Please critique this trade idea
Something I just came up with today, but I'd rather one of you point out the flaws in it than an interviewer.
Trade: Decreasing correlation between EM equities and the S&P
Position: Short March S&P put @ 900 strike Short S&P Index Long any highly correlated EM index (e.g. Brazilian Bovespa)
Why: S&P has come a long way since the March lows. Limited further upside compared to downside. MSCI EM index correlation with S&P shot up to 0.83
Profit: Collect premium on the put See near term downside to S&P but not extreme enough to cross 900. See plenty of upside for Brazilian stocks (for various reasons).
Problem: Is this really a decreasing correlation trade? The more I look at it, the more it looks like I'd be betting solely on the EM index outperforming the S&P.
What am I missing here? What would be a better way to play a reduction in both index correlatons? Perhaps some kinda delta and gamma neutral vol play on both indexes by replacing the long Bovespa index with a long March call on the Bovespa?
...u arent betting on correlation, you are just betting that EM will outperform the S&P. If S&Ps go up and EM goes down correlations will go down and you will lose alot of money so the trade has nothing to do with correlations. I also dont think your logic is that robust...you say you want to sell the S&P because "they are up alot"...that dosent really strike me as a good reason to put on a trade. And of course EM stocks are farther off their lows then US stocks are and you say you want to buy them. And lastly, why do u feel the need to sell those puts as opposed to sticking to the simpler construction of just selling spoos and buying EM? In short, if you gave me this as a trade idea in an interview i'd be fairly underwhelmed by the logic behind it.
I wouldn't say this in an interview. Thought I'd come up with a few trade ideas today and this was the only one which didn't make sense in my head ... as much as I wanted it to.
What about going short an S&P/MSCI EM correlation swap? Receive fixed and pay floating (annualized realized correlation). That would work right? But then again, with correlation already at 0.83, I guess the limited upside on long correlation will bear itself in increased transaction costs when going short the swap. I'm just speculating now as I have no idea how correlation swaps are even structured.
Intrigued to hear how else anyone would take views on stock index correlation.
I am not sure if you actually understand correlation swap.
For the record, a correlation swap entails receiving (paying) fixed and paying (receiving) floating where one leg deals with implied correlation and the other the realized correlation. In truth, you can't really long a correlation in the conventional sense of the word - contrary to your understanding of the limited upside (which you regard as 1) when it's already 0.83. What you're effectively trying to do is capitalize on the spread between the realized and implied vols - if you're long the spread, you're better off with the two vols (ie. implied and realized) being 0.70 and 0.30 than having them at 0.99 and 0.60 respectively!
Hence, you can't really long/short a swap - you're better off talking in terms of longing/shorting either the realized or implied correlation specifically...
It's senseless to talk about correlation trades if all you're doing is take two sets of data over the same time horizon and run a correlation test on that.
It would make much more sense to talk about correlation trades in the context of buying an EM equity constituent index and shorting some of the underlying stocks (depending on your exposure, market view and risk appetite) to manage your correlation risk. In short, that's dispersion trading (elaborating on this in detail would warrant a thesis on its own).
The difference between such correlation trades and the one you proposed is the exposure and hedge for your positions. As Bondarb has mentioned, there's no correlation to talk about if S&P and EM markets moved in opposite directions - you're doubly fucked! You've got to remember that you longing any "highly correlated EM index" is not a hedge for you shorting the S&P Index because the statistical correlation you get is not necessarily reflective of reality going forward.
In short, I'd suggest you forget about generating such complex trade ideas if you have no clue what you're talking about - to be honest, I don't even think you'd be grilled much on correlation risk or dispersion trading unless you're interviewing for exotics/hybrid desks (it took me sometime to figure some of this stuff out and suffice to say, some of the flow traders I work with know nuts about hedging correlation risk anyway)...
Funny enough, it was an article on dispersion trading that led me to the correlation swap idea, but like you said, I don't understand it very well. Knowledge never hurts though. I'm sure it can be learned.
I'd never be grilled on correlation risk or dispersion trading anyway as I'm interviewing for SA positions. Neither would I ever use such a complex idea in an interview. I was just wondering how one could express such a view though.
Thanks for your replies. Helped a lot.
Dubai's making things interesting for you right now eh
Short correlation? harder to do it for EM versus S&P rather than on the stocks comprising of the s&p alone, but you can buy variance swap or calls/puts on a basket comprising of S&P and EM indices.
Then, you can buy calls/puts on the S&P individually.
If correlation declines, the S &P and EM markets are much more likely to go in opposite directions, which means that the volatility on the basket of S&P and EM will be much lower.
However, the volatillity on the calls/puts on the s&p themselves will still be good.
essentially you sold vol on something that becomes less volatile, and you buy vol on one of the two.
(imagine S&p +10%, EM -10%, equally weighted basket ~0%
You said "buy calls/puts" on a basket comprising of S&P and EM indices, but I assume you meant sell calls/puts, not buy right?
So essentially, the position could be: Sell vol on a basket comprising of S&P and EM indices Buy vol on a basket of EM indices
Makes sense. Thanks.
btw, the above is kind of a like a dispersion trade
Deleniti sed vero porro est. Quos dignissimos nesciunt voluptas nesciunt eligendi et. Vel laboriosam nihil recusandae reprehenderit neque similique.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...