*You can find the code for the following blog post here, including example code on how to use Intrinio’s SDK in Python and how to apply transformations to feed data in Pandas.
In the last two decades, Exchange Traded Funds (ETFs) have become the single most important investment vehicle. They have completely changed the trading landscape by offering a low-cost, low-effort solution for retail investors to diversify their portfolios, while providing immense arbitrage opportunities for experienced traders, market makers, and Authorised Participants (APs).
An ETF can be distinguished from its mutual fund and index fund counterparts by the creation and redemption process. This involves an AP to act as an intermediary for the fund so that the fund itself does not have to actively rebalance its portfolio.
The creation/redemption mechanism of ETFs means that minimal to no fees are charged to shareholders, as opposed to mutual or index funds, where existing investors must cover the transaction costs for the fund when money leaves or enters the fund. Typically, most of the cost of owning a stake in an ETF is when an investor crosses the spread to purchase or sell said ETF (this is how APs make money).
Beta Slippage/Volatility Decay
As the popularity of ETFs has grown, a number of leveraged funds have popped up too. A leveraged ETF enables an investor to be exposed to a higher rate of risk with less capital. On the surface this seems like a great investment for young retail investors, who are risk seeking, have little capital, and have lots of time to reap the rewards of a high-risk passive investment. This could not be further from the truth. Leveraged ETFs are only suitable for experienced traders and should not be touched by investors who have either a longer-term horizon or want a passive income.
Aside from the obvious fact that leverage funds charge considerably higher management fees to incorporate the expense of leveraging and active rebalancing their funds, these ETFs, over longer time periods, suffer from something called beta slippage or volatility decay.
A leveraged fund typically advertises a multiple return on an underlying ETF. For example, TQQQ claims that it returns three times the profits of the SPY 500. While these claims are for the most part true amongst reputable leveraged ETFs, you must keep in mind that their claims only apply on a daily basis and DO NOT apply for long run investments. For example, if the S&P 500 increases 20% over two years, it is not reasonable to expect the TQQQ to increase 60% during the same period.
To understand why this is, let’s first look at some hypothetical scenarios. In this first scenario, let’s imagine that every day our underlying ETF increases by $1 consistently:
Assuming that our ETF continues to progress in the same trend, we can see that our leveraged ETF will return compounded returns on our underlying and would have actually been a better buy than our underlying.
In this case, our underlying ETF returned 50% over a 50-day time period, while our leveraged ETF returned 234%, significantly outperforming its leverage ratio (50% x 3 = 150%).
One of the traps rookie investors fall for is not accounting for the leveraged ETF being as punishing on downturns as it is rewarding on upticks. Just because an ETF had 50% returns over the past year does not necessarily mean that the leveraged ETF will return 150% or more returns.
To fully understand this, let’s take a look at another scenario where volatility comes into play. Imagine instead that the underlying ETF oscillates directionlessly, moving up $10 and then down $10 every day.
In a volatile environment, the compounding effects of a leveraged ETF now come back to haunt us, resulting in volatility decay.
In case you were wondering if a bear ETF would magically reverse the volatility decay, you would still be sorely mistaken.
As Jason Zweig states in his commentary of Benjamin Graham’s The Intelligent Investor:
“Once you lose 95% of your money, you have to gain 1,900% just to get back to where you started.”
This mathematical concept is at the heart of beta slippage, which refers to a leveraged ETF underperforming its leveraging factor over an extended time period.
Now let’s see how this theory plays out in the real world, using historical data from Intrinio’s US Core Market Data Feed. In particular, let’s focus on SPY and SPXL. SPY is an ETF that tracks S&P 500 one-to-one, with an extremely low management expense ratio of 0.09%. SPXL is a leveraged ETF, managed by Direxion, that aims to deliver 300% of the price performance of the S&P 500 index. It includes a significant expense ratio of 1.01%.
Firstly, let’s look at whether the leverage ratio actually promises the true ratio on a daily basis by graphing the daily ratio on days where the S&P 500 moves at least 10 basis points (0.10%).
As you can see above, SPXL does quite a good job of delivering its promise of 300% daily returns when the S&P 500 has moved significantly. Only on two occasions did it move in the opposite direction of the index, and the majority of the time the leveraging factor was between 2.8-3.05.
Now let’s see how it’s performed over the past five years:
You would be barely better off today if you had invested in SPXL (+149%) vs SPY (+148%) five years ago, but you would have risked those same returns for three times the volatility. The max drawdown of SPXL is a whopping 77% (SPXL lost 77% of its value from a peak to a trough) compared to only 34% for SPY. As we saw above, this has huge implications for the long-run performance of SPXL which would need returns of 435% to recover its peak value after a 77% drawdown.
This analysis may seem unfair, as we are currently amidst a financial phenomenon of almost unheard-of volatility. Let’s instead say we look at the returns of SPXL and SPY up till the start of 2020. In this scenario, SPXL and SPY would have return +307% and +169% returns respectively. Even though SPXL’s returns would have been significantly greater, it would still have performed below its leverage factor (169% * 3 = 507%) and therefore still offered a significantly worse Sharpe Ratio than SPY.
A closer inspection at the above graph would highlight how poorly SPXL performs during volatile downturns, such as early and late 2018 and most notably Feb-March 2020. You might also notice how exceptionally well SPXL performs during periods of flat growth, such as in 2017 and the fourth quarter of 2019.
To see the effect that directionless volatility has on SPXL’s beta slippage, let’s have a look at the rolling volatility of SPY and a moving average of SPXL’s returns using a one-month (20 trading days) window to see the negative correlation between the two.
To conclude, leveraged ETFs not only require more active management resulting in higher management fees and less reliable tracking of the index, they also expose you to volatility in the long run. They should therefore be used with a grain of salt by either day traders or experienced investors.
Interested in more content from me? Check out my blog.
This blog is part of our Blogger Program. If you’re a fintech, finance, or data science blogger who’s interested in getting free data and gaining exposure to our audience through guest blogging opportunities and other promotion, visit our website to learn more and apply.