I publish once a month the drifts of major leveraged ETFs on a trailing month and year. After the last issue, a reader asked if I could write a post on long-term drifts. Here it is, with calculations on 3 and 7 years. Only the biotechnology ETFs have disappeared from my usual list in a 7-year look-back. This may be useful for anyone using leveraged ETFs for investing, trading or hedging. Most of leveraged ETFs drifts are due to beta-slippage. Many investors assume the drift is always negative and think leveraged ETFs are systematic wealth destroyers. Some numbers below invite them to revise their opinions.
3-year and 7-year drifts on 2/24/2019
The definitions and formulas are the same as in my monthly dashboard. “Lev” is the leveraging factor. “Return” is the total return of an ETF (including dividends). “IndexReturn” is the total return of the underlying index, measured on a non-leveraged ETF (also with dividends). “ETFdrift” is the drift of the ETF relative to the leveraged index. “TradeDrift” is the drift relative to an equivalent position in the non-leveraged index. ETFdrift and TradeDrift are calculated as followed, where Abs is the absolute value operator.
ETFdrift = Return - (IndexReturn x Lev)
TradeDrift = ETFdrift / Abs(Lev.)
“Decay” is negative drift.
Only durations are different from the usual dashboard: “3 years” stands for 756 trading days, “7 years” for 1764 trading days.
A drift is a difference between 2 returns, so it can be below -100%.
| Index | Lev. | Ticker | 3-year Return | 3-year ETFdrift | 3-year TradeDrift | 7-year Return | 7-year ETFdrift | 7-year TradeDrift |
| S&P 500 | 1 | 52.07% | 0.00% | 0.00% | 136.25% | 0.00% | 0.00% | |
| 3 | 168.92% | 12.71% | 4.24% | 659.78% | 251.03% | 83.68% | ||
| -3 | -75.91% | 80.30% | 26.77% | -95.89% | 312.86% | 104.29% | ||
| ICE US20+ Tbond | 1 | 0.17% | 0.00% | 0.00% | 25.64% | 0.00% | 0.00% | |
| 3 | -17.21% | -17.72% | -5.91% | 24.87% | -52.05% | -17.35% | ||
| -3 | -14.16% | -13.65% | -4.55% | -75.39% | 1.53% | 0.51% | ||
| NASDAQ 100 | 1 | 72.25% | 0.00% | 0.00% | 193.72% | 0.00% | 0.00% | |
| 3 | 252.11% | 35.36% | 11.79% | 1122.10% | 540.94% | 180.31% | ||
| -3 | -86.87% | 129.88% | 43.29% | -98.60% | 482.56% | 160.85% | ||
| DJ 30 | 1 | 67.70% | 0.00% | 0.00% | 137.24% | 0.00% | 0.00% | |
| 3 | 257.85% | 54.75% | 18.25% | 680.21% | 268.49% | 89.50% | ||
| -3 | -82.30% | 120.80% | 40.27% | -95.91% | 315.81% | 105.27% | ||
| Russell 2000 | 1 | 62.17% | 0.00% | 0.00% | 111.56% | 0.00% | 0.00% | |
| 3 | 202.17% | 15.66% | 5.22% | 346.90% | 12.22% | 4.07% | ||
| -3 | -84.04% | 102.47% | 34.16% | -96.87% | 237.81% | 79.27% | ||
| S&P Select Energy | 1 | 23.20% | 0.00% | 0.00% | 5.01% | 0.00% | 0.00% | |
| 3 | 16.23% | -53.37% | -17.79% | -59.68% | -74.71% | -24.90% | ||
| -3 | -70.31% | -0.71% | -0.24% | -84.53% | -69.50% | -23.17% | ||
| MSCI US REIT | 1 | 27.91% | 0.00% | 0.00% | 82.83% | 0.00% | 0.00% | |
| 3 | 57.24% | -26.49% | -8.83% | 218.71% | -29.78% | -9.93% | ||
| -3 | -65.45% | 18.28% | 6.09% | -94.01% | 154.48% | 51.49% | ||
| ARCA Gold Miners | 1 | 26.48% | 0.00% | 0.00% | -55.84% | 0.00% | 0.00% | |
| 3 | -46.00% | -125.44% | -41.81% | -99.76% | 67.76% | 22.59% | ||
| -3 | -92.98% | -13.54% | -4.51% | -97.88% | -265.40% | -88.47% | ||
| MSCI Emerging | 1 | 47.51% | 0.00% | 0.00% | 13.55% | 0.00% | 0.00% | |
| 3 | 109.67% | -32.86% | -10.95% | -45.12% | -85.77% | -28.59% | ||
| -3 | -81.75% | 60.78% | 20.26% | -85.57% | -44.92% | -14.97% | ||
| Gold spot | 1 | 8.67% | 0.00% | 0.00% | -25.30% | 0.00% | 0.00% | |
| 3 | -2.05% | -28.06% | -9.35% | -79.68% | -3.78% | -1.26% | ||
| -3 | -31.04% | -5.03% | -1.68% | -0.32% | -76.22% | -25.41% | ||
| Silver spot | 1 | 3.60% | 0.00% | 0.00% | -54.10% | 0.00% | 0.00% | |
| 3 | -37.51% | -48.31% | -16.10% | -98.33% | 63.97% | 21.32% | ||
| -3 | -49.09% | -38.29% | -12.76% | -19.10% | -181.40% | -60.47% | ||
| Wells Fargo BDC | 1 | 45.61% | 0.00% | 0.00% | 59.54% | 0.00% | 0.00% | |
| 2 | 91.85% | 0.63% | 0.32% | 121.72% | 2.64% | 1.32% | ||
| PHLX Semicond. | 1 | 130.46% | 0.00% | 0.00% | 242.62% | 0.00% | 0.00% | |
| 3 | 578.38% | 187.00% | 62.33% | 1164.24% | 436.38% | 145.46% | ||
| -3 | -96.78% | 294.60% | 98.20% | -99.71% | 628.15% | 209.38% | ||
| VIX ST Futures | 1 | -92.09% | 0.00% | 0.00% | -99.56% | 0.00% | 0.00% | |
| 2 | -99.86% | 84.32% | 42.16% | -100.00% | 99.12% | 49.56% |
*BDCL and TVIX are exchange-traded notes (ETNs).
