Zero-Coupon Bonds: All the Return at the End

Removing the coupons sounds like a simplification, and in one sense it is: the instrument is a single future payment, which makes it the purest possible expression of the present-value arithmetic this pillar has been building on. But removing the coupons also removes the thing that moderates a bond's behaviour, which makes zeros the most interest-rate-sensitive instruments of their maturity, and it creates a tax situation that is genuinely awkward. This article covers the mechanics and accretion, the two properties that make zeros distinctive — maximum duration and no reinvestment risk — and the honest complications.

The mechanics: discount, accretion, and where zeros come from

A zero's price is simply the present value of one payment: face value discounted back over the remaining term at the required yield. So a $1,000 bond due in ten years at a 4% yield costs about $676 today, and the $324 difference is the interest — compressed into a single capital gain rather than paid out. As time passes and the maturity date approaches, the price rises toward par along a curve, a process called accretion (the bond's accreted value at any date being the price implied by the original yield and the time remaining). This is the pull to par from the pricing article, operating in its purest form: no coupon distractions, just compounding toward the endpoint. Three practical notes. Zeros arise in two ways. Some are issued as zeros from the start, particularly by governments at short maturities — a Treasury bill is functionally a zero-coupon instrument — while others are created by stripping: separating a conventional bond's individual coupons and its principal repayment into distinct tradeable securities, each a zero in its own right, which is what the US STRIPS programme does and which similar arrangements do elsewhere. Deep-discount bonds are a nearby category: low-coupon bonds trading far below par share some zero-like characteristics without being pure zeros. And the quoted yield is doing all the work: with no coupon to anchor a current-yield calculation, the only meaningful measure for a zero is yield to maturity — current yield is not defined in any useful sense, since the annual cash income is nil.

The two distinctive properties

First: maximum duration, therefore maximum rate sensitivity. Because all of a zero's value sits in a single payment at maturity, its duration equals its maturity — the longest possible for that term, since any coupon-paying bond of the same maturity returns some money earlier and therefore has a shorter weighted average timing. The consequence is that zeros move more than any comparable bond when yields change: a thirty-year zero is one of the most volatile instruments in fixed income, and a small change in yields produces a large change in its price. The duration article makes this precise. Stated as mechanics and nothing more: this magnifies both directions, and holders of long zeros have experienced both very large gains and very large losses from rate movements alone. Second: no reinvestment risk. The yield article identified the assumption buried inside yield to maturity — that coupons are reinvested at the YTM itself, which the real world does not guarantee. A zero has no coupons, so there is nothing to reinvest, and the assumption becomes vacuous: buy a zero at a given yield, hold it to maturity, and that yield is what you earn, with no dependence on future rates and no reinvestment shortfall. That is a genuinely unusual property in fixed income — the return is locked at purchase in a way no coupon bond's is — and it explains why zeros are the natural instrument for funding a known future obligation: one payment, one date, one known amount. Note the symmetry, though: the same property means a zero holder gains nothing if rates rise, having no coupons to reinvest at the better terms. Certainty cuts both ways, which is the honest way to describe it.

The complications: phantom income, liquidity, and credit concentration

Three issues, one of which is the reason many individual investors hold zeros only in particular account types. Phantom income. In a number of jurisdictions, the annual accretion of a zero is treated as taxable interest income as it accrues, even though no cash is received until maturity — so a holder can owe tax annually on income they have not been paid, for years. The label "phantom income" is the standard description of exactly this mismatch, it is a real and material consideration, and it is the same structural problem inflation-linked bonds present with their principal adjustments. Treatment varies substantially by jurisdiction, instrument, and account type, and this portal parks the specifics to Annex A: the existence of the issue is education, the resolution is a matter for a qualified tax adviser. Liquidity. Stripped securities and individual zeros are typically less liquid than the on-the-run conventional bonds they derive from, with wider costs — relevant to anyone who might not hold to maturity, per the bond-trading article. And concentrated credit exposure. With a coupon bond, a holder recovers some money along the way; with a zero, everything depends on a single payment years hence, so the issuer's creditworthiness at that distant date carries the entire outcome. For a high-grade sovereign zero this is a modest concern; for a corporate zero it is a substantial one, and it is why long-dated corporate zeros are an unusual combination of maximum rate sensitivity and maximum credit concentration. As always, the instrument is neither good nor bad: it is a single payment on a date, priced by arithmetic, with a distinctive risk profile — and whether that profile suits any particular purpose is a question for the reader and, given the tax dimension especially, for professional advice.