On this page
Volatility can be treated as a distinct risk exposure because an investor who sells protection against large market moves accepts a specific form of loss risk in exchange for a premium. That premium is neither a guaranteed yield nor automatically independent of equities. To evaluate it, compare implied and subsequent realised variance on the same horizon, then ask how the position behaves when both volatility and equity losses surge.
What the buyer and seller exchange
A buyer of volatility wants a payoff when realised movements or the market's pricing of future movements rise. Such exposure can help in an adverse market state, so the buyer may accept a cost for carrying it. The seller accepts the opposite side: a stream of apparently modest gains can be interrupted by a large loss when the market moves more violently than anticipated. The economic argument for a premium is compensation for bearing that difficult state, not evidence of a free return. The exact exposure depends on the instrument, its strike, maturity, rebalancing and financing.
The CAIA Association's current curriculum overview includes Volatility as a Factor Exposure within Level II Volatility and Complex Strategies. An earlier official Level II study guide sets out the conceptual distinction between long and short volatility and identifies opposite expected premium directions. The earlier guide is useful for conceptual evidence, while the current curriculum is the authority for current exam alignment. Candidates should check their own exam year's official material because CAIA Association updates its curriculum each year.
Measure the variance gap before naming a premium
Implied volatility is backed out from option prices and reflects pricing under market conditions at a particular moment. Realised volatility is calculated later from actual returns over a specified interval. A comparison becomes meaningful only if underlying asset, observation window, annualisation and variance convention match. Volatility is a standard deviation; variance is its square. Subtracting two volatility percentages is therefore not generally the same payoff as subtracting two variances. In a variance contract, the seller cares directly about the difference between a pre-agreed variance strike and variance realised over the contract's window.
Some writers call realised variance less implied variance the variance risk premium; under that sign convention the typical seller's reward appears as a negative number. Others define the gap as implied less realised variance and report a positive seller premium. The label by itself is insufficient. State the order of subtraction and whose payoff is being discussed. A candidate who knows that the seller benefits when realised variance is lower than the contracted strike can recover the correct sign even when a question uses unfamiliar terminology.
A deliberately simple worked scenario
Assume an invented variance contract has an annualised volatility strike of 20% for a matching measurement window. Squaring 0.20 gives a variance strike of 0.0400. If later realised volatility is 15%, squared realised variance is 0.0225. The seller's simple difference is 0.0175 variance units before any notional scaling, transaction costs or financing. The buyer has the opposite difference. These are chosen teaching values rather than an observed market premium. They show the mechanical source of a seller gain when the movement that occurred was smaller than the movement priced at inception.
Now change only the realised volatility in this invented example to 30%. Its squared variance is 0.0900, and the seller's difference becomes negative 0.0500. The adverse result is much larger in absolute size than the earlier positive difference. Squaring makes the payoff particularly sensitive to large realised moves. A candidate should not treat the initial 20% quotation as a cap on loss, or think that a seller can simply wait for an abstract long-run average to reassert itself. The contract settles on the realised outcome for its own period, and a forced exit or margin demand can matter before that period ends.
| Calmer outcome | 0.0400 | 0.0225 | positive 0.0175 |
| Stress outcome | 0.0400 | 0.0900 | negative 0.0500 |
Why the return may exist
The strongest intuition is insurance demand. An investor who is especially sensitive to losses in market stress may pay for an instrument that helps in that state. The party willing to supply protection needs compensation because the obligation becomes expensive precisely when cash and liquidity are scarce. The realised path in a quiet sample may make the protection look overpriced after the fact; that hindsight does not prove it was irrational when purchased. At inception, the seller does not know whether the future period will be quiet or dislocated.
Jump risk is another reason an average gap cannot be read as arbitrage. A jump may be rare in the observed window and still have a high economic cost if it occurs. Changes in correlation across assets can also amplify market-wide movement when supposedly separate risks occur together. Market makers may require compensation for warehousing option exposures that are hard to hedge continuously. These mechanisms are related, and none gives the candidate a universal numerical premium. The lesson is qualitative: an expected seller return comes with a contingent obligation whose severity is easy to hide in a short calm sample.
What makes the exposure distinct from equity beta
An equity holding participates in changes in the value of company cash flows. A short-volatility position takes the other side of a contract whose payoff depends on the scale or pricing of market moves. Those contractual drivers are conceptually different, even if both positions lose in the same stressed period. A delta-neutral option structure, for example, can reduce a first-order directional exposure at inception while retaining sensitivity to volatility, curvature and the cost of rebalancing. Neutral at inception does not mean neutral throughout the life of the position.
The practical test is to examine conditional behaviour, not only an overall correlation. Compare results in calm and stressed markets; identify whether losses arise from a rise in implied volatility, realised movement, equity direction, changes in the option surface, or the cost of keeping a hedge in place. Ask whether the return survives after accounting for broad equity exposure and whether the contract's payoff is consistent with the claimed source. CAIA Association's factor discussion emphasises looking through an asset label to its underlying return drivers and recognising that factor sensitivity can change over time.
Limits and exam mistakes
A high level of implied volatility does not by itself show that selling volatility is attractive. The future realised variance is unknown; a high strike can coincide with an even higher realised outcome. Equally, a low realised-volatility reading after the event cannot establish that an option was mispriced before the event. The appropriate comparison is between contemporaneous pricing and the later matching outcome across many periods, with attention to unusually costly tail periods. A claim that a premium is persistent should survive the full distribution, not depend on excluding the worst observations.
Do not confuse positive option vega with a complete description of long-volatility behaviour. An option's value can change for reasons other than implied volatility, including the underlying price, passage of time and changes in other contract inputs. Nor should a negative premium to holding protection be interpreted as a reason that protection is never useful: a hedge can have value because of when it pays. For the same reason, a positive expected seller premium says nothing by itself about suitability, position size or reliable diversification.
For a candidate reviewing this topic, the CAIA® programme page locates it in the wider course, and the study methodology provides a place to practise defining the payoff before interpreting a label. The testable sequence is straightforward: specify long or short exposure; put implied and realised measures on one basis; work out the payoff sign; then ask what happens when market stress changes both volatility and equity risk together.
Sources
- Explore the Curriculum | CAIA Association · caia.org · Retrieved
- CAIA Level II Study Guide September 2023 | CAIA Association · caia.org · Retrieved
- Strategic Asset Allocation: Practical Considerations for Alternative Investments | CAIA Association · caia.org · Retrieved



