Good Explanation for what causes the Volatility Smile


Not sure if this is the best place in this forum to post this and if this question has been asked previously on here (briefly checked but I could've missed something).

I was wondering whether there is any good resource (not too detailed and not too brief) that can give a good explanation for what causes the volatility smile? I've seen various explanations, ranging from economic-based ones (supply and demand dynamics associated with ITM, ATM and OTM options) to model-based ones (for example, simpler models e.g. non-jump diffusion vs. more complex ones e.g. jump-diffusion). Is there some consensus?
option premiums and implied vols are the same thing. your question can roughly be translated to why atm options are often cheaper than itm and otm ones (not in absolute sense).

in short, it's all supply and demand. u have to understand market microstructures and common behaviors of the participants of the market to really understand it. also things change all the time due to new macro and micro phenomenon, which ultimately influences who buy/sell what from who. different markets act differently. i'm in fixed income/credit market and what i observe/experience is not the same as equity and commodity markets.
Last edited:
Might not be a reliable source but this might help: Chapter 6 All about Volatility | The Derivatives Academy

Remember implied volatility is actually the Black-Scholes implied volatility. The big assumption behind BS is that log-returns are normally distributed. So if all european calls/puts on a stock had the same implied volatility (that is, flat vol smile) then the market would be predicting that log-returns are perfectly normally distributed.

But in reality we know that asset returns have fatter tails and negative skewness. So log-returns in reality are approximately normal but have a higher probability of large moves and higher probability of large negative moves than large positive moves. So if you have a smile it is saying that your probability distribution has fat tails (options away from ATM are more valuable) and a skew tells you that there is skew in the distribution.

Rather than coming up with a new model for the asset which correctly captures asset dynamics we just kept the same model and model the implied vol so that it matches market quotes