Implied Volatility

Implied volatility, or IV, is the annualized volatility input that makes an option-pricing model reproduce a selected market premium. It describes the variability priced into a contract under that model’s assumptions. Historical volatility is different: it is calculated from returns already observed over a selected past period. IV is neither a directional forecast nor a guarantee that subsequent realized volatility will match the quoted percentage.

Inferring volatility from a premium

An option model normally calculates value from the underlying price, strike, remaining term, volatility, interest rate, dividends, and exercise terms. To calculate IV, the other inputs are held fixed while volatility is varied until the model matches the chosen premium. If $V$ denotes the modeled option value, the implied volatility $\sigma_{\mathrm{IV}}$ satisfies:

This is an inverse-pricing equation. It does not determine a unique result independently of the model: the selected market price, dividend assumptions, and exercise treatment are part of the calculation.

For example, consider a European call with underlying price $750, strike $775, and 45 days remaining. With a 4% interest rate, a 1% continuous dividend yield, and $T=45/365.25$ years, the Black–Scholes–Merton model gives:

A premium near $12.03 therefore implies approximately 20% volatility under these assumptions. The table varies volatility alone; it is not a forecast of how the option will trade after a change in the underlying or the passage of time.

Using the bid, midpoint, or ask can produce different IVs because those prices differ. A solution also requires a premium within the model’s attainable price range. When the option responds weakly to volatility, a small pricing, rounding, or timing difference can produce a large change in calculated IV. Such instability is especially relevant when interpreting a quoted percentage without considering the premium from which it was inferred.

Interpreting the annualized percentage

Annualization places volatility estimates for different horizons on a common scale. In a constant-volatility model, annual volatility $\sigma$ over a horizon of $T$ years gives a standard deviation of log returns equal to:

The square-root factor converts the annual measure to the shorter horizon under the model’s assumptions. For 20% annual volatility over 45 days:

The resulting scale is about 7.02%. For relatively small returns, multiplying by the initial underlying price gives an approximate dollar standard-deviation scale. At $750, this is about $52.65. It describes dispersion rather than a required move, a directional target, or a boundary outside which outcomes are impossible. The log-return model and the dollar approximation should also be distinguished from a guaranteed real-world probability interval.

Events and changes in IV

A scheduled announcement can concentrate uncertainty into part of an option’s remaining term. Demand for exposure to a large move, or for protection against one, can increase premiums and the IV inferred from them. After the event, a reduction in unresolved uncertainty can cause IV to fall sharply, commonly described as an IV crush.

For a purchased call, a favorable rise in the underlying may increase intrinsic value while falling IV reduces time value. The call loses market value if the latter effect is larger and gains if the underlying move more than offsets it. The equivalent interaction applies to a purchased put after a decline. For a short option, lower IV generally reduces repurchase cost, but an adverse underlying move can increase the obligation by a larger amount.

Skew and term structure

Different strikes of the same expiration commonly imply different volatilities. A volatility smile has higher IV at strikes on both sides of its central region, while a skew is more pronounced on one side. Equity-index options often exhibit higher IV at lower strikes, reflecting the pricing of sharp declines and demand for downside protection. These patterns are features of relative option prices, not a requirement that all contracts share one volatility input.

Differences across expirations form the volatility term structure. A near-term announcement can account for a larger fraction of a short-dated option’s remaining uncertainty than of a longer-dated option’s uncertainty. It can consequently have a larger effect on the shorter contract’s annualized IV. Volatilities across maturities need not move together because the contracts cover different remaining horizons.

European calls and puts with the same strike and expiration are linked by put-call parity, which relates their prices to the underlying and financing terms. Prices that satisfy this relationship imply the same IV when calculated with the same consistent model inputs. Displayed differences can arise from bid-ask spreads, stale quotations, differing dividend assumptions, or early-exercise features. The European parity relationship should not be imposed unchanged on American options.

Vega and position exposure

Vega measures the local change in option value for a one-percentage-point change in IV. For an option at 20% IV with vega of $1.04 per share, an increase to 21% adds approximately $1.04 per share, or $104 for one purchased 100-share contract, with other inputs fixed. The corresponding short position has the opposite change in value.

This estimate uses the sensitivity at the initial inputs. Vega itself changes with volatility, time, and moneyness, so a large volatility move is better evaluated by repricing the option at the new inputs. The approximation does not include simultaneous changes in the underlying or elapsed time unless those effects are calculated separately.

In a spread, long and short vegas can partially offset the effect of a common IV change. That net sensitivity does not fully describe changes in skew or term structure. When one leg’s IV rises more than another’s, the position is responding to distinct volatility changes rather than to a uniform shift.