAI in their pocket from the CDP (?)
We assume they want to get the original REP back at some point. To do this, they will likely:
- Slowly take ETH collateral out of the Maker CDP as market goes back up
- Sell that ETH back to DAI
- Pay back the DAI loan on Compound
- Get REP back
Why would one do this? There are a few possibilities:
- The market depth for REP/ETH was too shallow and selling that amount would move the market, where the DAI/ETH market was deeper.
- They have other financial interest to keep the price of REP high (e.g., they hold way more REP than what they lent out, or they are affiliated with the Augur project in some way and wouldn’t want the market thinking they lost faith in Augur).
- “It would certainly be better to do that if risk reduction was the users only goal, but they could also be long on Eth at these prices. So they add some protection to CDP and enter into a leveraged long on Eth at the same time” — atyreefinch
- “My hunch is: they’re getting leverage on ETH from the REP collateral, while at the same time giving some (not desperate) extra protection to their CDP, and (maybe mainly) increasing their cost basis for ETH, in preparation for paying less capital gains taxes in the future.” — FollowTheChain
- “Regulatory Arbitrage” #1: Let’s say they bought the REP with an original cost basis of $1 USD. If they were to sell that REP, their gains would be realized and therefore they would have to pay taxes on those gains (15% of ~$10,136,000) — even if they are trading that REP for ETH and don’t have the USD to pay that tax bill (and don’t want to sell for USD in these market conditions).
- “Regulatory Arbitrage” #2: Let’s say they originally bought ETH with a cost basis of $1. Each time they sell any ETH, they pay huge taxes on their capital gains. However, if they buy ETH @ $125, it raises the cost basis to $125. When they sell those ETH at $130 or $150 or even $1000, the capital gains tax would be less.
So yeah, that’s a whole boatload of speculation and probably not the real reason for taking those actions. But basically: fancy financial trading magic that we don’t fully understand but probably somehow makes the person some money.
Our conclusion: you probably wouldn’t.
- It’s super easy, especially if you are using someone else’s money. There is literally only one call-to-action each step of the way, so if you just keep clicking the buttons and approving your MetaMask transactions, you’ll successfully supply a loan and start making interest the next block.
- There is not enough educational material on the Compound website or web app to let us know what the heck we were doing each step of the way. Also, why borrow?
- There is not enough information for any of us to make a decision on whether or not we should actually supply. Because everyone on the team was literally told to “supply a loan,” everyone did so. However, if this was our money and we were trying to decide whether or not to supply, we would want to know: “How much will I make? Over what period of time? How do I know how much I have already made?”
- Why do we have to “enable DAI” first and what is the MetaMask transaction that pops up for? Again, if this was our money we would have stopped before enabling.
- The people who read the additional information Taylor tracked down (https://medium.com/compound-finance/the-compound-guide-to-supplying-borrowing-crypto-assets-94821f2950a0 and https://medium.com/compound-finance/faq-1a2636713b69) were much more confident in their understanding of what was happening. Almost nothing found in these documents can be gleaned from the interface itself.
- ETH being separate from WETH. What? Why? What is Weth? Where do I get WETH from ETH? Ahhhhhh!
- We visited Investopedia a lot and asked questions in our Slack a lot.
Again, none of us are crazy about trading or have a deep understanding of these sorts of financial systems. If you see wrong information or can provide real-life examples that are easier to understand, especially the why behind these maneuvers, please share with us. We would love to learn more.