Duration: Why Your Safe Bond Fund Fell When Rates Rose

Why a bond's price moves at all

A bond is a contract to pay fixed amounts on fixed dates. Buy a $1,000 bond paying 3% annually for ten years and you have purchased a promise: $30 a year, then $1,000 back at the end. Those numbers never change. The bond does not know what the Federal Reserve is doing.

What changes is what that promise is worth to someone else. If new bonds of identical quality and maturity start paying 5%, nobody will pay $1,000 for your 3% stream when they could get 5% elsewhere. Your bond still pays $30 a year, so the only variable that can adjust is the price. It falls until the return a new buyer earns — the $30 a year plus the discount they get on the $1,000 repayment — works out to roughly 5%.

That is the entire mechanism. The cash flows are fixed, so the price absorbs every change in prevailing rates. Rates up, prices down; rates down, prices up. Always, and in that direction.

Duration is the sensitivity number

The obvious next question is: down by how much? That is what duration answers.

Duration is quoted in years, which confuses people into thinking it is a maturity measure. It is better understood as a price sensitivity measure. The working approximation is:

Percentage price change ≈ −duration × change in yield

A fund with a duration of 6 will fall roughly 6% if yields rise by one percentage point, and rise roughly 6% if yields fall by one. A fund with a duration of 2 moves about 2%. A long-term Treasury fund with a duration of 17 moves about 17% — which is equity-sized volatility from an asset most people file under "safe."

Run it concretely. You hold $100,000 in a bond fund with a duration of 6.2, yielding 4%. Yields across the curve rise by 1.25 percentage points over a few months. The price effect is approximately −6.2 × 1.25 = −7.75%, or about $7,750. Meanwhile the fund paid you roughly 4% in interest over a year, about $4,000. Net, you are down around 3.75% on an investment you bought for stability.

Nothing malfunctioned. The fund did precisely what a duration of 6.2 says it will do.

Duration is not a prediction. It is a specification — the same way a car's turning radius tells you exactly how sharply it will corner.

What makes duration long or short

Three factors drive it, and two are intuitive.

Maturity. The longer until you get your principal back, the longer you are locked into an off-market rate, and the more the price must move to compensate a buyer. A 30-year bond is far more rate-sensitive than a 2-year bond. This is the dominant factor.

Coupon. A higher coupon means more of your total return arrives early, in cash, rather than at maturity. That shortens the effective waiting period and reduces duration. A zero-coupon bond, which pays nothing until the end, has a duration equal to its full maturity — the maximum possible.

Yield level. A subtler effect: when prevailing yields are higher, future cash flows are discounted more heavily, which slightly shortens duration. This is why bond funds were more rate-sensitive during the very low-rate years than the same funds are today.

Useful rules of thumb: short-term bond funds typically run durations of 1 to 3, intermediate-term 4 to 7, and long-term 10 to 20+. Every fund publishes its average effective duration on its fact sheet, usually in the characteristics table. If you own bond funds and do not know their duration, that is the first number to go look up.

The part that rescues you: reinvestment

A price drop is the visible half of the story. The invisible half is that the fund is now buying bonds at the new, higher yields.

As older, lower-coupon bonds in the portfolio mature, the proceeds are reinvested at current rates. Every month the fund's income creeps upward. The immediate capital loss is followed by a rising stream of interest payments that gradually earns it back.

There is a genuinely elegant result here: for a portfolio holding constant duration, the breakeven horizon after a rate shock is approximately equal to the duration itself. A fund with a duration of 6 that drops on a rate rise will, holding rates steady thereafter, recover to where it would have been in roughly six years — because the higher reinvestment income offsets the price loss over about that span.

Which gives you a practical matching rule. If your duration is shorter than your time horizon, a rate rise is ultimately good for you. You take a temporary price hit and then collect higher income for years. If your duration is longer than your horizon, a rate rise is a real, permanent loss, because you will sell before the higher income can compensate.

"I'll just hold individual bonds to maturity"

This is the standard objection, and it is half right in a way worth being precise about.

Hold an individual bond to maturity and you receive exactly the yield you locked in at purchase, regardless of what happens to prices along the way. That is true. No interim price move can change your outcome if you never sell.

But the same is true of the bonds inside a fund — they mature at par too. What differs is that the fund continuously sells and replaces holdings to maintain its target duration, so it marks to market daily instead of letting you ignore the fluctuation.

And the individual-bond approach has a real cost that the framing hides: opportunity cost. Locking in 3% for ten years while the market moves to 5% means eight years of earning below-market income. You did not avoid the loss; you converted it from a visible price decline into an invisible stream of foregone interest. The economics are nearly identical. One is simply easier to not look at.

Individual bonds do have legitimate advantages — a known maturity date matched to a known liability, no ongoing expense ratio, no fund-level flows — but "you can't lose money" is not among them.

Choosing duration deliberately

Treat duration as a dial you set to match the job the money has to do.

Money needed within a year or two does not belong in a bond fund at all — a money market fund, Treasury bills or a high-yield savings account carry essentially no duration risk. Money for a goal three to five years out suits a short-term bond fund with a duration of 1 to 3. Retirement money twenty years out can hold intermediate duration comfortably, since the horizon exceeds the duration by a wide margin and rate rises are, on balance, helpful.

Long-duration funds are a genuine bet on falling rates, and they should be held deliberately by someone who wants that exposure — not accidentally by someone who bought "bonds" for safety and never checked the number.