QuoteOriginally posted by: CaesariaLets say that I am Goldman Sachs, and you want to buy a variance swap on a November future for Crude. You have a business need for this swap, so how should I price the swap and sell it to you? Should I never sell it to you, since my model for Nov Crude is based on its implied vol curve and is susceptible to black swan events like the Middle East turmoil? Should I charge a really high premium from my fair value, since I am willing to bet that my hedge error would be contained in that range? Or should I worry about a super fat tailed event and just never make a market for it? If I charge a really high price, then someone else (say JP) will sell you this swap for cheaper and will take the risk of a 3 standard deviations event. Damn, so should I be competitive and take the risk and sell it within premium of a 2 STD event? Or should I get the regulator to not allow JP sell that swap to you, since I know that JP's model is undercharging and they are being irresponsible and could potentially screw up their shareholders in case of a rare event? The insurance industry has been central to the protection of consumers, and for producers locking in profits and corporate management/planning for expansions. If derivatives that have business needs are not sold, then we would have to go back a few decades and since it would slow down the growth of many industries. So what should I do, should I wear my crown of thorns, and bear risks to encourage and promote growth while charging enough premium to know that I would "most likely" make a profit? Or should I say no to selling the derivative, since I want to protect my shareholders from the collapsing bridge? Or should I expand my trading desk to sell exotics for every type of underlying from technology to commodities to forex to other equities and sell exotics on all of them, so that my hedge errors on all those exotics comprise a classically diversified Markowitz portfolio? Or should I worry that in times of a bubble, all of my "diversified" assets will have positive correlation therefore breaking down the "diversification" relationship, and therefore busting up my entire trading desk? And all of the traders having their "call option" type bonuses do not get hurt, and only shareholders get hurt? What the hell do I do to help maintain a 21st century industry? Should I just accept that in our complex dynamic world, I have to come to terms with the fact that we go through boom and bust cycles that promote R&D for both the large customer who requires insurance and the underwriter? We learn from our mistakes don't we? We always recover, to an extent, from a bust right? Doesn't our mind work in a trial and error manner, then why wouldn't society? What should I do, should I sell you that Nov Crude Variance Swap?If you use static hedges (vanilla options), you are OK wrt tail-events. Perhaps a variance swap was a bad example?In general, though, the problem is not the "black swan" events, per se, but that many models don't take account of extreme tails because it's mathematically inconvenient. So long as your model is reasonably well-specified, and so long as you are imaginative about how things could go wrong, then your hedges are going to be a) reasonably robust and b) you will have some idea of what your downside is when things blow up. You'll also have carefully thought out risk-limits so, even if you are wrong, you won't go up in flames.In fact, just like Goldman and several other thoughtful players did before the credit crisis, you would keep on selling those derivatives. If you thought the market was mispriced, you'd sell the derivative, pocket the spread, and pass on the risk to the market. You'd do this as much as you possibly could, but you'd make damn sure that you didn't have too much of the muck on your balance sheet. If you thought you understood the mispricing, you'd position yourself the other way; if you suspected you might not, you'd stay as lean as possible and try to make money market-making.That's how derivative markets have always worked and how it will always work - and this is true from before we had Black-Scholes. Black-Scholes just allowed participants to think that they could be more certain of the value of the contracts they were writing, because delta-hedging allows you to reduce the variance of your portfolio PL in theory.And Goldman is most definitely not bearing risks "to encourage and promote growth"! They are doing it to make a buck. GS is not in the business of maintaining an industry. The decision for regulators, from a classical economic perspective is more subtle. If they believe there is an externality - e.g. bail-out moral hazard or systemic counterparty risk - then they should try to fix it; otherwise, they should not. But that's a broader question.