Skip to content

Volatility Carry: Conditional Return for Bearing Convexity Risk

Separate volatility carry from Theta, compare implied and realized variance correctly, and audit path, surface, hedge, cost, and tail-risk exposure.

Updated

Educational information only; not individualized investment, legal, or tax advice. Options involve risk and are not suitable for all investors; some written positions can lose more than the amount initially invested.

Direct answer

Volatility carry is the conditional P/L tendency from holding and, where relevant, hedging a volatility-sensitive position through time. A short-volatility position may earn positive carry when the volatility priced into options exceeds what is subsequently realized and monetized, and when decay or hedge gains exceed repricing and costs. Option prices can also place more weight on losses in stressed states than their physical frequency alone would imply. A long-volatility position often pays carry to own convexity, jump exposure, or insurance. Neither side receives a guaranteed yield.

Carry is not cash interest, premium received, or Theta alone. Theta is a local model derivative with other inputs held fixed. Actual P/L also depends on the realized path, Gamma, changing Delta hedges, implied-volatility level, skew and term structure, rates, dividends, borrow, execution, fees, funding, margin, exercise, assignment, and settlement.

Scope: this page uses ordinary U.S. exchange-listed, OCC-issued equity, ETF, and index options in a self-directed brokerage account, with public information checked on 2026-08-22. The decomposition is a local valuation model, not a quote, forecast, or backtest. Contract style, multiplier, deliverable, settlement, trading hours, account approval, margin, tax, and legal treatment vary by product, broker, account, date, and jurisdiction. Futures, OTC, exotic, employee, and crypto options are outside scope. Current contract documents, executable quotes, broker terms, and applicable law control.

What creates or consumes carry

For a signed, locally Delta-hedged option position, a diagnostic approximation is:

ΔV ≈ Theta_pos × Δt + 0.5 × Gamma_pos × (ΔS)^2 + Vega_pos × ΔIV + residual

residual can contain hedge timing and fills, skew and term-structure changes, rates, dividends, borrow, funding, fees, and model error. Greek signs and units must be position-scaled consistently; a vendor may quote Theta per calendar day or per year and Vega per 1 volatility point or per 1.00 decimal-volatility change. Full option marks plus the actual hedge-and-cash ledger determine P/L. Adding the Greek attribution to that result would double count it.

For ordinary long vanilla options, Gamma is usually positive and Theta negative; comparable short positions reverse those signs. A sufficiently large or favorable realized path can let long Gamma exceed time decay, while a quiet path can favor the short side. Discrete hedging, gaps, and surface repricing break the continuous, frozen-Greek intuition.

Compare matched horizons in variance units. If annualized implied volatility is 25% and a documented forecast of annualized realized volatility is 20%, then:

0.25^2 − 0.20^2 = 0.0225

This is an annualized variance difference, not a 5% return or a dollar profit. A single ATM IV is not a complete option-implied variance measure, and historical volatility is not the future realized path. Research finds that option-implied variance has often exceeded subsequent or expected realized variance in U.S. equity samples, but the sign, size, estimator, horizon, and investability are sample-dependent. The spread can compensate sellers for jumps, crash insurance, volatility and correlation shocks, illiquidity, and constrained capital rather than prove mispricing.

Two controlled examples

Assume an index level of 100 and a one-month ATM straddle premium of 6 index points. If both options are European-style, share the strike and expiry, and are held unhedged to cash-settled expiration, the short-straddle result before multiplier, fees, and funding is:

short expiration P/L = 6 − |final index − 100|

Final index Absolute move Short P/L
94 6 0
103 3 +3
106 6 0
112 12 −6

The breakevens are 94 and 106, but they apply only at expiration under these assumptions. Before expiration, IV and remaining time can make the close-out loss large even inside that range. The initial 6 points are consideration for an open obligation, not earned income, and the upside loss of the short Call remains uncapped.

Now isolate a one-day frozen-Greek estimate for one long option. Suppose per-share Theta is −$0.08/calendar day, Gamma is 0.04 Delta/$/share, initial Delta is hedged, and IV is unchanged. For a $1 move:

Gamma term = 0.5 × 0.04 × $1^2 = +$0.02/share

estimated P/L = +$0.02 − $0.08 = −$0.06/share

For a $3 move, the Gamma term is +$0.18/share, so estimated P/L is +$0.10/share. With a 100 multiplier, those figures are −$6 and +$10 per contract before hedge slippage, Bid/Ask, fees, funding, surface changes, and higher-order effects. The comparable short position has the opposite local estimate, not an opposite guaranteed outcome.

Measurement and risk controls

  1. Identify every series, signed quantity, multiplier, deliverable, exercise style, settlement method, currency, and account approval before calculating exposure.
  2. Record synchronized executable option and underlying quotes, rates, dividends, borrow, full volatility-surface coordinates, expected events, and exact timestamps.
  3. Declare the realized-variance return formula, sampling frequency, annualization, horizon, overnight treatment, and the physical forecast available at entry; never substitute a later realization without relabeling the test.
  4. State the hedge rule before observing the path. More frequent hedging can increase friction; less frequent hedging leaves larger directional and gap exposure.
  5. Reconcile option marks, actual hedge inventory, cash, funding, dividends, borrow, fees, and residuals. Do not count modeled Gamma gains on top of actual hedge cash flows.
  6. Full-reprice joint spot gaps, IV jumps or crushes, skew steepening, term twists, correlation shocks, Bid/Ask widening, reduced size, margin increases, and forced liquidation.
  7. Size against stress loss and available liquidity, not premium targets or historical win rate. Correlated short-volatility positions can fail together.
  8. Plan closing, exercise, assignment, cash or physical settlement, residual stock, and tax review. American-style shorts can be assigned before expiration, and brokers may liquidate when requirements are unmet.

Common misconceptions

  • “Premium received is earned income.” It is paired with a continuing liability and the cost of closing or satisfying it.
  • “Volatility carry equals Theta.” Theta freezes other model inputs; realized path, Gamma, Vega, hedges, surface, financing, and costs remain.
  • “IV above historical volatility means sell.” The windows may not match, history is not a physical forecast, and event or tail risk may justify the price.
  • “IV minus RV is the return.” Formal comparisons use variance and consistent estimators; an option strategy adds strike, path, hedge, and execution exposure.
  • “Delta-neutral removes crash risk.” Delta changes, gaps bypass hedges, and IV, skew, correlation, liquidity, and margin can move together.
  • “A high win rate means low risk.” Short-volatility results can combine frequent small gains with rare losses larger than accumulated carry.
  • “Long volatility is irrational because carry is negative.” Convexity and insurance can be valuable in states where capital, liquidity, and risk tolerance are most constrained.

Authoritative sources

Navigation

Search the wiki...