NFT Lending & Risk Management
Understand how NFT Lending works, which are the risks associated and how to mitigate them
Decentralized finance (DeFi) protocols for lending and borrowing are among the most active and rapidly evolving sectors in the cryptocurrency industry. In recent years, these protocols have begun to expand into new areas, such as yield farming, derivative based and accepting a wider range of assets as collateral, including non-fungible tokens (NFTs).
In this post, we will explore the growing NFT lending industry, where borrowers use unique and non-interchangeable tokens as collateral for their loans. If you are new to the world of DeFi and want to learn more about its core principles, we recommend checking out our previous post on the subject👇
In this post we will cover:
- What is NFT Lending?
- What to take into account when Managing Risk
- Agent-based simulation for Risk Management
- What’s next
What is NFT Lending?
NFT lending (or rather, NFT-based lending) is the process of using non-fungible tokens (NFTs) as collateral for a loan. This practice is facilitated by NFT lending protocols, which allow users to obtain liquidity from their NFTs. On one side, borrowers can use their NFTs as collateral to obtain a loan. On the other side, lenders can earn yields from the loans issued and can use the collateral as a form of repayment. This can take the form of peer-to-peer lending (between two individuals) or peer-to-pool lending (an individual interacting with a smart contract in the middle).
To illustrate how this works, let’s consider the following example:
A lender deposits a token (such as ETH, USDC, or SOL) in a lending pool, expecting to earn a yield. A borrower then introduces an NFT into the lending pool in order to obtain liquidity from it, borrowing the previously deposited token. Like other DeFi initiatives, NFT lending protocols typically make use of over-collateralized loans. However, as with DeFi in general, there are some complex questions that need to be answered, such as what loan-to-value (LTV) and liquidation thresholds and bonuses should be used.
After a predetermined period of time, one of two things can happen:
- The borrower repays their loan, including any interest that has accrued, and reclaims their NFT.
- The borrower is liquidated. Depending on the protocol, this can play out in different ways, so we will use the most common as an example. When a loan reaches its safety margin, it enters a grace period. During this time, the borrower can repay the loan or the NFT will be put up for auction. If the NFT goes to auction, different players can bid on it, and the winning bidder becomes the new owner and acts as the loan liquidator.
In summary, NFT lending involves the use of non-fungible tokens as collateral for loans facilitated by specialized protocols. However, like other trustless lending initiatives, NFT lending protocols can be vulnerable to extreme market conditions, which can make the process more complex.
What to take into account when Managing Risk
When managing the risks associated with NFT lending, there are several key factors that must be considered. These include the potential for changes in asset prices, market liquidity, protocol mechanics, and the ability to liquidate positions. These and other potential scenarios can be grouped into three main areas:
- Market Risk: This refers to the potential for changes in the value of assets that are used as collateral or as loan repayment, as well as the potential for a lack of liquidity in the market.
- Protocol Risk: This involves the potential for changes in the lending mechanics or other aspects of the protocol itself, which could affect the viability of existing loans or the ability to create new ones.
- Position Risk: This encompasses the potential for positions to become illiquid or otherwise difficult to manage, such as in the event of a borrower defaulting on their loan or the failure of a liquidation mechanism.
Managing these risks effectively is critical to the success of an NFT lending operation, as they can have significant implications for both borrowers and lenders.
In the following section, we will thoroughly examine each element individually.
Market Risk — Defining value, volatility and liquidity
Market risk is a key factor in the NFT lending industry, as it can have a major impact on the value of assets used as collateral or loan repayment. This risk is related to three key factors: valuation, volatility, and liquidity.
- Valuation: The unique nature of NFTs makes it difficult to accurately assess their value, which can create uncertainty and make it challenging to manage risk. The consensus within the industry is to estimate the worst-case value of an NFT based on the “floor price” of the collection it belongs to.
- Volatility: NFT lending is exposed to the risk of sudden price shocks that can leave loan positions inadequately collateralized. To account for this risk, we must consider the volatility of each asset (collection) based on its floor price and incorporate this into our price path generation engine.
- Liquidity: Sudden price movements are often linked to liquidity in the market. Therefore, the volume of sales for each collection of NFTs is a critical parameter to consider. In our simulations, we will define and monitor the slippage effect on NFTs as the change in price that the market experiences in response to a sudden increase in the number of NFTs listed on the secondary market.
Having clear definitions for price and volatility is essential for effective risk management in the NFT lending industry. These definitions help us to develop and apply robust methodologies for assessing and mitigating market risk. This is crucial for protecting both borrowers and lenders from the potential impacts of market volatility and other risks.
Protocol Risks — The parameters that generate debt
In the context of NFT lending, protocols are responsible for generating debt secured by NFTs. There are two key factors to consider when assessing the risks associated with this type of debt:
- The total amount of debt generated per asset, which is typically defined by the loan-to-value (LTV) ratio of the loan. The maximum LTV sets the maximum amount of capital that can be loaned against a given asset. A higher LTV can increase capital efficiency for borrowers, but it also increases the risk of default.
- The characteristics of each individual loan, such as the grace period and margin of safety. The grace period is the amount of time that a borrower is given to repay their debt once it reaches a certain limit (for example, the expiration of a flip loan or the breach of a liquidation threshold for a perpetual loan). The margin of safety is a parameter used to determine the loan’s health, and it can also be used to set the liquidation threshold and liquidation bonus.
As an example, we can look at FRAKT protocol, which is a client of Cenit Finance. In FRAKT perpetual loans, the start of the grace period is triggered when the debt reaches the liquidation threshold (equivalent to a loan health of 0). In flip loans, the grace period begins when the loan exceeds its time limit. The margin of safety in FRAKT has a dual role, acting as both a liquidation threshold and a liquidation bonus.
