[OHM] - OLYMPUS DAO Collateral Onboarding Application (MIP6)

Personally I feel that OHM trades at a very high premium to book value, but as mentioned above many assets have no “intrinsic backing” and can still make suitable collateral due to other factors like market liquidity and future growth potential.

Pros for accepting OHM as collateral:

  • There’s strong demonstrated borrow demand, as seen in Rari capital pools where users are frequently willing to pay double digit interest to borrow against their OHM
  • OHM is a big DAI user, and while this doesn’t outweigh risk considerations I think it makes sense to favor projects that are supportive of DAI in general
  • OHM has significant DAI liquidity that is unlikely to be removed in times of stress (because it is owned by the protocol rather than users), which helps reduce downside risk

Risks of onboarding OHM / blockers:

  • OHM currently only trades on a few DEX venues with no CEX coverage - this could make it more difficult to create a safe oracle price feed (generally Maker is not willing to use Chainlink feeds)
  • Significant admin priveledges are controlled by a multisig, with many of the signers remaining anonymous and without public reputation on the line. In a worst case scenario tokens could be minted or treasury assets removed, rendering the outstanding OHM tokens worthless
  • OHM trades at a very high multiple to book value (20x or more), which suggests greater downside risk if market enthusiasm for OHM declines. Maybe Maker could account for this risk by adjusting liquidation ratios or capping the OHM oracle price depending on price/book value ratio.

Personally I think it should be possible to onboard OHM as long as we have suitable risk parameters. For example one of the Rari pools gives sOHM a 33% LTV (equivalent to 300% liquidation ratio) - this gives a lot of room for accommodating downside risk before MakerDAO would lose money. I see the admin and oracle risks as the biggest issues standing in the way of integration for the time being.

If any OHM community members could clarify the multisig admin powers and any limits/safeguard in place (eg timelocks, mechanisms to cancel malicious actions, etc) this would be helpful!

24 Likes

In that case it makes sense to consider 1 DAI as the maximum amount you can mint using 1 OHM. Though it wouldn’t be possible to liquidate such loan, so perhaps something like 2-3 DAI per 1 OHM will work better.

Seeing a lot of new user accounts commenting here. Not hugely surprising considering the subject. Please keep the following in mind:

  • Keep the conversation constructive.
  • Focus on why onboarding OHM is good for Maker in a practical sense.
    • Why will people use this to mint DAI?
    • What risks are there, and how can they be mitigated?
  • Don’t assume that anyone responding speaks on behalf of MakerDAO, literally no one here does that. The proposal will follow the same process as all the others.
16 Likes

The relevant starting questions in my mind (and we likely don’t know the answers to all of them yet) are:

-What do we estimate for demand?

-Can we liquidate reliably in times of market stress?

-Can the issuer rug pull through extreme dilution or freezing/burning the asset?

I look forward to a good discussion.

4 Likes

Currently, the treasury backs every OHM by about$34 of what the protocol refers to as Risk Free Value meaning there are 79m$ worth of stable coins for a current circulating supply of 2.3m OHM. So, if the protocol were to go down the road of valuing OHM as something less than market value, I think this is the number to use.

1 Like

Hi folks, Ohmie and Maker user here (I use Maker to bond Dai borrowed from ETH to OHM, and keep several ALCX vaults with Dai). I see this as a great move that benefits both DAOs, reasoning below.

Pros for Maker

  • Helps Dai stability by allowing it to borrow backing from other decentralized algostables without direct exposure (OHM will continue to expand its bonding program to a variety of good collateral stables and reserve assets like LUSD, FRAX , wETH and eventually Fei, Rai etc).

  • Helps users like me who avoid keeping 100% realized profits in Dai due to its centralized backing risk from USDC.

I think this is a great opportunity for Maker to diversify backing of Dai away from centalized USDC, and think the objections raised so far around “ponzi schemes” and partial Dai backing make little coherent sense. OHM follows the well known central bank model to create a liquid reserve asset that other DeFi liabilities can be traded against, the people who have called it a ponzi scheme seem to not understand economics in general, let alone cryptoeconomics.

4 Likes

Its incredibly disingenuous to compare ohm as collateral to ETH saying ‘eth has no backing’, makes me wonder if you understand why ETH is valuable. In no way is OHM better collateral than ETH.

