Tom J. Pandolfi

Twitter / GitHub / Medium

Why nobody cares about NFTs x DeFi

Let’s talk about Honey’s competitors in NFTs x DeFi — what some call DeFi 3.0

“We’re bridging the gap between NFTs and DeFi”

Cool, but does this gap need to be bridged?

As a founder building a NFT x DeFi product, this is something that has been on my mind for a while now. So I decided to run a quick poll in our community of people who are somewhat exposed to the theme of NFTs and DeFi.

Obviously this sample is not representative of anything, but it was enough to make me think… how bad can liquidity really be, if these people aren’t even trying out alternatives.

Better yet, how come so many companies are raising millions of dollars in Series A’s using the exact same phrase, when nobody is using their products.

There’s two obvious possibilities:

1- Liquidity of NFTs isn’t actually a problem

2- None of these solutions are addressing the right pain points

Let’s explore option 1 for a minute…

Most of us have really just taken for granted the “fact” that liquidity is a problem in NFTs, if not the biggest problem. But let’s make sure that’s really the case.

The best way to check this is by testing the null hypothesis, which is: “liquidity in NFTs is not a problem” and then seeing if it can be disproven via reductio ad absurdum.

The following is a summary of this thought experiment, you can skip this to read the conclusion at the bottom.

First, we thought, “In the best case scenario, how liquid are the most liquid NFTs”, if even those failed to be liquid, we could argue that there is a lack of liquidity in NFTs overall. Then we find the consequences of this lack of liquidity to see if it’s a problem.

The definition of liquidity: Liquidity refers to the ease with which an asset, or security, can be converted into ready cash without affecting its market price — investopedia.

So there’s two elements to this definition, the first is time, the second is slippage.

Selling NFTs can be instant. A collection with enough demand will have bots waiting to snipe those who list far below the floor. In other words, NFTs can be sold instantly, but only with incredibly high slippage (loss in value compared to the market price).

How much slippage ? Let’s keep focusing on the best case scenario and find the lowest slippage collections. We’re looking for collections that are 1) very fungible 2) have a large number of NFTs 3) have large transactional volume indicating demand.

One of the best places to find these collections is LooksRare, a marketplace which allows users to bid on any given NFT in a collection. To fulfil the other requirements, the collection should have more than 10k NFTs, so as to allow a lot of different prices in our sample, and then large transactional volume to reduce the chances of the price being a fluke.

With this reasoning, we (not surprisingly) find that the top ETH collections are the most liquid NFTs.

Currently, the floor of BAYC is 97 ETH, and the highest “floor bid” is 82.3 ETH (Slippage is 15.2%). Mutant Apes floor is at 22.5 ETH while the highest “floor bid” is at 19 ETH (Slippage is 15.56%), Doodles is 13 ETH vs 11 ETH (15.38%), Meebits is 6 ETH vs 5 ETH (16.67%), Cryptoadz is 3.39 ETH vs 2.8 ETH (17.4%), Cool Cats is 8.5 ETH vs 7 ETH (17.64%), Sandbox Lands are 2.9 ETH vs 2.6 ETH (10.34%), Decentraland LAND 4 ETH vs 2.6 ETH (35%), Mavia land 1.1 ETH vs 0.9 ETH (18.2%) etc.

You’ll notice these collections all have around 15% slippage, except commoditised NFTs like land, which at best will hover at around 10% slippage.

If we generalised this best case scenario to all NFTs, they would incur an average slippage of 10% per token. Whether 10% is illiquid or not is subjective, however it would only take 10 trades with this kind of slippage for NFTs to lose 70% of their value.

I’ll let you judge if we’ve arrived at absurdity, but let’s get back on subject.

Conclusion: NFTs are most likely very illiquid.

Another reason I’m not convinced by option 1, is that not only is liquidity the biggest complaint of NFT investors, it’s a much bigger problem than most of them realise.

