A Case for Leveraged ETFs

by ChengSkwatalot

Warning:

  • this is not investment advice
  • past performance doesn’t guarantee future returns
  • potentially controversial, long post

Short Introduction to Leveraged ETFs

We all know that investing in a globally diversified equity portfolio is a great way to let your money grow. As long as your investment horizon is long enough, a diversified equity portfolio seems to consistently grant investors an (equity) premium over the risk-free rate.

This naturally raises the question of how we can further increase those returns. There are, generally speaking, two ways to increase expected returns:

  • decreasing diversification
  • taking leverage

Decreasing diversification will increase your exposure to idiosyncratic risk, which is not what we want in a passive investing strategy. Since this sub focuses so much on passive investing, I’ll focus on the second method: taking leverage (i.e., borrowing money that you can use to invest).

There are many ways to take leverage, but the one I like most is using leveraged ETFs. Most leveraged ETFs try to replicate the daily performance of a certain index multiplied by a certain number. This implies that they re-leverage every single day. A disadvantage of this daily re-leveraging is that it causes decay. What does decay mean? Suppose that a stock experiences a return of -10% on day 1, and a return of 11,11% on day 2. The resulting return over the two-day period, ignoring leverage, would be 0%. However, when applying 3x daily leverage, the return on day 1 equals -30% and the return on day 2 equals 33,33%. In this case, the return over the two-day period equals -6,67%. The fact that the unleveraged investment returned 0% whereas the leveraged investment returned -6,67% is due to the decay. Decay, along with large tail risk, are the biggest disadvantages of leveraged ETFs.

Historical Performance

So, now that you know about the basic characteristics of leveraged ETFs, let’s look at their historical performance. Or at least, what their historical performance would have been over the past 40+ years. For this analysis, I used daily returns in USD of the “Wilshire 5000 Total Market Full Cap Index“. The data starts at 30-11-1979 and basically runs until today. I applied the daily re-leveraging that leveraged ETFs use to the data. Note that I did not take any costs into account. However, as long as real-life leveraged ETFs succeed in replicating their benchmark, my main findings should still hold.

Annualized Returns

Period 1980-1990 1990-2000 2000-2010 2010-2020 Standard Deviation (Annual)
Wilshire 5000 16,62% 17,59% -0,17% 13,28% 16,33%
Wilshire 5000 x2 21,22% 32,57% -5,19% 34,94% 34,94%
Wilshire 5000 x3 27,17% 46,47% -14,41% 56,34% 56,34%

Worst Daily Return (19-10-1987) & Maximum Drawdown

Worst Daily Return Maximum Drawdown
Wilshire -17,31% -54,44%
Wilshire 5000 x2 -34,63% -83,06%
Wilshire 5000 x3 -51,94% -94,91%

I then did some tests using rolling-window periods for 10- and 20-year investment horizons. I calculated:

  • The median return over all 10- and 20-year investment horizons
  • The chance of a negative cumulative return over the total 10- and 20-year periods
  • The chance that the cumulative return for the leveraged investments over a 10- and 20-year period is worse than that of a normal investment over the same investment horizon
  • The highest and lowest possible cumulative return for 10- and 20-year investment horizons

Results Rolling-Window Analysis

Median 10-yr. Cum. Return Median 20-yr. Cum. Return Chance Negative 10-yr. Return Chance Negative 20-yr. Return Chance 10-yr. Return Worse than Un-leveraged Chance 20-yr. Return Worse than Un-leveraged
Wilshire 5000 182,57% 493,02% 3,38% 0,00% / /
Wilshire 5000 x2 477,43% 1596,49% 6,93% 0,00% 11,15% 0,00%
Wilshire 5000 x3 758,41% 2258,31% 11,18% 0,00% 15,46% 1,42%

Best 10-year Cum. Return Best 20-year Cum. Return Worst 10-year Cum. Return Worst 20-year Cum. Return
Wilshire 5000 498,05% 2 651,46% -31,10% 122,76%
Wilshire 5000 x2 2 825% 48 838% -70% 126%
Wilshire 5000 x3 11 590% 545 727% -92% 2%

Conclusion

As you can see, leveraged ETFs could potentially offer interesting long-term returns. The biggest disadvantage is clearly the high tail risk. If you lump sum, your returns will strongly depend on your buying point. To give some extra clarification, the worst results above are caused by starting your investment around the peak of the dotcom bubble. This problem could potentially be solved by making periodical investments instead of lump summing though.

The goal of this post is not to have you all allocate the majority of your investable capital to leveraged ETFs, but to start a discussion/conversation on the topic. I think leveraged ETFs are some of the most interesting but also commonly misunderstood investment vehicles out there.

I do believe that leveraged ETFs are useful for passive investors, as long as they track well-diversified indices. As previously stated, you want your idiosyncratic risk to remain as low as possible. The common rules that we all invest by also apply to these leveraged ETFs, but to a more extreme degree. Your investment horizon better be extremely long, you better not sell during a market downturn, you better ignore your emotions, don’t try to time the market, etc.

Thank you so much for reading and feel free to let me know what you think. If you have any questions, ask away.

EDIT: The data I used does not come from an actual leveraged ETF. I used daily returns from the Wilshire 5000 Total Market Full Cap Index and simply calculated what the performance would have been if daily leveraging were applied. The goal of this post is to spark a conversation about this, not to have everyone invest in leveraged ETFs.

 

Disclaimer: This information is only for educational purposes. Do not make any investment decisions based on the information in this article. Do you own due diligence or consult your financial professional before making any investment decision.

279 views

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.