What Is Positive convexity?
A position or trading strategy has positive convexity when its profit grows faster than the market move that causes it. Plotted against the price of the underlying asset, its value bends upwards: gains speed up as the market moves in its favour, and losses slow down as the market moves against it. Mathematically, the second derivative of value with respect to price (or yield) is positive. The opposite is negative convexity.
The term comes from two places in finance:
Options: a bought option has positive gamma, the rate at which its delta changes as the underlying moves. A long call becomes more like the underlying as the price rises and less like it as the price falls, so it keeps most of the upside and caps the downside at the premium paid. See options and gamma hedged.
Bonds: an ordinary fixed-rate bond gains more when yields fall than it loses when yields rise by the same amount. Bond analysts measure this with convexity, the second-order term after duration.
Positive convexity benefits from large moves in either direction. By Jensen’s inequality, a convex payoff is worth more on average than the payoff of the average move, and the gap grows with volatility. This is why positive convexity and long volatility describe the same exposure. The benefit is not free. Option buyers pay a premium and lose it through time decay, and bonds with more convexity offer lower yields, because the market prices convexity in. The typical cost is a small, steady loss in calm markets, and the typical return distribution is positively skewed: many small losses and a few large gains.
Positive convexity in trading strategies
In strategy and vault descriptions, “positive convexity” or “+convexity” usually means that the strategy aims to capture large market moves in either direction, without necessarily trading options. The common way to do this is trend-following: the strategy goes long in sustained rises and short in sustained declines, so its profit grows with the size of the trend. Fung and Hsieh (2001) show that trend-follower returns resemble those of lookback straddles. Dao, Nguyen, Deremble, Lempérière, Bouchaud and Potters (2016) show that the profit of a trend-following strategy equals the difference between long-term and short-term realised variance, which makes it convex in the trend-filter time scale. Man Group (2022) compares trend-following with rolling straddles on the S&P 500 and finds that straddles are more convex over one month, while trend is more convex over three months and costs less.
Strategy-based convexity differs from option convexity in three ways:
It depends on the price path. A trend-follower profits from a move that builds over its look-back window, not from a sudden jump. A one-day crash can hit it before it has turned short.
It shows only at the strategy’s own time scale. A six-month trend system can look convex on quarterly returns and linear on daily ones.
It loses in choppy, range-bound markets, where it keeps entering and exiting on false signals. This is its version of option premium.
Other ways to build convex exposure are stop losses and trailing stop losses, which cut losing trades and keep winning ones, breakout entries and anti-martingale sizing such as pyramiding, which adds to positions only while they are working.
How to check a convexity claim
Plot the strategy’s returns against the market’s returns over the same periods. A convex strategy shows a smile: its best results come in the largest up and down markets (Moskowitz, Ooi and Pedersen 2012). Treynor and Mazuy (1966) test this by adding a squared market-return term to a regression of fund returns on market returns: a positive coefficient means a convex profile. Henriksson and Merton (1981) use an option-like regression with separate up-market and down-market betas. Check the result at several return horizons, and check positive skew and the strategy’s results during past crashes. Many small wins and a few large losses point to negative convexity instead, whatever the strategy is called.
Vaults using related strategies: trend-following and options vaults.
Literature and references:
William Fung and David A. Hsieh, The Risk in Hedge Fund Strategies: Theory and Evidence from Trend Followers, Review of Financial Studies 14(2), 2001.
Tung-Lam Dao, Trung-Tu Nguyen, Cyril Deremble, Yves Lempérière, Jean-Philippe Bouchaud and Marc Potters, Tail Protection for Long Investors: Trend Convexity at Work, Journal of Investment Strategies, 2016.
Man Group, Creating Portfolio Convexity: Trend Versus Options, 2022.
Tobias J. Moskowitz, Yao Hua Ooi and Lasse Heje Pedersen, Time Series Momentum, Journal of Financial Economics 104(2), 2012.
Jack L. Treynor and Kay K. Mazuy, Can Mutual Funds Outguess the Market?, Harvard Business Review 44(4), 1966.
Roy D. Henriksson and Robert C. Merton, On Market Timing and Investment Performance. II. Statistical Procedures for Evaluating Forecasting Skills, Journal of Business 54(4), 1981.
Frank J. Fabozzi, Bond Markets, Analysis, and Strategies, Pearson (convexity of bond prices).
Nassim Nicholas Taleb, Dynamic Hedging: Managing Vanilla and Exotic Options, Wiley, 1997.
Nassim Nicholas Taleb, Antifragile: Things That Gain from Disorder, Random House, 2012.
See also