Investing in stocks, bonds, and other asset classes is the foundation of portfolio management. But occasionally, investors are seeking portfolio properties that cannot be achieved with these conventional instruments. This is where Complex or Leveraged Exchange-Traded Products (CLPs) come in.
These products are often stigmatized as being hazardous. At Bertram Solutions, we beg to disagree with this view, but at the same time recognize that using these products requires thoughtful consideration and constant supervision.
Leveraged ETFs, sometimes also called geared ETFs, are the simplest product category covered in this post. Typically, leveraged ETFs come with a nominal leverage of two or three, promising an amplified exposure to the index they are tracking. Assuming that markets keep climbing up, applying leverage sounds like a no-brainer. Unfortunately, things are not that easy.
|Day 1 Morning||100||$100||$200||$300||3.0|
|Day 1 Evening||110||$130||$200||$330||2.5|
|Day 2 Morning||110||$130||$260||$390||3.0|
|Day 2 Evening||100||$95||$260||$355||3.8|
|Day 3 Morning||100||$95||$189||$284||3.0|
To understand the mechanics of a leveraged ETF, let’s look at a hypothetical bull-market ETF with 3x leverage:
- On the morning of day 1, the underlying index is at 100, and we invest $100 in a 3x leveraged ETF. To achieve the leverage, the ETF borrows an additional $200 and spends $300 on assets.
- On the evening of day 1, the underlying trades at 110. With it and our assets are now worth $330. The borrowed amount didn’t change, leaving us with $130 in equity. Because our equity grew three times faster than the assets, the leverage shrunk from 3x to about 2.5x.
- Before the morning of day 2, the ETF performs its daily reset to adjust the leverage to 3x. It borrows an additional $60 and increases the assets held to $390.
- On the evening of day 2, the underlying trades at 100 again. The total assets are now worth $355. This figure includes a loan of $260, leaving us with $95 of equity. As before, the leverage is out of whack. It is now 3.75x, requiring further adjustments during the ETF’s daily reset.
To keep the example simple, we did not consider interest payable on the loan. We make the following observations:
- The ETF’s stated leverage only applies for a single day. Holding periods of more than one day can result in returns that are significantly different than the performance of the underlying index.
- In general, leveraged ETFs have too little leverage in rising markets, and too much leverage when prices are declining.
- This effect results in a path dependence or volatility decay, which becomes more pronounced with higher leverage, and increasing volatility.
With these observations in mind, it is time to look at some charts. For this post, we are looking at the Direxion Daily S&P 500 Bull 3x ETF (SPXL). This ETF has an inception date of November 2008. To better illustrate the dynamics of leveraged ETFs, we extended the price history back to January 2008 using a synthetic model.
In almost all years, the leveraged ETF did worse than its 3x leverage suggests: it gained less and lost more than triple the S&P 500’s return. Some examples:
- In 2008, the S&P 500 lost 36%, while SPXL lost 86%.
- In 2009, the S&P gained 26%, while the leveraged ETF gained 65%
- In the first four months of 2020, the S&P lost 15%, while our leveraged ETF lost 52%
Seldom did our leveraged ETF perform better than three times the S&P 500’s return. Examples:
- In 2013, the S&P 500 gained 32%, but our leveraged ETF gained 118%
- In 2017, the S&P 500 gained 22%, while our leveraged ETF gained 71%
- In 2019, the S&P 500 gained 31%, but the leveraged ETF gained 103%
In total, between January 2008 and April 2020, the return of a 3x leveraged S&P 500 ETF is disappointing: the leveraged ETF returned only about 4.6%, while the S&P 500’s total return was about 7.6%. While the geared ETF had a fulminant run after 2009, it lost an exaggerated amount during the 2008 recession.
|Market Regime||Accumulated Duration||Annualized Total Return of S&P 500||Annualized Return of SPXL||Leverage|
|Bull Markets||9.5 years||+24.6%||+78.7%||3.2x|
|Bear Markets||2.8 years||-34.1%||-83.2%||2.4x|
|Low Volatility||8.3 years||+36.6%||+138.6%||3.8x|
|High Volatility||4.1 years||-33.4%||-80.7%||3.4x|
In our previous post about the stock market, we have seen that returns may be quite different, depending on simple conditions. The table shows how historical returns of our 3x leveraged ETF varied, contingent on the market regime. These data lead us to the following conclusions:
- Leveraged ETFs have no place in buy-and-hold portfolios
- Instead, they require constant supervision and active management
- By segmenting the hold time according to market or volatility regime, the characteristics can be vastly improved
In summary, we believe that, if used responsibly, leveraged ETFs can indeed be useful portfolio components, even for conservative investors. We see the following relevant use cases:
- Modest leverage of conservative low-volatility portfolios: In this scenario, we mitigate the low yields of conservative and bond-heavy portfolios, e.g., Tony Robbins’ All-Seasons Portfolio, by applying a modest amount of leverage.
- Aggressive participation in the stock market: In this scenario, we aim to achieve excessive returns while using active management to protect the portfolio downside.
The next product category we want to look at is inverse ETFs. Typical inverse ETFs are available with leverages of one, two, or three. Inverse exposure suggests that we can make money in bear markets or hedge losses of existing long positions.
|Day 1 Morning||100||$100||$200||-$100||-1.00|
|Day 1 Evening||90||$110||$200||-$90||-0.82|
|Day 2 Morning||90||$110||$220||-$110||-1.00|
|Day 2 Evening||100||$98||$220||-$122||-1.22|
|Day 3 Morning||100||$98||$198||-$98||-1.00|
To start, let’s look at the mechanics of an inverse ETF with -1x leverage:
- On the morning of day 1, the underlying index is at 100, and we invest $100 in a 1x inverse ETF. To achieve the inverse exposure, the ETF sells $100 of the underlying and holds $200 in cash.
