Diversification is one of the first lessons in finance, and one of the easiest to lose sight of in practice. By not putting all of their eggs in one basket, investors can reduce their reliance on any single market outcome, reducing concentration risk while enhancing risk-adjusted performance through a wider set of return drivers. CTAs and trend following managers have long argued that trading more markets delivers more of that benefit. But is this fact or fiction?
Measure of Portfolio Diversification
Trend following is built on the belief that trends can emerge in any market, at any time. Because such trends are episodic and difficult to anticipate, a well-constructed portfolio is designed to be broad and ‘ready’, rather than trying to predict which single market will trend next. This principle underpins why diversified trend programmes trade large multi-asset universes across sectors such as agriculturals, bonds, currencies, energies, interest rates, metals, and stock indices.
A useful way to measure diversification is to ask: how many independent ‘bets’ does the portfolio take? The Portfolio Diversification Index (PDI) does exactly that, measuring how evenly risk is spread across the possible factors. If a portfolio is highly correlated and one factor dominates, the PDI ≈ 1, whereas if risk is evenly spread across N truly independent components, PDI ≈ N. Every real portfolio sits somewhere in between, and where it sits is far more informative than a headline market count: breadth counts, but only when in the form of diversifying, largely uncorrelated markets.
This distinction matters most in precisely the environments investors worry about: when a small number of narratives dominate headlines. During these environments, effects rarely stay contained within the sector where the story began. They transmit outward through supply chains, input costs, capital flows and funding conditions, pulling apparently unrelated markets into the same trade. Diversification measured on a quiet day can disappear on a loud one. Breadth answers this: when a narrative ripples across industries, a broad programme is positioned to capture the resulting trends wherever they surface, rather than only at the epicentre.
To test whether expanding the market universe enhances diversification, the PDI was measured as markets were added one at a time. However, ordering matters. Adding another stock index to a stock index-heavy subset may provide little incremental diversification, whereas adding an idiosyncratic agricultural market may add much more. To address this, the exercise was repeated 1,000 times using different random orderings (grey), then averaging the result (black). Figure 2 is consistent: diversification increases as markets are added.
Figure 2: The effect of the number of markets on the PDI (averaged across 2021-2025)
Why Does This Matter Now?
It is therefore a fact: adding more uncorrelated markets does increase diversification. In many strategies, diversification is whatever falls out of the process, visible only after the fact and liable to change without notice. In a diversified trend following programme it is the reverse: diversification is a structural input, engineered before a single position is taken – not something to be monitored nervously each quarter, but something to select for at the outset. That breadth is what converts into performance. Combining return sources and low correlation improves expected return per unit of risk. What limits a programme’s market count is liquidity and the cost of accessing markets well, not any exhaustion of the diversification benefit.
Insight Paper Disclosures
PAST PERFORMANCE IS NOT INDICATIVE OF FUTURE RESULTS. For Illustrative purposes only. Important performance information, investment risks and legal and regulatory disclosures applicable to Aspect Capital programmes can be found [here].
