Would RWA have helped DAI AVOID going to $1.25-$1.30 versus USDC?

Thanksgiving Morning in the U.S. 11/26/2020:

RE: “DAI price increase led to a massive $103 million worth of liquidations at DeFi protocol Compound” –TheBlockCrypto

With this mornings incident ( Jiecut ) of the Compound price oracle being exploited–if DAI was 50% collateralized by RWA, would the price of DAI have spiked as much as it did today, or would RWAs have saved the day and more importantly avoid the bad publicity the Media will portray on DAI over the next couple of days?

Also, how does this impact our ability to market DAI as an asset for RWA borrowers & lenders?

How much longer do we wait for RWA?

Do we wake up to another exploit on Christmas Morning? Seems like there’s an exploit every 48 hours…

"While Risk is Real, Fear is Not" --Mike Maples, Jr.

4 Likes

Price spike today’s wasn’t structural problem of DAI, but problem on Coinbase side. Market Makers must have had some issues with withdrawals and deposits and CB was down twice today already. Price was very calm elsewhere so I don’t perceive today’s price issue to be really related to Maker system at all.

11 Likes

You can start by reading the 6S risk assessment :slight_smile: Smart contract can be live on December 18th (according to last news). I guess @mrabino1 have a lot of work to finalize the legal stuff.

New Silver and ConsolFreight next week.

For ConsolFreight, the SPV it’s already on chain. We need the adapter on our side. New Silver just behind I guess.

Most operational processes (RWA team) to monitor/manage RWA will be done in December.

I discussed the subject somewhere else, but I don’t think RWA can help to offset a sudden spike in DAI. You can’t create a RWA loan in minutes. My discussion on accounting opens a perspective to be pro-active on the peg management (on top the PSM when it’s not enough). But I’m not sure it’s our job to make orderly markets on Coinbase.

3 Likes

Curve was okish I believe, with a dai probably cap at 1.01.
However more than RWA, PSM should fix this with a cap at 1.001

1 Like

PSM will tell us precisely how much RWA are needed to be onboarded (by looking at the stablecoin collateral).

RWA is relatively slow and lethargic. not exactly a “micro” market response tool… but it can scale … and therefore it is a superb “macro” market response tool.

2 Likes

after the holidays i am planning on doing an update twice a week with the implementation status …

4 Likes

Coinbase doesn’t have a lot of liquidity, it just isn’t the place where DAI price discovery is being done.

2 Likes

most certainly not with their fee structure on stablecoins

1 Like

Yea looks like Uniswap as well per Hayden: https://twitter.com/haydenzadams/status/1332062866191638535?s=10

Thank you for the feedback Primoz. But assuming one day DAI has 30-40% hard assets like Real Estate as a collateral—do you think the Coinbase Oracle would be smart enough to detect that DAI should not be 10% above the Peg? I mean I might be asking a crazy question but I want to imagine a future where a price Oracle understands that a pegged asset should not fluctuate 20-30%. Or, maybe I’m putting the cart in front of the horse…

Yea I guess I am looking for a circuit breaker. If any CEX or DEX sees the price rise 7% above the peg — they turn off all DAI trading… but obviously that’s prob not possible. And I guess protocols want to perform liquidations? Even if it’s bad publicity? Not sure how to view this…

And for protocols that due perform liquidations — where’s the stress test? Where’s the guidance based on scenarios? I guess this is where Gauntlet comes in. Can’t wait to see them stress test liquidations 2.0 @cmooney

Well CB oracle is based on DAI traded price on CB and for some reason nobody arbed the price at that moment today. There was free money on the table for some time. It was sort of a “flash pump” that took a bit longer (about 15min). Flash crashes happen to exchanges, but usually such spikes are shorter in time. Not sure what RWA has to do with with traded price on CB.

3 Likes

I’m assuming a CEX and the markets would price an Synthetic asset based on Collateral value. But yea I understand what you’re saying :+1:

@Primoz @SebVentures @mrabino1 This is what I was referring to, “Capping” of DAI because of RWA and Oracle feeds that know “Bricks” cannot trade higher than $1 USD–from the Compound Forum:

Broadly, potential changes could include:

  1. Modifying the DAI market parameters, including the Borrowing Cap, to reduce the size of the DAI market relative to trading venues
  2. Modifying the DAI price feed by either tightening the anchor bounds, capping the price (e.g. to 1.05), or utilizing additional reporters
  3. Removing the reporter in some/all cases and relying only on Uniswap, or taking an entirely different approach

Author: Robert Leshner

"Real Estate is liquid as a brick" --Unknown

1 Like

As purely a thought experiment (and let’s hope the following does not actually occur), if ETH were to collapse to $10 in a flash and then stay there for 6 months, what would the theoretical outcome look like?

Vault holders using ETH would need to acquire DAI in a rush to not cause a liquidation. The price of DAI would clearly be pushed to “elevated” levels (e.g. probably breaking over $2). Put bluntly, the entire Maker project would be at risk. There is no certainty that USDC could even keep up with such a scenario.

