Free Lunches with Vanishing Risks Most Likely Exist
Eckhard Platen, Kevin Fergusson · 2025
Read Free Lunches with Vanishing Risks Most Likely Exist by Eckhard Platen, Kevin Fergusson (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-07108.txt.
Introduction
Eckhard Platen and Kevin Fergusson challenge a founding assumption of modern finance: that no strategy offers a free lunch with vanishing risk. Standard pricing theory rests on that claim. The authors say the real market fails it, and they show a concrete strategy to prove the point.
Their portfolio starts from zero wealth. It trades dynamically between a broad stock index and a savings account, hedging a very long-dated zero-coupon bond with extreme accuracy. At the bond's maturity, the strategy ends with a strictly positive amount with positive chance: the forbidden free lunch. The authors note their benchmark approach to pricing can absorb this fact. If they are right, textbooks built on the old assumption need revision.
Catalog record
| CREATOR | Eckhard Platen, Kevin Fergusson |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-07108
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Free Lunches with Vanishing Risks Most Likely Exist Eckhard Platen, Kevin Fergusson Abstract The hypothesis that there do not exist free lunches with vanishing risk (FLVRs) in the real market underpins the popular risk-neutral pricing and hedging methodology in quantitative finance. The paper documents the fact that this hypothesis can be safely rejected. It performs extremely accurately the hedging of an extreme-maturity zero-coupon bond (ZCB). This hedge is part of a portfolio that starts with zero initial wealth and invests dynamically in a total return stock market index and the savings account to generate at the maturity date of the extreme-maturity ZCB a strictly positive amount with strictly positive probability, which represents an FLVR. The fact that FLVRs naturally exist in the real market can be accommodated theoretically under the benchmark approach. JEL Classification: G10, G11 Mathematics Subject Classification: 62P05, 60G35, 62P20 Key words and phrases: free lunch with vanishing risk, fundamental theorem of asset pricing, extreme-maturity zero-coupon bond, hedging, minimal market model, benchmark approach, benchmark-neutral pricing. 1 Introduction Quantitative methods in finance rely on market models. These models must be mathematically consistent and close to reality to ensure that the solutions to risk management problems can be reliably applied in practice. For this reason, it has been argued in the literature that it is not realistic to allow in a market model the existence of arbitrage opportunities . An arbitrage opportunity is a way of making profits with no initial investment without any possibility of loss . Arbitrage opportunities may exist briefly in real markets. However, most of the current literature argues that a useful market model must avoid this type of profit; see, e.g., ? ). For this reason, the classical mathematical finance theory, as described, e.g., in ? ), relies on its Fundamental Theorem of Asset Pricing (FTAP), which was derived in the seminal work of ? ). The FTAP has become extremely important as a theoretical underpinning of quantitative methods in finance because it provides the foundations of the widely implemented risk-neutral pricing methodology. The FTAP employs the notion of a free lunch with vanishing risk (FLVR), most generally defined in ? ). It is currently by far the most widely considered classical notion of arbitrage for continuous-time market models. The FTAP considers admissible portfolios , which are self-financing portfolios that are bounded from below . Informally said, a market model allows for an FLVR if there are admissible portfolios that can be chosen arbitrarily close to an arbitrage opportunity. This means, these portfolios start with no wealth, end up with strictly positive wealth (a free lunch) with a probability greater than zero, and the probability of ending up with negative wealth can be chosen arbitrarily small (the vanishing risk). The current paper aims to answer the question, do free lunches with vanishing risks most likely exist in the real market? The philosopher Karl Popper established a modern set of standards for the scientific methodology; see ? ). The current paper follows these standards by aiming to falsify the null hypothesis of the FTAP that FLVRs do not exist. For this purpose, it considers the hedging of extreme-maturity