Position Risk — How to evaluate the Risks of the debt generated
Once we have established a method for evaluating assets and creating secured debt on top of them, we must also find a way to measure the risk associated with these positions. For this purpose, we use two different risk metrics: Value at Risk (VaR) and Liquidator at Risk (LaR). These metrics are quantiles of the Profit and Loss (PnL) of each distribution, and they provide a useful way to assess the exposure and potential risks of our positions.
- VaR is defined as the 5th percentile of PnL for the lending pool at the end of the simulations (including any outstanding debt on non-liquidated loans). This metric helps us to monitor the potential bad debt that the protocol might incur due to non-incentivized liquidations.
- LaR is the 5th percentile of realized PnL for liquidators (the value of sales of NFTs on the secondary market minus the debt repaid for those NFTs). LaR measures the risk to liquidators, who may not be able to immediately sell the asset and therefore be exposed to market risks.
With clear definitions for market risk, protocol risks, and risk metrics, we can easily visualize how different market conditions and protocol parameters can either add or reduce risk in a given protocol. This helps us to manage and mitigate risk more effectively in the NFT lending industry.
But how can we create these risk measures and apply them in practice? Check out our methodology
Agent-based simulations for Risk Management
To assess the risks associated with NFT lending, we have developed a methodology based on agent-based simulations.
In this approach, each type of user that interacts with the protocol is represented by a mathematical function. The methodology consists of three main components:
- The simulation environment: This component simulates market conditions and determines the potential impacts on the protocol and its users. Different models can be used to simulate market price changes, such as Gaussian Brownian Motion. Based on these simulations, we can determine the changes in the number of NFTs listed on the market and their potential impact on the protocol. In the simulation, we also take into account the possibility that the increased supply of NFTs on the market could cause price slippage. By incorporating this model into the simulation, we can more accurately predict the potential impact on current loans on the protocol.
- The protocol replica: This component is a replica of the protocol itself, including its current mechanisms for generating debt and any other relevant parameters. This replica interacts with the simulation engine and follows the same rules and behavior as the real protocol, allowing us to simulate how the protocol would operate under different market conditions and user activity.
- Agent simulation: This component simulates the behavior of the various agents involved in an NFT lending operation, such as borrowers, lenders, and liquidators. By modeling their incentives, decision-making processes, and other factors, we can gain insight into how they will behave under different scenarios and how this will impact the overall performance of the protocol. There are four main types of agents in an NFT lending operation: lenders, borrowers, liquidators, and oracles.
a) Lenders: These agents provide liquidity to the protocol by depositing tokens into lending pools. They can choose which pools, flips, or perpetuals to add liquidity to, and they earn yield from the loans issued against their deposits.
b) Borrowers: These agents can obtain liquidity from their NFTs by using them as collateral for a loan. The conditions of the loan, including its duration and LTV ratio, are determined by the pool they borrow from. Borrowers must ensure that their NFTs are whitelisted by the protocol in order to use them as collateral.
c) Liquidators: When a loan is liquidated, these agents have the opportunity to purchase the NFT collateral at a discount. After a grace period during which the borrower can repay the loan with a penalty, daily raffle winners can opt to repay the debt and claim the NFT. Liquidators can also impact the price of the NFT collection if they attempt to resell the collateral.
d) Oracles: These agents provide the “floor price” for each NFT collection, which is used to determine the value of the collateral with a given loan. The accuracy and reliability of the oracles’ data is critical for the effective operation of the protocol.
Simulations are models that mimic the operation of an existing system. By replicating the system and generating new market data with different protocol conditions and user interactions, we can explore various scenarios and evaluate their potential risks.
We generate simulation steps of one hour that allow us to create probabilistic scenarios one month ahead. In each step, we gather the current floor price for an NFT collection from real market data and apply a change based on the volatility of the simulation. If a loan is eligible for liquidation, its ratio of competitiveness is calculated and a sale probability is determined. If the NFT is sold, we compute the PnL for the liquidator LaR. We also calculate the value at risk (VaR) for the lending pool by simulating the potential PnL for non-incentivized liquidation loans. This helps us monitor the potential bad debt the protocol might incur.
This process is iterated using our cloud-based agent-simulation tool, and we calculate the resulting risk metrics for different market conditions and protocol parameters. This allows us to identify potential risks and develop strategies for mitigating them.
If you want to know more we have a full detailed report 👇 https://github.com/CenitFinance/frakt-report/blob/main/frakt-report.pdf
Current state and important questions
The current state of the NFT lending industry is marked by a high degree of uncertainty and potential risk. While some have declared that “98% of NFT projects will fail and go to zero” (source), others see NFTs as “new digital primitives” with enormous potential (source). We agree with both perspectives, recognising that NFTs are indeed a new and rapidly evolving asset class with many challenges and opportunities.
As with any developing industry, there will be failures and successes in the NFT lending space. However, we believe that the potential rewards of building solutions in this space are well worth the effort. According to DeFillama (source), there are currently 11 NFT lending protocols with a total value locked (TVL) of nearly $100 million. While this may seem small compared to the larger DeFi lending industry, which has 193 protocols and a TVL of $11 billion, we expect that adoption and the introduction of new asset types will drive significant growth in the NFT lending space in the coming years. Music, Uniswap v3 positions, property from Real World Assets, and many others are on their way to tokenisation.
We are excited to be part of this hectic and dynamic industry, and we look forward to the challenges and opportunities that lie ahead.
Where there is an asset or value, there is a Lending industry on top.
