On the existence of shadow prices for optimal investment with random endowment

Lingqi Gu, Yiqing Lin, Junjian YangStochastics 2017 · DOI

Abstract

In this paper, we consider a numéraire-based utility maximization problem under constant proportional transaction costs and random endowment. Assuming that the agent cannot short sell assets and is endowed with a strictly positive contingent claim, a primal optimizer of this utility maximization problem exists. Moreover, we observe that the original market with transaction costs can be replaced by a frictionless shadow market that yields the same optimality. On the other hand, we present an example to show that in some case when these constraints are relaxed, the existence of shadow prices is still warranted.

Introduction

In classical financial theory, markets are often assumed to be frictionless, allowing investors to trade assets at a single price without cost. However, real-world markets are characterized by transaction costs, typically manifested as a bid-ask spread. In such markets, an investor buys at a higher "ask" price and sells at a lower "bid" price.

A "shadow price" is a theoretical construct that bridges the gap between frictionless and frictional markets. It represents a price process that lies within the bid-ask spread such that an investor trading in a frictionless market at this price would achieve the same optimal utility as in the market with transaction costs. While the existence of shadow prices has been established for various scenarios involving deterministic initial wealth, this paper investigates the more complex case where the investor holds a random endowment—a future payout contingent on market conditions—and faces no-short-selling constraints.

Problem Formulation

The study focuses on a numéraire-based utility maximization problem. By using a base asset (such as a bond) as the unit of account, the authors simplify the mathematical representation of the investor's utility. The primary constraints considered are:

  1. Proportional Transaction Costs: A fixed percentage cost applied to every trade.
  2. Random Endowment: A contingent claim that the investor receives at the end of the trading horizon.
  3. No-Short-Selling: The investor is prohibited from holding negative amounts of the risky asset or the numéraire.

A critical assumption in the main theorems is that the random endowment is strictly positive. This ensures the investor maintains a financial buffer, preventing utility from reaching negative infinity and ensuring the mathematical derivatives of the value function remain well-defined.

Methodology: The Primal Approach

Most existing literature on shadow prices utilizes "convex duality," solving a related dual problem to infer properties of the original primal problem. In contrast, this paper adopts a direct primal approach, working directly with the utility maximization problem.

The authors utilize several key mathematical tools:

  • Superreplication Theorem: They adapt this theorem to demonstrate that any payoff achievable in a market with transaction costs is dominated by a payoff in a specific frictionless shadow market.
  • Dynamic Programming Principle (DPP): The DPP is employed to show that the value function—representing the maximum expected utility—behaves as a martingale along the optimal trading path.
  • Direct Construction: Following the logic of earlier work by researchers like Loewenstein, the authors construct the shadow price process using the derivatives of the dynamic primal value functions, specifically adapted for the presence of random endowments.

Key Results

The paper presents two primary theoretical breakthroughs regarding the feasibility of optimal investment in frictional markets:

1. Existence and Uniqueness of the Optimal Strategy

In Theorem 3.5, the authors prove that under the constraints of proportional transaction costs, no-short-selling, and a strictly positive random endowment, there exists a unique optimal trading strategy for the agent.

2. Existence of the Shadow Price

In Theorem 3.8, the authors establish that a λ-supermartingale consistent price system exists. This system functions as a "classical" shadow price, meaning it is a frictionless price process that leads to the same optimal strategy and utility as the original frictional market.

Discussion and Heuristic Extensions

The paper highlights that the shadow price is not merely a property of the market but is intrinsically linked to the specific agent's holdings and constraints. The decomposition of a future payout into initial wealth and a random component can fundamentally alter the resulting shadow price.

Furthermore, while the main theorems rely on the positivity of the endowment, the authors provide a constructive example within a Black-Scholes framework. This example demonstrates a "buy-hold-sell" strategy and suggests that shadow prices may exist even when the endowment is a liability (negative) or when short-selling constraints are partially relaxed. This indicates that the conditions in the main theorems are sufficient but perhaps not strictly necessary.

Conclusion

This research provides a rigorous foundation for the existence of shadow prices in the presence of random endowments. By moving beyond deterministic wealth, the authors bring the theory of optimal investment under transaction costs closer to real-world applications where investors often hold complex portfolios of future claims. Future work may focus on relaxing the positivity requirement for endowments and extending the framework to multivariate utility functions or multi-currency settings.

Summary updated

Abstract

In this paper, we consider a numéraire-based utility maximization problem under constant proportional transaction costs and random endowment. Assuming that the agent cannot short sell assets and is endowed with a strictly positive contingent claim, a primal optimizer of this utility maximization problem exists. Moreover, we observe that the original market with transaction costs can be replaced by a frictionless shadow market that yields the same optimality. On the other hand, we present an example to show that in some case when these constraints are relaxed, the existence of shadow prices is still warranted.
BibTeX
@inproceedings{gu2017existence,
  author = {Lingqi Gu and Yiqing Lin and Junjian Yang},
  title = {{On the existence of shadow prices for optimal investment with random endowment}},
  booktitle = {Stochastics},
  year = {2017},
  doi = {10.1080/17442508.2017.1346656},
}
Figure