We see four main reasons behind this decision:

(1) The US economy is proving resilient, and concerns over a sharp deceleration of the cycle witnessed over the first half of 2025 are fading. The recent dip increasingly appears to have been only a soft patch – likely linked to elevated policy uncertainty, some of which now seems to be receding, particularly on the Sino-American trade front.

Kevin Thozet

(2) Going forward, we expect the US economy to reaccelerate, supported by a prolonged AI-led capex boom and a large fiscal boost from tax refunds that will kick in this spring (not to mention the risk that Trump introduces additional fiscal easing to support households ahead of the midterm elections). An increasing number of indicators point to bright spots in the US economy across the labour market (unemployment rate stabilising, fewer layoff announcements), consumption (the University of Michigan Consumer Sentiment Index showing improvement, robust retail sales) as well as housing and manufacturing.

The job market, which had been slowly deteriorating towards a 4.4% unemployment rate, could, in fact, be stabilising. When looking at a broad range of indicators (in public surveys as well as private surveys such as ISM employment, NFIB small business hiring plans or ADP and Indeed postings) most are pointing to a tightening of the labour market in an environment where wage inflation is still trending at 4%.

(3) With policy rates having been lowered by 75 basis points over the past four months, monetary policy is getting closer to the neutral rate of interest – estimated at 3%. As a result, the Fed is in less of a hurry to cut rates further. This is even more the case given the conundrum of the two-speed economy, which implies that the rate that neither stimulates nor restrains the economy (when inflation is at target and the economy is at full employment and both are big “whens”) is likely different between the upper leg of the K-shaped economy (i.e. higher-income and older households for whom 3.75% is likely accommodative) and the lower leg (for whom 3% is likely restrictive).

(4) Finally, the hard line adopted by the Trump administration and its open attacks on Fed members (Powell’s subpoena, the (likely failed) firing of Lisa Cook) have led to a real escalation in tensions between the executive and monetary authorities. As a result, any decision – whether to hold or to cut – will be seen as politicised (triggering renewed pressure from Trump in the former case, and concerns over independence in the latter), thereby putting additional pressure on long-term yields.

In this context, we maintain a negative view (and positioning) on US sovereign bond yields, with a bias towards steepening. Markets expect the Fed to hold rates unchanged on Wednesday, but they still expect the easing cycle to resume thereafter, assigning a roughly 20% probability to another cut before Jerome Powell leaves his role as Chair (his continuation as a member of the Board of Governors remains uncertain), and nearly two full cuts priced in by year-end – a scenario we do not think is warranted absent a strong deterioration in the labour market.

Longer-term bond yields are at risk from the Fed’s desire to support the labour market even though inflation has been above target for 57 consecutive months, as well as from a backdrop of stronger growth and higher inflation. This is all the more relevant given that, unlike in most other major economies, long-term US interest rates trended lower last year as many investors were worried about recession risk – a dynamic likely to reverse in the context of a reaccelerating economy.