The 10-Year Treasury Yield at 5% Becomes a Budget Bill

The 10-year Treasury yield closed at 5.01% on Sept. 16, the first finish at that level on the Federal Reserve’s constant-maturity series since 2007. It printed 5.00% the day before, fell to 4.94% after the Fed raised rates, then finished Friday at 5.01% on Treasury par quotes.

That round number is what screens flash. The cost shows up in the budget books, in this week’s note sales, and in the mortgage rate that moved with it.

Two Closes at 5% in a Single Week

The Fed series, drawn from the H.15 release, put the 10-year at 4.96% on Sept. 11, 4.97% on Sept. 14, 5.00% on Sept. 15 and 5.01% on Sept. 16. Thursday’s 4.94% print was the post-meeting dip. Friday’s rebound put Treasury par yields back on 5.01%.

The rest of the curve did not move as one. Two-year yields led the week, up 13 basis points to 4.76% by Friday, while the 30-year eased 1 basis point to 5.34%. The 10-year ended 25 basis points above the two-year. Mohamed El-Erian, the Allianz adviser, wrote that the 10-year “is back at 5%” and the 30-year “is above 5.30%.”

TREASURY PAR YIELDS, SEPT. 11 VS. SEPT. 18

Private forecasts now sit on both sides of the tape. ING has a next target of 5.25%. KKR’s year-end mark is 5.10%. TD Economics is still at 4.70% for the fourth quarter. The 52-week range on one widely watched quote ran from 3.923% to 5.045%, and the yield is about 21% higher than the 4.13% reading a year earlier.

Five percent is acting like a magnet, not a one-session spike. The market tagged it, backed off after the Fed, then printed it again into the weekend, which is why the coupon sales that start on Tuesday matter more than the round number itself.

CBO Built Its Budget on 4.1% Yields

In its February outlook, the Congressional Budget Office, under Director Phill Swagel, projected a 4.1% yield this year on 10-year notes and 4.3% in 2027. That forecast was locked in before the Iran war pushed oil and inflation views higher. Friday’s 5.01% close sits 91 basis points above the 2026 figure CBO used to build the books.

CBO also put net interest at $1.0 trillion for fiscal 2026, rising to $2.1 trillion by 2036, or from 3.3% of GDP to 4.6%. Its monthly budget review counted $857 billion of net interest in the first nine months of this fiscal year, already most of that full-year forecast with a quarter of the year left. Debt held by the public was $31.7 trillion in June. Total public debt outstanding passed $40 trillion in mid-August, according to the Bipartisan Policy Center, against a $41.1 trillion statutory limit.

THE TAB ON A $40 TRILLION DEBT STACK

• This year’s interest line: CBO’s February outlook put net interest at $1.0 trillion for fiscal 2026, more than it projected for defense.
• Year-to-date actual: The same office’s monthly review counted $857 billion through the first nine months, about $100 billion more than the same stretch of fiscal 2025.
• The deficit: CBO projected a $1.9 trillion gap for fiscal 2026; the Bipartisan Policy Center had the cumulative deficit at $2.0 trillion through August.
• Debt ratio: CBO has debt held by the public rising from 99% of GDP at the end of 2025 to 120% in 2036.

J.P. Morgan’s private bank has argued there is little evidence that fiscal news, on its own, explains the latest jump in yields. The Iran war, a Fed that just hiked, and heavy issuance from other governments and from companies building data centers are all bidding for the same buyers. That may be right about the last two months. It does not shrink the bill. A larger stock of debt means each extra basis point costs more once old notes roll off and new ones price in this market. Fiscal Lab on Capitol Hill, using CBO’s own rule of thumb, estimated that if yields stay elevated through the decade, outlays would run about $1.3 trillion higher over 2026 to 2036 than the February baseline.

$183 Billion of Notes Go on Sale This Week

The next test is not a speech. It is three coupon auctions. Treasury’s August refunding kept September two-, five- and seven-year sizes unchanged from the summer run, and those sales land Tuesday through Thursday.

THIS WEEK’S COUPON AUCTIONS

• Tuesday, Sept. 22: $69 billion of two-year notes, settling Sept. 30.
• Wednesday, Sept. 23: $70 billion of five-year notes, same settlement.
• Thursday, Sept. 24: $44 billion of seven-year notes, bringing the three-day total to $183 billion.

Those sizes have been the standard print all year: $69 billion, $70 billion and $44 billion, month after month. What changes is the yield buyers demand. August’s two-year sale stopped at a 4.204% high yield with a 2.60 bid-to-cover. The two-year now sits at 4.76%, so this week’s stop-out will tell whether the new 5% neighborhood in the 10-year is leaking into the front of the coupon curve.

