
Fed Chair Warsh praised the bond market for finally looking at data, not the Fed, as the 30-year yield surged to 5.28%, the highest since 2006.
The 30-year Treasury yield jumped 7 basis points Friday and 12 for the week, hitting 5.28% – the highest since July 2006. That puts it 165 basis points above the effective federal funds rate, the rate the Fed targets with its policy tools.
Fed Chair Kevin Warsh used the FOMC press conference Wednesday to praise the bond market for finally looking at economic data and not at the Fed. He said ending “forward guidance” was already working. Treasury yields had surged since the June meeting, when forward guidance ended, as markets began to focus on data. Buyers and sellers were raising rates and tightening financial conditions, Warsh said, and this “has provided us some comfort that we’ve got the ability and capability to deliver.” In his telling, the bond market was finally doing its job.
That job has been brutal for holders of long-dated Treasuries. The market value of 30-year bonds sold at auction in mid-2020 has plunged about 50%. Investors who bought then can hold to maturity and get their money back in 24 years, but they collect only 1.3% a year in interest. Current buyers earn 5.28%. The difference is the cost of believing the old forward guidance.
Warsh scuttled forward guidance early in his tenure. He had blasted the Fed for locking itself in as inflation surged toward 9% in 2021 while rates sat at zero and QE ran full tilt. When the Fed finally tightened, it was too late. Some banks that had loaded up on long-term Treasuries and MBS in 2020 and 2021 collapsed in 2023. Warsh told the bond market to figure things out on its own. Now buyers and sellers react to inflation and supply data, not to the Fed.
The 10-year yield rose 7 basis points Friday to 4.75%. It briefly hit 5% during the debt scare in October 2023, and that level opened the floodgates of demand, pushing yields back down. There is no guarantee the same floodgates open at 5% again. The 10-year at 4.75% is not historically high – it was pushed to very low levels by the Fed's interest-rate repression and QE starting in 2008. Now inflation is out of the bottle, and the Fed cannot do QE in this environment. Warsh wants to shrink the Fed's balance sheet further to help bring down inflation, the opposite of QE, which could put upward pressure on long-term yields. At Wednesday's meeting, he did not have a majority for anything beyond maintaining the status quo.
Short-term yields moved the other way. The three-month Treasury yield fell 13 basis points during the week to 3.83%, about 20 basis points above the EFFR. Markets had already priced in a rate hike for July or September. The July hike did not come, and conviction on a September hike has softened.
The yield curve has steepened and is starting to look healthy, but the steepness is still modest. The spread between the 2-year and 10-year yield is only 45 basis points. During periods of economic growth, that spread spent lots of time in the 100-to-250 basis point range. The spread between the 3-month and 10-year yield is 92 basis points – also low for growth periods. Both suggest the 10-year yield at today's level is well below where it could end up.
The yield curve inverted in mid-2022 as the Fed hiked short-term rates and long-term yields lagged. That inversion triggered endless recession calls. When the curve un-inverted temporarily in early 2025, it triggered more recession calls on the theory that the un-inversion itself predicts a recession. Then in the second half of 2025, another sag appeared in the middle as rate-cut mania pushed down yields one to three years out. All that is behind now. The curve finally looks normal – just not steep.
For a broader view of how yields are transmitting through risk assets, see our market analysis.
The next FOMC meeting is in September. Markets are pricing in a rate hike with less conviction than before.
Prepared with AlphaScala editorial tooling from the source reporting linked above. Indexable analysis may include a cited Alpha Score value. Publishing checks screen each story before release. Educational coverage, not personalized advice.