The Worst and the Best Drifts
In 3 years:
-
The leveraged ETF in gold miners (NUGT) has the worst decay with a drift close to -42% normalized to 1x the underlying index exposure.
-
The worst absolute loss is for the leveraged volatility ETN (TVIX) losing 99.86% in 3 years. The non-leveraged volatility ETF (VIXY), inverse ETFs in semiconductors (SOXS), miners (DUST) have also lost more than 90% of their values in 3 years.
-
The highest positive drift with a gain is for leveraged semiconductors (SOXL), with a drift above 62% relative to 1x the underlying exposure. This is real outperformance over the leveraging factor.
-
The highest positive drift with a loss is for the inverse leveraged Nasdaq 100 product (SQQQ), with a 43% drift relative to 1x the underlying exposure. This is an asymptotic loss due to compounding negative leveraged daily returns.
In 7 years:
-
The inverse leveraged ETF in gold miners (DUST) has the worst decay with a drift about -88% normalized to 1x the underlying index exposure.
-
The worst absolute loss is for the leveraged volatility ETN (TVIX), losing 100% in 7 years (rounded, meaning at least 99.995%). VIXY, SOXS, DUST have also lost more than 99% of their values.
-
The highest positive drift with a gain is for the leveraged Nasdaq 100 ETF (TQQQ), with a positive drift about 180% relative to 1x the underlying exposure. This is outperformance.
-
The highest positive drift with a loss is for inverse leveraged semiconductors (SOXS). This is an asymptotic loss.
A Lesson on Trend and Volatility
All leveraged ETFs on diversified stock indexes (S&P 500, Dow Jones, Nasdaq 100, Russell 2000), long and inverse, show positive drifts on 3 and 7 years. This is typical of a steady bull market, despite 2018 volatility. It means not only that long ETFs have outpassed their leveraging factors, but also that inverse ETFs have been cheap hedging tools, even losing most of their values on the way. However, the real cost of hedging depends on when the hedging position is rebalanced. These diversified indexes products would likely show negative drifts in a bear market, not because of a change in trend, but because bear markets usually have a high daily volatility bringing a high beta-slippage.
None of these past returns and drifts should be taken as an indication of future performance. However, we can see a general pattern: leveraged ETFs generate outperformance in trending markets and fall in decay when volatility comes. As a consequence, the more diversified is an underlying, the less volatile it is, and the less risky are leveraged ETFs as mid-term or long-term holdings. Keep also in mind that shorting an asset or buying an inverse product may imply a systematic decay due to inflation, and that leveraging amplifies the inflation effects on the long and the short sides.
Disclosure: I am/we are long QQQ,SPXU. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I hold SPXU as partial hedge of my stock portfolio.
======================
Just in the last year, (April 2018 to April 2019), SPY returned 6%, UPRO returned 11% ---- and we had volatility -- we had a 20% correction in December 2018.
Explain that?Still think Leveraged funds are not meant for the long-term?
Thanks for the multi year summary.
At the risk of being called lazy for not doing it myself, perhaps a dot plot or graph(s) of the data would bring it to life a bit more, than the table with numbers..
can you please elaborate a bit more on the following:As a consequence, the more diversified is an underlying, the less volatile it is, and the less risky are leveraged ETFs as mid-term or long-term holdings.
In retrospect I should have added one more question - maybe for next time - as I always find it confusing not being able to compare -3x leveraged products to the inverse (-1x) non-leveraged counterparts. I think it would help if we could compare, say, SPXU to SH.
Thanks again,
drftrThe high positive drift of SOXL is not typical of an industry ETF and might not continue this way.
On the other hand, do you consider QQQ/TQQQ as diversified enough.