SVNexus · Contagion
Cross-asset contagion
Also called: financial contagion, contagion analysis, shock transmission
Cross-asset contagion is a rise in linkages between markets after a shock, beyond normal interdependence, so losses spread further than usual.
Definition
Contagion is the spread of a shock from one market, asset class or region to others with a strength that exceeds normal interdependence. Forbes and Rigobon (2002) define it as a significant increase in cross-market linkages after a shock; links that are always strong are interdependence, not contagion.
Cross-asset contagion analysis looks for those breaks in linkage across equities, rates, credit, currencies, commodities and crypto.
Intuition
When a large market falls, forced selling, margin calls, funding stress and shared investor bases can drag down assets with little fundamental connection. Diversification that worked in calm periods then fails when it is most needed.
Separating contagion from normal co-movement matters for risk: if links always tighten in stress, portfolio risk models fitted on calm data understate the real exposure.
How it's calculated
A known pitfall is that correlation rises mechanically when the source market's volatility rises, even with unchanged underlying links. Forbes and Rigobon proposed an adjustment:
ρ* = ρ / √[ 1 + δ · (1 − ρ²) ]
where ρ is the correlation measured in the high-volatility period and δ is the relative increase in the source market's variance (high-period variance divided by low-period variance, minus one). Contagion is indicated if ρ* is significantly higher than the calm-period correlation.
Other tools include tail dependence and copulas for joint extremes, DCC models for time-varying correlation, Diebold–Yilmaz connectedness for system-wide transmission, and network measures for how shocks propagate across many assets.
Example: suppose the correlation between two markets is 0.3 in calm periods and 0.5 in a crisis during which the source market's variance quadruples (δ = 3). The adjusted correlation is 0.5 / √(1 + 3 × 0.75) = 0.5 / √3.25 ≈ 0.28, close to the calm level: the rise is explained by higher volatility rather than by stronger linkage.
How to read it
Compare linkage measures across calm and stressed regimes, adjusting for volatility. A sharp rise in total connectedness, tail dependence or correlation, beyond the adjustment, signals contagion rather than normal co-movement.
Identify transmitters and receivers: the same aggregate rise can reflect one market exporting stress or a common shock hitting all.
Common mistakes
- Calling any rise in correlation during a crisis contagion without the volatility adjustment.
- Choosing crisis dates after seeing the data.
- Relying on correlation alone and missing joint tail risk.
- Ignoring common shocks that hit all markets at once.
- Forgetting that the adjustment assumes no omitted common shocks and no feedback between the markets.
In Saqrid
SVNexus shows an animated force-directed network of how assets in a universe move together and spill over, and lets you analyze a chosen group of assets in isolation.
It measures total market connectedness and labels a stress regime when it runs well above its own history.
Open SVNexus.
Go deeper
Saqrid Institute lesson: Dependence Beyond Correlation: Copulas. Lesson text is open to read without an account.