3x Leveraged ETFs Outperform Only If the Stock Rises 170% First
Leveraged ETFs track daily returns, not annualized multiples: a 3x product on an asset with 5% daily volatility needs the asset to rise 170% just to outperform the stock itself—thresholds jump exponentially with volatility.
The video won't play here. Listen to the audio instead:
The argument · tap a timestamp to hear it
Leveraged ETFs track daily returns, not annualized multiples
A leveraged ETF's prospectus states it tracks a multiple of the underlying index's daily return, not a multiple of long-term annualized returns. The fund gains leveraged exposure through swap agreements with banks—for example, a $1 billion 2x fund must enter a $2 billion swap on the underlying asset and rebalance the swap size every day after close to maintain its target leverage ratio. This leverage is not free; its cost fluctuates with borrowing rates, and the sharp rise in rates over recent years has directly increased the cost of holding these products.
— Chris MurrayUp 10% then down 10%: stock loses 1%, 2x ETF loses 4%
If an underlying asset gains 10% on day one and loses 10% on day two, the arithmetic average return is zero, but after compounding it actually loses 1%—this is the compounding effect all assets experience, unrelated to leverage. With a daily 2x leveraged ETF, day one gains 20% and day two loses 20%, but the result is not a 2% loss; it's a 4% loss. Of the extra 2%, roughly 1 basis point comes from convexity bonus, and about 2.01% comes from volatility drag. These two are often conflated, but are actually two separate effects.
— Marco SammonDaily reset actually keeps you from going to zero
An underlying asset falls from 100 to 75, then to 50—a cumulative 50% decline. If you apply 2x leverage only once initially and never adjust again, after paying back the debt you'd be at zero. But with a daily-reset 2x leveraged ETF, the first day's decline takes your position from 100 to 50, and the second day you lose two-thirds of that lower base, leaving you with $16.67, not zero. Part of this difference comes from convexity bonus, and part is eroded by volatility drag—the convexity bonus itself comes from daily (or any periodic) reset, which static leverage never provides.
— Marco SammonLeveraged ETF trading costs hide in the closing price
Leveraged ETFs trade only at close, and don't pay transaction costs through bid-ask spreads. Instead, they place large orders into the closing auction, directly pushing the closing price itself higher or lower. Suppose a stock should have closed up 8% on a given day; if a leveraged ETF's purchase pushes that closing price up to 9%, this amplified daily return feeds directly into the volatility drag formula (the sum of squared daily returns), making the drag larger. This mechanism becomes significant only when the leveraged product is large relative to the underlying's liquidity and the day's volatility is pronounced—for example, certain Korean single-stock ETFs.
— Chris MurrayBy arithmetic mean return, infinite leverage is optimal
If you evaluate leveraged products using ‘expected terminal wealth’or ‘expected arithmetic mean return’, you reach a counterintuitive conclusion: as long as the underlying has positive expected returns, the optimal leverage is always infinite, because this metric favors volatility—a fund with 10% annualized returns and 40% volatility, even if it only turned $100 into $102 last year, will rank above a fund with 8% returns, low volatility, that turned $100 into $107 last year. Switch to ‘expected log return’or ‘expected log terminal wealth’, and the optimized leverage falls back into intuitive ranges like 2x or 1.5x.
— Chris MurrayBreak-even thresholds jump exponentially with volatility
The ‘break-even return’metric measures how much an underlying must rise for a 2x or 3x leveraged ETF to outperform the underlying itself. Taking a broad market index with a 4% risk-free rate, 0.5% funding spread, and 1% daily volatility as an example: the threshold is an 8.4% rise in the underlying; at 3% daily volatility, the threshold jumps to 32%; at 5% daily volatility (close to Nvidia-scale), a 2x product's threshold becomes 100%; add another layer of leverage for 3x with the same 5% daily volatility, and the underlying must rise 170% before a 3x ETF outperforms the stock itself.
