US long-term yields (T-Note 10Y) have risen by 70bp since the start of the year (including 35bp in the last month), mainly as a result of downward revisions to expectations of future Fed rate cuts. At the start of the year, futures were forecasting more than 6 Fed funds rate cuts by 2024. We are now down to 2, which seems about right. In this sense, the upward movement seen in recent months corresponds to a form of normalisation from excessively low levels at the end of autumn 2023.
Nevertheless, there is reason to believe that long rates will continue to fall gradually over the coming months. Why should this be?
Firstly, because disinflation is set to continue, even if only gradually. Unless, of course, there is a major geopolitical deterioration that affects Iranian oil production and/or traffic through the Strait of Hormuz in a lasting way.
The stronger dollar should limit the rise in the price of non-energy imports to between 0 and 2% on an annual basis, with relatively little impact on the price of non-energy industrial goods.
And let’s not forget that rents will continue to slow (see chart below).
Above all, however, robust productivity and wage moderation are helping to slow unit labour costs, whose rise is in line with an increase in core CPI inflation of around 2.5% on an annualised basis (see chart below).
In addition, household inflation expectations have not risen, although they have stabilised at a higher level than before the pandemic (see chart below).
Finally, there are 2 main measures of consumer prices in the United States: the consumer price index (CPI, calculated by the BLS) and the consumer price deflator (PCE, Personal Consumption Expenditures price index, calculated by the BEA), see chart below.
While investors largely focus on the CPI, which is published before the PCE, the Fed prefers to use the PCE (and more precisely its underlying component). This is the indicator on which it publishes its inflation forecasts.
Because of the difference in the way rents are taken into account, core PCE has fallen faster than core CPI. In particular, the weight of the housing component is much greater in CPI than in PCE (around 30% vs. 15%).
The underlying PCE index came out at 2.8% year-on-year in February, compared with 3.8% for the CPI.
Overall, a reasonable scenario is for disinflation to pause in Q1 at around 3.5% (CPI); inflation will then resume a downward trend (at around 2.5% annualised sequential rate in H2). Taking into account an unfavourable base effect for CPI, this would result in CPI headline inflation of 3.3% in Q2 2024 (3.5% for core) and 2.9% in Q4 2024 (3.1% for core). For PCE, a plausible forecast would be 2.5% in Q2 2024 (2.3% for core) and 2.5% in Q4 2024 (2.5% for core).
Another argument in favour of moderating nominal long rates is that the current real long rate appears too high.
Indeed, the rise in US nominal long-term rates since January has mainly reflected the rise in real rates, while inflation expectations have remained virtually stable.
However, real long-term rates of over 2% (see chart below) appear too high and unsustainable in the medium to long term, particularly given the levels of debt.
It should be added that, from a more factual point of view, the trend in nominal long-term rates is above that suggested by nominal GDP growth, which is gradually decelerating (see chart below).
And we should not forget that there is still potential for a downward adjustment in bond yields, given what we have seen over the last 40 years after the key rate peak (in this case, last July), even if we look only at phases that did not result in recessions (see chart below).
It is also possible that the recent bond correction is somewhat disconnected from macro fundamentals and fuelled by technical factors (selling duration to cover convexity, VAR model adjustments imposed by US institutional investors, etc.).
Even so, the easing in nominal long-term rates will only be moderate. In addition to the exceptional resilience of the economy and the laborious nature of disinflation, investors are still taking a bearish interest rate stance, and competition from cash remains significant:
- Surveys continue to be bearish on rates, albeit less so than in autumn 2023 (see chart below). This is moderately unfavourable for long interest rate products.
- At the same time, the yield differential between the 3-month T-Bill (5.4%) and the 10-year T-Note (4.4%) remains substantial (despite narrowing in recent months); see chart below.
As a result, short-dated debt products still compete strongly with long-dated interest rate products.
What about European long-term yields?
Of course, the directional trend in long-term yields will continue to be dictated by the trend in US yields. But we believe that the ECB’s more rapid (cut from June vs. September for the Fed) and more intense (100 bp cut predicted by the market, which we believe to be correct) stance justifies 2 bets: widening the T-note/Bund spread and increasing the weight of Europe in equity portfolios relative to the US.