- On the evening of day 1, the underlying trades at 90, and we owe $90 on our shorted assets. Our cash position did not change, resulting in $110 of equity. However, our leverage is no longer -1x, but reduced to -0.8x.
- Before the morning of day 2, the ETF performs its daily reset. It increases the short position by selling another $20 of the underlying, adjusting the leverage back to -1x.
- On the evening of day 2, the underlying trades at 100 again. We now owe $122 on our short position. Together with $220 of cash, we have $98 of equity. Our leverage is now -1.2x, requiring adjustments during the next reset.
Because we are short-selling the underlying asset, we carry a positive cash balance on which we may receive interest. To keep this example simple, we did not consider any interest received. We make the following observations:
- The ETF’s nominal leverage only applies for a single day. Holding periods of more than one day may result in returns that are significantly different than the inverse performance of the underlying index.
- In general, inverse ETFs have too little leverage in falling markets, and too much leverage when prices are rising. This even applies to inverse ETFs with a leverage of 1x.
For this example, we are taking a closer look at ProShares Short S&P 500 ETF, SH. This ETF is available since June 2006, so we did not need to use a synthetic model to extend the price history. The charts above show how the inverse ETF is declining while the S&P 500 is gaining.
Overall, the inverse ETF is tracking surprisingly well. Here are some examples:
- In 2008, the S&P lost 36%, and our inverse ETF gained 37%
- In 2019, the S&P gained 31%, and the inverse ETF lost 22%
- In the first four months of 2020, the S&P lost 12%, but our inverse ETF gained 5%
|Market Regime||Accumulated Duration||Annualized Total Return of S&P 500||Annualized Return of SH||Leverage|
|Bull Markets||9.5 years||+24.6%||-20.9%||-0.9x|
|Bear Markets||2.8 years||-34.1%||+34.0%||-1.0x|
|Low Volatility||8.3 years||+36.6%||-27.1%||-0.7x|
|High Volatility||4.1 years||-33.0%||+35.0%||-1.1x|
In the period between January 2008 and April 2020, the S&P gained 7.8% annually, while the inverse ETF lost 10.8% per year. The table segments the annualized returns based on market and volatility regimes.
Based on these data, we make the following observations:
- Overall, the 1x inverse ETF tracks its benchmark quite well
- Tracking to the benchmark is mostly independent of the market regime
Encouraged by these results, curious investors might wonder about the characteristics of levered inverse ETFs. For this example, we use ProShares UltraPro Short S&P 500 ETF, SPXU.
This chart looks dramatically different from the 1x inverse ETF. As expected, the 3x inverse ETF is steadily declining. But the steepness of the decline leaves only brief periods of gains in the ETF. While SPXU was turning a profit in 2008, it suffered unexpectedly large losses in 2011, 2015, or the first four months of 2020.
Return of S&P 500
Return of SPXU
|Bull Markets||9.5 years||+24.6%||-52.53%||-2.1x|
|Bear Markets||2.8 years||-34.1%||+54.42%||-1.6x|
|Low Volatility||8.3 years||+36.55%||-62.43%||-1.7x|
|High Volatility||4.1 years||-33.4%||+72.9%||-2.2x|
The conditional returns show that across all market regimes, the 3x inverse ETF fails to deliver the desired triple-inverse performance. Instead, the achieved leverage is less than 2x where it counts: in bear markets. In the long term, the product leads to a stunning average loss of 38% per year.
Reviewing these data, we arrive at the following conclusions:
- Because the stock market is strongly biased towards growth, inverse ETFs are no buy-and-hold instruments.
- Instead, they require constant supervision and active management
- While 1x inverse ETFs show mostly docile behavior, inverse leveraged ETFs fail to achieve their nominal leverage when held over extended periods
- Leveraged inverse ETFs require a much faster reaction to shifting market conditions than unleveraged products
- The wide spread of conditional returns suggests that active management is possible
Based on these findings, we identify the following use cases for inverse ETFs:
- Opportunities in bear markets: In this scenario, we aim to achieve gains while markets are declining. The characteristics of 1x inverse ETFs seem better suited for this use case than their counterparts with higher leverage.
- Hedging portfolio exposure: In this scenario, we aim to counteract the losses incurred by other portfolio positions, mainly to avoid realizing capital gains. For this use case, inverse leveraged ETFs might be useful, but, due to their high costs, only for brief holding periods.
Because of their sometimes finicky characteristics, CLPs are considered high-risk investments. Consequently, their use often requires prior trading experience and the signing of additional disclosures. Here is the policy used by Bertram Solutions’ custodian:
Approval for Complex or Leveraged Exchange-Traded Products requires a minimum of two years of stocks and two years of either options or futures trading experience, or two years of either stocks, options, futures experience and the completion of the CLP Teaching Exam.
This policy provides another indication that these products should be used sparingly and in a thoughtful manner. At Bertram Solutions, we have strict limitations regarding the percentage of assets that may be allocated to these products.
We hope this post provided a helpful overview of complex or leveraged products. Bertram Solutions offers comprehensive wealth management services to retail investors, specialized in quantitative analysis and active management. Please contact us to learn more.