Pricing and hedging of decentralised lending contracts

Lukasz Szpruch,Marc Sabaté Vidales,Tanut Treetanthiploet,Yufei Zhang

2024-09-06

Abstract:We study the loan contracts offered by decentralised loan protocols (DLPs) through the lens of financial derivatives. DLPs, which effectively are clearinghouses, facilitate transactions between option buyers (i.e. borrowers) and option sellers (i.e. lenders). The loan-to-value at which the contract is initiated determines the option premium borrowers pay for entering the contract, and this can be deduced from the non-arbitrage pricing theory. We show that when there are no market frictions, and there is no spread between lending and borrowing rates, it is optimal to never enter the lending contract. Next, by accounting for the spread between rates and transactional costs, we develop a deep neural network-based algorithm for learning trading strategies on the external markets that allow us to replicate the payoff of the lending contracts that are not necessarily optimally exercised. This allows hedge the risk lenders carry by issuing options sold to the borrowers, which can complement (or even replace) the liquidations mechanism used to protect lenders' capital. Our approach can also be used to exploit (statistical) arbitrage opportunities that may arise when DLP allow users to enter lending contracts with loan-to-value, which is not appropriately calibrated to market conditions or/and when different markets price risk differently. We present thorough simulation experiments using historical data and simulations to validate our approach.

Pricing of Securities,Computational Finance,Risk Management

What problem does this paper attempt to address?

This paper attempts to solve the problems of pricing and hedging of loan contracts in Decentralised Lending Protocols (DLPs). Specifically, the author studies the loan contracts provided by DLPs from the perspective of financial derivatives and raises the following key questions:

1. **Are loan contracts mispriced?**
- Given a fixed initial loan - to - value (LTV) ratio \(\theta_0\) and a liquidation loan - to - value (LLTV) ratio \(\theta\), under different market conditions, does the premium \(P_0(1 - \theta_0)\) paid when entering a loan contract correspond to the no - arbitrage price?

2. **What is the initial capital required to establish a portfolio that replicates the returns of a loan contract after considering market frictions?**
- How can modern deep - learning tools be effectively used to learn the corresponding trading strategies?

### Specific Problems and Solutions

#### 1. Pricing of Loan Contracts
The author first explores the situation when the borrowing and lending rates are the same and there are no market frictions. They derive the fair - price formula for loan contracts under such ideal conditions and prove that the optimal strategy at this time is to never enter a loan contract. Specifically, according to the non - arbitrage pricing theory, for the case where the borrower decides to repay the loan at any time \(t\in[0, T]\), the non - arbitrage price of the loan contract is:

\[ \sup_{\tau\in T} E_Q \left[ e^{-r_c^D \tau} (P_\tau e^{r_c^E \tau} - \theta_0 P_0 e^{r_c^D \tau}) 1_{\{\tau < \tau_B\}} \right]=P_0(1 - \theta_0) \]

where:
- \(T\) is the set of stopping times taking values in \([0, \infty)\).
- \(\tau_B:=\inf \{ t\in[0, \infty) | \theta P_t e^{r_c^E t}\leq\theta_0 P_0 e^{r_c^D t} \}\) is the liquidation time.

#### 2. Pricing and Hedging after Considering Market Frictions
To deal with market frictions in reality (such as different borrowing and lending rates, trading costs, etc.), the author introduces an algorithm based on deep neural networks to learn the trading strategies of the external market in order to replicate the returns of loan contracts. This not only helps to hedge the option risk sold by the lender to the borrower but also enables the use of statistical arbitrage opportunities. For example, when a DLP allows users to enter loan contracts with an LTV that is not properly calibrated, or when different markets price risks differently.

### Simulation Experiments
The author uses historical data and simulation experiments to verify the effectiveness of their method. The experimental results show that, in the case of considering trading costs, the relative error of the deep - learning hedging strategy is far lower than that of the traditional Delta - hedging strategy (within one order of magnitude). In addition, the experiment also shows the comparison between the initial wealth \(v_0\) and the loan price \(P_0(1 - \theta_0)\) under different market conditions (such as volatility, market trends), as well as the change in training error.

### Summary
The main contributions of this paper are:
- Proposing a framework for evaluating whether the pricing of loan contracts in decentralized lending protocols is reasonable.
- Developing a deep - learning - based hedging strategy that can effectively replicate the returns of loan contracts and hedge risks in the presence of market frictions.
- Verifying the effectiveness and robustness of the proposed method through extensive simulation experiments.

These research results can provide important theoretical support and technical means for designing more robust decentralized finance (DeFi) systems.