The protocol owned liquidity is a large portion of the DAI that Olympus ‘owns’ and I love the bonding mechanism I think that is a real innovation for defi but much of that DAI can easily be lost if everyone sells their ohm after realizing they paid 900$ for something backed by 30$ or whatever the RFV currently is. This ‘liquidity’ can dry up fast especially considering the amount of unrealized gains that stakers have right now and how much of the borrow demand from rari is to just buy and stake more OHM. Add to this the fact that a multisig of anons can mint ohm at will, I think this poses a large risk to maker for not much benefit. On the other hand this proposal would massively benefit olympus, by expanding further and making them a systemic risk to defi, they become too big to fail.

If OHM is onboarded as collateral to maker the LTV should be extremely low (as low as the RFV) and we should also onboard renDoge as collateral since were backing dai with memes now

10 Likes

Someone tell me more about this in a neutral, detailed manner. Or link to where I can read the relevant mechanics

4 Likes

This.

Adapt or Die. DeFi is about collaboration not centralisation.

If MakerDAO tries to fight DeFi it will lose.

I hold DAI, sOHM, FRAX, LUSD (a small bit of USDC :sweat_smile:)

I’m a stablecoin wench.

disclaimer: I invest in OHM

I think under the right parameters using OHM as collateral could be a good idea.

My only reservation is that not all the OHM contracts are multisig and the platform as a whole I think can use more security audits. Taggging Zeus because he would know @ohmzeus.

I think we should address the multisig issue before onboarding OHM as collateral since that might require contract changes.

Again don’t know much so tagging Zeus to explain, but fairly sure there something that happened that made one of the contracts non-multisig. Doesn’t affect treasury as far as I know, but my understanding is that if the single key were compromised it could be used in an exploit which is bad.

1 Like

Disclosure: I’m not affiliated with the OlympusDAO project, but I do hold a significant amount of staked OHM.

This is in regards to the OHM ERC-20 contract itself. I’ll just copy a few of my own messages from the OlympusDAO server that I posted when I noticed this during my own review of the contracts:

Original message

I really don’t want to be annoying but I think the issue I previously raised with ownership of the OHM ERC-20 contract (0x383518188C0C6d7730D91b2c03a03C837814a899) is more severe than I really conveyed. Since its owner (0x763A641383007870ae96067818f1649E5586f6DE) is a single key (not a multisig), if this single key is compromised, it can call setVault to an address it controls to be able call mint (which is behind an onlyVault guard) to mint as much OHM as it wants to any address as the recipient.

It would probably be a very good idea to update the owner of the OHM contract to one of the dev-controlled multisigs (like 0x245cc372C84B3645Bf0Ffe6538620B04a217988B which controls the treasury) very soon.

Alternately, the owner of this contract could be set to the null address which would mitigate this, but it would be come impossible to call setVault if the vault needed to be migrated to a new contract for some reason

Message #2

The risk is stated in my original message, but to summarise:

  • The treasury itself is a lot safer, it is controlled by a developer multisig, as detailed in the docs (and I confirmed that this is the case on chain too)

  • The sOHM contract is controlled by a single key, but the damage it can cause if compromised is limited, largely limited to messing up the index in rebase events logged which will make analytics harder, but as far as I can tell, the index value in this contract is purely informational, it’s not used by the actual rebase calculations

  • The OHM contract is controlled by a single key (different from the one that controls sOHM). If compromised, setOwner can be called to update the owner to another address, and to call setVault to change the treasury contract address to an address the attacker controls. The main point of concern here is the address set as vault in the OHM contract is able to call mint to create new OHM and assign them to any address of it’s choosing. Were this key compromised it could not steal treasury funds or token balances, but it could dilute the OHM supply outside the defined parameters

The follow on from this, is since an attacker could mint OHM for free, they could use this free OHM to crash the price and drain any LP pairs containing OHM.

This means the LP tokens in the treasury could be devalued, but the dai, frax & weth in treasury are still safe in this scenario

The response I received to this is that the key in question was irretrievably lost (faulty hardware wallet without backup) and that there would be a migration to a new token contract at some point in the future, as any methods beyond an onlyOwner guard on the current contract can no longer be called, if the key in question is truly lost.

Regarding the treasury itself

As per the documentation, the treasury is currently under the control of a four-of-seven developer multisig (a Gnosis Safe instance). As such, the project is not currently trustless - the developers retain a privileged position until such a point where the project migrates to on-chain governance.

I personally feel the risk of malicious actions to be extremely low, in large part due to all the efforts to integrate with other projects and to build relationships & partnerships in the wider defi ecosystem. These are not actions I would expect a project intending to rug pull or otherwise scam to undertake.

