Leveraged ETFs for Long-Term Investing? Smart or Reckless?

I’ve been thinking about leveraged ETFs (LETFs) recently and their viability as long-term investment vehicles—particularly for someone like me who still has a long investment horizon.

LETFs, for those who don’t know, are passively managed funds that use financial derivatives and debt to amplify the returns of an underlying index. A popular one right now is TQQQ, which is leveraged to achieve 3x the performance of QQQ. This means if QQQ goes up 10%, TQQQ goes up 30%. Of course, the reverse is also true—if QQQ goes down 10%, TQQQ drops 30%. It's a classic high-risk, high-reward situation.

Opinions are split on whether LETFs like TQQQ should only be used for short-term strategies or if they can work for long-term buy-and-hold strategies. The biggest argument against holding TQQQ long-term is the concept of "volatility drag." In simple terms, volatility drag means you can lose money even when the average return is 0%. For example, let’s say you invest $100 in TQQQ. If the underlying index (QQQ) goes up 10%, your investment increases to $130. But if QQQ then drops 10%, you lose 30% of that $130, leaving you with $91. Even though the average return is 0%, you lost $9 due to volatility drag.

Critics of long-term LETF holding also point out the risk of losing your entire portfolio during a bear market. If QQQ lost 33% in a single day, theoretically, TQQQ could lose most, if not all, of its value. While fund managers may rebalance to avoid a total collapse, the risk of massive losses remains. This is why timing your entry is especially critical with a LETF. Invest at the start of a recession, and you might lose everything—but invest at the beginning of a bull run, and you could significantly beat the market.

I ran some backtesting to see how your entry date into a LETF impacts returns, holding all other factors constant. I wrote some Python code that allows you to input a ticker, a start/end date, and a holding period. The code calculates what your returns would be for each possible entry date and exports it to a CSV for analysis. If you’re comfortable with Jupyter notebooks, you can try out the code here: https://colab.research.google.com/drive/1L1b_6A4-9nSIr35pADX4QCmw9Z_wAslx?usp=sharing

Note that this is a very simple simulation, it simply compares how different entry days would've performed assuming a lump-sum investment and a set holding period. These results would change significantly if you employ a dollar cost averaging strategy or rebalanced your portfolio overtime.

Here’s what I found: For TQQQ, with a 5-year holding period, the worst entry date was 1/5/18, yielding a positive total return over 5 years of 33.42%. The best day to enter was 2/9/16, where your total return would’ve been a whopping 1727%. The average total return for all possible 5Y holding periods was 582%. These results are remarkable, but it’s important to consider that TQQQ’s inception date was in 2010—after the Great Recession and just in time for the largest bull run the stock market has ever seen.?

The next LETF I looked at was SSO, a 2x leveraged ETF based on the S&P 500. I chose this LETF because it is one of the oldest LETF’s currently tradable, with an inception date of 6/19/2006. With SSO, the worst entry date (with a 5-year holding period) was 10/04/06, resulting in a total return of -50%. The best entry date was 3/09/09, with a total return of 655%. The average total return was 156%. Since SSO’s inception, there were 3,463 possible entry days for a 5-year holding period. Of those, 425 days would have resulted in negative returns, while 3,038 days would have resulted in positive gains. In other words, 86% of the days since SSO’s inception would have led to positive returns over a 5-year holding period.

This data has me rethinking the traditional view that LETFs are only suitable for short-term strategies. However, we should keep these factors in mind: The results don't account for the higher expense ratios of LETFs, which can erode gains over time, especially in flat or sideways markets Survivorship bias also plays a big role here, as many LETFs that launched during volatile periods have completely failed and were delisted. Recency bias is also important, as past results do not guarantee future results.

I'm not taking a stance in either direction right now but I am curious about what other people think. Should TQQQ only be used for short-term bets on the market? Why or why not should someone invest in TQQQ for the long run?

If you'd like to see the data for the TQQQ and SSO examples, you can view the file here: https://docs.google.com/spreadsheets/d/1_3as-HX7kMVSi6NO2A-zZQJ-O_QcdVFwmm2e6PbNynU/edit?usp=sharing

P.S. I plan to write a follow-up article on this subject after hearing your thoughts. In the follow up I will discuss the history of LETFs and inverse ETFs and how many have been delisted since their creation. I will also delve into the idea of timing entry into TQQQ based on technical factors such as 200 day EMA, and review the fundamentals behind QQQ.

?