Get Credit for Rare NFTs
Parallel will enable owners of rare or "grail" NFT's to access greater credit when used as collateral.
Parallel's money market allows NFT holders to access instant liquidity against the value of their NFT's via its Peer 2 Pool lending model. We strongly believe P2Pool lending presents significant advantages, but it is likewise clear that the Pool-based model requires all NFT's within a collection to carry the same value.
Our Trait Boost research shows we can nonetheless offer collateral value premiums for specific NFT's and thus offer users greater flexibility as they access credit against their NFT's.
Peer 2 Pool Lending and Rare NFT's
The nature of Peer 2 Pool lending against NFT’s means we need to give the same valuation for a given NFT within a collection—the “Floor” NFT price. Plainly however there are NFT’s within each collection with traits that make them rarer/more desirable and thus empirically more valuable.
Users ask whether such traits should account for greater valuations than the simple floor price calculation, and intuitively users are less likely to use the protocol with their “grail” NFT’s as a result.
Notably we have a flexible liquidation model to account for higher-valued NFT's, but it is likewise clear many users would like to access greater value for their rare NFT's than a basic P2Pool model can provide.
Trait-Boosting as the Solution for Users who Own Rare or "Grail" NFT's
Inspired by JPEG’d we will award certain “trait boosts” which multiply the Floor Price evaluation of the collateral by the on-chain metadata/traits for that specific NFT. JPEG'd provides a CDP-like NFT-Finance solution which is comparable in several ways to Parallel's P2Pool lending model.
Research and Methodology for Trait Boosts
We process all NFT sales the top-tier NFT collections within the past 365 days. We then calculate the “Floor Multiple” for each successful NFT sale for which the NFTGo API has captured data. We use Floor Multiple, or the total USD sales value over the contemporaneous Floor Price value in USD, to normalize variation over time and collections. Finally, we compare average Floor Multiple for top traits and focus on the lasting relationships for each.
All of this data then shows us the relationship between specific NFT traits and their sales values as compared to the average value of NFT's within a collection.
We likewise use Dune on-chain data for sales to run similar studies and as a sense-check against NFTGo API data. Notably, however, specific NFT metadata and traits are less visible on-chain. For collection Floor Prices on-chain we look to use Parallel's oracles in the future. In the interim, however, we depend on BendDAO's Floor Price Oracles which have greater on-chain history.
The Dune results corroborate what we observed via the NFTGo data on a macro level. On a micro level specific traits are not easily visible on-chain, and we lean on NFTGo’s API for the final summary.
Eligible Traits for Boosting
In our calculations we highlight which traits across top-tier NFT collections have empirically gathered an average Floor Multiple sale price of at least 2x. We use 2x as the minimum for two reasons:
2x Floor price is most often a statistically significant difference in average sales price versus the average NFT price. Though not immune from random variation, a minimum of 2x premium gives confidence this particular trait garners a meaningful premium.
The liquidity premium discount (read below) is such that smaller trait premiums would not receive trait boosts.
There is a level of discretion we must use to ignore potentially problematic transactions and/or outliers. Further there are certain traits which are subsets of parent traits which we also must ignore.
Liquidity Premium Discount
Clearly we want to give the borrower the biggest trait boost possible, but as the protocol Parallel needs to balance risks for lenders and ultimately ensure users remain solvent. Intuitively and empirically we know there is a significant liquidity premium in NFT purchase prices; the highest “sell it now” bid price is almost invariably below the “buy it now” ask price.
Further we can gauge overall liquidity premiums in NFT markets over time by the proportion of NFT sales which are the result of accepted “sell it now” bid prices versus the traditional “buy it now” offer prices.
Using Dune we analyzed total “Floor Multiple” per sales on whether the sale was a result of a purchase at the “Buy it Now” ask price or an accepted offer at the “Sell it Now” bid price. For the top collections measured this liquidity premium measured as an 8% average discount in Floor Multiple between “Offer” sales and “Purchase” sales.
Source: Dune Analytics Dashboard, BendDAO Floor Price Oracle On-Chain Data
As a key caveat, the chart above measures average differences in only the past 10 months. Anecdotally and empirically we know that the proportion of sales resulting from accepted offers or outright purchases can vary significantly depending on market conditions. It’s when NFT traders and collectors suddenly command a much higher liquidity premium that accepted offers drive volume, while outright purchases are much more common in times of stronger NFT demand.
