Testing Out The Hypothesis on Shorting Paired Leveraged ETFs

This post is going to serve as a placeholder for a spreadsheet I’ve set up to monitor an investment strategy I mused about in the post “The Equal Short Bull-Bear As The Ultimate Negative Correlator“. I had posited that instead of putting 100% of your money into an index, you would’ve been better off (since 2000 anyways) to have put 50% in the index and then put 50% into shorting (in equal amounts) a 2x Bull and 2x Bear ETF (on that same index).

The first spreadsheet is displayed below and has an inception date of June 3rd, 2009, and you can click on the tabs at the bottom for a more detailed version. UPDATE: A second spreadsheet is located at the bottom which has back-tested data.

The Performance Tabs show the values adjusted for paying the interest owed on holding your short positions, and the Detailed Tabs show the non-interest-adjusted strategy returns.

NOTES:

  1. I’m using $100,000 of market exposure. Technically you could take the proceeds from the shorts to invest in the long part of the portfolio. So essentially, if you had $100,000 in cash, you could get yourself more than $100,000 of total market exposure, but I’m not going to look at this. I’m strictly comparing $100,000 of market exposure in the long only strategy (100% invested in XIU) versus $100,000 of market exposure in “the strategy” which is 50% long XIU, and 25% short HXU and short 25% HXD.
  2. On the Performance tab, I have subtracted the running interest total for holding $50,000 short, calculated at 3.75% (prime + 1.5% as quoted by one of my discount brokerages).
  3. The spreadsheet updates automatically every 5 minutes.
  4. The above portfolio has an inception date of (intraday) June 3, 2009.
  5. The BELOW spreadsheet shows the data for as far back as all three ETFs have history (to January 9, 2007).
  6. Feel free to make suggestions for improvement!

Preet Banerjee
Preet Banerjee
...is an independent consultant to the financial services industry and a personal finance commentator. You can learn more about Preet at his personal website and you can click here to follow him on Twitter.
Related Posts
Showing 13 comments
  • Jordan

    Could you enable copying/sharing on your google docs spreadsheet so others can view the formulas, or make changes?

    Is this spreadsheet setup in such a way that it could test the strategy out on a variety of other indexes, or did you have to enter the figures manually?

    I’d like to hear how your EDC/EDZ paper trade went along with:

    IVV / RSW / RSU
    EFA / DZK / DPK

  • Returns Reaper

    Preet,

    Would hard-to-borrow fees (as you covered in a recent post) affect this strategy? Do you think they’d impact the results of this strategy very much? Can they be accounted for? Or are they difficult to determine?

  • Preet

    @Jordan – the spreadsheet is setup so that all you would have to change is the ticker symbol and purchase price per share (and the start date). It will calculate the rest. I’ll send you an invite to the doc that will let you edit it (a copy of it anyways! it’s live-linked to this post!)

    EDC/EDZ – I set up a portfolio with that one and also FAS/FAZ. The two equal short pairs together are up 0.7% as of right now and since May 26th, 2009 (lagging the S&P which is up 5% during that time), but note that they are picking up steam as the market reverses course. It will need more time to play out as the value add comes over time, and assuming lots of volatility.

  • Preet

    @Returns Reaper – Hard to borrow fees would kill the strategy in the real world. But my broker has no hard to borrow fees on HXU and HXD (at this time anyways). 2X ETFs make up close to 90% of the leveraged ETF market, so from a practical point of view using 3x ETFs won’t work, and 2x might.

  • karash

    Preet,
    Could you also send me the ‘invite’ for downloading the spreadsheet.!

  • Jordan

    Preet, thanks for sharing!

    So I’ve made some modifications to your spreadsheet, it’s now fully automated so you only have to put in the start date and symbols and it will look up the historic start prices.

    You can also now change the amount of equity, interest rate, and % allocated to each equity on the first sheet.

    I’ve also tried it out on a few more indexes and made public spreadsheets for anyone else to see and edit themselves:

    S&P 500 : Index: IVV, Bull: RSW, Bear: RSU

    This is 2x leverage, the result 29% benchmark out performance:
    http://spreadsheets.google.com/ccc?key=rnmTshMlWhy02Q07badodIQ

    MSCI Developed : Index: EFA, Bull: DZK, Bear: DPK

    This is 3x leverage, the result +6.5% benchmark out performance:
    http://spreadsheets.google.com/ccc?key=r7XuVM83tBaxWW7S-HeeW2g

    Emerging : Index: EEM, Bull: EDC, Bear: EDZ

    This is 3x leverage, the result -9.91% benchmark UNDER performance:
    http://spreadsheets.google.com/ccc?key=rpHLG0ze1a8CX1T6mOdK5qA

    And also the original XIC, HXU, HXD
    http://spreadsheets.google.com/ccc?key=r8ds7ZybVaihyrU02hN5ghA

  • Jordan

    Here is the S&P 500 with IVV, SSO, SDS which has a longer history, starting in July 2006.

    It’s 2x leverage, resulted in 28% benchmark out performance:
    http://spreadsheets.google.com/ccc?key=rZdXKltF-vuT5V4LZYOXPkA

    The Russel 2000 : Index IWM, Bull: UWM, Bear: TWM
    This is 2x leverage, an amazing 40% out performance? Jesus!

    http://spreadsheets.google.com/ccc?key=r8xiYdqd27Lurd4jAQjMuIw

  • Preet

    @Jordan – right back at ya: thanks for sharing! :)

  • cannon_fodder

    I think what this really shows is that shorting the 2x ETFs generates good returns over the long haul. The strategy to me would be to short the TSX/60 bear as I would expect in 3-5 years the underlying index would be higher.

  • BobJ

    This is a very interesting concept. I made the mistake in the past of holding a 2X ETF, and saw the effects of the equity dripping away daily.

    I have a problem with the spreadsheets above. They open, but the data doesn’t fill. When I go to “details” the fields just say “loading”. Can you tell me what I’m doing wrong.

    Thanks

  • BobJ

    As usual, you figure something out as soon as you ask for help. The end date came up as June 6, (Saturday) so didn’t have a price. As soon as I changed it to the 5th, everything filled.

    This is going to be fun to play with changing starting/ending dates for the different stock cycles.

    Thanks Preet/Jordan!

  • moosey

    Anybody have suggestions on newsletter that tracks and recommends Canadian ETFs?

  • Preet

    @moosey – sorry I can’t find anything really. I don’t subscribe to any investment newsletter at all actually… Let us know if you find anything.