MKR security against BT events - an idea

TLDR: reward people (a form of staking) for placing buy orders on MKR, locked for a fixed time period, to prevent flash crashes.


during the BT crash, the value of MKR dropped from 600usd to 200usd in a few hours. This corresponded to a 400m loss in MKR marketcap, and was caused by “only” 12m usd lost (caused by the 0-DAI liquidations).

I just had the following idea. What if:

  1. We ask people to place bids on MKR at p discount (e.g., p=-20\%) wrt market price. These bids are locked for a fixed time frame T (e.g., T = 1 week).
  2. We reward these bidders at the end of time T, like with a fixed percentage q (E.g., q =0.1\%… basically q should be a funciton of p) of the value of their bid.


  1. Today 1MKR = 2000 DAI.
  2. A bidder may place a bid for 100MKR for 100*(80% of 2000) = 160k DAI.
  3. The bid is locked and stays active for 1 week.
  4. No matter what is the outcome, at the end of the week the bidder gets a reward of 160k DAI* 0.1% = 160 DAI.

There are two cases:
5a) If the prices of MKR does not drop in the market below 80% of 2000 DAI = 1600 DAI, then the bid is never used. And at the end of the week, the bidder gets back the 160k DAI.
5b) There is a crash in the value of MKR. In this case the bid is used to buy 100MKR for 160k DAI.

TLDR: reward people (a form of staking) for placing buy orders on MKR, locked for a fixed time period, to prevent flash crashes.

OasisDex might be just the perfect platform to develop this idea. @bartek ?
Currently (due to the high fees) OasisDex is not practical. But if we talk about 1week (minimum) long bids, rewarded by the protocol, it makes again a lot of sense.


That’s selling a put option. Usually, a market-maker manages the position to convert it in MKR sell-buy orders.

The main issue is that will cost a lot.

To counter such market reactions you need strong MKR holding. For that, you need strategic investors.

It will not help with the price but at least it avoids governance attacks during those times.


yes that is an option.
Options for crypto are still highly problematic due to illiquid markets, even for BTC and ETH. Options for MKR are in other words highly unlikely for now.

1 Like

Yes, I agree that’s the right way to phrase it. Thanks.
Just that we can do it in-house with Oasis I think.

Yeah it must be calculated. But again during BT last year we suffered a -400m drop in Market cap…

EDIT: one remark: the ‘option’ is actually usable by everybody, not just us. It’s an open bid.

In general I agree.

But for the specific task (MKR), we have OasisDex already in place and it seems just perfect for this. We just need to insert a module that deliver rewards and locks the bid for a timeframe T.

EDIT: I think a lot of the complexity in projects trying to do Defi Options is to tokenise such options (so to trade them). We don’t need this here. The specific need is much simpler than the general “DeFi Option” holy grail.

I think this could be implemented w/out any changes to current OasisDEX. Similar scheme can be used as a backstop liquidity for Liquidations 2.0. The reward structure can be computed off-chain entirely with MKR/DAI bonus for people providing such backstop liquidity. This is no dissimilar to many platforms providing airdrops to liquidity providers - any condition can be off-chain coded to make paricipants eligible for such airdrops. So if the bonus is higher than what you normally get on your stables on yield-aggregating platforms, this could be actually an interesting way to deploy your capital if you are a long-term long on MKR and want to buy the dip. The main problem with such scheme that I see is that dips of 20% happen quite regularly even in a bull market and have nothing to do with MKR.

1 Like

Can’t we open a private MKR vault, that will buy at credit MKR at a certain price using MKR to back it.

This is great news imho!
Writing SC code is always the bottleneck.

Building a buy wall on OasisDex might help reducing the downward fluctuations on the MKR price.

In any case we have some kind of “measurable assurance” that, if a 20% drop happens, it will have to eat at least N-millions of buying orders.

This can be a useful information for the risk teams (@Primoz ?)

Yes, this is what I’d love to do with my DAIs, instead of leaving on Aave, for instance.

I suspect this will not get approved. People have rejected the idea of MKR-vaults (for good reasons). Furthermore, buy MKR with DAI minted with MKR as collateral feels a bit like printing DAI out of thin air?

They rejected MKR as vault not a private vault.
This is more or less the same as selling MKR.
Instead you take a loan the same as Yearn did.

Tho if MKR vault doesn’t sound good we can open an aave position instead.
I don’t think it is thin air.

1 Like

What is the purpose of preventing flash crashes? Is it to ensure that we don’t mint MKR at the “wrong” time as we did on BT or something else like avoiding governance attacks?

yes, basically to attenuate events like:

Just to point out, MKR lost a huge amount of market cap compared to the amount lost because the protocol didn’t quite work as expected. Additionally, the markets as a whole were falling anyway so it’s not quite fair to compare the 12M USD lost to the 400M USD market cap lost.

Second, if it’s minting MKR rather than governance attacks that you’re worried about, it’s worth understanding why this more complicated mechanism is better than the “hit it with a hammer” solution of just having a larger surplus buffer.

Simpler solution here is to create a DSR deposit facility that has a fixed rate and a term (basically a bond backed both by MKR and revenue generation). In effect when the system has losses don’t sell MKR raise DAI using a new DSR.

I calculated that Maker could have offered a 6.2M 10%/yr bond type return DSR-L/B (liquidity/backstop whatever you want to call it) and not sold 1MKR to cover the DAI shortfall. This would have been paid back in ~6-8months out of the surplus return costing the protocol and MKR users 300-350K DAI of additional interest, no MKR dilution sales.

All that would be needed is a DSR facility to be set up that has a ‘term’ (say 6 months or 1 yr). It could even be as short as 1-3months where people could opt to roll out and into the next if people don’t want their cash locked up.