Vanna and Charm: The Second-Order Flows That Move Markets Quietly

Dealer gamma is now widely discussed, which means its informational edge has decayed. The second-order sensitivities behind it are still poorly understood, and they generate flows that are larger and more predictable than gamma alone.

Vanna measures how an option delta changes when implied volatility changes. When volatility falls, the deltas of out-of-the-money options shrink, and dealers who hedged those deltas must unwind part of their hedge. A broad decline in implied volatility therefore produces mechanical buying of the underlying, entirely independent of anyone view about value. The reverse holds when volatility rises.

Charm measures how delta changes with the simple passage of time. As expiry approaches, out-of-the-money options lose delta and in-the-money options gain it. Dealers hedging a large book must rebalance continuously as this decay proceeds, and the flow concentrates in the days immediately before large expiries.

Why the two compound around expiries

The reason expiry weeks behave differently is that vanna and charm flows frequently point the same direction at the same time. A calm week ahead of a large expiry produces falling implied volatility, which generates vanna buying, at the same time that time decay generates charm buying. The combination can lift an index steadily on no news whatsoever, which is precisely the price action that generates the most confident and least accurate fundamental explanations.

After the expiry, the options that generated those flows cease to exist. The mechanical support disappears overnight, and the market becomes free to move on other drivers. This is the structural basis for the observation that the days following major expiries behave differently from the days preceding them.

What this framework cannot tell you

These flows explain mechanics, not direction over any horizon longer than the flow itself. Vanna and charm describe why an index drifted upward through a quiet expiry week. They say nothing about whether the index is expensive, whether earnings will disappoint, or where the level should be in six months. Treating a flow framework as a valuation framework is the most common way it gets misapplied.

Estimating the Flows From Public Data

Reconstructing dealer exposure requires assumptions that cannot be verified directly. The standard approach infers dealer positioning from options open interest by assuming that customers are net buyers of puts and net sellers of calls, which makes dealers net short puts and net long calls. The assumption is reasonable in aggregate and wrong in specific cases, particularly around large structured product issuance that flips the customary direction.

The practical consequence is that these estimates are best used as a description of the likely mechanical environment rather than as a precise quantity. A framework that says the environment is currently one where declining volatility should generate buying pressure is useful. A framework that claims to know the exact dollar amount of that pressure is claiming a precision the underlying data cannot support.

Quantifying the flow around a large expiry

The size of a vanna or charm flow depends on the notional exposure sitting in the options concerned. For a monthly index expiry, that notional runs into the hundreds of billions of dollars in aggregate. A one percentage point decline in implied volatility during the week of expiry can generate underlying purchases in the low tens of billions purely from vanna hedging, and time decay through that same week can produce comparable buying from the charm side. Neither is an opinion. Both are the mechanical consequence of dealer books being hedged at all.

The direction reverses immediately after expiry as the options generating the flows cease to exist. This is the structural basis for the well-documented pattern where indices drift higher into monthly expiries in calm markets and behave differently in the days that follow.

Where the framework breaks

The framework assumes dealers are net short customer options and therefore need to hedge in the same direction as the flow implies. This assumption holds in aggregate for equity index options and fails in specific cases. Large structured product issuance can flip dealer positioning, and periods of heavy institutional call buying against corporate hedging programmes can reverse the customary flow entirely.

The honest way to use the framework is as a description of the mechanical environment when the assumption holds, alongside checks on options open interest and positioning surveys that would confirm the assumption is holding in the current period. Any flow model that produces a directional forecast without those checks is claiming to know something the underlying data cannot support.