Oil crossed $100. Your bonds paid for it.

This article first appeared in The Nest Egg, my daily email, on 2026-09-11. Read the original issue. It is educational, not a recommendation for you. The household in it is a hypothetical example, and market figures are as of the date shown and will have changed.

Every figure below names its document. Every calculation is shown. Nothing here is a forecast.

Good morning. It is Friday, and at 8:30 this morning the Bureau of Labor Statistics publishes the August consumer price index. That number is the last piece of data the Federal Reserve sees before it meets next week. You have about ninety minutes.

The tape

S&P 500 7,591.70, down 0.58% · Dow industrials 52,064.10, down 0.60% · Nasdaq 100 29,103.51, down 1.08% · Russell 2000 2,890.95, down 1.04% · VIX 17.84, up 8.38% · Gold $4,357.80, down 2.31% · WTI crude $103.87, up 8.14% · 10-year Treasury 4.963%, up 12.3 basis points · 30-year Treasury 5.369%, up 8.3 basis points

Closing levels and session changes, CNBC US markets, September 10, 2026.

The story

The headline yesterday was oil. U.S. crude closed at $103.87, up 8.14% on the session, after intensified hostilities in the Middle East revived the question of what happens to the tankers moving through the Strait of Hormuz. Gasoline futures rose 6.93%. Diesel rose 7.23%.

That is the story every outlet ran. It is not the story in your account.

Look at what else moved. Gold fell 2.31%. Silver fell 6.59%. Copper fell 5.25%. Those three do not fall together on a war scare. They fall together when the market decides real interest rates are going up.

And they are. Traders raised the probability of a Federal Reserve rate increase at next week’s meeting to 70% in Thursday morning action, according to CME Group’s FedWatch gauge, after the producer price index rose 0.4% in August and pushed the annual wholesale inflation rate to 5.4%. Odds of a second increase in December moved to near 60%. The European Central Bank raised its own policy rate a quarter point the same day and lifted its inflation forecast.

“The ongoing spike in oil, combined with low jobless claims, make it hard for the Fed to not hike next week,” David Russell, global head of market strategy at TradeStation, told CNBC on Thursday.

U.S. Central Command forces begin setting conditions to clear mines in the Strait of Hormuz, April 11, 2026. NAVCENT Public Affairs, public domain, via DVIDS.

So the oil price is the visible thing. The repricing of the Federal Reserve is the thing that touched your balance sheet. And it touched it in the half of the portfolio you were not watching.

The position

Take a $2,000,000 portfolio, 60/40, held by a man of 63 who stopped working in June. The $800,000 fixed income sleeve is the part he thinks of as the safe part. Say he holds it in three equal pieces of roughly $400,000: short Treasurys, intermediate Treasurys, and a long bond position he bought two years ago because the yield looked generous.

On Thursday the equity sleeve lost about 0.6%. He will see that. What he will not see, because his statement will not break it out, is what the bond sleeve did.

The arithmetic

The approximation every bond desk uses is one line: price change is approximately minus modified duration times the change in yield.

Modified duration on a par bond, semiannual coupon, at Thursday’s closing yields: the 2-year is 1.890, the 10-year is 7.808, the 30-year is 14.825.

Now put Thursday’s actual moves through it, on $400,000 in each.

2-year. Yield rose 15.9 basis points. Minus 1.890 times 0.159% is minus 0.301%. On $400,000, that is $1,202.

10-year. Yield rose 12.3 basis points. Minus 7.808 times 0.123% is minus 0.960%. On $400,000, that is $3,842.

30-year. Yield rose 8.3 basis points. Minus 14.825 times 0.083% is minus 1.230%. On $400,000, that is $4,922.

Yield changes, CNBC US markets, September 10, 2026 close. Hypothetical illustration.

Read those three lines again. The 2-year yield rose 1.92 times as far as the 30-year yield. The 30-year still lost four times as much money.

That is not a paradox. It is the whole point. The size of a yield move tells you almost nothing about the size of the loss. Duration does. The 30-year’s modified duration is 7.84 times the 2-year’s, and that ratio overwhelms everything else.

The bond sleeve lost $9,966 on Thursday. The $1,200,000 equity sleeve, if it tracked the S&P 500, lost $6,960. The safe half lost more than the risky half.

And if the CPI print at 8:30 this morning runs hot and the whole curve adds another 25 basis points, the same three positions lose $1,890, $7,808 and $14,825. That is $24,523, from a single morning’s data release, in the part of the portfolio that exists to not do that.

What the four of them would say

The portfolio manager. Duration is a choice, not a side effect. If you cannot say what the modified duration of your bond sleeve is, you have not chosen it, you have inherited it from whatever looked attractive on the day you bought.

The CPA. There is one consolation buried in that number. An unrealized loss in a taxable bond fund can be harvested against gains and against $3,000 of ordinary income a year. A Treasury you intend to hold to maturity cannot be, because you will get par back and the loss never becomes real. The flaw and the tax feature are the same fact.

The financial planner. A 63-year-old drawing income is not indifferent to when the loss happens. If the sleeve funds the next four years of withdrawals, the maturity dates matter more than the yield. A rung that matures in 2028 pays par in 2028 no matter what Thursday did to its quoted price.

The estate attorney. Every one of these positions gets a basis adjustment at death, which quietly means an unrealized bond loss held to the grave is a loss nobody ever deducts. If harvesting is on the table, the sequence of who holds what, in which account, and for how long is not a trading question.

One more thing

The Treasury Department publishes the daily par yield curve every business day, and it is the document the entire bond market is quoted against. Thursday’s row was not posted when this was written, so the durations above were computed from the closing yields as quoted, and the published curve for September 9 shows the 30-year at 5.28%. Every figure here can be checked against a public document, which is the only reason to trust any of them, including ours.

Tomorrow

The CPI number lands at 8:30 this morning. We will show you what it did to the same three positions.

Sources

CNBC US markets, closing levels and Treasury yields, September 10, 2026. CNBC, “The likelihood of a Fed interest rate hike next week just got a lot higher,” September 10, 2026. U.S. Department of the Treasury, Daily Treasury Par Yield Curve Rates, September 2026. CME Group FedWatch, as cited by CNBC, September 10, 2026. Photograph: U.S. Central Command, NAVCENT Public Affairs, taken April 11, 2026, public domain, via DVIDS.

The conversation a portfolio manager, a CPA, a financial planner and an estate attorney would have about your money if the four of them were ever in the same room. None of them ever talk to each other. This is that room, every weekday at 6:45am, sized to a portfolio between $500,000 and $5,000,000.

The Nest Egg is educational material published by a registered representative. It is not tax, legal or investment advice, and it is not an offer or solicitation. Every case shown is a hypothetical illustration of published rates, built to demonstrate arithmetic, and is not a projection of any reader’s result. Duration figures are approximations that ignore convexity and are less accurate for large yield moves. Consult your own tax and legal advisors before acting.