How is implied correlation extracted from dispersion?

Index variance equals the weighted sum of component variances plus a correlation term. Inverting the equation given listed implied vols on the index and its components gives the implied average correlation.

Why does dispersion matter for traders?

Dispersion trading - long single-name vol against short index vol - profits when realized correlations fall below implied correlations. The trade is structurally short correlation risk.

When does dispersion widen?

Dispersion widens (implied correlation drops) during stock-specific catalyst windows like earnings season, when single-name vol bids up faster than index vol. It compresses during macro-driven sell-offs when correlations spike to one.

What are the risks of dispersion trading?

Correlation crashes during stress (everything moves together) wipe out long-dispersion positions. Margin requirements are non-trivial because the multi-leg structure carries gamma and vega across many names.

What Is Dispersion in Options?

Q: What is dispersion in options?

Dispersion is the gap between index implied volatility and the weighted average of index-component implied volatilities. Index vol is dampened by component correlation; dispersion measures the implied correlation embedded in that gap.

Last reviewed: May 17, 2026 by Options Analysis Suite Research.

Dispersion in options is the gap between index implied volatility and the weighted average of index-component implied volatilities. Index vol is structurally lower than component vol because correlation between components is less than 1; the dispersion gap measures the implied correlation embedded in that difference. Dispersion trades exploit this gap.

What Is Dispersion?

For an equity index (SPX, NDX), the index vol is:

vol_index^2 = sum_i w_i^2 * vol_i^2 + 2 * sum_(i<j) w_i * w_j * rho_ij * vol_i * vol_j

where w_i are component weights, vol_i are component vols, and rho_ij are pairwise correlations. If all correlations were 1, the formula reduces to the weighted average of component vols. With correlations less than 1, index vol is dampened. The gap between weighted-average component vol and index vol is the dispersion margin: it is the priced average pairwise correlation.

Dispersion trades go long single-name vol (buy strangles or variance swaps on the components) and short index vol (sell strangles or variance swaps on the index). The trade is structurally short correlation: it profits when realized correlations are below implied correlations.

Why Does Dispersion Matter?

Structural short-correlation alpha. Empirically, equity-index implied correlation has averaged above realized correlation (the correlation-risk premium documented by Driessen-Maenhout-Vilkov 2009). Selling implied correlation via dispersion has historically generated positive carry.
Diversification stress test. When dispersion compresses (implied correlation rises), the market is pricing reduced diversification benefit. This is itself a stress signal.
Single-name vs index vol decomposition. Dispersion is the structural bridge between individual-stock options and index options. Calibrating models on one and using them for the other requires dispersion adjustment.

Worked Example

SPX with 30-day implied vol = 14%, dollar-weighted average component implied vol = 22%. The gap is large because correlation is far below 1. Implied correlation extracted:

rho_implied ≈ vol_index^2 / weighted_avg_vol^2 = 0.14^2 / 0.22^2 ≈ 0.40

Implied 30-day SPX correlation is 0.40. Selling index vol and buying component vol bets that realized correlation will fall below 0.40; if realized is 0.30, the dispersion P&L is positive. If realized correlation spikes (e.g., 0.65 in a market crash where everything sells off together), the dispersion trade loses heavily.

Dispersion Trade Structure

Variance-swap form. Long N variance swaps on N component names, short 1 variance swap on the index (with notional and weights matching). The cleanest dispersion trade.
Strangle form. Long strangles on each component, short a basket-equivalent strangle on the index. Operationally simpler but less precise correlation exposure.
Top-name approximation. Index dispersion is dominated by the top 50-100 weighted components; trading those plus a hedge on the long tail captures most of the structural exposure.

How Do Models Treat Dispersion?

Black-Scholes: single-asset model. Dispersion arises across instruments rather than within a single BSM calibration. Cross-asset BSM is not a complete framework.
Correlation-augmented stochastic-vol models. Multi-asset Heston, multi-asset SABR with explicit correlation structures. Used by institutional desks for dispersion pricing and risk management.
Local-correlation models. Inspired by local-vol but applied to correlation - implied correlation as a function of spot levels. Limited adoption; mathematically demanding.
Static decomposition (Bossu 2007). Treats dispersion as a static portfolio of variance swaps. Pricing follows from variance-swap valuation; correlation enters via the index-vs-components variance decomposition.

The Correlation Risk Premium

Driessen-Maenhout-Vilkov (2009) documented that index-implied correlation has consistently exceeded realized correlation across multi-decade samples. The premium compensates dispersion sellers (long index vol, short component vol) for bearing correlation-spike risk: implied correlation jumps to 1 in market crashes, generating large losses on dispersion-short positions exactly when the rest of the portfolio is also losing money. The structural similarity to variance risk premium is no coincidence: both compensate sellers for correlated tail risk.

Limitations and Caveats

Operational complexity. Trading variance swaps on 50+ single-name components plus an index hedge has significant operational overhead. Proxy-trade approximations introduce tracking error.
Crash risk. Dispersion trades lose worst exactly when markets crash. Stress-test sizing accordingly.
Single-name liquidity. Component variance swaps may not be tradable on the long tail of index members. Strangle proxies or index-component-list focus partially solves this.
Component-set drift. Index rebalances change the component set; dispersion positions need re-weighting.

Related Concepts

Volatility Smile · Volatility Skew · Variance Risk Premium · VIX · Realized Volatility · Heston Model · Pricing Model Landscape

References & Further Reading

Driessen, J., Maenhout, P. J. and Vilkov, G. (2009). "The Price of Correlation Risk: Evidence from Equity Options." Journal of Finance, 64(3), 1377-1406. Empirical evidence for the correlation risk premium.
Bossu, S. (2007). "A New Approach for Modelling and Pricing Correlation Swaps." Working paper. Practitioner-oriented dispersion pricing framework.
Carr, P. and Madan, D. (1998). "Towards a Theory of Volatility Trading." In R. Jarrow (Ed.), Volatility: New Estimation Techniques for Pricing Derivatives, Risk Books, 417-427. Static replication of variance via option payoffs - the foundation for variance-swap-based dispersion.
Jacquier, A. and Slaoui, S. (2010). "Variance Dispersion and Correlation Swaps." Working paper. Mathematical structure of dispersion and the correlation-swap formulation.

View live VRP and dispersion screeners ->

This page is part of the Pricing Model Landscape and the canonical reference set on options market structure. Browse all documentation.

As of the latest snapshot, SPY has an ATM implied volatility of 12.8%, IV rank 12% (percentile 17%); 20-day realized vol 9.8%. 25-delta skew is +3.6%, meaning OTM puts trade richer than OTM calls. The IV here is the input that pricing-model walkthroughs (Black-Scholes, Heston, SABR, local-vol) take as their starting point: each model decomposes the same observed quote into different latent dynamics (constant vol, stochastic vol, surface-fitted vol, etc.) which is why two models can agree on price but disagree on Greeks and on how vol will evolve.