A few weeks ago I had mused that holding a double short pair of leveraged ETFs might be a great negative correlating investment to hold in tandem with a regular long position on a plain vanilla ETF. The double short pair would be expected to have a positive rate of return if the volatility of the underlying index was high enough, and a hypothetical backtest had shown that it could smooth overall portfolio volatility.
I still want to re-iterate that you shouldn’t be playing around with this stuff unless you understand it.
Having said that, I set up a live mock portfolio on June 3rd with $100,000. The strategy has added about roughly 3% of net alpha for the period to date (June 3rd to June 23rd). Simply holding XIU and you would be down 5.84%. By holding 50% XIU and 50% in the paired short, you were only down 2.84% (and that includes subtracting the interest you would pay on the short based on prime + 1.5%).
Unfortunately I can’t plot the path of the portfolio as Google Docs doesn’t support macros (so I can’t pluck daily values out), but just from eyeballing it, the return is higher and the overall volatility has indeed been lower as expected. Below are the live results (updates every few minutes during trading hours) – I don’t know if this will show up in emails, so click on the title of this post to see the results on the website.
Preet I just started to wonder, why would you use XIU for the long position to track the Canadian market instead of CRQ? You’ve shown before pretty good evidence that the RAFI Fundamental indexes outperform a standard market cap index. I understand the 2 shorts are perfect hedges of each other but is there any reason the long has to be a perfect match as well?
CRQ outperformed your return but the time span is too short to make any conclusions. If you compare XIU/CRQ alone YTD, or pretty much any period over the last 3 years it has outperformed. Any thoughts how it plays into this strategy?
Simply because the cap-weighted versus fundamentally weighted index debate is a separate issue that I didn’t want diluting the main point.
There is no reason why the long has to be a match – just cleaner for the analysis again.
Personally, I would use a fundamental index as the long myself.
Here’s another random idea about this strategy, let me set it up with a few assumptions:
So assume this strategy does accurately track the performance of the underlying indexes, but is able to add some alpha long term.
Now say a fund manager knew of it and managed a fund with a mandate to track the same underlying index.
Would holding shorts on these 2 leveraged ETFs somehow synthetically qualify as meeting their mandate? I imagine with competition between similar funds that managers might look for something new to juice their own fund’s results without breaking any rules.
I’m kind of thinking back to a story I heard about shorts against mortgage backed securities in the US enabling the bundling firms to some how create synthetic CDOs after they ran out of “legitimate” mortgage holders. I don’t fully understand it, but I’m curious if there are similar implications for this strategy.
Preet,
This strategy is really interesting. Are there any new developments?