Variance Swap
A variance swap was an over-the-counter derivatives contract that allowed two counterparties to exchange the difference between the realised variance of an asset over a specified period and a pre-agreed fixed variance strike, with the buyer profiting if actual realised variance exceeded the strike and the seller profiting if it fell short.
Variance swaps provided the cleanest possible exposure to volatility as a standalone asset class. Unlike options, which had delta exposure that needed to be continuously hedged to isolate the volatility component, a variance swap had no directional (delta) exposure by design — its payoff depended purely on the difference between the variance that actually occurred over the contract period and the variance level agreed at contract initiation, multiplied by a notional variance amount called the vega notional.
The payoff formula was straightforward: (Realised Variance − Variance Strike) × Vega Notional. If Nifty's daily returns over the swap period produced an annualised realised variance of 400 (corresponding to 20% realised volatility) and the variance strike was 324 (corresponding to 18% implied volatility at initiation), the buyer received (400 − 324) × Vega Notional. The buyer had effectively paid the implied variance and received the realised variance — a long volatility position without directional contamination.
In India, variance swaps were not exchange-listed products available to retail participants. They existed in the OTC derivatives market and were accessible to institutional counterparties — foreign portfolio investors, large domestic financial institutions, and qualified entities meeting the requirements of SEBI's OTC derivatives framework and RBI's regulatory perimeter for forex and interest rate derivatives. NSE and BSE had explored exchange-listed volatility derivatives, with NSE having previously explored Nifty volatility futures, but a standardised variance swap on Indian exchanges was not an active product in the mainstream retail segment.
The replication of a variance swap through a portfolio of options was a theoretical result that underpinned the product's valuation and India VIX's methodology. India VIX was calculated using a variant of the fair-value formula for a variance swap — specifically, the VIX methodology aggregated a weighted portfolio of OTM Nifty call and put prices across multiple strikes to produce the model-free implied variance, which was then square-rooted to express as an implied volatility number. In this sense, understanding variance swaps was foundational to understanding why India VIX was computed the way it was.
Convexity was the key difference between variance swaps and volatility swaps (which paid on realised volatility rather than realised variance). Because variance was the square of volatility, a variance swap buyer benefited disproportionately from very high-volatility periods — large positive returns on variance multiplied faster than the linear volatility equivalent. This convexity made variance swaps more expensive to buy (higher fair value) than the equivalent volatility swap, a relationship known as the convexity adjustment.