TL;DR
This paper develops a general framework for understanding equilibrium pricing in financial markets that are incomplete, meaning not all risks can be perfectly hedged. The authors introduce a novel approach using "translation invariant preferences" to prove the existence and uniqueness of equilibrium prices, offering a more robust theoretical foundation for pricing complex financial instruments.
What Problem Does This Paper Address?
Financial markets are rarely "complete" in the theoretical sense. This means that even with a wide array of assets, there are still risks that cannot be perfectly hedged away. This incompleteness poses a significant challenge for pricing derivatives and other complex financial products, as standard models often rely on the assumption of complete markets. Existing approaches to incomplete markets often struggle with issues of existence and uniqueness of equilibria, making it difficult to determine a definitive equilibrium price. This paper aims to bridge this gap by providing a general framework for equilibrium pricing in incomplete markets, specifically addressing the complexities introduced by heterogeneous agents, unspanned random endowments, and trading constraints.
Methods
The core of the paper's methodology lies in the introduction and application of "translation invariant preferences." These preferences, which are updated in a time-consistent manner, allow the authors to recursively construct equilibria using backward induction. This approach is flexible enough to handle heterogeneous agents, unspanned random endowments, and convex trading constraints.
Key techniques include:
- Backward Induction: The equilibrium is constructed step-by-step, moving backward in time from the final maturity date.
- One-Step Representative Agent: At each step, a representative agent is constructed with preferences over one-step gains, simplifying the analysis of complex market interactions.
- Convex Duality and Fixed-Point Arguments: These standard mathematical tools are employed to prove existence and uniqueness results.
- Backward Stochastic Difference Equations (BSΔEs): For specific noise models (finitely many Bernoulli random walks), the equilibrium dynamics are characterized by coupled systems of BSΔEs, which in the continuous-time limit become Backward Stochastic Differential Equations (BSDEs).
Datasets
Not discussed in the paper.
Key Insights and Ideas
- Translation Invariant Preferences as a Unifying Framework: The paper demonstrates that translation invariant preferences provide a powerful tool for analyzing equilibria in incomplete markets, offering a more tractable and general approach than previous methods.
- Existence and Uniqueness Guarantees: Under certain conditions (e.g., differentiable preferences and open trading constraints), the framework guarantees not only the existence but also the uniqueness of equilibrium prices. This is a significant theoretical advancement.
- One-Fund Theorem in Incomplete Markets: In a special case where agents have similar preferences and endowments are replicable, the paper shows that a one-fund theorem can hold even in incomplete markets, simplifying portfolio allocation.
- BSΔEs as a Natural Language for Incomplete Market Dynamics: The use of BSΔEs to describe equilibrium dynamics highlights their suitability for modeling complex, path-dependent processes in discrete-time incomplete markets.
Key Results
- The paper provides theoretical results on the existence and uniqueness of equilibrium prices in dynamically incomplete financial markets under translation invariant preferences.
- It shows that if preferences are differentiable and trading constraints are open, equilibrium prices are unique.
- In the case of finitely many Bernoulli random walks, the equilibrium dynamics can be described by coupled BSΔEs, which become multi-dimensional BSDEs in the continuous-time limit.
- The paper offers an example simulating option prices under stochastic volatility, demand pressure, and short-selling constraints, illustrating the practical implications of the theoretical framework.
Limitations and Future Work
- Computational Complexity: While the paper provides theoretical results, the explicit computation of equilibrium prices and strategies, especially for complex noise models and constraints, can be computationally intensive. The authors note that simulating the BSΔE system is a non-trivial numerical task.
- General Existence of Solutions for BSDEs: The paper conjectures the existence of solutions for the coupled system of BSDEs in the continuous-time limit, but a general result is still lacking.
- Heterogeneity and Unspanned Endowments: While the framework accommodates heterogeneity and unspanned endowments, the complexity of these features can make explicit solutions challenging. The paper suggests that the situation becomes different when agents are truly heterogeneous or endowments are unspanned, implying further research is needed in these areas.
Summary updated