Bond Ladders: Building One and What It Solves

Fifty thousand dollars. Five bonds, ten thousand each, maturing one year apart. That is the whole structure, and everything interesting about a ladder comes from what happens on the day the first rung comes due and the money has to go somewhere.

The point of the structure is timing. A ladder stops one morning from setting the terms for all of your money, and that property turns out to matter a great deal more than the small difference in headline yield between a ladder and whatever else was being considered alongside it, because interest rates spend most of their existence somewhere nobody predicted.

Five rungs, fifty thousand dollars

Take a hypothetical curve. These figures are round numbers chosen to make the arithmetic visible, and the shape of the curve you actually face will differ.

Rung	Maturity	Amount	Yield	Annual income
1	1 year	$10,000	4.60%	$460
2	2 years	$10,000	4.40%	$440
3	3 years	$10,000	4.30%	$430
4	4 years	$10,000	4.35%	$435
5	5 years	$10,000	4.40%	$440

Total income is $2,205 a year on $50,000, a blended yield of $2,205 / $50,000 = 4.41%. Every figure in that column is locked the day you buy. The bonds have to pay it, and the only thing that can change it is a default or a call.

Notice what the ladder does to the shape of the curve. It owns a piece of every point on it, so a dip in the middle costs you one rung and a bulge at the front earns you one rung, and the blended yield lands close to the average of the five whatever shape the curve happens to hold that week.

Spacing is the other decision. Annual rungs are the default. Six month spacing doubles the number of positions and halves the wait for cash, which suits somebody living off the income, while annual spacing stretched across ten years reaches further out the curve for a little more yield and commits the last rung for a decade.

Building it, rung by rung

Equal dollar amounts per rung is the standard. Equal income per rung is the alternative, and it requires buying more of the low yielding maturities, which almost nobody does deliberately.

Treasuries are the easy instrument here. They come in $100 increments, they cannot be called away from you, the credit question does not arise, and a brokerage account reaches both the auction calendar and the secondary market where a bond maturing in the precise month you need the cash is usually sitting there waiting to be bought. Brokered certificates of deposit work as rungs as well. Federal deposit insurance covers them to the published limit for each depositor at each bank, so a ladder spread across several banks can stay inside the guarantee, and a great many of them are callable, which is the detail on the second screen of the offering page.

Corporates and municipals pay more. They also bring credit analysis, call schedules, wider markups and minimum lot sizes.

There is one trap worth naming at the start. Building all five rungs on a single afternoon means all five were priced against the same curve, so the averaging across time that the whole structure exists to deliver has not started yet and will not start until the first rung rolls. A new ladder is one purchase date wearing five maturity dates. The property you paid for arrives over the following five years, gradually, and there is nothing to be done about it except wait.

Rolling the maturing rung

At the end of year one, rung one matures. Ten thousand dollars lands in the account. You buy a new five year bond with it, which sits behind the old rung five, and the ladder is a five year ladder again with one year knocked off every remaining rung.

Repeat that four more times and the ladder reaches a steady state where every bond in it was bought as a five year bond, on five different dates, in five different interest rate environments. That is the useful property. The portfolio earns something close to a rolling average of five year yields while returning a fifth of your money in cash every twelve months, so you get long-end income with a short-end schedule of liquidity, and the price of that arrangement is that you will never have all of your money at the highest yield on the curve and never have all of it at the lowest either.

What it does to reinvestment risk

Reinvestment risk has a plain definition. Cash comes back on the maturity date and has to be put to work at whatever the market is offering that week, which can sit well below the rate printed on the bond that just paid off. Watch it work with numbers.

Rates fall. By the time rung one matures, five year paper yields 3.00%. The ladder reinvests $10,000 at 3.00%, giving up the $460 that rung was paying and picking up $300, so income drops from $2,205 to $2,045. That is a fall of 7.3%.

Now do the same thing to somebody who put the whole $50,000 into one year bills and rolls the lot every twelve months. At 4.60% that portfolio earned $2,300. At 3.00% it earns $1,500. The fall is 34.8%, and it happened on a single morning, to every dollar, with no part of the portfolio holding an older and better rate.

The ladder holder gave up something for that. In the world where rates rise, the bill roller repriced everything upward at once and the ladder repriced one fifth of itself. Averaging cuts both ends off.

What it does to price risk

Average maturity across the five rungs is (1 + 2 + 3 + 4 + 5) / 5 = 3 years, and duration sits a shade below that because the coupons arrive before the principal does. So a one percentage point rise in yields across the curve takes something close to 2.7% off the market value of the ladder. The same move would take about 4.4% off a single five year bond and far more off anything longer, which is the arithmetic laid out in duration and interest rate risk.

Here is the part that matters more than the percentage. The ladder holder does not have to realise any of it. Rung one pays $10,000 on its date whatever the market thought of it in the meantime, and so does every rung behind it, because a bond held to maturity returns face value and the price in between is an opinion the holder is free to ignore. The mechanism behind that pull to par is worked through in how bonds work.

A ladder against a bond fund

Both structures can hold identical bonds. The difference is the date.

Every rung in a ladder has one. A bond fund sells its holdings as they age and buys new ones, so the portfolio stays permanently at its target maturity and there is no day on which you are handed face value. If the fund falls 6%, recovery has to come from income piling up and from yields falling back, and it can take years. This is the distinction drawn at greater length in bond ETFs explained.

The fund wins on almost everything else. It holds hundreds of issues for a few basis points a year, it trades in one click, it reinvests coupons automatically, and it never leaves you calling a bond desk for a price on $10,000 of an odd lot that nobody particularly wants to make a market in. A household with $50,000 in bonds and no interest in the administration should probably own the fund. A household funding five specific years of spending should probably own the ladder.

Barbells and bullets

Two other shapes get compared with the ladder, and both are defined by where the money sits on the curve.

Structure	How $50,000 sits	Average maturity	What it is for
Ladder	Equal amounts at 1, 2, 3, 4 and 5 years	3 years	A maturity every year, income that averages across rates
Bullet	The whole amount at 5 years	5 years	One obligation falling due on one known date
Barbell	$25,000 at 1 year and $25,000 at 9 years	5 years	Cash soon and long-end yield, with more convexity

The bullet is the right answer when the date is known. A tuition bill in five years, a house deposit in three: match the maturity to the obligation and the interest rate question disappears from the problem entirely.

The barbell and the bullet in that table have the same average maturity and behave nothing alike. A barbell gains more when yields fall than it loses when they rise, because the long leg carries the convexity, and it profits when the middle of the curve cheapens against the two ends. It is a position on the shape of the curve, taken deliberately or taken by accident.

Where ladders go wrong

Callable bonds break the structure. The issuer redeems when rates fall, handing your money back on precisely the morning reinvestment is worst, which converts the rung you built for year seven into a rung maturing now. Treasuries have no such problem, which is one reason Treasury ladders are the common household version.

Costs bite hardest on small corporate and municipal lots. The dealer’s markup lives inside the price and never appears as a commission, an odd lot gets a worse bid than an institutional block, and a ladder of five corporate names leaves you holding five specific balance sheets, which is a different species of risk from anything a Treasury ladder carries. Price what you are being shown with the bond yield calculator before agreeing to it.

And a ladder answers nothing about inflation. Every rung pays nominal dollars on a schedule, so a decade of high inflation erodes the lot while every payment arrives exactly as promised. Building the same structure out of inflation-linked securities is possible and behaves very differently, for the reasons set out in TIPS explained.