A Bond Doesn't Fall the Way a Stock Falls
 |
"Buy the dip" came from equities, and it travels badly.
When a stock falls, you are being offered the same asset at a lower price. The business did not change overnight; the number attached to it did. Whether that makes it a bargain is arguable, but the question at least is simple — same thing, cheaper.
A bond does not work that way. When a bond's price falls, something specific has happened to a contract, and it can be one of two entirely different things. They look identical on a screen. They mean close to opposite things about what happens next.
Most of the confusion around fixed income lives in that gap.
Price and Yield Are One Sentence Said Twice
A bond is a fixed set of payments on fixed dates. Once it is issued, those payments never change. The only variable left is what someone will pay today to receive them.
So when the price falls, the yield rises. Not as a consequence — as the same fact, stated the other way around. A bond is a loan you own, and the price is just what the loan is currently worth to somebody else.
Which means "the dip" in a bond is not a discount. It is a higher yield. The asset did not get cheaper the way a stock gets cheaper; the terms got better. That sounds like the same sentence and is not — because why the terms got better is the whole question.
Two Falls That Look Identical
The first kind: the discount rate moved.
New bonds are being issued at higher rates, so an older bond paying less has to fall in price until it competes. Nothing happened to the borrower. Every promised payment still arrives exactly as written, on the day it was written for.
What you lost is the gap between what you are earning and what you could now be earning. That is a real cost. It is also opportunity cost, and it has a natural end: hold the bond to maturity and you receive precisely what was promised. Time works in your favour, because every day brings the repayment closer.
The second kind: the market changed its mind about being paid.
Here the price fell because the probability attached to those payments moved. Waiting does not repair this one. Time works against you, because the closer maturity gets, the closer the borrower comes to the day they must actually produce the money.
Same red number on the same statement. One heals by sitting still. The other may not heal at all.
What a Credit Spread Measures
The instrument for telling the two apart has a dull name and a simple job.
A credit spread is the extra yield a borrower must pay above a government bond of the same maturity. The government bond stands in for the payment least in question. Everything stacked on top of it is the price of doubt.
It is quoted in basis points — one hundredth of a percentage point each. When a spread widens, doubt got more expensive. Often that has nothing to do with any particular company; the whole market has simply revised what it charges for the possibility of not being repaid.
The scale of that revision is easy to underestimate. The ICE BofA US Corporate index option-adjusted spread, a standard measure for high-quality American company debt, reached a record 6.56% in December 2008. Solid borrowers were paying more than six percentage points a year above the government for the same money.
No arithmetic inside any of those bonds had changed. What changed was the price of the doubt attached to them.
The Largest Bond Loss on Record Was the Forgiving Kind
In 2022 the Bloomberg US Aggregate Bond Index returned about −13%, its worst calendar year since the index began in 1976. The Bloomberg US Treasury Index fell about 12.5%, its worst since 1973. Twenty-year US government bonds lost roughly 29%.
The headline written on top of that was bonds are not safe anymore.
What the numbers describe is narrower, and more interesting. The damage was concentrated in government and high-quality debt — the corner of the market where repayment was never the question. It was a rate move from start to finish. The United States did not become less likely to pay. The discount rate applied to its promises rose very fast.
Which means the worst year in the history of the index was, mechanically, the recoverable kind of fall. Anyone holding an individual Treasury to maturity through 2022 received every payment on schedule. The loss was real for anyone who sold, and real as opportunity cost for everyone else — but it was not a loss of the promise.
That distinction went almost entirely unreported, and the reason is structural rather than sinister. Bonds had their worst year ever is a complete sentence. Bonds had their worst year ever, for the less alarming of the two possible reasons is not.
The Difference Between a Hedge and a Bet
This is where the distinction stops being trivia.
People hold bonds for two different jobs and rarely separate them.
One job is behaviour — owning something that tends to hold up when equities do not. High-quality government debt has historically done that, because a growth scare pushes money toward the payment least in doubt. It is bought for how it acts, not for what it yields.
The other job is yield — earning more than a government bond pays. That is corporate credit, and the extra yield is not a bonus. It is the compensation for the extra doubt. The spread is the payment for that risk, stated in public.
The awkward part is that the second one tends to fall at the moment the first one is needed. The growth scare that hits equities is the same event that widens spreads. Credit is therefore not a hedge against equity risk. It is a milder version of the same risk, filed in the calm section of the portfolio.
That distinction is the difference between a hedge and a bet — and both arrive in the account labelled bonds.
Why a Fund Statement Hides Which One Happened
A bond fund reports one number, total return. Both kinds of fall show up inside it, in the same colour, and the statement will not say which.
Three published figures separate them, and none of them requires a forecast.
The government yield for the same maturity, then and now. If it rose, part of your fall is a rate move. That part is arithmetic, and it comes with a matching rise in what the fund now earns.
