Asset Allocation using Leverged ETFs

Asset allocation, risk, diversification and rebalancing. Pros/cons of hiring a financial advisor. Seeking advice on your portfolio?
Post Reply
jay
Contributor
Contributor
Posts: 547
Joined: 19 Feb 2008 21:13

Asset Allocation using Leverged ETFs

Post by jay »

Let's assume the asset allocation is made of two entities, one for fixed income and another for Equities. Fixed income is a short term bond ETF or money market fund and the equity portion some equity index ETF, say DIA. It turns out that using those 200% leveraged ETFs (like ProShares Ultra Dow30) would produce a better return, for the same risk. The advantage simply lies in having a higher allocation of fixed income when taking the leveraged approach. So a balanced 50 FI / 50 Equities portfolio would be replaced by a 75/25 one.

Here is the simple math: Assume a fixed income return of X% and an equities return of Y%, and assume that fixed income allocation is m% (hence equities 1-m %):

Portfolio Return = mX + (1-m)Y

Portfolio Return the leveraged way
= (m+((1-m)/2)X + ((1-m)/2)* (Y*2)
= (m+((1-m)/2)X + (1-m)y

resulting in the leveraged approach having an advantage of (1-m)/2 * X, which can be non trivial. Here are some examples using a fixed income rate of 4%:

for a 50 FI / 50 Eq portfolio, the advantage is (1-0.5)/2 * 0.04 = 1%
for a 40 FI / 60 Eq portfolio, the advantage is (1-0.4)/2 * 0.04 = 1.2%
for a 30 FI / 70 Eq portfolio, the advantage is (1-0.3)/2 * 0.04 = 1.4%

I am convinced that I haven't invented anything and that this is probably discussed in many places, probably somewhere on FWF as well, but I thought of sharing it. It seems like an easy extra return, but only if one is OK with using leveraged ETFs, and that the return of the leveraged ETF is indeed always 200% of the underlying index.
User avatar
martingale
Veteran Contributor
Veteran Contributor
Posts: 1099
Joined: 20 Feb 2005 14:13
Location: On The Customers' Yacht
Contact:

Post by martingale »

You've at least not accounted for fees. The fees on the leveraged funds are in the range of 1.15% versus 0.17% for the non-leveraged version of the ETF's for example in the case of the TSX S&P 60 index. That claws back a substantial chunk of the difference in performance; the question is then whether the tracking error inherent in the leveraged version eliminates the remaining difference.

(The tracking error is inherent because the 200% is "before expenses" and presumably the derivatives strategy used by the leveraged fund is not without additional cost.)

It is still possible that this strategy is going to have nearly the same return but with less risk. If the fund guarantees that you will never lose more than the money you put in (and it seems to) then by dropping your equity position to 25% you've guaranteed that you could never lose more than 25% of your portfolio to an equity slide.

Some interesting questions arise on the extent to which you would defeat that benefit via regular rebalancing, though.
[url=http://www.efficientmarket.ca/]Canadian Mutual Fund, ETF, and Self-Directed RRSP Advice[/url]
User avatar
adrian2
Veteran Contributor
Veteran Contributor
Posts: 13333
Joined: 19 Feb 2005 08:42
Location: Greater Toronto Area

Re: Asset Allocation using Leverged ETFs

Post by adrian2 »

jay wrote:Here is the simple math
I've mentioned it more than a few times, this is not how double leverage works: for anything but a straight line, the constant leverage trap works as a double digit MER equivalent.

If you pay minimum or no commissions and you rebalance every single day, your strategy might work. For anything longer than a day, buy and hold double leverage does not work.
jay
Contributor
Contributor
Posts: 547
Joined: 19 Feb 2008 21:13

Re: Asset Allocation using Leverged ETFs

Post by jay »

I knew I posted about this a very long time ago :).

I recently discovered http://www.portfoliovisualizer.com for backtesting. Great site. And I also always thought leveraged products intriguing, so I ran some tests.

This backtest compares 100% equity allocation (SPY in this case) against a 50%/50% in 2X equity (SSO) and short term treasuries (SHY).

And this backtest is for the same allocation as above, but starting from Mar 2009 (major low).

So the 50/50 approach seems to track fairly well overall. Annual re-balancing is obviously key for without it, the 50/50 approach will underperform. Do you find this surprising as I did?

So why would anyone implement such a strategy? I guess on the cash side it is easy to outperform treasuries. With annual re-balancing, only some of the cash may get called to bring the mix back to 50/50, so instead of 50% in cash one can have 10% in accessible cash and the remaining 40% in longer duration bonds or a ladder.

This is just fun research. I don't own leveraged products. But maybe leveraged products are not all that bad after all.

Am I missing something here?
User avatar
adrian2
Veteran Contributor
Veteran Contributor
Posts: 13333
Joined: 19 Feb 2005 08:42
Location: Greater Toronto Area

Re: Asset Allocation using Leverged ETFs

Post by adrian2 »

jay wrote:I recently discovered http://www.portfoliovisualizer.com for backtesting. Great site. And I also always thought leveraged products intriguing, so I ran some tests.

This backtest compares 100% equity allocation (SPY in this case) against a 50%/50% in 2X equity (SSO) and short term treasuries (SHY).
Backtest to 1985? Of course they must be using theory, not real world.

Quoting myself from upthread:
I've mentioned it more than a few times, this is not how double leverage works: for anything but a straight line, the constant leverage trap works as a double digit MER equivalent.

If you pay minimum or no commissions and you rebalance every single day, your strategy might work. For anything longer than a day, buy and hold double leverage does not work.
One data point from TD Waterhouse:
SPY 10 year return = +6.86%
SSO 10 year return = +7.06%
Does it look like double to you?
Imagefiniki, the Canadian financial wiki
“It doesn't matter how beautiful your theory is, it doesn't matter how smart you are. If it doesn't agree with experiment, it's wrong.” [Richard P. Feynman, Nobel prize winner]
jay
Contributor
Contributor
Posts: 547
Joined: 19 Feb 2008 21:13

Re: Asset Allocation using Leverged ETFs

Post by jay »

adrian2 wrote: Backtest to 1985? Of course they must be using theory, not real world.
the backtest actually goes as far back as SSO does, which is June 2006.
One data point from TD Waterhouse:
SPY 10 year return = +6.86%
SSO 10 year return = +7.06%
Does it look like double to you?
Yes I'm aware of that. Simple math of compounded decay/growth, esp. in side-ways markets. Over some periods SSO might even produce negative returns when the underlying index produced positive returns. I get it.

But it seems you missed the whole point of my post; or perhaps you didn't read it or checked the backtest in detail. I wasn't suggesting going all in into a leveraged ETF in my last post but rather try to track a 100% equity allocation yet with a better risk adjusted return.. Use a 50% allocation to SSO to get a 100% allocation to equity, try to beat treasuries on the cash portion, and make sure to rebalance.
Post Reply