Pre-IPO Mirrored Assets

One of the advantages of having a synthetics ecosystem is the ability to come up with new frameworks for asset classes and investment products that do not currently exist or have high barriers of entry in traditional markets. One of the main candidates is pre-IPO asset trading. Recently, FTX has listed various pre-IPO assets on their exchanges to be traded, and it has seen some success but with the downside of an external settlement party (cf. here). However, keeping true to the core of ‘decentralized’ finance, let us give a rough description of how the Mirror Protocol could offer pre-IPO mAsset trading without the need of a centralized counterparty.

A major problem encountered with dealing with pre-IPO assets on a synthetic assets platform is the lack of a trustworthy oracle price feed. We attempt to remedy this problem through a typical auctions-type process.

The following is a step-by-step process for the minting, trading, and price adjustment of mirror Pre-IPO assets through out the whole IPO process.

User Flow

  1. When news about a specific company seeking to IPO comes out, any user can upload a Whitelist & Register PreIPO asset proposal. This will function exactly like a combination of the whitelist and register parameter proposals that already exist (with additional parameters such as collateral ratio for pre-IPO vs post-IPO, etc.)
  2. Given that the proposal passes, users are able to submit a both an offer_price used as the reference for minting and the collateral necessary to mint the assets at the given pre-IPO collateral ratio. The amount of time allowed for submission is set by mint_period.
  3. When a price range is published, all price submissions that fall within +/- tolerance of the range will automatically go through with the mint at the posted price (let us say average price if given by a range). Upcoming IPO price ranges can be found form a variety of sources such as here and here.
  4. Trading of the pre-IPO asset is done through the existing AMM DEX just like the current mAssets.
  5. When the underlying asset actually IPOs, the IPO’d asset becomes automatically whitelisted on Mirror. The pre-IPO mAsset is deprecated and can be exchanged 1:1 for post-IPO mAsset, and all mint positions are recalculated at the post-IPO collateral parameter.

Example

Let us assume that the governance for listing has passed for pre-IPO mTerraform with pre-IPO collateral ratio of 10x and a tolerance of 10%.

  • During the mint period, I submit a price of 10 UST and collateral of 100,000 UST. This effectively means, that given that my price is correct, I will mint (100,000)/(10*10) = 1000 shares of the pre-IPO mAsset.
  • Let’s say the estimated IPO price is published to be $11 (= 11 UST). Then all submissions with price in the range of [9.9, 12.1] are accepted and automatically minted at a reference price of 11 UST. In the case above, 10 falls in the range above, so 100,000 UST will be used to mint the pre-IPO mAsset at a collateral ratio of 10x and reference price of 11 UST. Then the real amount of shares minted is (100,000)/(11*10) = 909.090909 shares.
  • Then the typical process of trading on the AMM DEX (or perhaps OTC markets) happens until the underlying asset officially IPOs.
  • When the underlying officially IPOs, all pre-IPO mAssets become post-IPO mAssets which trade exactly like the already existing mAssets. Assuming current parameters, the post-IPO new mint CDP ratio would drop from 10x to 1.5x, so minters can remove up to 85% of their UST collateral.* If the price post-IPO is still 11 UST, then the total amount of collateral required would only be (11*909.090909)*1.5 = 14999.999998 UST, so I could remove up to 85000.000002 UST from the mint contract.

*Percentage varies depending on submitted price vs actual mint price.


During this process, there are a few things more to be considered by the community. The following list is not exhaustive.

Unresolved Questions

  1. Do we allow the cancellation of submitted prices/collateral during the mint period?
  2. Should minting always be open until the actual IPO happens? If not, the supply is fixed after the estimated prices are published.
  3. As a follow-up to previous question, the initial public offering price can be revised several times, so how would this be reflected on-chain? Given that pre-IPO mAssets would have already been calculated and minted, do we simply ignore the new information?
  4. Do we allow for the staking of pre-IPO asset LP tokens? Empirically, the price of an asset right after IPO tends to rise in price, while minting (and then selling the pre-IPO asset) is a short on the asset. To incentivize a larger supply of the pre-IPO asset, perhaps we can consider increased MIR rewards for staking to indirectly incentivize supply (read: minting).
  5. Given that outside prices are used once during the whole process (minting calculation stage), how do we reckon the prices used to determine the CDP ratio in the automatic liquidation process during the pre-IPO trading period? If we use the DEX prices, there probably will need to be some sort of grace period where CDPs cannot be liquidated so that price discovery can be adequately determined on the DEX.
  6. High collateral ratio

A simple comparison of the original IPO process vs. Mirror IPO process is attached below.


