We have updated all models.
We have updated all our models, incorporating for the first time in-sample data for the fourth quarter of 2025. Overall, the quarterly models were only mildly surprised by our nowcast, and revisions remain limited for the time being. We have also updated our monthly model forecasts. From a bottom-up perspective, the forecasts have been revised slightly lower, mainly reflecting softer complementary indicators, particularly from the Adobe data.
Although the models differ in the exact timing and shape of the disinflation path, they continue to converge on the same message at the end of the forecast horizon: inflation remains above—or well above—target.
Caution is warranted, however. Persistent uncertainty surrounds both the data themselves—given the current gaps in data availability—and the quality of recent releases, as a large share of CPI components has been subject to imputation.
As for the Fed, our baseline remains unchanged. Despite Chair Powell’s uncertainty in October, we continue to expect the Fed to deliver a rate cut in December (too scared by the markets in our view), followed by a pause thereafter.
All models forecasts
Table 1. All models’ forecasts for core, headline CPI and PCE.
Please, find below all models’ forecasts for core, headline CPI and PCE. The colors in the table indicates the left-hand-side variable. The numbers in the table can be found here. Given the volatile environment, we will update these models more frequently. Please, get in touch in case you need more details.
Note: “ML” refers to “machine learning” model (see documentation here), “Bottom up” refers to our set of sectoral models, “main” model refers to our Phillips curve model (see here), and “FRB-US” stands for “Federal Reserve Board – US model” (see here). All MoM refer to the seasonally-adjusted figures.