During "ETF Month", I plan to survey a number of ETFs that have a lot of followers on Seeking Alpha, but for some reason don't have much coverage on the site. Hopefully, this will improve the visibility of Cambridge Income Laboratory and draw more members to our service, allowing me to make the newsletter better than ever. I will also be taking ETF suggestions, so do let me know if you have any ETFs on your horizon.
What are the advantages and disadvantages of ETFs (exchange-traded funds) versus CEFs (closed-end funds)?
- Most ETFs are passively managed. The advantage of this is lower fees compared to CEFs, which are actively managed. The disadvantage is that a passive fund will simply buy everything in the index, indiscriminately.
- ETFs will nearly always trade close to their net asset value [NAV]. The advantage of this is that one does not have to worry about premiums or discounts. The disadvantage is that one does not have the opportunity to buy funds at a discount, nor to exploit the concept of premium/discount mean reversion.
- ETFs usually do not employ a managed distribution policy, in other words, they pay out as dividends what they receive as income from their underlying investments. The advantage is that one doesn't have to worry about an ETF overpaying from its earnings, leading to destructive ROC. The disadvantage is that ETFs are generally lower yielding than CEFs.
For the inaugural edition of ETF Month, where I go into more detail about the rationale behind this series, please click here.
ETF Month #3: Direxion Daily Gold Miners Index Bear 3x Shares ETF (NYSEARCA:DUST)
All data from the article are current as of March 5, 2018.
Gold is one of the most hotly followed topics on Seeking Alpha. 203,727 users follow the ticker for SPDR Gold Trust ETF (GLD), while the VanEck Vectors Gold Miners ETF (GDX), often considered to be a leveraged play on gold, has 64,764 followers.
Those who are more daring may consider leveraged plays on gold or gold miners. I was somewhat surprised to see that not a few number on Seeking Alpha apparently have such an inclination, with 39,583 following the Direxion Daily Gold Miners Bull 3X ETF (NUGT) and 19,507 following the Direxion Daily Gold Miners Index Bear 3x Shares ETF.
Unfortunately, NUGT is a very poor way to invest in gold miners for the long term due to leverage decay, also known as beta slippage. For example, the gold miner ETF has returned -2.42% in the past year, but the 3x-leveraged NUGT has fallen by -32.8%, much worse than three times of -2.42% (i.e., -7.26%).
Enter DUST
Can we put leverage decay to good use? Let's consider DUST, the 3x-leveraged inverse play on gold miners.
Basic details of DUST are shown in the table below. DUST is a highly liquid fund, with 6.6 million shares changing hands daily and an AUM of $242 million. Annual expenses are 1.08%.
DUST Yield [ttm] 0.00% Expense ratio 1.08% Inception Dec. 2010 AUM $242m Avg. Volume 6.6m Morningstar rating n/a No. holdings 50* Annual turnover 26%* Investment mandate DUST seeks to track 300% of the inverse of the performance of the NYSE Arca Gold Miners Index.
(Source: Morningstar. *Denotes data of GDX.)
Since GDX went down by -2.42% last year, we'd expect DUST to have gone up by +7.42% (negative three times of -2.42%), right?
Wrong! DUST went down by -21.9%, thanks again to leverage decay.
In other words, shorting DUST (equivalent to going long gold miners) would have given a much better performance over the last year than actually going long gold miners with GDX, thanks to the fact that we are now benefiting from leverage decay.
The outperformance remains even if we normalize for exposure. For instance, say an investor put $30,000 into GDX one year ago. After 1 year, he would be down $726, or -2.42%. Alternatively, the investor could short $10,000 of DUST (since DUST is 3x the inverse of GDX) and keep the remaining $20,000 in cash. After 1 year, the short DUST position would have gained $2,190, or +21.90%, for a total return of +7.30% on the overall portfolio (which is the DUST position plus cash).
Strategy
Let's now do a backtest of this strategy. The rules are simple:
- Buy $30,000 of GDX, -or- short $10,000 of DUST and keep $20,000 of cash.
- Compare results one year later.
That's it!
Results
DUST was incepted on Dec. 8, 2010, so there have been 1,567 observations of 1-year returns since then.
