SGOV vs BND vs AGG: Cash, Core, and the Difference That Matters

These three tickers turn up together in every bond-fund search, and the reason is obvious once you see them lined up: they cost almost nothing, they hold high-quality debt, they pay within half a percentage point of each other, and they sit on the same shelf at every broker. So the question gets asked as though it were a ranking — which is best? — when the funds are not competing for the same job at all. Two of them are, to a degree that is almost funny, the same fund. The third is a different product with a shared surname.

The three, side by side

Read the duration column and ignore the rest for a moment. Duration is roughly how many percent a fund's price falls when interest rates rise by one percentage point — it is the standard rule of thumb, and Vanguard states it in those words on BND's own fact sheet. SGOV's 0.10 against the core funds' 5.80 is not a difference of degree. It is a 58-fold difference in how hard the same event lands, and nothing else on the page comes close to mattering as much.

BND vs AGG: stop trying to rank them

This is the most-searched pairing of the three and it has an unsatisfying answer. BND tracks the Bloomberg U.S. Aggregate Float Adjusted Index; AGG tracks the Bloomberg U.S. Aggregate Bond Index. Same family, same market, same investment-grade US bonds. Both charge 0.03%. Both carry 5.8 years of duration. Their yields are 4.03% and 4.05%, a gap of two hundredths of a percentage point, which on $10,000 is two dollars a year and will have changed by the time you finish reading this sentence.

The holdings counts differ, and it is the one place a real difference exists — but both numbers are enormous and the extra bonds are the same kind of bonds. Neither fund is meaningfully more diversified than the other in any way you will ever experience. So the honest tie-breaker is administrative: whichever trades commission-free at your broker, or whichever you already own and would trigger a tax bill by selling. What you should not do is buy both and feel diversified. You have bought the US investment-grade bond market twice and paid two lots of spread for it.

SGOV vs BND: this is the actual decision

Here is the question worth spending time on, and the yields are a trap in it. SGOV pays 3.74% and BND pays 4.03%, so BND looks like a slightly better version of the same thing — a bit more income for what the fund page describes, in both cases, as high-quality bonds. That reading is wrong, and the number that shows why is not the yield.

Divide each fund's yield by its duration and you get the rate rise that wipes out exactly one year of income. For BND that is 0.69 percentage points. For AGG, 0.70. For SGOV, 37.4. Rates would have to do something that has never happened in the history of the United States to cost a bill fund a year's income, and they only have to twitch to cost the core funds theirs.

That is the whole argument. Both funds hand you an income; only one of them can have it taken back by an ordinary move in interest rates. A 0.69-point rise is not a crisis — it is a couple of meetings where the central bank does something mildly unexpected. In 2022, rates moved several times that, which is why people who had bought the safe thing opened their statements to a double-digit loss.

What one point of rates actually costs

Take $10,000 in each fund, hold it a year, and assume rates rise by a single percentage point. The income arrives either way. The price move is what separates them.

Put the price move against the income and the scale becomes plain. One point of rates costs BND about a year and five months of its own income. It costs SGOV about ten days of it. The extra 0.29 percentage points of yield you picked up by choosing the core fund was, in that year, worth $29, against a $570 difference in what the price did.

Notice which number is doing the work, because it is not the one most comparisons lead with. SGOV charges three times what BND charges — 0.09% against 0.03% — and that headline ratio sounds significant right up until you price it: six dollars a year on $10,000. Fees matter enormously in funds, and this is the rare case where a fee difference is the least interesting thing on the table.

Which yield are you even reading?

One complication worth knowing before you compare any two bond funds, because it is the most common way people end up comparing numbers that are not comparable. Fund pages publish two different yields. The trailing-twelve-month distribution yield says what the fund has paid out over the past year. The 30-day SEC yield is a standardised estimate of what it is earning now, and it is the forward-looking one.

On a stable rate path they are close. When rates are moving they are not, and the gap is largest on exactly the fund whose holdings turn over fastest. SGOV's trailing yield was 3.74% in mid-August 2026 while its 30-day SEC yield was 3.60% — the trailing figure is describing bills bought when rates were higher, and those bills have since matured and been replaced. The forward number is the truer one for a fund of 3-month paper.

When SGOV is the wrong answer

Everything above makes SGOV sound like the sensible default, and for money with a date attached it is. But the property that protects it is the same property that limits it, and the limitation is genuinely costly in one specific scenario: rates falling.

If rates fall a point, BND's price rises about 5.8% — roughly $580 on $10,000, on top of its income. SGOV gains about $10 and then starts paying less, because it is continuously relending at the new, lower rate. The core funds lock in today's yield for years; the bill fund locks in nothing. Anyone who parked in a money-market-like fund through a rate-cutting cycle has felt this, usually as the slow realisation that their monthly income has quietly halved.

• Money you need inside two years: SGOV, or a savings account paying as much. Rate risk is the thing to avoid here, not the thing to be paid for.
• Money that is staying put for years as ballast against your stocks: BND or AGG. The duration is the point — it is what makes a bond fund rally when a stock crash pushes rates down.
• Money you might need, on a timetable you cannot name: honestly, both. There is no rule against holding a cash fund and a core fund at the same time, and it is a far more sensible pair than BND plus AGG.
• A long horizon and no need for the cash: possibly none of the above. Plenty of sensible investors hold no bonds at all at thirty years out, and that is a real position rather than a mistake.

So which one should you buy?

Ask when you need the money and the answer falls out. Inside two years, SGOV — the 0.10-year duration is the entire reason it exists, and giving up 0.29 points of yield to be sure the money is there is a trade worth making every time. Beyond that, BND or AGG, chosen by whichever is free at your broker, because on every figure that matters they are the same fund. The only genuinely poor answer is buying BND and AGG together and believing you have done something.