Dispersion: When the Index Is Calmer Than Its Parts

An index is a portfolio, and the volatility of a portfolio depends on the volatility of its holdings and on how those holdings move relative to one another. Two markets can produce identical index volatility through entirely different mechanisms: one where every constituent moves modestly together, and one where constituents move violently in opposing directions and largely cancel out.

The difference between index implied volatility and the average implied volatility of its constituents is therefore a measure of implied correlation. When that gap is wide, the market expects constituents to move independently. When it is narrow, the market expects them to move as one, which is the condition that removes the benefit of holding many names instead of a few.

What the measure reveals about regime

Implied correlation compresses toward its upper bound during systemic stress, which is the options market pricing the same phenomenon described in the diversification note on this site: correlations converge exactly when their divergence would have been most valuable. Watching implied correlation rather than realised correlation gives an earlier reading, because it reflects what participants are paying to position for rather than what has already happened.

Low implied correlation describes a market being driven by idiosyncratic factors, where individual company results dominate and macro is quiet. High implied correlation describes a market driven by a single common factor, usually liquidity, rates, or a systemic risk event. The transition between the two states is often more informative than either state in isolation, because it marks the point where stock selection stops working and factor exposure starts dominating returns.

Why this matters to an allocator who never trades options

An investor with no intention of trading a dispersion structure still benefits from reading it, because it answers a question that directly affects portfolio construction: is diversification currently doing anything. In a high implied correlation environment, adding names to an equity sleeve reduces risk far less than the position count suggests, and the defensive allocation has to come from a genuinely different exposure rather than from more of the same one.

The Structural Bid That Keeps Index Volatility Rich

Index implied volatility has historically traded rich relative to realised, while single-stock implied volatility has traded closer to fair, and the persistent gap has a structural explanation. Demand for index-level protection comes from institutions hedging whole portfolios, which is a large and consistent bid. Supply of single-stock volatility comes from systematic call overwriting programmes, which is also large and consistent but points the other way.

The result is a durable imbalance rather than a mispricing that arbitrage should close. Anyone treating the dispersion spread as free money is assuming the flows that create it will persist unchanged, which is an assumption about institutional behaviour rather than about mathematics, and institutional behaviour changes when mandates change.

The tell in the local market

The ASX 200 shows a starker version of this dynamic than the S&P 500 because it is dominated by two sectors, financials and resources, which each carry substantial correlated exposure to a small number of macro drivers. A rate shock hits the big four banks together. A China growth scare hits the iron ore majors together. In both cases the observed implied correlation on the local index compresses toward its upper bound quickly, and stock selection within those sectors delivers very little for the risk taken.

The practical implication is that the diversification benefit inside a domestic equity sleeve is smaller than the position count suggests, and it shrinks precisely when it would be most valuable. Adding names inside financials or inside resources is largely additional exposure to the same factor. Adding exposure across a genuinely different driver, whether that means offshore equities, a defensive allocation, or a different asset class entirely, is where diversification actually happens.

Watching implied correlation rather than realised

Realised correlation is calculated after the fact, from returns that have already happened. Implied correlation is derived from current options prices and describes what the market is currently paying to position for. The two diverge at turning points, which is exactly the moment where the divergence matters. A realised measure will confirm a correlation shift only after the shift has damaged portfolios that were sized on the previous number. An implied measure will register the same shift as market participants pay up for hedges, which typically leads the realised move by weeks.