In my eyes, I don’t think capping the DAI price that is fed into the system is the solution; however, I do think you are on to something about how we could use RWA (or more specifically the RWA structure utilizing a Trust) as a mechanism for such a black swan event.

While it is too early to chart out specific plans for the following, the discussion to start the thinking has merit. As contemplated, the Trust is a legal entity for the benefit of the DAO that is “pre-wired” with a secured conduit for getting DAI specifically to a destination that includes instruction to sell DAI to the USD. If we ignore slippage (buy and sell) and the fees service providers will charge (based on AUM), the RWA structure itself could be utilized in such an emergency situation. While I am not advocating we use a specific LendCo, the minting of a large amount of DAI that we KNOW will be sold into the market with certainty and then the USD be transferred to sit in an FDIC regulated US Bank, could present a novel way to mint DAI that is still in favor of the DAO that is “backed” with legal enforceability. Think of it as the DAO’s own version of USDC but without the whole token issuance part.

The above is purely a thought experiment and not for implementation anytime soon.

2 Likes

Coinbase is one of the most trusted companies in the crypto space and a major part of our mission is growing the cryptoeconomy. A highly reliable price feed anchored into Coinbase’s secure infrastructure can help make the DeFi ecosystem safer, reduce systemic risks and unlock the next wave of growth and adoption.

It seems that Robert Leshner believed them, as he said at the time;

“Coinbase price oracle will increase the security and decentralization of Compound’s price feed, which is mission-critical to the protocol and the ecosystem of applications built on top of Compound. We’re not alone — the rest of DeFi will benefit with faster development, consistent data, and shared standards.” —Rekt Ghost :ghost:

That is a very extreme scenario.
On a more realistic case, that of -95% in about 1year, MakerDAO has already shown that it can survive (see year 2018).

In fact, when we embark RWA real estate, it is unclear if MakerDAO would react equally well when it is necessary to margin call RWA holders with real-estate going down like 50% in 1 year (like 2008/2009) in certain regions).

Anyway, if ETH goes to 10usd (not to 0, so Ethereum is still working and alive) but bitcoin and stablecoins survive, we might still be able to survive using WBTC and stablecoins (using a basically uncapped PSM).

Agree. All of these price brackets and fixes just widen the attack surface in unpredictable ways. 1 DeFI project fixes price @ 1.05, another at 1.1 yet another at 1.001 etc. One can’t just magically price fix problems.

Again back to the same issue I started with here. Liquidity, liquidity, liquidity. I want to be fair here in that it looks like these prices were highly confined. Unless you had DAI on CBP you weren’t able to take advantage of the price. I don’t think in this case Liquidity would help as one has to have liquidity available at the market where the price is disjointed.

One thing this shows is that having more diversified price oracles is probably important. Taking the difference between the volume weighted vs. just the weighted to see if markets are skewed is probably an important characteristic of a price oracle. I also have NOT YET seen a price oracle that gives not just a price but say a 3 sigma price bound over the time period of interest. The point here is we would rather focus on what a nominal price is vs. reacting to sharp moves that are short lived. Utimately when one sees high price discrepencies between exchanges this is a time to be highlighted to a brewing potential problem (operational alarm) so that live operators can decide whether/how to take action. The idea that these markets can all be machine automated - while a great goal - is rather unlikely because open code is most easily susceptable to modelling and then gaming. Humans in conjuction with code/machines tend to do better than machines or humans alone.

There was discussion a while back whether the protocol could utilize either extra vault collateral to back DAI but I think that fizzled somewhat. I am back to having a second DSR that has terms such that it could be used to provide DAI liquidity in some emergencies for a piece of the emergency profits (or even a fixed DSR-R rate). My work envisioned a number of ways to create a DSR-R that would have DAI deposited which could be used in emergencies to provide DAI liquidity when needed and then would sop it up when it wasn’t to provide a boost in DAI DSR-R returns.

Maker is pretty robust if the price moves slowly. It is when we get a 50% drop in 24hrs that we have serious problems. Or a sufficient drop or change in price on ETH so that liquidations start kicking out. The problem here is strong correlation between markets (BTC, ETH and many other coins). It is something we just don’t get away from easily. Is RWA as strongly correlated. I don’t have data on this.

If ETH were to go to 10USD I am skeptical Maker would survive simply because so much of the outstanding DAI is dependent on deposited ETH. Maker would contract pretty dramatically. RWA though is unlikely to drop 96-98%. Worst case in realestate btw the drop(s) I have personally witnessed were on the order of 50% (give or take depending on specific market) and that took at least a year to unfold. From the perspective of pricing risk and variability I believe RWA is a much better asset class to loan against than crypto generally. It is also likely to hold much greater value than crypto generally. If I was to hazard a look at the future I can see one where RWA actually dominates the collateral backing of DAI and honestly I think we should prefer that to crypto, and stablecoins from a market risk perspective.

1 Like