The spread. If it widened, some of the fall is a change of opinion about repayment, and that part does not resolve by waiting.
Your fund's duration. Duration is roughly how much the price moves for a one-percentage-point move in rates. A fund with a duration near 17 and a fund with a duration near 2 are different instruments with different jobs, and both are sold as bond funds. It is printed on the fact sheet, and most people who own one have never looked.
Three numbers. They will not tell you what happens next. They tell you what already happened to you, which is the more answerable question.
What This Looks Like From Korea
A saver holding a dollar bond from Seoul carries a third variable, and it is usually the loudest one.
Three separate things move the figure in the account: the yield, the spread, and the exchange rate. The bond can be entirely intact — nothing changed about rates, nothing changed about credit — and the won number still falls, because the won moved.
It runs the other way too. Through 2022 the dollar strengthened sharply against the won, so a Korean holder of US government debt collected a currency gain during the worst bond year on record. The two partly cancelled. A great many people here experienced that year as something considerably milder than the headlines described, and learned correspondingly less from it.
There is a sharper version of this worth keeping. A hedge denominated in a currency you do not spend is only partly a hedge. It may still do its job — the money moves toward safety, and you own some of the safety — but the amount that reaches you has passed through an exchange rate that answers to something else entirely. That is not an argument against holding it. It is an argument for knowing that the calm part of your portfolio has a moving part in it that the fact sheet does not mention.
What This Page Used to Say
This URL previously held a different piece of writing, and the way it was wrong is the reason this version exists.
It was a timing call. It named specific spread levels as entry points — investment grade around one figure, high yield around another — and advised waiting until those levels printed. It closed by telling the reader to wait for the stress and then buy the dip.
Two things were wrong with that, and the smaller one is that the levels never printed. Anyone who took the instruction literally spent the following stretch holding cash and waiting for a signal that did not arrive.
The larger problem was the frame. Wait for a level, then act assumes the level is the information. In credit, the level is not the information — the reason behind it is. A spread of any given width can mean the market has correctly priced a genuine deterioration, or that it has overcharged for a scare that passes. The number alone does not distinguish those, and a number repeated out of somebody else's research note least of all.
Borrowed conviction is the specific failure there. A bank's threshold is a conditional judgement, published on a particular day, about a particular market, with an institution's own constraints behind it. Reprinting it as a trigger strips out every condition and keeps only the digit.
Where This Stops Being Useful
It says nothing about the direction of rates. Which of the two falls you are looking at is a question about the past, answerable from published data. Where rates go next is not, and no amount of clarity about the first buys any about the second.
It does not make credit a bad holding. Being paid for doubt is a legitimate arrangement, and someone has to hold that risk. The point is knowing that you are paid for doubt rather than believing you bought calm.
And holding to maturity does not transfer to funds. An individual bond matures; a fund does not. It sells and rebuys continuously, so there is no date on which you are made whole by the terms of a contract. A fund's recovery comes from reinvesting at the new, higher yields — which is a real mechanism and a slower one, and it is not the same promise.
Where You Are Standing
If you own bonds and the number went down, the first question is not what to do. It is which of the two things happened.
If it was rates, and your horizon is longer than the fund's duration, the arithmetic is already working for you. The same move that cut the price raised what the fund earns from here — the mechanism behind real versus nominal returns applies in both directions.
If it was spreads, the useful question is backward-looking: were you being paid enough for that doubt before it widened, and did you know you were carrying it?
If you hold bonds as the steady part of the portfolio, check that they are the kind that behaves steadily when everything else does not. That is a different holding from the kind that pays a little more.
And if you are holding cash while you wait for a better entry, that decision has its own arithmetic, running quietly the whole time — the subject of why "safe" cash quietly shrinks and of what cash actually buys you. Waiting is a position. It is simply one whose cost does not appear on a statement.
There is also a longer arc under all of this. When the rate paid on government debt sits below inflation for years at a time, bondholders are the ones financing the gap, which is how debt burdens have historically shrunk without anyone defaulting on anything.
A Closing Note
Buy the dip asks about price. A bond answers a different question — what was promised, and how sure the promise is.
A falling bond price is not a discount on a promise. It is either a better rate on the same promise, or a worse opinion of it, and those two arrive in the same colour on the same screen on the same afternoon.
Naming which one you are looking at is most of what fixed income asks of an ordinary saver. It takes no forecast, and three numbers someone else has already published.
Data as of 2026-08-09. Sources: Bloomberg US Aggregate Bond Index and Bloomberg US Treasury Index calendar-year returns, 2022; ICE BofA US Corporate Index option-adjusted spread, December 2008 record, via the Federal Reserve Bank of St. Louis. This post is observation, not investment advice.
Related Posts:
What a Bond Actually Is — The Loan You Own
Real vs Nominal — Why Your Money Can Grow and Shrink at the Same Time
