This article takes about 6 minutes to read

The Fundamental Principle of Dollar-Cost Averaging (DCA)

Dollar-Cost Averaging (DCA) is the practice of investing a fixed amount at regular intervals. When prices are high, fewer shares are purchased; when prices are low, more shares are purchased, resulting in an average cost basis lower than the time-weighted average.

DCA into conventional ETFs (VOO, QQQ, etc.) is widely recommended. It eliminates the need to time the market and removes emotional decision-making. However, DCA into 3x leveraged ETFs is often considered 'unconventional' due to the prevailing wisdom that leveraged ETFs are unsuitable for long-term holding.

This article tests that conventional wisdom with hard numbers. Is DCA into 3x ETFs truly pointless, or is it an effective strategy under certain conditions? We use 15 years of data to reach a conclusion.

Theoretical Rationale for Applying DCA to 3x ETFs

The greatest enemy of 3x leveraged ETFs is volatility decay. Daily rebalancing causes returns to fall below 3x the underlying index in choppy markets. This decay accumulates with longer holding periods.

The theoretical mechanism by which DCA mitigates this problem is as follows: during crashes, more shares are purchased at low prices, amplifying returns during recovery. With lump-sum investing, there is risk of deploying all capital at pre-crash highs, but DCA distributes this risk across time.

Furthermore, 3x ETF crash magnitudes far exceed conventional ETFs (-70% to -90%). DCA enables purchasing large quantities at deeply discounted prices post-crash, making the 'buying cheap' effect far more pronounced with 3x ETFs than with conventional ETFs. This is the theoretical basis for DCA's compatibility with 3x ETFs.

Simulation Parameters

The simulation was conducted under the following conditions. Investment targets: TQQQ (NASDAQ-100 3x) and SPXL (S&P 500 3x). Period: January 2010 through March 2025 (approximately 15 years). DCA: 50,000 yen invested at the beginning of each month. Lump-sum: 9 million yen (equal to 15 years of DCA contributions) invested in January 2010.

For comparison, DCA into QQQ (NASDAQ-100 1x) and VOO (S&P 500 1x) under identical conditions was also calculated. Dividends are reinvested, and currency effects are excluded (USD-denominated).

Note: Since TQQQ's inception date was February 2010, the January 2010 data uses simulated values based on 3x the NASDAQ-100's daily returns.

Simulation Results

TQQQ DCA (50,000 yen/month x 15 years = 9 million yen total investment) produced terminal wealth of approximately 120 million yen, an annualized return of approximately 28%. Under the same conditions, QQQ DCA yielded approximately 32 million yen (annualized approximately 14%), and VOO DCA approximately 24 million yen (annualized approximately 11%).

SPXL DCA produced approximately 55 million yen (annualized approximately 22%). TQQQ lump-sum (9 million yen in January 2010) reached approximately 350 million yen (annualized approximately 35%), far exceeding DCA. This is because 2010 was near the market bottom, making lump-sum timing fortuitous.

Comparing maximum drawdowns: TQQQ DCA experienced -65% (2022), TQQQ lump-sum -72% (2022), and QQQ DCA -28% (2022). DCA reduced drawdown by approximately 7 percentage points compared to lump-sum.

Why DCA Is Particularly Effective with 3x ETFs

The simulation results reveal the mechanism by which DCA functions particularly well with 3x ETFs. When TQQQ crashed -70% during the March 2020 COVID shock, DCA purchased over 3x the normal number of shares at deeply discounted prices.

Specifically, 50,000 yen purchased approximately 50 shares of TQQQ in February 2020, but the same 50,000 yen purchased approximately 170 shares in March 2020. This 'buying large quantities cheaply' effect generated explosive returns during the subsequent recovery.

Conventional ETFs (QQQ) experience the same effect, but with declines limited to approximately -30%, the 'buying cheap' benefit is modest. A 3x ETF's -70% decline represents a 'once-in-a-generation opportunity to buy massive quantities at deep discounts' for DCA. This asymmetry is the essence of DCA's compatibility with 3x ETFs.

Risks That DCA Cannot Eliminate

DCA is not a panacea. During prolonged downtrends, DCA means 'continuously buying a declining asset,' accumulating losses. If TQQQ were to decline for five consecutive years, the majority of DCA-invested capital would be destroyed.

If a dot-com bubble collapse scenario (2000-2002) were to repeat, the NASDAQ-100 would fall -80%, and at 3x leverage the product would approach zero. Even with monthly DCA contributions, recovery could take over 10 years.

Additionally, DCA always underperforms optimal-timing lump-sum investing. When markets rise over the long term (as U.S. equities have for the past 100 years), deploying all capital early is statistically superior. DCA's advantage is limited to the premise that 'timing cannot be predicted.'

Testing Optimal Contribution Frequency

Comparing daily, weekly, and monthly contribution frequencies, the difference in terminal returns was surprisingly small. For 15 years of TQQQ contributions, daily yielded 28.3% annualized, weekly 28.1%, and monthly 27.8%, with differences within 0.5%.

Theoretically, higher frequency allows the 'buying more on cheap days' effect to operate more precisely, but the actual difference is negligible when considering transaction costs and effort. Monthly contributions deliver sufficient effectiveness.

However, a 'modified DCA' with additional contributions during crashes shows significant benefit. Adding an extra 100,000 yen whenever TQQQ falls -40% or more from its recent high, on top of the regular 50,000 yen monthly contribution, improved annualized returns by approximately 2-3%.

The Relationship Between Compounding and DCA's Synergy

The relationship between DCA and compounding is particularly interesting with 3x ETFs. Standard compound interest grows wealth as 'principal x (1 + r)^n,' but with DCA, new principal is added monthly, with each month's contribution compounding over a different time horizon.

Capital invested early compounds over 15 years, while the final contribution compounds for just one month. At 25% annualized return for a 3x ETF, the first month's 50,000 yen grows to approximately 1.4 million yen after 15 years, while the final month's 50,000 yen remains at 50,000 yen.

For those wanting to deeply understand the relationship between systematic investing and compounding, systematic investing guides on Amazon can strengthen the mathematical foundation for designing leveraged ETF DCA strategies.

Designing a Practical DCA Strategy

When implementing DCA into 3x ETFs, the following design is recommended. First, limit investment targets to TQQQ or SPXL. Sector 3x ETFs have insufficient liquidity for long-term systematic investing. Second, keep contribution amounts within 10-20% of the overall portfolio.

Third, do not set stop-loss rules. DCA's essence is 'buying cheaply during declines,' so selling during downturns negates the strategy. Exceptions apply only when structural problems arise (ETF delisting risk, fundamental index changes).

Fourth, establish profit-taking rules in advance. Mechanical rules like 'take profits on half when terminal value reaches 5x contributions' or 'halve contribution amounts after three consecutive years of 30%+ returns' remove emotion. While long-term holding is fundamental for maximizing compounding, an exit strategy to avoid excessive concentration risk is also essential.

Finally, DCA is a strategy premised on 'markets rise over the long term.' As long as the U.S. equity market continues satisfying this premise, DCA into 3x ETFs has the potential to dramatically outperform DCA into conventional ETFs. However, if this premise breaks down, losses are also amplified 3x.