What it is
Implied volatility (IV) is the level of volatility that, when plugged into an option pricing model (typically Black-Scholes), produces the observed market price of an option. Unlike historical volatility — which is calculated from past price movements — implied volatility is forward-looking: it is the market's aggregate expectation of how much an asset will move over the life of the option.
IV is quoted as an annualized percentage. An IV of 20% implies the market expects the asset to move approximately 20% in either direction over the next year (more precisely, that annual log returns have a standard deviation of 20%).
What IV is and is not
Implied volatility is not a forecast. It is the price of hedging. When you buy an option, you pay the implied volatility; when you sell an option, you collect it. The difference between implied volatility (what you pay) and realized volatility (what actually happens) is the option seller's profit or loss.
Historically, implied volatility has systematically overstated realized volatility by approximately 2-4 percentage points on average for equity index options. This is why volatility-selling strategies — systematically writing calls and puts — have positive expected value over long periods. The premium compensates option sellers for providing insurance when it is most needed.
The VIX is the most famous measure of implied volatility: it measures the 30-day implied volatility of S&P 500 options and serves as the market's "fear gauge." When the VIX is elevated, options are expensive; when it is low, they are cheap.
The volatility smile and skew
In a world perfectly described by Black-Scholes, all options on the same underlying with the same expiration but different strike prices would imply the same volatility. In reality, they do not.
Equity index options exhibit volatility skew: out-of-the-money put options (which protect against crashes) trade at higher implied volatility than at-the-money options, which trade at higher IV than out-of-the-money calls. The smirk shape reflects the market's demand for downside protection — it is pricing in tail risk that the normal distribution would assign near-zero probability.
This skew is directly informative about the market's fear of specific scenarios. A steep skew says the market views downside crashes as far more likely (or more painful to be unhedged against) than upside surges of the same magnitude.
Why IV spikes in crises
When volatility actually materializes — markets sell off sharply, correlations spike, uncertainty increases — implied volatility rises dramatically. During Lehman Brothers' collapse in September 2008, the VIX reached 80. During the initial COVID crash in March 2020, it reached 85. Normal levels are 12-20.
The spike has a mechanical explanation: as markets fall, existing option buyers who hedged get paid and sell their options (reducing demand), while new buyers desperate for insurance drive up prices. The spike also has a behavioral explanation: fear amplifies price discovery, and option markets are where fear is priced.
One thing most people get wrong
Implied volatility is not a forecast — it is the price of hedging. It systematically overstates realized volatility on average because option buyers are paying for insurance, not just for probability. The option seller who consistently collects IV premium and hedges dynamically earns the variance risk premium — the difference between what the market fears and what actually happens.
But this strategy is asymmetric. The variance risk premium is collected gradually in small amounts and given back suddenly in large amounts. A vol-selling strategy that earns 2% per month for 20 months can give back a year's profit in one week when the underlying crashes. The positive expected value of selling implied volatility comes attached to catastrophic tail risk. This is why the correct mental model for IV is not "expensive or cheap relative to realized vol" but "what insurance are you providing, and can you bear the cost if you have to pay out?"