Starting on November 20, the 30-year Treasury bond ETF fell for two consecutive weeks, with a cumulative drawdown close to 5%. Over the same period, market yields rose from roughly 2.14% to 2.26%.

At the time, I interpreted that decline as a short-term and mostly technical fluctuation, so I added to the position against the move.

By December 4, my 30-year Treasury ETF position had grown to more than 40% of the entire portfolio.

What happened afterward forced me to rethink that allocation. Even though my equity holdings did not experience a major drawdown, the portfolio kept giving back profits because the ultra-long Treasury position continued to fall. For a while I was in the awkward situation of:

stocks going up while the account was still losing money.

This article is not written to vent emotion. It is an honest investment review: looking back, where exactly did I go wrong?

1. My original logic: a starting point that looked correct

The allocation was not impulsive.

Over the previous few months, A-shares had continued to rise in both sentiment and valuation. As an investor with a long-term equity bias, I felt that the portfolio's overall risk exposure was increasing. So I started asking whether I should introduce an asset that could hedge a potential drawdown in advance.

My reasoning chain was roughly:

• ultra-long Treasuries are safe assets;
• if the macro environment weakens, interest rates may fall;
• long-duration bonds respond more strongly to risk events;
• therefore, they can serve as a hedge against equities.

Under that framework, buying a 30-year Treasury ETF looked perfectly reasonable.

Looking back, the problem was not the direction of the idea. The problem was the incompleteness of my understanding:

I picked the right category of instrument, but severely underestimated its conditions of use and its own volatility structure.

2. Mistake one: treating a 30-year Treasury ETF as a stable asset

This was my first major mistake.

Investors often fall into a reflexive classification:

• stocks mean high volatility;
• bonds mean stability.

That intuition completely breaks down for ultra-long Treasuries.

A 30-year Treasury ETF is not a stable bond fund in the everyday sense. It is much closer to:

a highly duration-sensitive instrument whose price is extremely exposed to interest-rate changes.

In practice:

• duration is often around 18 to 22 years;
• if yields rise by 0.5%, price declines can reach 8% to 11%;
• annualized volatility may sit in the 15% to 20% range.

3. Mistake two: overestimating long-term return significance

If the long-run market yield of 30-year government bonds stays around 2% to 3%, then the long-term annualized return of such an ETF should also be anchored around that yield-to-maturity range.

That means ultra-long Treasuries are not primarily there to deliver strong long-run compounding. Their main attraction is the price convexity that appears when rates fall sharply.

My mistake was to buy after a drawdown while waiting for a macro reversal that had neither been confirmed nor sufficiently tested.

4. Mistake three: turning a hedge into a directional bet

There is one sentence I keep using to remind myself:

Once a 30-year Treasury ETF position exceeds roughly 25%, what you have is no longer a hedge. It is a clear rate-direction bet.

My actual portfolio was:

• a growth-tilted equity book;
• more than 40% in a 30-year Treasury ETF.

At that point, overall portfolio performance started depending on one variable: whether long-term interest rates would continue falling.

5. What I learned

This experience left me with several practical lessons:

• do not confuse asset label with risk behavior;
• do not use a low-yield, high-duration instrument as if it were cash-like ballast;
• do not call something a hedge when the position size has already turned it into a macro bet;
• always distinguish directional thesis from portfolio insurance.

Final remarks

The mistake was not purely technical. It was cognitive.

I thought I was introducing balance into the portfolio. In reality, I introduced a new single-factor exposure that I did not understand deeply enough.

贡献者: Junyuan He