Please note that this is just a rough description of my personal thought process, so I would appreciate any community feedback and discussion.

6 Likes

Thanks @Sihyeok for engaging with us on this topic, this is very exciting, and will drive a lot of adoption if we happen to make it right.

I have a few clarifications I would like to ask you:

How do you see this process working here ?

Ideally I would like to be able to submit multiple proposals under a single wallet address as long as I have enough collateral to commit for each proposal. This would give me better odds to mint mAssets.

If the IPO’d asset becomes automatically whitelisted - at which point do we decide on the oracle feeder, and the min CDP ratio needed post IPO ?

Unresolved questions:

1. Do we allow the cancellation of submitted prices/collateral during the mint period?
I would prefer if we did, but the risk of gaming the auction makes it impossible, so no.

2. Should minting always be open until the actual IPO happens?
It can’t stay open, because if you allow minting after the auction has concluded, then you’re preventing price discovery from happening and the pre-IPO asset will always trend towards the auction price.

3. The initial public offering price can be revised several times, so how would this be reflected on-chain?
I think we should ignore these inputs, we already know that most of the time, the post-IPO stock will trade at a different value than the auction price - if anything , this could help minters to make a decision as to whether they should hold on to their assets or sell them in anticipation of an arbitrage.

4. Do we allow for the staking of pre-IPO asset LP tokens?
Yes, I think this is a must.
The whole pre-IPO mechanism is not attractive for minters as they are de facto shorting the asset as you said so we will have to find a way to incentivize them to create inventory.
Boosted staking rewards are necessary, along with a grace period for CDP fees (ideally in your example, the 85,000 UST removed from the CDP should not be taxed at 1.5%)

5. and 6. Liquidation process and high collateral ratio
I do not really have an opinion on this , we should probably run some simulations on the past few years IPOs and see at which multiple of the IPO price these stocks have been trading to get an idea of the expected volatility and required collateral ratio.
But the model you’re proposing is going to be unique in the sense that there won’t be much selling pressure happening in the pools before we have an oracle feed and minting can resume, so the min C-Ratio could be hit very quickly for anyone with a CDP position.

Thanks for your thoughtful response. I’ll try to add my additional thoughts below.

I think if it is technically capable to allow multiple to allow multiple proposals under a single wallet address, then I don’t see any reason to not allow multiple submissions as long as the corresponding collateral is locked up for each proposal. If this is not technically feasible, one can simply just split his/her funds and vote using multiple wallet addresses.

This is a good point. The minimum CDP ratio post-IPO would be voted together in the beginning with the pre-IPO parameters. The remaining problem is that it would be unlikely that one could immediately the asset post IPO as adding oracle support for asset would require time (probably few days minimum if the Band/Chainlink put it as their top priority).

Other potential structural changes could include:

  • Completely freeze any post-IPO activity until oracle support is provided
  • Disable only functions that require oracle support until available (minting)

There are probably more elegant solutions to this, but the above are the first two that came to mind.


What would the optimal strategy be here if you could cancel your vote during the mint period?

Makes sense. Perhaps if there was a parameter to adjust mint price as a function of time and on-chain trading prices, we could allow unlimited minting. However, I think the reward:risk of detailing this process is probably too low.

That sounds like an interesting idea. What would you think the minimum reward threshold would need to be in order for minters to take on this risk from a back-of-the-envelope calculation? I’m amenable to the idea of offering a grace period for newly IPOd mAssets wrt to CDP closure fees.

From my experience as well as light searching here and here, it seems like an initial collateral of say 10x will be sufficient to cover forced liquidation cases. However, there is always the tail risk of defaulting if price change multiple > collateral ratio. Perhaps we could take some sort of linear smoothing of the MVWAP AMM DEX price and initial minting price where more weight is given to the former as time goes on.

We need this ASAP that will blow up mirror exposure

I did not spend enough time thinking about this but will try to reply by tomorrow.
In the meantime, I wanted to post this Coinbase to offer secondary market for private shares ahead of public stock listing

1 Like

Excellent product to be added into Mirror Protocol. Will vote for it in governance :slight_smile:

Hi there,

I think I have a simpler alternative solution to what you guys are trying to achieve here.
We’ve set up a fund (https://utc.fund/), that issues derivatives, backed by shares of Pre-IPO companies with 1:1 ratio, pretty much like what Stablecoins do with fiat currencies.

