This paper introduces a new NFT fractionalization primitive: RICKS (Recurrently Issued Collectively Kept Shards).
When you fractionalize an NFT into RICKS, the protocol mints new shards at a constant rate — say, 1% per day or 5% per month — and sells them. The proceeds are distributed to existing RICKS holders as staking rewards.
This design solves the reconstitution problem, ensuring that RICKS can always be converted back to their underlying NFT, while avoiding the liquidity and coordination problems of all-or-nothing buyout auctions.
Fractionalizing NFTs is hard because owning them can be an all-or-nothing proposition.
If you want to sell 25% of a plate of eight cookies, you can sell two of the cookies. If you want to sell 25% of a business, you can sell the rights to 25% of its future cash flows. In either case, the 75% you have left is still useful to you.
On the other hand, owning 75% of an in-game asset may not entitle you to use even part of that asset in a given game. If you sell 25% of such an asset to a buyer, and they refuse to sell it back, or even lose their private keys, you’re in trouble. With no way to reconstitute the NFT, even if you nominally own 99.99% of it, that ownership could be worth nothing.
As a result, fractionalization protocols must provide some way to reconstitute fractions back into the original NFT, a design constraint we refer to as the reconstitution problem.
The most popular solution so far, pioneered by fractional.art, is the buyout auction.
Buyout auctions exist to ensure that Alice’s and Bob’s fractions retain their fair market value by solving the reconstitution problem.
To see why, imagine that Alice had fractionalized her NFT, representing an in-game asset, into 100 shards using a protocol that did not offer buyout auctions. Instead, only someone who owned all 100 shards could reconstitute the NFT.
If someone accidentally burned one of the 100 shards, it would then become impossible for anybody to reconstitute the NFT, and the remaining shards would lose all of their value. Because of this risk, even at the time of fractionalization each of the 100 shards is worth well below 1/100 of the value of the NFT.
With buyout auctions, losing one of the shards no longer destroys the value of the others. For example, if Alice lost one of her shards immediately after minting, she could initiate a buyout auction and submit the winning bid to get back the NFT, with 99% of the proceeds going to her.
In this sense, buyout auctions are more of a necessary evil than a feature. They are certainly not for the benefit of potentially interested buyers like Clara, who are not stakeholders in the fractionalized NFT to begin with and therefore do not merit special consideration by the protocol.
Buyout auctions are also not meant to encourage the reconstitution of the NFT. In order to solve the reconstitution problem, it must be somehow possible to reconstitute a fractionalized NFT, but Alice and Bob went to the trouble of fractionalizing their NFT in the first place and may be content to keep it fractionalized forever.
Unfortunately, buyouts can run into issues due to capital constraints: if an NFT is valuable enough, it’s possible that nobody can round up enough money to pay a fair price for it once an auction is initiated.
To avoid situations like this, fractional.art includes a reserve price in its buyout auction mechanism, which specifies the minimum price at which a buyout can be initiated. In the above example, if the reserve price were set to 100,000 ETH, Clara wouldn’t have been able to initiate an auction at 10,000 ETH.
The difficulty comes when trying to set a reserve price. In the above example, Alice didn’t want to sell the NFT for much less than its fair value of 100,000 ETH, but Bob didn’t mind. Coming to agreement here can be quite difficult and contentious, especially as the parties involved may change as shards change hands.
In practice, setting reserve prices requires active participation by shard owners. As a result, they are not updated frequently due to the attention demand on participants. Nobody has yet found a reserve price mechanism that solves the problem of undesired buyouts.
The Party of the Living Dead was a group of NFT enthusiasts who joined up to bid on a rare Zombie CryptoPunk, which they collectively won for a price of 1,200 ETH. They then fractionalized it on fractional.art and distributed its fractions to contributors in proportion to their contributions.
At the time of initial fractionalization, five “whales” (owning 5% or more) collectively owned 56% of the NFT, while the remaining fractions were spread among 451 other participants.
Fractions of the Zombie traded on Uniswap, where an anonymous collector realized they were underpriced relative to the value of other Zombie Punks. This collector bought enough fractions to increase their individual reserve price voting power, lowered the reserve price, and initiated a buyout auction.
The buyout auction started at a price of 1,100 ETH, lower than the initial sale price to the Party, and ended up closing at 1,900 ETH.
Many of the non-whale owners were unhappy with the buyout, which they believed was at too low a price.
Unfortunately, they found themselves essentially powerless to do anything about it. Individually, none of them could source enough liquidity to beat the bid and buy the NFT outright. Even if they had wanted to band together to buy the NFT as a single bidder, the coordination overhead and limited time available made that path unfeasible.
Recurrently Issued Collectively Kept Shards, or RICKS, solve the reconstitution problem while sidestepping the liquidity and coordination problems of a full buyout.
Instead of one all-or-nothing buyout auction, the protocol issues new RICKS for a given NFT at a constant rate — for example, 1% per day, or 5% per month. These new RICKS are sold for ETH in an auction. The proceeds go to existing RICKS holders as staking rewards.
As we will explain below, liquidity-constrained buyers who wish to increase their ownership can always trigger auctions for less than a full day’s quantity of RICKS.