The below chart shows the proportion of sales resulting from accepted offers or purchases. Notably the sharp spikes in accepted offers coincides with times of strong crypto coin and NFT market volatility.
Source: Dune Analytics Dashboard
Here we see at least two potentially significant risks:
The liquidity premium is heavily correlated with market sentiment and dynamics. Fear is a stronger emotion than greed and retrenchments in risk appetite can exacerbate this liquidity premium. This can be seen clearly by the sharp rise in accepted “sell it now” bid prices as a share of total transaction volume.
“Grail” NFT’s by their nature are themselves a function of bull markets. During times of strong market stress/turmoil we empirically see premiums for these grail traits fall and exacerbate risks with liquidity premiums.
The chart below shows the volume for Azuki NFT’s widely considered as the “grails” of the collection in ”Spirit” types. The line in blue shows the average premium over floor prices in Average Floor Multiple, while the orange bars show aggregate volume in Spirit sales.
Average Premiums and Total Volume for Azuki’s Top Trait — “Spirit”
Source: NFTGo API Sales and Traits data
The Azuki collection floor price dropped sharply in May, and we saw the premium for “Spirit” Azuki NFT’s drop in kind. This was essentially a double whammy—the overall price of Azuki NFT’s fell while the demand for the collection’s top NFT’s fell even faster. The average USD price for a Spirit NFT fell from approximately $580,000 USD in April, 2022 to $168,000 in May. Any loans made at the former value would certainly have been liquidated and likely forced losses on any loans made against such collateral.
The Azuki NFT price declines were certainly among the most extreme we have witnessed in the past 12 months among blue chip collections. But the especially large declines in the “grail” Azuki NFT’s serve as clear caution against assigning too-large a premium for specific traits. What is a reasonable compromise?
Using On-Chain Transaction Prices as a Benchmark
According to our own on-chain studies of Floor Price data and multiples for accepted offers (”sell it now” prices”) versus purchases (”buy it now”), the liquidity premium for the average NFT should be between 10% and 27% for Bored Ape Yacht Club NFT’s. Or in other words, sellers will on average accept a “sell it now” offer 10-27% lower than their asking price.
When we factor in the “double whammy” of selling pressures on top NFT’s within a collection it becomes slightly more challenging—by their nature these top-trait NFT’s seldom see forced sales. But according to our data we may see a further 25% discount for top traits in top collections.
The combination of the liquidity premium and the so-called grail premium results in an average discount of 40% for Bored Ape Yacht Club’s top traits and almost exactly the same for the Mutant Ape Yacht Club.
With widely-recognized “grail” traits this effect is diminished; the Otherdeeds from the Otherside collection shows an approximate 25% aggregate discount for “Koda” NFT’s. We will have to subsequently set trait boost premiums within these ranges.
Estimates for Grail Trait Boosts
Below find our estimates for grail trait boosts across the top Bored Ape Yacht Club ecosystem as well as Azuki. We focus on these collections to start as they cumulatively have generated some of the highest volumes of any NFT collections. These collections likewise have specific “grail” traits which have consistently garnered substantial premiums in sales against the collection’s floor price.
See the full list of Token ID's for each collection which correspond to the below traits via this Google Sheet.
Bored Ape Yacht Club
Solid Gold Fur
6x
46
Trippy Fur
2.5x
77
Blue Beams Eyes
2.7x
49
Laser Eyes
1.2x
69
Mutant Ape Yacht Club
M2 Solid Gold Fur
2.6x
51
M2 Blue Beams
1.8x
59
M2 Trippy Fur
1.4x
87
M2 Black Suit
1.2x
44
Bored Ape Kennel Club
Solid Gold Fur
1.8x
98
Trippy Fur
1.2x
110
Blue Beams Eyes
1.2x
64
Otherdeeds from Otherside
Koda
8x
10,000
Azuki
Spirit Type
8x
97
Special Water
2.7x
49
Special Lightning
2.5x
48
Special Fire
2.5x
58
Special Earth
2.3x
53
Special Fireflies
2x
88
Last updated