— Marco SammonLucid's intraday liquidation wipes out the 2x ETF to zero
Lucid's stock gapped down sharply one day due to negative news, prompting market-making banks to worry about insufficient margin in their swap accounts. They chose to liquidate the swap position intraday. The selling pressure from their liquidation pushed the stock price lower still, but by the time the banks finished unwinding, the stock closed down only 16% that day. Investors holding 2x long Lucid leveraged ETF would have expected a loss proportional to 2 times the 16% close—negative 32%—but instead got wiped out to zero. This happened because the liquidation trigger was the intraday low, not the closing price, which is nearly impossible to replicate in historical backtests using closing prices.
— Chris MurrayInverse ETF inflow of $93 billion dwindles to $13 billion
When you account for ‘when the money actually flowed in’, the results contradict intuition: during the sample period, investors in broad market leveraged ETFs, compared to those who put the same capital into the underlying at the same times, collectively made an extra $50 billion—because they didn't all enter at the worst time (like March 2020). Single-stock leveraged ETFs show much smaller differences. But inverse (short) leveraged ETFs are another story: during the sample period they saw cumulative inflows of $93 billion, but only $13 billion in AUM remains, with 86% of principal vanished.
— Marco SammonIn their own words · checked verbatim
And if that term isn't coined, we're going to coin it right here at 947 a.m. Pacific time on the podcast, for the record.
Chris Murray11:16
So if the pathwise volatility doubles, the drag will quadruple. If the pathwise volatility triples, the drag will multiply by a factor of nine.
Chris Murray19:23
So I imagine I have a lottery ticket. It costs a dollar and it has a 1% chance of paying $108 and a 99% chance of paying nothing.
Marco Sammon35:58
If the underlier is positive, the answer is infinity. Infinity leverage is always the answer. Infinity leverage always gives you maximized expected terminal wealth.
Chris Murray41:06
So for your 3x return to beat the underlier, there has to be a return of at least 170%.
Marco Sammon51:24
And you had this really rare situation where you walk in the day, you're longing 2x Leverity ETF on Lucid. By the end of the day, Lucid only closed down 16%. … But if you own the 2x Leverity ETF, you got a big zero.
Chris Murray55:41
My dad was 26 years old. If he had bought one share of VOO, the mutual fund share clash version of it, for $10,000 in 76, he'd have $3 million today.
Marco Sammon1:26:01
Figures
| Leveraged ETF industry total AUM (sample period) | Slightly over $100 billion | 3:04 |
| 2x ETF break-even return (1% daily volatility, broad market) | About 8.4% | 50:21 |
| 2x ETF break-even return (5% daily volatility, mega-cap stock) | About 100% | 50:21 |
| 3x ETF break-even return (5% daily volatility) | About 170% | 51:24 |
| Single-stock leveraged ETF total annualized cost (financing spread + fees) | About 15%–16% | 1:04:09 |
| Broad market leveraged ETF total annualized cost | About 5%–6% | 1:04:09 |
| 3x Nasdaq-type fund terminal value lost to costs over 5 years | About 40% | 1:00:00 |
| Inverse leveraged ETF cumulative inflow vs. current AUM (sample period) | $93 billion inflow, $13 billion remaining (86% decline) | 1:14:35 |
| Broad market leveraged ETF investors' excess earnings vs. long stock (sample period) | $50 billion | 1:16:37 |
| Newly issued leveraged ETF underlying stocks' historical percentile (median) | 94th percentile volatility, 92nd percentile 6-month return | 1:06:10 |
Glossary
- Convexity bonus
- Daily reset generates additional positive returns by squaring rather than doubling gains—more pronounced with larger swings.
- Volatility drag
- Larger volatility on the underlying erodes leveraged product wealth exponentially over time.
- Break-even return
- The return needed from the underlying for a leveraged ETF's performance to exceed the underlying itself.
- Swap
- A contract between the ETF and a bank to gain leveraged exposure beyond the fund's own assets.
- Expected log wealth
- A return metric using logarithmic returns to avoid distortion from extreme paths.
How to listen
Individual investors seeking to understand how leveraged and single-stock ETFs actually work, financial advisors conducting product due diligence, and industry professionals interested in the mathematics of compounding.
The highly technical section on expected log returns and autocorrelation (around 37:00–44:00) can be skipped if you only want the conclusions.