Treasury Secretary Scott Bessent has already tried to cushion the long end. In August the department said it would double purchases of off-the-run 10- to 30-year securities, taking each operation up to $4 billion, after long yields hit their highest levels in almost two decades. The refunding still assumes a $950 billion cash balance at the end of September. Buybacks can smooth a bad day. They do not retire the $183 billion that has to be sold this week.

Mortgage Rates Follow the 10-Year to 6.95%

Housing feels the 10-year with a lag and a spread. Freddie Mac said the 30-year fixed rate averaged 6.95% in the week ended Sept. 17, up from 6.76% the week before and 6.26% a year earlier. The 15-year averaged 6.26%, up from 6.09% and from 5.41% a year ago.

A 0.19 point jump in one week is a large move for that survey. Against the Fed’s 4.94% 10-year close on Sept. 17, the 30-year mortgage sat 2.01 points over Treasuries. That gap is the extra yield investors want for prepayment risk and for packaging the loans. It is also why a 5% 10-year does not produce a 5% mortgage, and why buyers who sat out the 2023 spike are meeting another expensive tape.

A year ago the 30-year was already 6.26%, so this is not a return from 3% money. It is a second leg higher after a stretch when rates had eased into the mid-6s. Purchase applications in Freddie’s sample are conventional, conforming loans with strong credit and a 20% down payment, which means the advertised average is the clean quote. Weaker credit and smaller down payments price richer still.

Rate-sensitive stocks already showed the split on days when the 10-year held 5%. Utilities and other long-duration names lagged while a handful of growth names still advanced, the usual pattern when discount rates reset and earnings further out get marked down.

Why the Fed Tightened Into a Bond Selloff

Kevin Warsh, the Federal Reserve chairman, oversaw the first rate increase since July 2023 on Sept. 16. The Federal Open Market Committee voted 12-0 to raise the funds range to 3.75% to 4%, reversing a cut from last December. Twelve of 18 officials in the projections see one more hike this year, and four see two. The median funds-rate path is 4.1% at the end of 2026 and still 4.1% at the end of 2027. CBO’s February forecast had assumed a 25 basis-point cut this year.

We must be confident that underlying inflation is moving to our objective, clearly and at sufficient speed. Today, the FOMC decided that this standard has not been satisfied.

Kevin Warsh, Federal Reserve chairman, Sept. 16 press conference

Warsh also said “the plain fact is that inflation is too high and has been for too long.” The median projections put PCE inflation at 3.7% this year and 2.3% next, with unemployment near 4.1% and real GDP at 2.3% this year. Core PCE was 3.3% in the 12 months through July. The Iran war, which began in late February, has kept energy prices elevated, and the 10-year has risen about a full percentage point since then.

Ed Yardeni, the economist, has called a 4% to 5% 10-year “a vote of confidence in the economy’s strength.” That reading fits a week when the two-year led the selloff and the 30-year barely budged, a flattening that looks like a repricing of the Fed path more than a buyers’ strike on long debt. The oil shock still sits underneath both views. If energy stays expensive into winter, Warsh’s inflation test gets harder to meet, and the 10-year has less room to fade even if growth slows.

The Last 5% Close Dates to July 2007

Traders treated 5% as a line last week because the Fed series had not settled there in 19 years. The yield did trade through 5% in October 2023, then fell back. This time the close held, then held again on Friday’s par quotes.

THREE TIMES THE 10-YEAR MEETS 5%

1. July 19, 2007: Settles above 5% for the last time before the financial crisis ushers in a long stretch of ultra-low long rates.
2. October 23, 2023: Trades through 5% during the day, with an intraday high of 5.005%, then fails to hold a close on the Fed series.
3. Sept. 15-16, 2026: Fed series closes at 5.00%, then 5.01%, the first back-to-back finishes at that level since 2007.

The 2007 close sat on the eve of a crash that sent the 10-year to 2% and, years later, to 0.52%. The 2023 spike faded as inflation cooled and the Fed stopped hiking. This episode is different in one respect the last two did not share at the same scale: a $40 trillion debt pile, a $2.0 trillion deficit through August, and a Treasury that still has to clear $69 billion, $70 billion and $44 billion of coupons in three sessions while the 10-year sits on 5%.

Tuesday’s two-year sale is the first of those three. The stop-out yield and the bid-to-cover on that $69 billion will say whether 5% on the 10-year is a level buyers will fund, or a level they will only visit.

Disclaimer: This article is news reporting and analysis of Treasury yields, federal budget figures and mortgage rates, and it is for information only. It is not investment advice, a recommendation to buy or sell bonds, notes, mortgage products or any other security, and it is not tax or financial-planning advice. Readers should consult a licensed financial adviser or investment professional before making decisions about bonds, mortgages or other rate-sensitive holdings. Yields, auction sizes, deficit totals and policy rates reflect the official series and statements cited here and can move with the next close, the next sale or the next Fed meeting.