The Tale of Two Tails — Why Convexity Isn’t Insurance

The Best Brakes Win the Race: A conversation with Antoine Guillon, Derivatives Trader & Portfolio Manager

"Risk attribution is much easier than risk forecasting."

Antoine Guillon has spent over two decades trading nonlinear risk, from credit correlation and structured credit at Brevan Howard and Credit Suisse to equity volatility and systematic portfolio research today. He was also the Co-CIO at Fourth Quadrant Investment Group. His starting point is a statistic: remove the ten best days from decades of S&P 500 returns and performance drops sharply, remove the ten worst and it jumps. A handful of observations can dominate decades of compounding.

In this conversation, he dismantles the idea that convexity is just expensive insurance.

On convexity: Antoine reiterates that convexity is a property of how a portfolio responds when conditions change, not a trade and not alpha. The edge is in the price of that payoff geometry, how it's financed, and how well it fits the portfolio underneath. He judges a hedge by four questions: what failure mode it changes, what you pay while you wait, whether it responds to your actual losses rather than a generic sell-off, and whether you can monetise the gain before your portfolio hits its constraint.

On carry: Ask what you're being paid to be short. Antoine walks through the negative CDS-bond basis heading into the financial crisis, a trade that looked like carry plus convergence but was really a short option on funding and liquidity. Apparently free convexity, he argues, usually means you've sold a tail somewhere else.

On owning the right convexity: A hedge can be perfectly convex and still be a poor hedge if its payoff doesn't line up with what actually hurts the portfolio. He separates payoff convexity, portfolio convexity, and monetizable convexity, and explains why index puts do little for a book concentrated in names that fall twice as hard as the index.

On tail risk: There is no single tail. 1987, 2008, 2018, 2020, and 2022 were different events, so Antoine prefers to talk about portfolio failure modes. Known risks map to specific hedges, and unknown risks call for robustness in liquidity and optionality. You can't hedge everything, because you still have to generate a return.

On correlation: Correlation is a tricky statistic, and its sign and size depend on your window. He explains why low index volatility can mask violent single-stock moves, and shares his own study suggesting changes in standalone volatility drove most of the change in a simple multi-asset portfolio's risk, with correlation explaining far less.

Topics covered: the S&P 500's best and worst days, credit correlation and tranches, convexity vs insurance, the cost of bleed and the St. Petersburg merchant, CDS-bond basis, payoff vs portfolio vs monetizable convexity, David Dredge's Formula One and forest fire analogies, correlation regimes, skew and term structure, the limits of backtests, systematic vs discretionary judgment, hedge sizing, who sells convexity and why, a broken-wing butterfly that lost money, and advice for starting a career in markets.

Timestamps

0:15 Introduction and the S&P 500 best/worst days study

4:00 What credit correlation teaches about nonlinear risk

7:08 Where "convexity as insurance" misleads investors

9:28 Four questions for a convex hedge, and the Formula One analogy

13:26 How much bleed is acceptable, and the St. Petersburg merchant

18:02 Carry vs compensation for tail risk, and the CDS-bond basis

22:32 Owning convexity vs owning the right convexity

25:16 The forest fire analogy and systemic risk

26:53 Why correlation is a tricky statistic

30:24 The danger of treating tail risk as one phenomenon

34:39 Why hedging second- and third-order effects is too late

35:23 Skew, term structure, and the problem with backtests

39:14 What can be systematic and what needs judgment

40:35 Sizing a convexity overlay

41:55 Who sells convexity, and why it matters

46:49 A trade that went wrong

53:19 Why convexity is gaining traction

58:33 How his beliefs about markets have evolved

1:01:07 Advice for people starting out

📌 Follow Antoine Guillon

LinkedIn:https://www.linkedin.com/in/anguillon/

Tale of Two Tails — Substack: https://taleoftwotails.substack.com/?r=8dm61l&utm_campaign=referrals-subscribe-page-share-screen&utm_medium=web

https://taleoftwotails.substack.com Antoine's

Substack explores convexity, tail risk, volatility, and the geometry of derivative markets.

FOLLOW & LISTEN

Website: https://www.1bps.us

Disclaimer: This podcast is for informational and educational purposes only and does not constitute financial, legal, tax, or investment advice. The views expressed by the hosts and guests are their own and do not necessarily reflect the views of any current or former employer, firm, or affiliated organization. Nothing in this podcast should be construed as a solicitation, recommendation, or offer to buy or sell any security, fund, or financial instrument. Listeners should consult with a qualified professional for specific advice tailored to their individual circumstances. We are not responsible for any losses or liabilities incurred by acting on information from this podcast.