This means that ownership of the NFT is always incrementally flowing to whoever is willing to pay the most for it, while existing owners benefit.
RICKS allow a motivated buyer to obtain the extreme majority of ownership in an NFT over time. We finish solving the reconstitution problem by adding a mechanism for extreme majority owners to complete their ownership and reconstitute the NFT.
One of two things will happen:
Either way, if this activity persists over time, it will draw attention and buyer liquidity, improving the likelihood trades occur at a fair price for all parties involved.
Let’s say Clara is dedicated to owning the NFT and repeatedly bids for RICKS at a valuation of 100,000 ETH, a price nobody else is willing to match. Eventually, she owns 99% of the RICKS, perhaps after 458 days of winning the auction, since . She would now like to claim the NFT. To allow this, an additional mechanism will be needed in the protocol.
One route would be to acknowledge that RICKS have some inner Morty and use a lottery. For example, if a majority owner controls 99% of the shards for an NFT, they could trigger a coin flip. If the coin comes up heads, they get the whole NFT (so they gain an additional 1%). If it comes up tails, the other owners have their positions doubled (so the majority owner loses 1%). This procedure is perfectly fair in the sense of expected value.
To avoid weirdness around the 99% boundary, we could allow a majority owner to trigger a coin flip at 98%, or 90%, or even 75%, with the caveat that they would receive worse odds the farther away they were from 99% ownership.
If the fractionalized NFT becomes expensive enough, even bidding for 1% of it might be prohibitively expensive for most. Furthermore, it is possible that, on some days, there will be no interest for an auction, so that holding an auction would be a waste of gas.
As a result, instead of holding an auction every day, the RICKS protocol can implement an on-demand auction system: if it has been days since the last auction, and the issuance rate is per day, then the protocol would auction off shards. We take the minimum of the elapsed time and 1 day to avoid issuing too many RICKS all at once.
For example, if the issuance rate were 1% per day, so that , and it had been half a day since the last auction at the time a new one is triggered, then the protocol would issue and sell of its supply as new RICKS.
Just like current fractionalized NFTs, we expect RICKS to trade on AMMs like Uniswap. This provides a convenient arbitrage mechanism to ensure that RICKS auctions do not complete for too low of a price: if an auction’s closing price is significantly below the RICKS price on Uniswap, an arbitrageur can profit by buying the RICKS in the auction and then immediately selling them on Uniswap.
We could consider taking this logic a step farther and specifying that RICKS auction bids must be at a minimum of, say, 5% or 10% above the Uniswap TWAP price for those RICKS.
Because a sufficiently motivated buyer would still be able to accumulate ownership of the NFT over a long-enough time frame, the reconstitution problem would still be solved. And, because the auctions would trade at a premium to Uniswap, they would likely cause minimal sell pressure.
On the other hand, this modification would reduce the consistency of staking rewards for RICKS holders. It also makes reconstitution more difficult, and it is hard to tell what the market effects of that will be a priori.
RICKS offer a natural mechanism for launching new fractional shards of an NFT. Rather than having to provide shards on Uniswap and choosing a price, NFT owners can simply fractionalize using RICKS, start as 100% owners themselves, and let the automatic auctions take care of the rest.
RICKS holders will need to stake their RICKS in order to claim auction proceeds.
However, this poses challenges for composability. In particular, it is computationally infeasible to determine how many RICKS are held by each concentrated liquidity position on Uniswap V3 at all times, meaning auction proceeds cannot be given directly to Uniswap V3 liquidity providers.
Instead, the RICKS protocol will keep track of the aggregate RICKS owned by all Uniswap V3 LPs, and will direct the auction proceeds due to all of them to liquidity mining rewards for the Uniswap V3 pool, as described in this article. In this way, market participants are incentivized to provide liquidity to the pool as efficiently as possible.
There are other potential solutions to this problem, including (1) creating a wrapped RICKS token that holds both RICKS and ETH auction proceeds and (2) redirecting auction proceeds to buybacks of RICKS, but both have significant drawbacks.
We hope RICKS will make NFT fractionalization even more fun and useful.
They also open up an entirely new design space. For example, we could let RICKS stakers automatically use their rewards to bid in future auctions. RICKS could also be pooled together into on-chain “councils” consisting of RICKS with similar attributes, like Zombie Punks or Wizard Hat and Scarf Ocelots.
9.14.2021 | Dave White
Summary This paper introduces a new NFT primitive: Martingale shares, or "Mortys." Mortys are synthetics representing fractional ownership of classes of NFTs. They do not require buyouts or oracles, but [→]
Disclaimer: This post is for general information purposes only. It does not constitute investment advice or a recommendation or solicitation to buy or sell any investment and should not be used in the evaluation of the merits of making any investment decision. It should not be relied upon for accounting, legal or tax advice or investment recommendations. This post reflects the current opinions of the authors and is not made on behalf of Paradigm or its affiliates and does not necessarily reflect the opinions of Paradigm, its affiliates or individuals associated with Paradigm. The opinions reflected herein are subject to change without being updated.