How should Maker respond to the black swan crisis?

Content

In the history of encryption, there is a character that cannot be ignored. He is Ross Ulbritcht, the founder of the black market website "Silk Road". He was arrested and imprisoned in 2015. More than 9.5 million bitcoins were in circulation on the "Silk Road", which accounted for most of the bitcoin circulation at that time. This is also a major reason why Bitcoin was criticized in the early days.

Interestingly, Ross Ulbricht did not forget to pay attention to the development of the encryption field while in prison. He wrote a lot of articles about Bitcoin.

Recently, he has also conducted research on the Maker protocol and put forward his own views on improvement. He believes that the current Maker protocol has not fundamentally solved its ability to respond to crises. In order to solve the problem fundamentally, he believes that iterative improvements must be made from the underlying mechanism.

Re-recognize the role of Vault owner

Ross Ulbricht thinks Maker’s idea is cool. It creates a stablecoin that can track the value of the dollar. However, the 3.12 Black Thursday crisis in 2020 revealed the loopholes in Maker, and the current mechanism does not fundamentally solve this crisis response problem.

Ross Ulbricht was surprised that in the overall system of Maker, the owner of the vault was defined as a "borrower". He believed that the cornerstone of the Maker system lies in the owner of the vault, because they are only talented. It is the person who provides support for the value of DAI.

In Ross Ulbricht's view, the root cause of the crisis in the Maker3.12 Black Swan incident was the misunderstanding of the role of the vault holder. Vault holders are not borrowers and should not pay interest (stability fee). Instead, they should charge interest. They should be regarded as "lenders."

Ross Ulbricht mentioned the model of housing mortgage loans. He believes that the current Maker model compares ETH to a house and DAI as a loan from a bank. Vault holders need to pay interest (stability fee) for the borrowed DAI, and if the value of the collateral falls, it will also cause the vault to be auctioned and liquidated. This is like the homeowner cannot repay the loan and his house will be auctioned off by the bank.

Ross Ulbricht believes that this model is difficult to operate effectively. Because if DAI is separated from the collateral, it has no value. The value of DAI stems from the ability to redeem its collateral in a liquidation auction. The vault itself is similar to a bank. The vault holders issue DAI and use their real money to support DAI, and they should receive interest instead of paying interest.

Vault is similar to traditional banks before the introduction of modern central banks. Banks own gold in their vaults. Gold supports the issuance of currency. The issuance and lending of DAI are also supported by the collateral of the vault. Vault is not only a bank, but also more transparent than a bank. Because modern banks often adopt a partial reserve system, and the vault on Maker cannot cheat, all happens on the transparent blockchain. Vault holders cannot issue too much DAI.

Therefore, Vault should be regarded as a lender rather than a lender.

Repair measures to deal with black swan

A remedial measure in the Maker Agreement is to set up DSR. Users can lock DAI to get storage interest, so that the vault holders who issue DAI can also get interest. As long as the interest is higher than the stability fee, it is equivalent to the Vault owner earning interest.

Maker’s original intention in doing this is similar to the Federal Reserve, adjusting the amount of DAI in circulation through interest rate adjustments, so as to achieve its goal of anchoring DAI to the U.S. dollar. However, this also hides a problem: This will incentivize people to lock in collateral such as ETH, and then earn interest. This does not match DAI's goal of building a blockchain-based decentralized stablecoin. It should not encourage people to lock DAI in smart contracts.

In Ross Ulbricht's view, a better way is not to adjust interest rates from top to bottom, but to achieve price stability through market forces. As long as there are arbitrage opportunities, the value will tend to be balanced. Attempting to set important parameters through a top-down committee model is a weakness in itself, not a strength.

In practice, as in the practice of the 3.12 black swan, as the price of ETH dropped by 50% rapidly, the price of DAI surged by 10%, causing many vaults to be under-collateralized, forced to liquidate, and to pay stability fee penalties. At the same time, DAI holders are unwilling to give up the stable and high-yield DAI during the period of asset decline, which leads to a shortage of DAI. In order to solve this problem, Maker introduced more stable USDC and other collaterals, but this is hard to see as a decentralized solution.

If you ultimately want to become an independent decentralized stablecoin protocol, its better choice may be to achieve it through market incentives, while rejecting the temptation of top-down control.

Vault-based solutions

The core of Roth's solution is to revolve around the interest rate and mortgage rate of the vault. If according to the current maker's settings, according to his plan, the stability fee can be set to a negative number, which can benefit the vault owner. Vault provides DAI holders with a decentralized stable currency service, and DAI holders should pay for it. This is a reversal of thinking.

