This article takes about 5 minutes to read

TZA Basics and Its Relationship with TNA

TZA (Direxion Daily Small Cap Bear 3X Shares) is an inverse ETF targeting -3x the daily return of the Russell 2000 index. Managed by Direxion; the firm's fact sheet as of June 30, 2026 lists an expense ratio of 0.99%. It is the 'mirror image' of TNA (Direxion Daily Small Cap Bull 3X), profiting when the Russell 2000 declines.

The Russell 2000 comprises 2,000 U.S. small-cap stocks with higher volatility than the S&P 500 or NASDAQ100. Daily volatility is said to be approximately 1.3% (annualized ~21%), well above the S&P 500's 1.0% (annualized 16%). This elevated volatility accelerates TZA's decay.

TNA and TZA are 3x bull and -3x bear on the same index, yet their long-term returns are completely asymmetric. TNA can potentially generate long-term profits, while TZA tends to decay over the long term. The mathematical basis for this asymmetry follows.

Bull vs Bear Asymmetry - Mathematical Proof

A leveraged ETF's long-term return can be approximated from the base index's return (mu) and volatility (sigma). The 3x bull's expected log return is 3*mu - 9*sigma^2/2; the -3x bear's expected log return is -3*mu - 9*sigma^2/2.

The critical difference is clear. The 3x bull has 3*mu (trend-direction profit) minus 9*sigma^2/2 (volatility decay). When mu > 0 (long-term uptrend), the bull can produce positive returns as long as 3*mu > 9*sigma^2/2.

The -3x bear, however, has -3*mu (loss from fighting the trend) AND -9*sigma^2/2 (volatility decay), both negative. As long as mu > 0, the bear's expected return is always negative. The double punishment of trend loss plus decay makes bear products unsuitable for long-term holding.

Why Russell 2000's High Volatility Is Especially Punishing for the Bear Side

Russell 2000's annualized volatility of 21% is lower than NASDAQ100's 24% but higher than the S&P 500's 16%. However, Russell 2000's long-term return is said to have trailed the S&P 500.

For the 3x bull, the return-to-volatility ratio matters. Assuming +7% a year for the Russell 2000 and +10% for the S&P 500, Russell 2000's mu/sigma^2 = 0.07/0.044 = 1.59, far below the S&P 500's 0.10/0.026 = 3.85. This means TNA has less return cushion against volatility decay compared to UPRO (S&P 500 Bull 3X).

For the -3x bear TZA, higher volatility means greater decay. With Russell 2000's sigma^2 = 0.044, the term 9*sigma^2/2 = 0.198, meaning approximately -20% annual volatility decay. Adding trend loss of -3*mu = -21% under the +7% assumption, TZA's annual expected return is approximately -41%.

Concrete Calculation - Russell 2000 at +10% Annual Return

Consider a year where Russell 2000 returns +10%. TNA (bull 3x) theoretical return: 3 x 10% minus volatility decay = +30% - approximately 20% = roughly +10%. Despite decay, positive returns are maintained.

TZA (bear 3x) theoretical return: -3 x 10% minus volatility decay = -30% - approximately 20% = -50%. A mere +10% Russell 2000 rise halves TZA's value. Two consecutive such years produce -75%; three years, -87.5%.

Actual data confirms this. Direxion's fact sheet as of June 30, 2026 puts TZA's annualized return (NAV) since its November 2008 inception at -49.02%. That works out to losing almost all of one million yen. Meanwhile, TNA returned +10.83% annualized over the same period (though it went through very deep declines along the way). This asymmetry is the fundamental difference between bull and bear.

Correct Use of TZA - Ultra-Short-Term Only

TZA generates profits only when the Russell 2000 crashes sharply over a brief period. Small caps tend to fall harder than large caps during crashes; in March 2020, the Russell 2000 plunged, and TZA rose sharply during this period.

However, small-cap recoveries are also rapid. From the March 2020 bottom through March 2021, the Russell 2000 rebounded strongly, and TZA lost most of its value. Even if you profit from TZA during a crash, missing the exit timing can wipe out those gains.

TZA holding periods should be limited to 1-2 weeks maximum. Due to Russell 2000's high volatility, decay is even faster than SQQQ or SPXS. The longer you hold, the more decay piles up.

What Happens When You Hold Both Bull and Bear Simultaneously

If you hold equal amounts of TNA and TZA, intuition suggests a net-zero position with no gain or loss. In reality, both decay from volatility, so combined assets tend to decline.

Concrete calculation: on a day Russell 2000 gains +2%, TNA rises +6% and TZA falls -6%. With 1 million yen each, TNA = 1.06M, TZA = 0.94M, total 2M unchanged. Next day Russell 2000 falls -2%: TNA = 1.06 x 0.94 = 0.9964M, TZA = 0.94 x 1.06 = 0.9964M, total 1.9928M. Two days, -0.36% loss.

This loss accumulates daily. Annually, approximately -20% of combined assets vanishes. Holding bull and bear simultaneously is not 'hedging' but a combination that tends to erode assets. If hedging is needed, reduce position size or move to cash.

TZA Investment Conclusions and Lessons

TZA is one of the products that embody bull-bear asymmetry in leveraged ETFs in an extreme form. Despite both being 3x on the Russell 2000, TNA can potentially profit long-term while TZA is unsuitable for long-term holding.

The root of this asymmetry is that equity markets have risen over time. Bull products riding the uptrend can potentially earn returns exceeding volatility decay. Bear products fighting the trend suffer the double burden of trend loss and decay.

The only rational scenario for TZA is a short-term trade when you are confident small caps will crash sharply and can exit within days. Every other use case violates the mathematics of compounding.