zero-coupon bonds (ZCBs) that pay approximately one unit of the savings account at maturity. For each extreme-maturity ZCB a portfolio will be formed that starts with zero initial wealth and invests dynamically in a total return stock index and the savings account. This hedge will be shown to generate very reliably, at the maturity date of the extreme-maturity ZCB, some nonnegative and mostly strictly positive amounts. The empirical study will investigate potential FLVRs for the S&P500 total return stock market index and many-extreme maturity ZCBs. The study will falsify in a hypothesis test with high confidence the null hypothesis that FLVRs do not exist. In its second part, the paper reviews parts of the benchmark approach, as described in ? ). This approach can accommodate the fact that FLVRs exist in the real market. By using its benchmark-neutral pricing, as described in ? ), and the information-minimizing dynamics of a stock market index, as derived in ? ), the theoretical prices and hedging strategies of ZCBs are derived. These prices and hedging portfolios explain the presence of FLVRs in the real market under the benchmark approach. The paper is organized as follows: Section 2 studies empirically the existence of potential FLVRs and Section 3 explains how FLVRs naturally exist under the benchmark approach. 2 Potential Free Lunches with Vanishing Risk 2.1 Free Lunch with Vanishing Risk Most papers and books currently published in the mathematical finance and quantitative finance literature base their results on the Fundamental Theorem of Asset Pricing (FTAP); see ? ). To illustrate the statement of this important theorem let us consider a continuous-time financial market model on a filtered probability space (\Omega,\mathcal{F},\underline{\cal{F}},P) , satisfying the usual conditions; see, e.g., ? ). The filtration \underline{\cal{F}} =(\mathcal{F}_{t})_{t\geq t_{0}} models the evolution of information relevant to the market model. The information available at time t\in[0,\infty) is captured by the sigma-algebra \mathcal{F}_{t} . The market model that we consider describes the dynamics of a savings account S^{0}_{t}=1 and a vector {\bf{S}}_{t} of n\in\{1,2,...\} risky securities {\bf{S}}_{t}=(S^{1}_{t},...,S^{n}_{t})^{\top} for t\geq t_{0} . For convenience, we discount these securities by the savings account. The latter is a locally risk-free portfolio and, in practice, typically approximated by a roll-over short-term zero-coupon bond account. We model an idealized market with continuous trading, instantaneous investing and borrowing, short sales with full use of proceeds, infinitely divisible securities, and no or small realistic transaction costs. In ? ) the FTAP has been formulated as follows: Theorem 2.1 (Fundamental Theorem of Asset Pricing) For an {\bf{R}}^{n} -valued semimartingale {\bf{S}}=\{{\bf{S}}_{t}=(S^{1}_{t},...,S^{n}_{t})^{\top},t\geq t_{0}\} , the following are equivalent: 1. There exists a probability measure Q_{S^{0}} equivalent to P under which {\bf{S}} is a sigma-martingale. 2. {\bf{S}} does not permit an FLVR. The precise definitions of the notions involved in the above formulation of the FTAP are given in the above-mentioned paper ? ). It is not the purpose of the current paper to go into any details of the formulation of the FTAP or its technically demanding proof. Instead, it aims to provide empirical evidence concerning the validity of the key assumption of the FTAP, which is the assumption about the absence of FLVRs. Roughly speaking, the FTAP states that the absence of FLVRs in a market model is equivalent to the existence of an equivalent risk-neutral probability measure Q_{S^{0}} , where the savings account S^{0} is the numéraire and the equivalent risk-neutral probability measure Q_{S^{0}} is the pricing measure. This is a crucially important statement for pricing and hedging in real markets because the existence of an equivalent risk-neutral probability measure Q_{S^{0}} provides a system of unique prices for the pricing and hedging of replicable contingent claims. Risk-neutral pricing is currently the dominant pricing method for the pricing and hedging of derivative contracts; see, e.g., ? ). We define extreme-maturity contracts as contracts with terms to maturity that reach at least over a period of 15 years. Large volumes of extreme-maturity contracts exist in the pension and life insurance industry. The costs of these contracts are important for millions of individuals in many countries. Risk-neutral pricing is usually