6 Likes

Just want to repeat this for emphasis :point_up:

A lot of negativity in this thread from newly registered users. Don’t mistake them for the MakerDAO crowd.

7 Likes

negativity because its 3% backed and wants to be collateral, and the backing is collectively cornering small caps, result is inevitable pump backed by its own liquidity, its false numbers, its a ‘buy my bags’ smart contract to pump external projects, all the dai here will go into MIM CRV vault and will be vaulted by yearn, then people leverage the minted dai into Mai to keep the peg when they leverage up their Mai vaults, lets recreate the conventional finance leverage pyramid, maybe then we can next time collapse in a real 2008 fashion

2 Likes

This is unfortunate because there’s no way to verify a key is permanently lost and no recoverable copies exist. Until contract is migrated there would always be black swan risk of the key being compromised leading to infinite minting.

12 Likes

Lets start with some basics.

  1. Admin key ‘lost’. Honestly this is a key sticking point. Get off that contract. Don’t care what anyone says 1 key here and the entire system compromised. This one point alone is sufficient to deny this MIP6 application.
  2. Oracles going to be a key issue. I don’t see how this is overcome in the short term.

I actually looked over the documentation linked here.

Some interesting quotes to take into account the realities of this.

  1. " OHM is backed, not pegged.

Each OHM is backed by 1 DAI, not pegged to it. Because the treasury backs every OHM with at least 1 DAI, the protocol would buy back and burn OHM when it trades below 1 DAI. This has the effect of pushing OHM price back up to 1 DAI. OHM could always trade above 1 DAI because there is no upper limit imposed by the protocol. Think pegged == 1, while backed >= 1.

You might say that the OHM floor price or intrinsic value is 1 DAI. We believe that the actual price will always be 1 DAI + premium, but in the end that is up to the market to decide."

Hence by definition the system is ‘capturing’ value when OHM is > 1 anyone who buys OHM above 1DAI in price is basically giving their money away to the system.

  1. OHM to pay out its high rewards is a rebasing token. So for each 1% you think you earn by the rebase the price in the LPs is dropping. So far money is moving in filling the coffers, LPs and causing postiive rebases. Just look at the reasons for buying and the whole return statements in the document. 2%/day APY and up to 100000% APYs - seriously folks this is the new monetary policy that you want the world to be based on?

  2. Lets forget about every other asset in this system and just look at using DAI to back OHM 1:1 - if you thought about this system from just being one kind of stablecoin OMH being backed by another stablecoin DAI - you’d just say this is a rehypothication of DAI to OHM.

Ask yourself what is the product being offered here.

The document has the answer:

" Is OHM a stable coin?

No, OHM is not a stable coin. Rather, OHM aspires to become an algorithmic reserve currency backed by other decentralized assets. Similar to the idea of the gold standard, OHM provides free floating value its users can always fall back on, simply because of the fractional treasury reserves OHM draws its intrinsic value from."

‘fractional treasury reserves’ and no it isn’t a stablecoin because unlike DAI and Maker that did not seek to profit from DAI PEG running high (which we do to some extent with PSM tin/tout but not to the extend OHM is doing) we try to keep the PEG at 1+/-tin/tout this protocol seeks to profit from this price disparity. BTW: This whole 1OHM = 1DAI idea flies flat in the face of another statement made in that document.

" Dollar-pegged stablecoins have become an essential part of crypto due to their lack of volatility as compared to tokens such as Bitcoin and Ether. Users are comfortable with transacting using stablecoins knowing that they hold the same amount of purchasing power today vs. tomorrow. But this is a fallacy. The dollar is controlled by the US government and the Federal Reserve. This means a depreciation of dollar also means a depreciation of these stablecoins.

OlympusDAO aims to solve this by creating a free-floating reserve currency, OHM, that is backed by a basket of assets. By focusing on supply growth rather than price appreciation, OlympusDAO hopes that OHM can function as a currency that is able to hold its purchasing power regardless of market volatility"

Yet as quoted OHM is backed by 1 DAI + ‘something’ which related to a staking, bonding model of behavior that is supposed to manage value accrual to the system beyond 1DAI per OHM.

The big argument for OHM is that Bitcoin and ETH basically are backed by ‘nothing’ so hell why not OHM.