We use Ubeswap (Celo blockchain based fork of Uniswap) for secondary market trading on the backend, so you now have a live price feed for private equity that you can pull directly from smart contracts on Celo’s public blockchain. If anyone is interested in setting up an oracle - I’m happy to provide assistance & clarifications.

Our tokens are redeemed upon IPO of the corresponding companies, so this is not a competitor to Mirror.

Examples:
SpaceX: Ubeswap Info
Robinhood (we still have some locked up shares): Ubeswap Info
Airbnb (before it went public): Uniswap Info

Cons & limitations:

  1. We can only tokenized companies, whose shares our fund is able to buy
  2. The total market cap of our derivative is limited by the amount of the actual shares we are able to acquire on the private market. This indirectly affects the frequency of price movement.
  3. Most of this is very new, takes a lot of education, and trade volumes are still very low. Price moves once per day or so.

We’ve been slowly scaling this up for the last 3 years and expect Ubeswap-based secondary market to become more dynamic in the coming months.

I’m the founder of this project and CEO of Ambisafe(https://ambisafe.com/) and happy to answer any questions.

This is completely off the table – we just watched multiple platforms shutdown exactly what you described following pressure from regulators. Real share backing is a total non-starter.

It’ll always be a weak point of the system because the vehicle that owns the shares will be subject to real world controls.

This is completely off the table – we just watched multiple platforms shutdown exactly what you described following pressure from regulators.

Some specific examples would be appreciated.

We’re doing everything legally to the best extent possible. We’ve got a proper fund’s license in the country of incorporation (BVI) and SEC have also already questioned me about the project. So far it’s set up in a way that it’s out of their jurisdiction.

Our solution has risks, but I think this is about the only way you can do it for private equity for now (talking about the stage before the IPO has been announced).

Follow-on thought / more decentralized proposal: since there is no official price for the private market securities (unlike public), and multiple brokers can offer the same product at different prices, Mirror’s oracle can actually use an average of all known offerings within certain time range.

Synthetix, Binance. Uniswap “voluntarily” removed.

Being “out of their jurisdiction” doesn’t suddenly make it legal. You’d be selling securities to US citizens, thus in their jurisdiction.

Private equities on Mirror will be a little weird for shorting (what’s the APY?). Interested to see how SpaceX goes.

I think your point on the multiple brokers makes sense, issue will be getting the oracle data

Even if you solved the oracle issue, what is the incentive for someone to mint an illiquid, volatile, difficult to access asset (other getting paid unstainable levels of MIR rewards)?

One of the main incentives for someone to mint an mAsset is that they can buy the real asset and be market neutral but get paid in MIR rewards for providing the liquidity. If you can’t easily access the real asset, and the real asset is the “next hot IPO which is going to pop” who is going to be interested in minting that?

Well if the planned price is $20 and the mAsset is $150 then that’s a huge potential return if the IPO doesn’t pop ($130 of profit if it trades at $20 launch price)

Of course if you can mint at a high price and then the price goes way down, you will make money. That’s just being short. But why would the oracle price be that high if the IPO price was that low?

If the private market was valuing it at $150, the IPO price won’t be $20. IPO’s generally go up because the availability of the asset has increased and the pent up buying demand can access the shares.

Just think about it from the minter’s perspective. The least attractive stocks to mint are illiquid, volatile, and difficult/impossible to access the real asset. Every mAsset needs minters since they are the liquidity providers. For any stock/asset there will be some people who think it is over valued and will be interested in taking on the short exposure, but you won’t get a functioning LP on that source of liquidity alone. Long interest in virtually every stock exceeds short interest by a wide margin, and that is particularly true for the desirable pre-IPO companies being discussed. You need “professional minters” who are willing to provide liquidity by opening the CDPs that create the mAssets, but if they can not access the underlying asset to get delta neutral, they are not going to do that and will just mint mAssets where they can access the underlying asset.

I get it won’t actually open at $20. The point is people might be overbuying it on Mirror and thus others decide to short it.

Overall I agree that there’s a mismatch with new IPO assets, especially if they are priced low.

I think the real world arbitrage is significantly overstated. You can’t close the loop on that trade so you’re hoping the Mirror price changes to make it work

Did we consider adding Stripe Inc. ?