What’s a Black Swan?
And how a Black Swan Event played out on Maker’s Black Thursday
With the Corona pandemic raving havoc on lives and economies worldwide, the term “Black Swan” is back in fashion. However, there might be a bit of a misunderstanding of the word. Even according to the author first popularizing the term, the pandemic does not qualify as a Black Swan Event.
But what exactly is a Black Swan Event?
The term Black Swan is derived from the following happening. When humans first learned about swans, for millennia, they’d only be confronted with white swans. It seemed reasonable to assume that all swans were white — that it was one feature of a swan. Until one day in Australia, the first human stumbled across a Black Swan. This simple sighting of one black swan wasn’t just a big surprise but also invalidated the general statement built over millennia that swans were black.
The first to really popularize the term “Black Swan” to refer to a specific type of event was Nassim Nicholas Taleb, who initially used it in his book “Fooled by randomness” in 2001, and then dedicated a whole book to the topic in the “Black Swan.” Curiously, he published that book in 2007, right before the financial crisis.
The three signs of a Black Swan event:
- They are entirely unpredictable. No matter how much modeling you do, the likelihood of predicting a Black Swan Event is extremely low, as it’s just out of the imaginable.
- Bring severe consequences with them, often with a dramatic impact on lives and livelihoods.
- They are often tried to be explained as foreseeable in retrospect.
(Note, Black Swan Events can be positive as well. Sometimes it also depends on your perspective. Sure, the big financial crisis was terrible for the world economy, but those who shorted before the crash went out a lot richer.)
In the Black Swan, Nassim Taleb explains that our ability to learn from experience or experience is minimal, as illustrated by the first sighting of a black swan. Sometimes all it takes to invalidate what we thought we knew is one single observation. If we could foresee Black Swan events, things would be easier.
It might be less severe in the swan’s case but think of the terror attacks on the Twin Towers in 2001 that shook the world. Had on the 10th of September that year anyone considered that terrorists could hijack a plane and fly it into the World Trade Center, military jets would have flown around to protect the building. Planes would have locked bulletproof doors keeping the pilot from the main cabin, and the attack would never have happened. But even with all these measures hypothetically in place, the terrorists might have come up with a completely different way to attack.
That’s the scary part of Black Swan events. We do not know, nor do we have the tools to predict them. This uncertainty has led the author of the Black Swan to conclude that:
“Inability to predict outliers implies the inability to predict the course of history, given the share of these events in the dynamics of events.”
Standard forecasting fails to predict black swan events and even increases our vulnerability towards them as we’re left unprepared. While we’re all familiar with the financial crisis of 2008, in crypto, we experienced a similar unexpected Black Swan Event just last year.
Maker Black Thursday
While many have preached the gospel of crypto being completely trading separately from macro-economic movements, last year, the beginning of the Coronavirus pandemic proved us wrong. Not only did stock markets plummet, as the world economy came to grasp with being put on hold, panic also spread to crypto markets, and we saw the price of ETH plunged more than 50%.
Maker DAO, a decentralized borrowing and lending platform, suffered its worst day in history due to the ETH price dive on March 12th, 2020. To understand why the price dive had such an impact on the Maker ecosystem that the day became known as Maker’s Black Thursday, it’s important to understand how Maker works.
Put simply, on Maker; users can take out a stablecoin loan against their cryptocurrency holdings. They do so by depositing their volatile ETH, and in return, they receive the Maker stablecoin DAI.
Unlike other stablecoins, DAI doesn’t rely on dollars held in a bank account for its value but for the collateral deposited by anyone who mints it. To hedge the volatility of crypto collateral, borrowers have to over-collateralize their loans. That means they deposit more in collateral than they take out in loans. In the below illustration, when depositing $100, the user takes out $66 worth of DAI. This is known as a collateral debt position (CDP).
When the value of the deposited collateral falls, a liquidation mechanism is triggered that sells off the collateral to recover the loan position in an auction. This works similarly to a Margin call that traders get when trading with leverage.
Before March 2020, Maker had grown to be one of the biggest lending and borrowing platforms in DeFi, facilitating millions in loans. Yet, no one expected what would happen once the price of ETH began diving.
