Sep 24, 2012

BAC Dividend Trade

BofA Unit Said to Lose Millions on Options Error

WSJ reported on Friday that Bank of America Merrill Lynch lost about $10M on SPY dividend trade. I cannot comment on veracity of the story but explanation provided in the article is incorrect in their description of the dividend trade.

The article writes "Usually shares move lower after this date, so call options that give the holder the right to buy the shares also decline in value. Market makers familiar with the process may look to profit from the scenario by selling borrowed call options with the intention of buying them back later, when the contracts are cheaper." , and later "Critics of the strategy charge that it takes advantage of investors that haven't exercised their call options before the ex-dividend date, and don't have the resources to transact big options plays."

That is not quite how the dividend trade works ...
First, let's start with options pricing - dividend is an expected and deterministic stock move, and it is completely priced into the option price. Not only theory of pricing options on dividend paying stock is known and well developed, but also every professional options trader knows dividends and prices options accordingly. If they were not - it would create an arbitrage opportunity, and market-makers are smart enough to prevent them.

Second, while short selling stocks works as selling borrowed shares and buying them back, an opening trade in options creates a long position for a buyer, and a short position for a seller. There is no transfer of security like it is with stocks.

Third, dividend trade works not by just selling an option that is expected to drop in price after dividend, but rather as quoted above - by hoping that holders of long options (which market maker is short) will not optimally exercise them. Market makers usually create Δ-neutral spreads in deep call options (edit: or even on the same strike by trading back and forth - the position is not zeroed out for clearing purposes until it settles at t+1), and hope that short leg will not get exercised. In case of correct exercise market maker breaks even;  if option is not exercised the market maker gains dividend without stock risk. Some exchanges - ISE - are particularly against the strategy, because it creates significant risks if the underlying makes a sharp fall, and options spread suddenly gains gamma. Other exchanges - PHLX - are welcoming the practice; from what I know dividend trade accounts for most of the volume on that exchange.

It is true that the strategy takes advantage of investors who don't exercise their options - which is true of most trading - being smarter than the other guy. The other piece of criticism - about not having resources - that is also true - dividend trade is a low-margin activity, and only traders with low costs can take advantage of them.

P.S. ISE paper with very clear explanation of dividend strategy.

Sep 17, 2012

Backtesting ETFs

Institutional Investor magazine recently published an article titled "Study Finds Many ETF Indexes Misleading". For me and for other readers of this blog I'm sure this hits home, especially in terms of performance of VIX-related ETFs. Over the last two years that I wrote about several ETFs and ETNs and tried to provide some sort of guidance on their future performance based on past data. While some were relatively easy to reconstruct and explain, other products were more complicated.

For example, in Jan 2011 I posted my analysis of XVIX with quite optimistic projections. The performance since publication over the last year and a half was disappointing -8%, with maximum drawdown of ~22%. While the balancing rule for the ETF was simple: -0.5 * VXX  + VXZ, rebalanced daily, after all things taken into account (including 0.85% fee from the issuer) the volatility risk premium, or term structure trade simply did not work out.

Another forecast I made turned out better - XIV and other daily inverse ETFs performed quite well in the last 1.5 years - XIV rose 71% , although with 49% drawdown between March 2012 and June 2012.

Predicting future is fundamentally challenging, and even strategies that seem robust do not always perform as expected. Having said that I want to bring another point: there are different trading styles: the ones mentioned above, and others like Mebane Faber's GTAA ETFs (that also so far has disappointed in its performance) are rule-based. My intuition tells me that these type of strategies are most susceptible to data mining bias / overfitting. Strategies that are based on valuation, and particularly on relative valuation seem to be more robust to model error.  However I have no idea how to quantify either one. 

Sep 5, 2012

Thomas Peterffy NPR Interview

Thomas Peterffy - founded and chairman of Interactive Brokers recently gave a short interview on NPR (here is a partial transcript and 5 minute condensed interview, here is complete 26-minute podcast) . My first job after college was at IB over 10 years ago, and I have all the respect for him. However I find his commentary about "no social value" to be tendentious - high frequency trading probably has as much social value as reselling Juicy Fruit gum. And Peterffy himself certainly would not be able to become such a success living under social regime.

Peterffy and his firm Timber Hill (market-making branch of IBG) were real pioneers of trading, who in the 70s, 80s, 90s, and most of 00s were years ahead of competition. Exchanges and other options firms were engaging in all kings of anti-competitive tactics trying to slow down Timber Hill - the typing robot is just one of many stories like that. But now that Timber Hill is not a leader in speed (and Timber Hill profits have been in decline)  does it really make sense to start complaining about social value?