The 10-year Treasury yield closes above 7%

Where things stand (as of 14 Oct 2026)

• Latest level. Treasury's par yield curve shows the 10-year at 5.21% on 13 Oct 2026. FRED DGS10 was 5.24% on 9 Oct. The highest close since the opening date is 5.31% on 5 Oct, a 24-year high. To resolve YES the series needs a print of 7.01 or more, about 180bp higher. The last DGS10 close above 7% was 7.05% on 8 Jul 1996 (I computed this from the full FRED file).
• Drivers of the rise. The US–Israel war with Iran began 28 Feb 2026. Hormuz traffic is still a trickle and Brent is about $105. The Fed hiked to 3.75–4.00% on 16 Sep, and the minutes say most participants expect another hike by year-end. Futures price about 3.6 hikes through Dec 2027. Other pressures are a roughly $2T deficit and heavy borrowing to fund AI build-outs. The 10-year is up about 120bp this year, and Q3 was the worst quarter for Treasuries since 1994.
• Breakdown of the yield (FRED). The 10-year TIPS (inflation-protected) yield is 2.91% and 10-year breakeven inflation is 2.33%, so the move has been in real yields, not inflation expectations. The Kim-Wright term-premium estimate is 1.07%, high for the post-2011 period but below the 1990s range of 1.5–2.5%.

Markets and forecasters

• Prediction markets. Polymarket's "How high will the 10-year go before 2027" prices reaching or exceeding 5.4% at 61%, 5.5% at 33%, 5.7% at 12.4% and 6.0% at 6.75%. Kalshi prices hitting 5.50% by 31 Oct at about 13–15%.
• Strategists. The Reuters poll of 5–7 Oct has a median of 5.00% at year-end and 4.75% in a year, but 28 of 30 respondents see upside risk, and the poll has been too low nine months running.
• Named views. Pimco's Ivascyn says 6% is "feasible". BofA says 7% is not its base case, while warning the Fed funds rate could pass 5%.
• Fading after spikes. RSM notes that after the six post-1990 episodes where the 10-year rose at least 75bp in six months, yields gave back half within three quarters; only 2022 kept climbing.
• Treasury response. Treasury doubled its buybacks and may cut 20-year issuance at the 4 Nov refunding.

Base rates I computed

1. US history (DGS10, monthly start dates). Share of start dates followed by a maximum rise above 180bp within 5.2 years:

   • 1962–2021: 41.5%
   • Post-1985: 17.4%
   • Post-1985, starting level 4.5–6.5%: 15%
   • Post-1985, after a rise of at least 80bp over the prior 12 months: 9%

2. 17-country OECD panel, post-1990 (FRED IRLTLT01xxM156N, monthly averages):

   • All start dates: 15.5%
   • Starting level 4–6.5%: 7.4%
   • Starting level 4–6.5% plus a 12-month rise of at least 80bp: 1.9–2.8% (only euro-crisis cases)
   • Including the 1960s–80s, at 5-year highs with that momentum: 31–44%

   The result depends heavily on whether the inflation regime looks like the one before or after 1985.

3. Simple statistical model. I simulated a yield that drifts back toward a long-run average (volatility 0.8–0.9 points a year, which matches realized volatility of 70–90bp; half-life 3–5 years; long-run mean 4.5–5%). It gives 2–4% by end-2027 and 17–28% by end-2031. I treat that as an upper-side anchor, because policy responses and the economic drag from high yields thin the top tail.

My synthesis

Getting to 7% needs something like a Fed funds rate near 6%, or a funds rate above 5% plus a 1990s-size term premium. Alternatively, inflation expectations would have to break loose, or the market would have to start treating Fed policy as driven by government debt.

I weight about 85–90% on a regime like the one since 1990, where I put the chance near 4% (somewhat above the panel base rate because of this year's unusual shocks). I weight 10–15% on a 1970s-style regime, where the chance is about 55%. That gives roughly 10–13%, which sits between the panel base rate and the simple model.

Shape over time. Q4 2026 would need a move of about 180bp in 2.5 months, which has almost no precedent outside 1987 and 1994; I put it at 0.2%. The hazard rises through 2027–28, when the war and oil path, more hikes and possible Fed balance-sheet cuts play out (task forces report early 2027), at roughly 0.8–0.9% a quarter. After that it eases to 0.5–0.6% a quarter, mainly for fiscal or Fed-independence shocks.

Strongest case against. The post-1990 developed-market base rate is about 2%. Breakevens are anchored. Payrolls are softening (+29k in September). The Treasury and the Fed have tools to cap long yields, and a Hormuz reopening would pull oil and yields down. On the other side, bond bear markets have run 3 points or more (1986–87 and 2020–23), and this one is up only about 1.3 points from its February low.

Sources: FRED DGS10, DFII10, T10YIE and THREEFYTP10 (fred.stlouisfed.org); Treasury daily par yield CSV (home.treasury.gov); Polymarket gamma API; Kalshi API; CNBC 7 Oct 2026 (FOMC minutes and auction); CNBC 28 Sep 2026 (BofA); Reuters poll 7 Oct 2026 (galvnews mirror); RSM yield scenarios (Oct 2026); Savvy Wealth Oct 2026; CNBC 25 Sep 2026 (Warsh balance sheet); BigGo 13 Oct 2026 (20-year issuance).

Question. By the last day of each quarter, has the US 10-year Treasury yield closed above 7% on any day?

Definitions.

• US 10-year Treasury yield: the nominal 10-year constant-maturity par yield, in percent per annum, reported in the US Treasury’s Daily Treasury Par Yield Curve Rates, column 10-year (or 10 YR). The corresponding Federal Reserve H.15 series is Treasury constant maturities — Nominal — 10-year, distributed by FRED as DGS10, “Market Yield on U.S. Treasury Securities at 10-Year Constant Maturity, Quoted on an Investment Basis.”
• Close: the official daily observation in that series, derived from indicative closing market bid quotations collected at approximately 3:30 p.m. New York time; not an intraday high or the last trade in a particular Treasury note.
• Above 7%: a published value strictly greater than 7.00%; 7.00% does not qualify. Use the published precision without inferring additional decimal places.

Resolves YES if a qualifying source publishes at least one daily observation above 7.00%, for an observation date after 2026-09-16, by the quarter-end. The Observation period and Resolution window conventions apply. An official revision that first raises an eligible observation above the threshold qualifies on the revision’s publication date. Once YES, subsequent revisions or lower yields do not reverse it.

Resolves NO if no such observation has been published by the quarter-end. Intraday quotes, yields on individual securities, auction results, inflation-indexed yields, other maturities, forecasts, and observations dated on or before the opening date do not count. Missing observations and nonbusiness days are not filled by carrying forward a previous value.

Resolving source. The US Treasury’s Daily Treasury Par Yield Curve Rates, as described in its Interest Rate Statistics; the Federal Reserve Board’s H.15 Selected Interest Rates; or the Federal Reserve Bank of St. Louis’s FRED DGS10. Treasury takes precedence for conflicting values available at the same time, followed by H.15, then FRED. An explicitly designated official successor measuring the same nominal 10-year constant-maturity yield counts. These official statistical publications, rather than other A2 sources, resolve the question. Apply Resolution date (A3): use the date the qualifying value was published, not its observation date.