oh man I wouldn't be surprised if ppl were just "arbitraging" the difference between var and vol swaps where they are struck thinking they were picking up free money without realizing they were selling convexity, since vol really didn't go anywhere for several years pre-2007 sle I have been following your thread on the exchange-listed variance swaps. Do you think this convexity effect is why VA hasn't taken off? Why do you think the VX is cash-settled on VIX instead of just being a volatility swap? That seems like a cleaner contract spec to me.
selling premium (usually shorting otm puts) when it feels so wrong is in hindsight the most profitable activity i've done. i.e. sell otm puts on good quality companies mcd, pg, coh, tif, ko if they miss earnings and/or into a constantly declining market (aug 2011). i sold otm leap puts as far as 24 months out and bought them back within 2-4 months at less than 50% of price sold at. it felt wrong at the time as i just lost money day after day and kept selling puts but it seems the most profitable strategies i've had make me feel like puking at the time. worst case in above strategy is you own decent companies at lower prices (most people say whatever; you own ko at $30 when its at $25). it just feels like "they" are screwing you as you keep raising your ask price and "they" just keep buying all the way up stacking the losses up on your ledger. 90% of those strategies are losers for the first month and 90% are winners within six.
I think people understood that they are selling convexity but could not find anything to do. It is a pretty frequent occurrence - people can't find anything worth doing, but have to do something so they throw shit at the wall and see if it sticks. Most times it sticks, but sometimes it falls on their heads. There is other "awesome" stuff people have done, like sell conditional variance - "well, you are selling var at levels where it has never been before, so how could you lose". No, I think variance futures have not taken off simply because its complex (some of it by nature and some of it by how CBOE decided to do it to mirror variance swaps in OTC markets), transactions are very costly (CBOE are idiots, should have made it trade in 10s of var units). Probably the key reason is that they did not really put much effort into grabbing the market share away from the OTC dealers - I was hoping that hedge funds like mine would preferr to get crossed on the exchange, but the fees are high and clearing costs per contract are even higher. Volatility swaps are not fungible, (vol swap mark to market and residual vega are dependent on the prior volatility and is not consistent with the current strike - you can prove it to yourself), so it's not really possible to make a futures contract out of it.
Yes, which is not capital efficient at all. For example, to short $1000 worth of vega in 1 month ATM option you would need approximately 850k of notional. If you use reg-T margin, your capital requirement is like 250k day one (approximately - i think it all boils down to about 30% of notional as day one margin). And that's to have an equivalent of a single VIX futures. Which is why I was hoping variance futures take off - there is no real access for retailer traders to the realized volatility market. Maybe someone should make an ETF that delta-hedges front straddles or something along these lines instead.
Well, I am sure some days you will regret it. CBOE should make an index (rolling straddles plus delta hedge at close) and some dude should issue an ETF. Overall it's not that different from VXX