Key Highlights

• Inflation expectations drive 2y yields, while the 10y is more nuanced.
• Foreign holdings of USTs have declined significantly since April.
• Despite strong labor data, we maintain our call for two rate cuts by year end.

View chart pack

Not all yield changes reflect the same concerns

EXHIBIT #1:  DIFFERENT DRIVERS OF 2Y AND 10Y YIELD CHANGES

_{Source: BNY Markets, Federal Reserve Bank of New York, Bloomberg}

EXHIBIT #2: ONLY A SMALL INCREASE IN THE EXPECTED POLICY RATE

_{Source: BNY Markets, Federal Reserve Bank of New York, Bloomberg}

Bond yields are up significantly since the beginning of the U.S.-Iran conflict. To wit, the 2y note yield has risen from 3.38% on February 27 to 3.86% as of Monday morning, a move of 48bp, while the 10y yield went from 3.94% to 4.34%, or 40bp. It’s worth noting that the drivers of these moves at different ends of the curve differ. Exhibit #1 shows the change in various components for the 2y and 10y notes’ yields. We look at the change in the nominal yields as well as those for the breakeven (or expected) inflation rate at each maturity, the change in real rates, and the change in term premium.

Almost the entire 48bp increase in the 2y nominal yield has been through rising inflation expectations, or an increase of 44bp on the breakevens. The real rate is up less than 1bp. Through a separate set of calculations, the Federal Reserve Bank of New York computes an 11bp increase in the term premium, reflecting an increase in uncertainty regarding the expected policy rate path. The fact that the 2y real interest rate is barely higher than it was before the conflict suggests that investors don’t put a lot of weight on a significant – if any – increase in the federal funds rate.

At the 10y maturity, only 11bp of the 40bp increase in the nominal yield came from higher breakeven inflation, while nearly 30bp of the increase are attributed to higher real rates. This suggests somewhat higher policy rates over the next decade, but still relatively stable federal funds rates. As with the 2y note, the term premium has stayed relatively stable at the 10y maturity, up just 6bp.

In short, the 2y yield has moved due to higher short-term inflation expectations, while further toward the back of the curve, investor focus seems relatively unconcerned with inflation over a long horizon, but higher base rates instead. Long-term inflation expectations are indeed well-anchored, which probably explains the relatively muted behavior of implied policy rates.

In its calculation of the term premium, the NY Fed also produces an estimate of the market’s expectation for policy rates at each maturity for which it calculates the term premium. Exhibit #2 shows the evolution of the “risk neutral expected short rate” for the 2y and 10y maturities. There has been a modest increase in these expectations, almost equal for the two maturities, but hardly a sea change in central bank expectations. This is understandable and suggests that markets are either looking through the current disruption of the war in the Middle East – either to prices or central bank policy in the long term.

Checking on cross-border U.S. Treasury demand

EXHIBIT #3: REDUCING UST EXPOSURE SINCE LAST APRIL

_{Source: BNY Markets, iFlow}

EXHIBIT #4: IS THIS THE BEGINNING OF A LONG-TERM DECLINE OR JUST A REALLOCATION?

_{Source: BNY Markets, iFlow}

Cross-border holdings of U.S. Treasurys have been falling almost inexorably since last April and “Liberation Day,” as seen in Exhibit #3, which plots proprietary iFlow data, BNY’s custody-based data on real money investment activity. Lower holdings of short duration USTs (i.e., less than 1y in maturity) account for a significant portion of the decline, as foreign cash holdings move back to home shores.

In Exhibit #4, we show a longer-term look at UST holdings, both overall (i.e., both domestic and overseas holdings) going back to 2018, a few years before the COVID pandemic. Note that over this nearly decade-long sample period, total holdings of USTs – outside early 2020 when the global shutdown began – have been relatively steady, while cross-border holdings have moved with much more volatility. Interestingly, between 2024 and the beginning of 2025, overseas investors actually increased their exposure to USTs, only to more than reverse their positions since April 2025.

It’s too early to tell if this shedding of USTs by foreign-based investors is the beginning of a secular decline in demand for USTs or merely a response to a higher inflationary environment and U.S. policy uncertainty. Over the eight-year period shown in the graph, we note that the level of cross-border holdings is still close to the period average, and that the recent rundown in holdings since early 2025 is merely a reallocation out of their previous overweight into more diversified sovereign bond holdings. 

Rates outlook

March’s strong jobs market data could appear to have put our forecast for two rate cuts in some doubt, but we aren’t considering changing it. The March data represent to us a snapback from a particularly weak February reading, which was influenced by strikes and poor weather. We still believe the job picture will deteriorate progressively from here through midyear, and once the conflict eventually ends, price pressures will likely recede, and the Fed will lean into its labor mandate in Q4.

The analysis we present above signals that the market is either unsure of rate policy and therefore unwilling to take too severe a directional view, or it’s convinced that current conditions are temporary, and the policy outlook on the other side of the conflict shall return to status quo ante.

Despite our medium-term view in favor of rate cuts, this week is likely to add to inflation concerns instead, with February PCE inflation on Thursday (i.e., before the conflict) and March CPI on Friday. However, we think the market should look through what we expect to be a wartime blip and maintain its steady rates view. 

Chart pack