Bond yields rose. That is not the return a bondholder earned.

The comparison begins with two different quantities

A gold-market note often says that higher yields make bonds more attractive. That can describe the terms available to a new buyer while leaving an existing holder with a mark-to-market loss. This article examines that difference rather than repeating today’s Daily Gold Brief or the weekly review of nominal and real references. No new bond quotation or gold return is asserted.

TreasuryDirect distinguishes a security’s fixed interest rate from the yield determined by its purchase price. When required yield is above a fixed coupon rate, the price can be below face value; the reverse relationship can place it above face value. A yield is not a cash profit already earned by every investor who held the instrument during a market move.

Source notes: [1] TreasuryDirect — Understanding Pricing and Interest Rates

Duration translates a small yield move into a price estimate

FINRA describes duration as a measure of interest-rate sensitivity, despite its expression in years. Higher duration generally means a larger price response to a given rate move. For a simple fixed-cash-flow bond, a first-order estimate uses modified duration multiplied by the yield change, with the opposite sign. The yield change must be entered as a decimal.

Consider two hypothetical bonds with modified durations of two and eight. If each bond’s own yield rises by 0.25 percentage points, or 0.0025, the first-order price estimates are approximately minus 0.5% and minus 2.0%. These are calculated scenarios, not observed returns or forecasts. The same yield move therefore does not imply the same capital loss for every bond.

Source notes: [2] FINRA — Duration and interest-rate risk

Convexity describes the bend in that relationship

CFA Institute explains that duration is a linear approximation to a curved price-yield relationship. For an option-free fixed-rate bond, positive convexity makes the estimated gain from a yield decline larger, and the estimated loss from a yield increase smaller, than duration alone would suggest. The adjustment becomes more relevant for larger yield moves.

Using deliberately hypothetical modified duration of eight and convexity of eighty, a one-percentage-point yield increase gives a duration estimate of minus 8%. The second-order term, one half times eighty times 0.01 squared, adds 0.4 percentage points, producing approximately minus 7.6%. An equal decline gives approximately plus 8.4%. This simplified calculation is an approximation under the stated assumptions, not a valuation of a named security.

Source notes: [3] CFA Institute — Yield-Based Bond Convexity and Portfolio Properties

Price return still is not the entire holding-period result

The SEC’s investor guidance explains why fixed-rate bond prices generally fall when market rates rise, and why maturity and coupon affect sensitivity. It also distinguishes sale before maturity from holding an individual obligation to its maturity, subject to payment risk. That distinction is especially important when a gold comparison uses a tradable fund on one side and an individual bond’s quoted yield on the other.

A consistent holding-period comparison needs beginning and ending market values plus cash distributions received over that same interval. Accrued interest, reinvestment assumptions, fees and currency treatment also need a declared convention. Comparing gold’s one-day percentage price move with a bond’s annualized yield mixes horizons and quantities. The attractive-looking numerical contrast can be generated entirely by that mismatch.

Source notes: [4] SEC Investor.gov — Fixed-rate bonds and rising interest rates

The relevant curve move may not be parallel

A single duration estimate compresses a complicated exposure into one sensitivity. My practical interpretation is that the compression must be matched to the question. If short yields rise and long yields fall, applying one observed ten-year move to every cash flow in a diverse portfolio can misdescribe the change. A portfolio review should examine which maturities actually carry the exposure.

The simple convexity scenario also excludes embedded options and changing cash flows. It should not be transplanted unchanged into callable securities, mortgage portfolios or products whose effective duration can shift materially. Nor does a bond-price approximation reveal the investor’s financing cost or need to sell. Those are separate records, not details that can be recovered from a yield chart.

What this changes in gold commentary

I would separate three statements: the yield newly available on a defined bond, the return earned by a defined holder over a defined interval, and the interpretation of how that opportunity set may influence demand for gold. The first can improve while the second is negative. The third remains a hypothesis about behaviour, not a mathematical consequence of the first two.

A stronger comparison therefore names the instrument, duration measure, observation window, cash-flow treatment and currency. If the required data are unavailable, leave the return comparison open. This article provides no claim that bonds or gold will outperform next. Its calculations explain a measurement problem; the premium illustration is conceptual and does not represent a historical yield curve, trade or investment result.