Correlations Between Major Asset Classes

7 min read

Markets rarely move in isolation. Understanding how major asset classes tend to relate to one another - and why those relationships aren't fixed - adds useful context to any single-instrument analysis.

These relationships are best treated as tendencies shaped by whatever is currently driving markets, not permanent laws - which is exactly why they're worth understanding rather than simply memorising.

Commonly observed relationships

The US dollar and gold have historically tended to move in opposite directions, since gold is dollar-priced internationally. Equities and government bond yields often move together during growth-driven rallies, since both can rise on optimism about the economy, while yields and equities can diverge sharply when inflation, not growth, is what's pushing yields higher.

Commodity-linked currencies, such as those of major commodity exporters, often show a positive relationship with the commodities they export, since a stronger commodity price improves that country's terms of trade.

• Dollar-gold: Historically tend to move in opposite directions, though not in every period.
• Equities and yields: Can move together in growth-driven rallies, or diverge when inflation drives yields.
• Commodity currencies: Often positively linked to the commodities the exporting country sells.

Why correlations are not constant

A relationship that holds for months can break down when the underlying driver changes - for example, a stable dollar-gold relationship can shift if a geopolitical shock pushes both higher at once as investors seek safety in multiple places simultaneously.

Treating correlations as a general tendency, not a fixed rule, avoids being caught off guard when one breaks down - the goal is to understand why a relationship usually holds, so you can recognise when the reason no longer applies.

• Regime dependent: Correlations reflect the current driver of markets, which can change over time.
• Breakdown risk: A usually reliable relationship can invert during an unusual event.
• Understand the why: Knowing why a correlation holds helps spot when it's no longer valid.

Using correlation as context, not a signal on its own

Cross-asset relationships are most useful as a sanity check on a single-instrument view - if your analysis expects the dollar to weaken but gold isn't responding as the usual relationship would suggest, that's worth investigating rather than ignoring.

It's a lens for asking better questions, not a standalone trading signal - correlation alone rarely justifies a position, but it can flag when a view is missing something.

• Sanity check: Use correlation to test whether a single-instrument view holds together.
• A lens, not a signal: Correlation informs the questions to ask rather than triggering a trade on its own.
• Worth investigating: An expected correlation failing to show up is a prompt to dig deeper.