Few investors factor in the opportunity cost of holding JPEGs. 1$ in an NFT is a dollar that could be yielding 20% or more on UST stablecoins. That 20% compounds, so not only are you missing out, the cost of you missing out grows exponentially.

So what solutions already exist ?

Peer-to-Peer (P2P) loans on NFTs

There is a lot of merit to peer-to-peer loans, just like there is merit to using real world art as collateral in a pawnshop. In the case of very rare assets, this is almost a necessity.

My issue with P2P lending protocols, is the way they have been praised by investors as the panacea for over 2 years, stopping any real innovation from taking place. It’s tempting to think that peer-to-peer is the limit of what can be done for non-fungible loans, and we’ve seen very little innovation since the birth of NFTfi in May 2020.

It’s been almost 2 years now since peer-to-peer loans were introduced. Most people you see complaining about liquidity in NFTs probably joined the NFT space after peer-to-peer loans were introduced. The jury is out, and the lack of product market fit is obvious.

It’s all the friction of a marketplace, with less volume and high interest rates.

NFT fractionalisation

This could be very interesting one day, but what problem is it currently solving?

Whether we like it or not, NFTs currently derive most of their value from their cultural or artistic significance within a group.

Nobody cares if you own 20% of a Bored Ape, you’re not a part of their community. You’re not even 20% part of it.

It’s naive to think that owning 50% of an SMB offers 50% of its value, and that value goes down exponentially the less of it you own.

With fractionalisation, the sum of the parts is less than the whole.

The technology may one day be very interesting as NFTs become complex financial assets, whose risk can be diversified through fractionalisation.

Pawning (aka Treasury Backed Loans)

The next common liquidity solution is pawning. I’ll use the term pawning to describe projects who issue loans on the basis of their treasuries.

These projects don’t just willingly decide to offer a non-scalable product. The reason they do this is because they’re forced to make a tradeoff between scalability and speed. They choose speed at the cost of scalability.

You can think of it like AirBnB vs a chain of hotels.

AirBnb is scalable because it brings together supply (landlords) and demand (tenants) similar to how NFTfi might bring together lenders and borrowers. There’s no shortage of these two and as long as you have an even number of users on both sides, your platform will continue to match supply and demand, generating profit each time.

Hotels, despite offering the same service as AirBnB (housing), have a very different business model. They supply housing, just like an NFT pawnshop supplies loans, but how much can they really grow relative to a network of landlords supplying the same commodity?

The platform business model (AirBnB) scales. The centralised entity (the hotel) doesn’t, but reduces friction and has far better margins.

Another problem with pawning: issuing loans as a single lender in a market leads to receiving entire NFTs that you don’t have to instantly sell (in a market with many lenders, you would need to sell the NFT to pay back a group of lenders in the asset they supplied). Their liquidation mechanism is simple. Receive the NFT, list it on a marketplace below floor, wait for it to sell. In the next (first) black swan event for NFTs, these treasury backed lending projects who rely on marketplaces to liquidate will be left with no treasury and lots of JPEGs.


These projects do issue instant loans.

So why don’t borrowers take advantage of this ? The answer is flat interest rates.

Interest rates are like prices, literally, it’s the price of borrowing money in a market.

The 20th century provides more than enough examples to illustrate the following point: centralised / planned pricing doesn’t work.

Unless prices are naturally determined by supply and demand, they will always lead to shortages and/or surpluses.

In our example of prices being interest rates:

Pawning protocols have a choice to make: Charge too much or charge too little.

In both cases, the growth rate is very limited.

A protocol that can charge variable interest rates, will thus (in the long run) always provide better interest rates than one who charges flat rates, as the price of borrowing with flat rates is not a true reflection of supply and demand.

As NFT users, our team intuitively knew the problem had yet to be solved, or else we’d be first in line using the solution.

We naively decided to undertake what peer-to-peer projects spent two years saying couldn’t be done, creating a peer-to-contract model.