The chart below shows the 1-year trailing returns of either going long GDX or short DUST (with 33%, the remaining 67% in cash) for all ending dates for which we have data. We can see that for most of the time since inception of DUST, the short DUST strategy (red line) would have outperformed the long GDX strategy (blue line). To be precise, short DUST outperformed long GDX on 84% of the 1-year observation periods.
(Source: Stanford Chemist, Yahoo Finance)
The table below shows the summary statistics of the two strategies. The average 1-year return of short DUST was 3.73%, which was over 12 percentage points higher than for long GDX, which had an average 1-year return of -8.86%. The median for short DUST (9.15%) was 25 percentage points higher than the median for long GDX (-15.88%). However, the max gain of short DUST was only 32.90% compared to 134.99% for long GDX. This is because the short DUST strategy has two-thirds of its portfolio in cash, and since the gain from the one-third short position is always capped at 100%, this means that the maximum gain of the blended portfolio can only be 33.3%. The maximum loss of the short DUST strategy (-85.96%) was also much higher than for the long GDX strategy (-56.74%).
short DUST long GDX average (1-year return) 3.73% -8.86% median (1-year return) 9.15% -15.88% max gain 32.90% 134.99% max loss -85.96% -56.74%
(Source: Stanford Chemist, Yahoo Finance)
Let's take a look at the histogram of 1-year returns. We can see clearly that the 1-year returns of the short DUST strategy are generally clustered towards a higher return value compared to the long GDX strategy. Moreover, we can also see from this chart that two of the features that we pointed out in the table above, namely, that (1) the short DUST strategy had more instances of greater losses (exceeding -60%) compared to long GDX, and (2) the long GDX strategy had more instances of greater gains (exceeding 40%) compared to short DUST.
(Source: Stanford Chemist, Yahoo Finance)
Risks
Of course, with the short DUST strategy you have all of the risks of shorting, including having your position called at any time, and paying short fees. A short position also has the possibility of suffering an unlimited loss. We don't have to worry about paying the dividend in this case because DUST has no yield. According to IBorrowDesk, DUST has been fairly easy to short over the past several years, with an average borrow rate estimated to be about 8%. As the borrow is only paid on 1/3 of the position in the short DUST strategy, this works out to be around ~2.7% per year on the overall portfolio, which is still far less than the outperformance of the short DUST strategy vs. the long GDX strategy. In other words, the leverage decay of DUST far exceeded the expenses required to short the note.
(Source: IBorrowDesk)
Additionally, as I pointed out above, the short DUST strategy had a number of instances of far greater losses than the long GDX strategy. This occurred during periods where the underlying index suffered a large drawdown, causing DUST to more than double in value inflicting large losses on the short DUST strategy.
On the other hand, the short DUST strategy also underperformed when the underlying index rose strongly over a 1-year period. As explained above, this is because short DUST strategy (with 2/3 of the portfolio in cash) has a maximum upside of 33.3% in any 1-year period.
Therefore, we can reach the conclusion that the short DUST strategy will outperform the long GDX strategy in flat or volatile markets, but will underperform when gold miners are trending strongly upwards or strongly downwards.
Summary
This article examined the short DUST strategy as an alternative to going long GDX. The short DUST strategy aims to exploit leverage decay, which degrades the return of any leveraged investment in a flat or volatile market. However, short DUST may underperform GDX in a strongly trending market.
I've previously discussed shorting strategies on leveraged ETFs to take advantage of leverage decay here:
- Shorting Leveraged ETF Pairs: Easier Said Than Done (Apr. 2015)
- Another Look At The NUGT-DUST Double Short (Mar. 2016)
If you have enjoyed my article, please click the "Follow" button next to my name to be alerted to my new free content! The Cambridge Income Laboratory is my Marketplace service on Seeking Alpha focused on income and arbitrage strategies for closed-end fund (CEF) and exchange-traded fund (ETF) portfolios. Members receive exclusive subscriber articles and an early look at public content with more actionable recommendations and ideas.
To find out why one subscriber calls us a "one-stop shop for CEF research," join us by clicking on the following link: Cambridge Income Laboratory.