Furthermore, in Ross's solution, vault holders can set their own interest rates, and vault holders compete with each other. The lower the interest rate, the easier it is to win interest from DAI holders. This competition can keep interest rates low. Once there is a shortage of collateral, interest rates will increase through market forces, encouraging more collateral to enter the system and generating more DAI. At the same time, rising interest rates will also suppress people's demand for hoarding DAI.

For Vault owners, they can get a positive return due to mortgage generation DAI instead of paying a stability fee, which gives them greater motivation to create DAI. This is in contrast to the lack of motivation for some people to use collateral to generate DAI. At the same time, if it is self-generated DAI, there will be no interest income. Then, the treasury holders will want to put the DAI created by them into the market, allowing others to hold assets, and thus get a return.

This is conducive to generating sufficient fluidity. If the DAI price is lower than the anchor price, DAI holders have the opportunity to redeem the collateral, and people have an incentive to buy discounted DAI and redeem it to obtain additional collateral. Then sell the collateral to make a profit. Therefore, as the supply of DAI decreases, the price of DAI tends to be anchored. If the price of DAI is higher than the anchor price, vault holders have the incentive to issue more DAI, which can obtain higher returns and increase the market supply of DAI, which will eventually make the price of DAI tend to be anchored. In this way, the DAI market achieves greater depth and liquidity.

As mentioned above, Vault can set its own interest rate, and lower interest rates can win the competition, which is beneficial to DAI holders. However, this also needs to ensure the reimbursement of DAI. This involves the issue of the vault mortgage rate.

In Ross Ulbricht’s plan, a vault with a higher mortgage rate can obtain interest from a vault with a lower mortgage rate. But only to obtain a part of the interest of the already issued DAI treasury, which is mainly to encourage the treasury with a higher mortgage rate to participate in the generation and sale of more DAI. This scheme also involves the cut-off rate, because there is excess collateral in the system. With the cut-off rate, the vault that has issued DAI will give part of its interest to the vault with over-collateralization and below the cut-off rate, and both parties earn a lower interest rate.

The purpose of these mechanisms is to maintain excess collateral in the system. When the collateral is too high, the collateral can be reduced; when it is too low, the collateral is provided as an incentive. This also requires a mechanism that allows DAI to automatically transfer from a vault that is close to insolvency to a vault that is over-collateralized. This can be done through liquidation auctions. The vault uses the excess collateral to generate DAI and uses it to bid. If it is during a crisis, in order to respond quickly, automatic execution can be used to speed up the processing, and the assets and liabilities (collateral and DAI) can be simply transferred to a well-funded vault. The treasury can mark part of its excess collateral and charge a certain fee, and the treasury will be liquidated first to the low-cost treasury. This is similar to the "debt transfer" mechanism of the Liquidity agreement mentioned earlier by Blue Fox Notes.

The mortgage rates set by different treasuries in the system tend to be a simple "market interest rate", which is the cut-off rate. Vaults below the cut-off value will "leave the money on the table" because as long as they are not higher than the cut-off value, they can charge more without giving the money to a cheaper vault. On the other hand, vaults above the cut-off value lose money to vaults that are below the cut-off value and have excess collateral.

When the value of the collateral drops, the cut-off value will automatically rise, because there is less excess collateral below the cut-off value, so that the vault with a higher mortgage rate can charge the full amount of interest. The capture rate will also increase because a small portion of DAI is paying interest on excess collateral. These two things will also encourage new vaults and new collateral to flow into the system, alleviating shortages, and these are market forces.

Ross's views and the project mentioned by Blue Fox Notes before "Can Liquity Challenge Maker?" 》There are some similarities, the essence of which is to achieve the stability of stablecoins through market forces. In addition, in the practice of stablecoins, there are two other projects mentioned in the Blue Fox Note: "The Holy Grail of Stablecoins: Exploration of New Paths" and "Meter's Path to Stablecoins: Explorations Completely Different from Libra" , These different paths constitute the continuous exploration of decentralized stablecoins.

DeFi governance will become more and more important

Compound's release of COMP is not only a token distribution model and incentive mechanism, but also a governance token. It has completely activated the enthusiasm of governance, and as the operation of Compound has deepened, more and more proposals for improvement have been put forward. Here are the collision of plans and the game of interests, which promote the progress of Compound and reflect the flexibility of governance.

If more DeFi projects can upgrade the agreement through governance and adjustment of algorithms and game mechanisms, this may be the key to a project's long-term competitiveness and moat.

Governance will become heavier and heavier. In the development of blockchain, the top-down governance model has an efficiency advantage at the beginning. However, in the long run, this is incompatible with the spirit of the blockchain itself, and may not be the optimal model. A reasonable governance mechanism may give rise to late-comer advantages and surpass the leaders.