Maker stands in stark contrast to this and other models by being fully liquid in the collateral, and DAI backing value is overcollateralized. Maker offers clients the ability to borrow a token that yes trades with $1US value (I know there probably were a lot of debates why this value was chosen and not some other defining value basket and defer this argument for another thread) but is fully backed by collateral value exceeding the DAI notional value in $$. This system is not fully backed especially at these 20x book valuations.

What I see this OHM system doing is presenting the user with massive returns, to get them to buy OHM at prices far over what the backing value is at the time, so the system can bank some of this value difference which is only going to buy back the tokens when the OHM value drops below 1DAI. This is straight up a hype value sink designed to trap people high and exit them low.

Not one single person in their right mind should pay more than 1DAI/OHM and if I were to add if Maker decides to onboard OHM it should basically treat it as a stablecoin. 1.2OHM can maybe borrow 1DAI because that is what the actual OHM system says the intrinsic value is, in fact that system is designed to move that value eventually to that 1OHM = 1DAI + ‘this not so defined’ something. Right now you could basically look to buy OHM at or below book value and maybe try to play a speculative game but in the end the way this system is constructed is that if you are paying more than 1 DAI per OHM you will eventually find yourself diluted so that eventually 1OHM is basically worth 1DAI. Now will the rebasing returns you got actually offset the OHM reduction in value. Depends on when you bought and whether the system is still getting inflows or outflows.

Ultimately the individuals earning the maximum value are the early stakers while as time goes on there will be less and less pie to share and the OHM price will start dropping. Look at the parabolic rise in those charts at some point all liquidity would run out and the fuel to sustain the growth and hence the returns is going to stop and this will lead to a supernova type collapse that the system is designed to handle,

BUT ONLY when the OHM price is below 1DAI/OHM.

So I am not going to say ponzi. I am going to say unsustainable by design. I am also going to say there are significant issues not to onboard OHM at a minimum that would need to be cleared up.

Also like @monet-supply suggests we can certainly onboard this and put a very high LR and even a higher LF to buffer the system here against losses. In the end I have another concern and this is regulatory.

I am not saying OHM is some illegal ponzi, but what if MakerDAO didn’t see through an addition that basically rug pulled using DAI and trapped a whole bunch of people in something. I can see such an event would be like a light to the regulator mosquito - urging them to bite us.

I have often had this concern about what happens if a blacklisted address were to drop wBTC or ETH into the protocol and mint DAI. As far as I know Maker can’t do one thing about this and in this sense Maker becomes complicit in criminal activities and brings the protocol under huge scrutiny. It is why when we weigh collateral additions via MIP6 the community and DAO has to have significant scrutiny.

One thing I have wanted to change with MIP6 proposals is that at least one CU should have to co-sponsor them. This way we have some accountability and control within the protocol CU structure over what proposals are even submitted and these create a lot of churn and discussion eating up valuable community productivity. I want to see these but I also want to know at least one CU is in favor of them because they are good additions already basically vetted by at least one person in the protocol before put up for consideration by the community.

21 Likes

Something people should consider is the ethics behind them controlling their own liquidity, with over 92% staked. Since the value accrual is happening, at the new user’s expense, is it even ethical for the premium in the first place? People making a bet this is going to 20X to match book, is a scary proposition. Ultimately, it’s just a liquidity drain until the tokenomics from a growth plateau kick in, and what happens to the rest of the market, when it’s a 40 billion dollar monster and the liquidity begins to flow again?

1 Like

Have to be frank here… All of this astro turfed back and forth re OHM by new accounts seems a thinly veiled attempt to build some narrative about “Maker being against DeFi” (which is ridiculous). FWIW, I like OHM, am a fan of Zeus and am in the project’s Discord but I don’t get where all of this whipped up faux aggression comes from, unless, as I surmise, it’s a silly internet game played by people trying to “score points” on crypto twitter.

/rant done

18 Likes

Agreed, there is no way to prove they key is lost, and it remains a risk that cannot be mitigated until the relevant contracts are replaced.

I’m still comfortable enough with the funds I have in OlympusDAO, but I can understand how this risk may impede potential partnerships.

7 Likes

Yeah same thoughts here. I would be more comfortable with OHM if they would fix known exploits even if the risk is low. I think it would make more sense to update the contracts to be safer before listing on Maker and other protocols.

Again, fully support the Olympus project, but I need to be 100% honest in saying that all known vulnerabilities should be fixed immediately. We’re talking about hundreds of millions of dollars invested in OHM so it’s not okay to leave this hanging for a long time. It needs to be patched ASAP.

10 Likes