Mortgage rates moved sharply higher this week, ultimately reaching their highest levels since early 2025. Most of the damage took place on Wednesday and Thursday, but Friday ultimately made things slightly worse. The average top-tier 30yr fixed rate rose almost a quarter of a percent in total. 

Oil prices remained one of the biggest problems. Bonds have been highly attuned to oil price movement throughout the Iran war because higher energy prices imply higher inflation. That matters to longer-term rates directly, and it can also affect expectations for Federal Reserve rate policy. If you'd like to skip the rest and simply view rising rates in the context of rising oil prices, that's probably good enough.

For those who want the full story, read on.

Tuesday offered a brief glimmer of hope when bonds initially held their ground despite another increase in oil prices. That resilience proved temporary. Oil continued higher. Bond yields (which correlate with interest rates) followed. 

Wednesday brought a separate source of volatility involving the Treasury Department's bond buyback program. Treasury buybacks allow the government to purchase older bonds that are less heavily traded. The program can improve liquidity and temporarily add demand, but it does not reduce the government's overall financing needs. Treasury must ultimately issue debt to fund the purchases.

Treasury Secretary Scott Bessent has recently spoken about buybacks as a way to send a signal to the bond market. On Wednesday morning, he went so far as to say, "I am the house now," followed by "you can bet against me if you want." Traders apparently accepted the invitation (see the spike in the blue line in the chart above).

The Treasury announced a $6 billion long-end buyback, which was larger than the previously announced $4 billion minimum but smaller than some traders had anticipated. Bonds sold off immediately, pushing 10yr yields to another long-term high just under 4.86%. The market may have made too much of the announcement, but the reaction added momentum to an already difficult week.

Thursday was worse. Oil surged overnight and broke above $100 per barrel before the Producer Price Index (PPI) was released. PPI was roughly in line with forecasts, but several internal components suggested an unfriendly result for the Federal Reserve's preferred inflation index, the Personal Consumption Expenditures (PCE) price index.

That distinction matters because the Fed officially targets PCE inflation rather than CPI or PPI. Both reports contain components that help economists estimate PCE before it is released. Thursday's PPI details added to the concern that core PCE inflation will remain too high for the Fed's comfort, ultimately pushing 10yr yields to 4.95%, just 0.05% below the highest level since 2007.

Mortgage rates, again, are still under their early 2025 highs. They've performed slightly better than 10yr Treasuries over the past year for a variety of reasons. One reason is simple: the increased MBS purchases by Fannie and Freddie have helped compress the spread between mortgage rates and Treasuries. The other reason is more esoteric as it has to do with mortgage debt behaving more like a 5-7yr Treasury combined with the fact that 5-7yr Treasuries have outperformed the 10yr since 2023.

Friday's Consumer Price Index (CPI) created one of the week's more interesting reactions. The most important metric, monthly core CPI, was slightly hotter than expected. At first glance, this is bad news for rates, but there was a paradoxical reaction. The CPI reading caused an immediate increase in Fed rate hike expectations at next week's meeting. This too would normally be bad for mortgage rates, but there is a twist in the present narrative. 

Part of the recent rate spike has been driven by the market's concern that the Fed has not been aggressive enough in fighting inflation. The inflation data was high enough to increase rate hike odds, but not so high as to cause inflation panic. As such, it threaded the needle and helped longer-term rates move lower even as the very shortest-term rates moved higher. 

Unfortunately, the paradox was short-lived. A reversal in oil prices only explains part of the reversal in bonds/rates. The rest of the explanation relies on the esoteric concept of short covering. In not so many words, traders who had open bets on higher rates simply closed those positions quickly (i.e., they "covered" their "short" positions). From that point on, the market was free to trade as it pleased. This is a common pattern as short covering often produces a meaningful rally, but it tends to run out of steam unless new buyers step in.

The afternoon reversal eventually became large enough for multiple lenders to reprice. The average lender ended the day roughly 0.05% higher than Thursday morning. While this took rates to the highest levels since early 2025, it was a much smaller jump than Wed/Thu. 

Next week's focus is squarely on Wednesday's Fed announcement. A quarter-point hike is now mostly priced into financial markets, but many traders are waiting to see if Warsh has the will to actually pull the trigger. Of course, there are multiple Fed voters, but it's still the Fed Chair and Board who set the tone for Fed meetings. As such, Wednesday afternoon is also at risk of paradoxical movement. For instance, a decision to hold rates steady could revive the concern that the bond market will be left to do more of the inflation-fighting work without sufficient help from the Fed. In addition, the Fed's economic projections and Warsh's press conference introduce additional volatility potential.

Last but not least, and completely independent from Fed policy, war headlines and oil price volatility remain constant risks for rates. As this week demonstrated, even a manageable inflation report can only do so much when energy prices, Treasury supply and policy uncertainty are all pushing in the same direction.