The two-year Treasury yield closed Friday at 4.6254%, up 25.93 basis points on the week, after the September 11 CPI report briefly lifted it to 4.6551%. The 10-year ended at 4.9669% after a W-shaped session that peaked at 4.9792% and reversed to 4.9016%, leaving it 18.48 basis points higher on the week and still below the 5.0187% intraday high from October 23, 2023.
Traders spent Friday adding a quarter point of tightening to their expectations for the next 12 months, and the front end of the Treasury curve showed it: the two-year yield finished at 4.6254%, a weekly advance of 25.93 basis points, after the September 11 consumer price report briefly drove it to 4.6551%.
"The front end is doing the talking now, and it is telling you the market no longer believes the Fed can look through this inflation impulse," said Priya Raghunathan, chief rates strategist at Meridian Macro Partners. "A 26-basis-point weekly move in twos is what a repricing looks like when positioning was leaning the wrong way."
The long end told a messier story. The 10-year benchmark yield added 0.43 basis point to 4.9669%, having spiked to 4.9792% at 20:30 Beijing time when the CPI data crossed, then reversed to a session low of 4.9016% by 21:03 — a W-shaped session that left the yield 18.48 basis points higher on the week and still short of the 5.0187% intraday peak set on October 23, 2023. The 30-year yield reached 5.37%, its highest since July 2004, and a $22 billion 30-year auction cleared at 5.308%, the highest auction yield since August 2001.
The divergence between the two maturities is the signal. Two-year yields track the expected path of the fed funds rate, so a 25.93-basis-point weekly rise means the market has added roughly a quarter point of tightening to its expectations for the next 12 months. Ten-year yields also carry inflation expectations, term premium and Treasury supply, which is why the CPI spike faded within 33 minutes: buyers stepped in at 4.98% and capped the move.
That buying interest is the same dynamic that ended the October 2023 episode. The 10-year briefly traded above 5% on October 23, 2023, drew a wave of demand, and then fell more than 100 basis points over the following four months as inflation cooled. The difference this time is the composition of the move. In 2023 the entire curve was repricing a peak in policy rates; now the front end is repricing the possibility of additional hikes while the long end is being held down by investors locking in yields last seen in 2007.
The repricing accelerated on September 10, a day before the CPI print, when the August producer price index showed year-on-year inflation at 5.4%, up from 4.8% in July, with energy prices up 4.2%. CME FedWatch data put the probability of at least a 25-basis-point hike at the September 16 meeting above 70% after that release, a reversal from earlier in 2026 when futures priced cuts.
The October 2023 peak is not a round number, and that is precisely why it matters. Round-number psychology gets the attention, but the 5.0187% intraday high is the last confirmed swing high on the 10-year chart, which makes it the reference point for every duration-sensitive position sized off technical levels. A close above it would leave no overhead resistance until levels not traded since 2007, and systematic strategies that scale exposure to realized volatility would be forced to reduce duration.
The transmission to risk assets runs through the discount rate. Higher long-end yields compress the present value of future earnings, which hits long-duration equities hardest. The 2-year move matters more for the near-term path of policy, and a market pricing a hike in September has to also price the possibility that the Fed does not stop there.
Kody Sherlund, a New York-based certified financial planner, framed the asymmetry in an interview with NerdWallet: a hike on September 16 would reinforce the Fed's credibility on inflation and could stabilize or ease long-end yields by reducing the inflation risk premium, while holding rates steady could be the rockier outcome if the Fed is seen as comfortable with above-target inflation. That is the inverse of the usual reaction function, and it is why the September 16 decision and the accompanying Summary of Economic Projections carry more weight than a single meeting normally would.
The last time the Fed held rates steady against rising inflation expectations, in late July, long-term yields surged. The September SEP will show where individual Fed officials expect the benchmark rate to end 2026, 2027 and 2028; the June version projected rates falling slightly, and any upward revision would confirm the front end's repricing rather than contradict it.
For positioning, the arithmetic is straightforward. A portfolio holding 10-year duration at a 4.97% yield loses roughly 4.5% of value on a 100-basis-point rise in yields, and gains a similar amount if the 2023 pattern repeats and buyers cap the move. The 2-year at 4.6254% offers a locked-in return with far less duration risk, which is why the front end absorbed the week's selling while the long end found bids at 4.98%.
The next data points that decide the direction are the September 16 Fed decision and the SEP, followed by the next CPI release. A break above 5.0187% on a closing basis would signal that the 2023 playbook no longer applies; a failure there would mark the second time in three years that buyers defended the 5% area.
This article is for informational purposes only and does not constitute investment advice.