Background: Before AAVE was AAVE, it was called ETHLend. ETHLend was the first lending protocol on Ethereum in the very early days of DeFi, and just like all new financial systems, it started with peer-to-peer. The protocol became famous when it moved to a peer to contract model, where all the liquidity from lenders was pooled together in a smart contract, and distributed to borrowers. The ratio between how much of the liquidity was being supplied vs borrowed determined a variable interest rate, and AAVE became the scalable juggernaut that it is today. Credit to Robert Leshner from Compound for creating these types of interest rates.

So why not do peer-to-contract loans on NFTs ?

Two big issues need to be solved before we can consider that:

  1. Liquidations

  2. Oracles

When lenders pool money together to supply liquidity for people borrowing ETH with BTC collateral, they know that if borrowers default, the BTC collateral can easily be sold, and they will get their ETH back.

When lenders pool money together to supply liquidity for people borrowing SOL with NFT collateral, they have no guarantee that if the borrowers defaults, the NFT collateral can easily be sold, and that they will get their SOL back.

That’s because selling NFTs as we saw before is usually a tradeoff between time and slippage, sell it fast and you incur a loss, sell it at a better price and it takes too much time to liquidate.

First solution ? Selling NFTs in liquidity pools.

Great protocols such as Solvent, NFTx or NFT20 allow users to create liquidity pools for NFTs. Deposit an NFT, get 100 tokens worth 1% of the floor’s value. Deposit 100 tokens, claim an NFT in the pool.

These pools are fantastic for getting instant liquidity, and don’t require the demand to be as high for the liquidity to be instantly available. NFT Projects can provide liquidity to these pools with treasury money, earn trading fees and drastically reduce slippage of selling NFTs.

This solution should really be used in tandem with other liquidation methods, such as allowing DAOs and collectors to bid on collateral. The more liquidation methods can be included in a dapp, the less holes there will be in the liquidation net.

It became clear to us early on that flexibility and modularity was the best way to approach the diversity of different situations in the world of NFTs. Some collections would need randomised liquidations for gamification, while others can use liquidity pools.

It’s with this modular approach that Honey Labs has tackled the second issue, oracles.

Because Honey allows markets to be created on its platform, rather than issuing loans from its treasury, it can give a lot of flexibility to market creators.

An example of this is the fact that market creators get to pick what oracle best suits their needs. By leveraging many oracles at the same time, Honey benefits from different teams building sophisticated tools for NFT pricing, and is agnostic about which one market creators select.

We’re partnering with Tensor to provide high quality on-chain information, which uses statistical processes to remove potential price manipulation. However some projects may decide to use a TWAP oracle from the liquidity pools on Serum used in our liquidations. Others may decide to use switchboard, Magic Eden, or any source of truth that best suits their market.

That is the beauty of an open lending market approach, where we allow others to create markets on our platform.

Our DAO is committed to helping these teams build resilient NFT price feeds and infrastructure on top of Honey, so that we may become the number one hub for cheap, instant, and decentralised NFT loans.

So all that to say, the problem with NFTs is very real, and very costly.

However we’ve yet to find a protocol with product market fit. As NFTs and the metaverse continue to grow, the market is still on the lookout for who will conquer this problem. The issue, is that one protocol can’t do it alone, there are too many moving parts for one project to solve on its own.

Honey’s approach is focused on modularity and composability with other protocols. Our best hope for providing users with a liquidity solution that actually works is leveraging the composability of DeFi. Allowing different teams to focus on their piece of the puzzle.

With our open lending market approach, we hope to foster this composability on our platform and form an ecosystem around our community that is focused on solving the big problems of tomorrow’s metaverse and financial systems.

The power of Honey Finance comes from the Honey Ecosystem, full of talented developers and innovative projects building on top of our protocol and our $HONEY token. They will be announced in next week’s Sunday newsletter.