Case 002Options and volatility tradingHard
An index's 25-delta put trades at 28% implied volatility, the 25-delta call at 20% and at-the-money at 23%. Price a zero-cost risk reversal, explain what the skew is paying for, and say who is on the other side.
1The situation
A client of Hemadri Volatility wants three-month upside on an equity index at spot 20,000 without paying premium, and is willing to give up downside protection to get it. The desk's screen shows the three-month smile: the 25-delta put at 28% implied volatility, at-the-money at 23%, and the 25-delta call at 20%.
Use Black-Scholes with zero rates and dividends for simplicity. Solving for the 25-delta strikes gives the put at about 18,377 and the call at about 21,503. Interpolate the smile between the three quotes where you need another strike.
2Your task
Structure the zero-cost risk reversal, say what the skew is paying for, and who takes the other side of the trade.
Quick check
Selling the 25-delta put and buying the 25-delta call: what is the net premium?
Worked solution
Try it on paper, then open one step at a time.
30-second answerThe answer to give first
Sell the 18,377 put and buy the 20,986 call for zero premium. The 25-delta put is worth about 449 points at 28%, and a 25-delta call only 284 at 20%, so the put pays for a call about 34-delta. The skew is the price of crash insurance, bought by hedgers of long equity books. The client takes the other side of that insurance and loses heavily in a crash.
Step 1What does the skew say before you price anything?
Read the three quotes as prices of insurance. A put five percent out of the money protects against a fall; a call five percent out of the money pays on a rally. The market charges 28% volatility for the downside and 20% for the upside, so protection against a fall costs far more than a ticket to a rise of similar size. A risk reversalBuying an out-of-the-money call and selling an out-of-the-money put, or the reverse. Quoted as the call volatility minus the put volatility at the same delta. quoted at 25 delta is 20 minus 28, minus 8 vols: a steep downside skew.
Step 2How do you make the structure cost nothing?
Price the legs you know. The 25-delta put at 28% is worth about 449 index points. The 25-delta call at 20% is worth about 284. Selling the put and buying the same-delta call leaves a credit of about 165 points, so the client can afford a call closer to the money. Walk the call strike down the smile until its premium equals the put's: that happens at about 20,986, where the interpolated volatility is 20.9% and the call's delta is about 0.34.
| Leg | Strike | Implied vol | Premium, points | At flat 23% |
|---|---|---|---|---|
| Sell 25-delta put | 18,377 | 28.0% | 449 | 296 |
| Buy 25-delta call | 21,503 | 20.0% | 284 | 383 |
| Buy zero-cost call | 20,986 | 20.9% | 449 | |
| Credit on the 25/25 package | 165 | -87 |
Step 3What is the skew actually paying for?
Compare the put with what it would cost at the at-the-money volatility: about 296 points instead of 449. The extra 153 points are the premium for crash insurance, and they exist for two reasons: index volatility rises when prices fall, and institutions holding equities keep buying puts to protect their books. A household buys fire insurance for more than the expected cost of fires because it cannot afford the fire; pension funds and insurers pay over the odds for index puts for the same reason. Covered-call writers sell upside calls at the same time, which pushes call volatility down.
Step 4Who is on the other side, and what can go wrong?
The client is selling that insurance: short the put the hedgers want, long the call the overwriters are selling. Most of the time this earns the skew premium, which is how the upside comes free. The cost arrives all at once: a 15% fall to 17,000 loses about 1,377 index points on the sold put, with no premium cushion because none was received. Hemadri's desk, having sold the call and bought the put, is on the hedgers' side and should say plainly that the zero cost is paid for in crash risk.
Close with the limit of the model. Black-Scholes with a quadratic smile is a convention for converting quotes into prices, not a forecast of the crash probability. The fair question for the client is whether the skew overpays for crash risk over the coming three months, and that is a judgement about the market, not something the model can answer.
Where candidates lose it
The common loss is assuming a 25-delta put and a 25-delta call cost the same because their deltas match. They only do on a flat smile; with 8 vols of skew the put is worth well over one and a half times the call.
The second is calling the structure free. It costs nothing up front because the client has sold crash insurance, and the interviewer wants to hear who bought it and what the client owes if the crash comes.
What the interviewer asks next
- The skew steepens to 32% on the put overnight. What happens to the client's mark?
- How would you delta-hedge this risk reversal as the dealer, and what does vanna do to that hedge?
- Why is single-stock skew often flatter than index skew?
- Price the same structure with six months to expiry. Does the zero-cost strike move closer or further?
Company names and figures are illustrative.