March 12th, 2020
In case you were unsure where exactly March 12th is on this price chart, it’s the big red candle showing the price of Ether crashing by more than 50%. Fuelled by a sell-off in markets across the board (stonks were having a bad time as well), Maker’s liquidators started getting to work when the price started falling.
Initially, everything was working just fine, as borrowers have to deposit at least 150% of the loan value, which gives liquidators a buffer to seize high-risk loans and maintain the system's solvency.
Liquidators in crypto are individuals and businesses taking advantage of liquidations that happen on borrowing and lending platforms. They essentially buy the collateral that borrowers deposited with their own funds to ensure that the loan can be recovered. While potentially highly lcurative, liquidations come with a high opportunity cost for liquidators. They never know what amount of stablecoin they might need to keep (and can’t deploy otherwise) to be able to seize the opportunity.
However, the system started failing when at 5 am pacific time, the price of ETH hit $130 hit, and the system begins to build up debt. Unlike other lending protocols or centralized exchanges that rely on insurance funds to cover losses, Maker assumes the possibility of debt and addresses it with the issuance of its native token: Maker (MKR). In the case of system debt, a new Maker is issued and sold for DAI — despite the danger of diluting the Maker supply.
Liquidators bidding $0 on ETH
What accelerated the growing system debt was that the price after liquidations started to fall below market level while ETH continued its downfall. During auctions, liquidators managed to bid $0 on ETH and win. In total, $8.23 Million worth of ETH was liquidated for 0 DAI in return. This obviously wasn’t foreseen by the Maker team or by borrowers (vault owners) who now didn’t receive any collateral back.
Yes, that’s right. Some liquidators got ETH for free. To be a liquidator on that day. But this was an absolute disaster for Maker. It could only happen due to the set up of auctions designed to withstand normal market conditions.
In a normal auction, liquidators bid against each other during the first stage, increasing amounts for a specified amount of ETH. Once the minimum value is reached, the auction enters the value maximization stage, during which liquidators bid less and less until, in the end, the winner leaves with discounted ETH. In a normal market, even if there were a bid for $0, someone would logically bid at least $1 for ETH; even when it’s trading around $100, that is still a great discount. However, 2 factors made matters worse:
- Network congestion: whenever there is a rush on Ethereum (remember Cryptokitties?🐈), the price of gas increases. In this case, more than 10–fold. On top of that, Maker itself was relying on the price feed of Chainlink — a decentralized Oracle that only executes when paying it in gas. This slowed down the liquidation process and made low-value transfers on time nearly impossible.
- DAI liquidity dried up: usually, DAI is a very liquid stablecoin that you can easily buy and sell on exchanges and DeFi protocols. Not so much during this 12th March 2020. As users feared being liquidated, fewer users deposited ETH into the protocol to mint new DAI. The supply of DAI shrank while more traders redeemed their DAI. It became so scarce that the price of 1 DAI went up to $1.12 (it usually trades around $1), making it hard for liquidators to get enough DAI to do their job.
Maker had accumulated a system debt of $5.3 million at the end of this crash—a lot for a DeFi protocol.
In reaction to this Black Swan event, maker has addressed the DAI liquidity issue by lowering the saving rate to 0% (what holders of DAI would receive) in order to reduce demand and lower the price of DAI to its soft-peg again.
They also added USDC as an option to mint DAI to the platform, which gives users a very low-risk way of taking out a loan against their crypto, that even during market turbulence will maintain stability. The team pledged to increase the robustness of auctions and reduce the number of transactions for bidders, so that transaction fees on Ethereum wouldn’t play as big of a role.
Additionally, Maker minted new MKR and sold it to DAI in order to reduce the system debt.
After the event, some vault owners (borrowers) sued the maker foundation for not providing them with accurate information on the risks associated with opening a CDP.
Maker has done a lot to address problems detected curing this incident including adding multi-collateral DAI and making its platform more accessible for all kinds of users.
While there is no sure-fire way to predict black swan events (that’s just their nature), creating resilient systems and diversifying can offer some hedge against them. And, lastly, it might also depend on your perspective. While it was a Black Thursday for Maker, it wasn’t for those few liquidators lucky enough to receive Ether for free.