REAL-TIME GLOBAL RESEARCH
European Daily: ECB—Hike or Hold?
Research evidence excerpt
European Daily: ECB—Hike or Hold?
Goldman Sachs European Daily
Exhibit 2: A Simple Taylor Rule to Size the Threshold for Another Hike
Source: Goldman Sachs Global Investment Research
This exercise suggests that our own forecast (in which growth runs 0.2pp ahead of
potential and core inflation is at 2.5% in 2027) would call for 35bp of total tightening—a
weak case for another 25bp hike from here (Exhibit 2).1 Both the staff’s baseline
projections and milder scenario would only call for 15bp of total tightening—implying
reversing some of the 25bp hike delivered to date (as lower inflation in the “milder”
scenario relative to the baseline would be offset by higher growth).2
Second, we explore how changes in the staff projections might affect the policy decision.
To do so, we compute the changes in vintages of the staff projections for growth and
headline inflation in the current and next year, going back to 2000.3 Exhibit 3 shows that
the Council typically hiked the policy rate when the projections for inflation or growth
were revised up. The two instances when the staff revised down the inflation projections
and the Council still hiked are December 2006 (when oil prices fell but growth was
revised up and unemployment continued to decrease) and March 2023 (when energy
prices fell but core inflation data surprised to the upside).
1 We forecast growth of 1.2% in 2027 against our estimate of potential growth of 1%.
2 The ECB staff projects growth of 1.2% and 1.4% in their baseline and milder scenario in 2027, respectively,
against their 2028 projections of 1.4%.
3 The ECB started publishing projections for core inflation in 2013 and started including a two-year ahead
projection horizon in 2014. We also use the midpoint of the projection ranges until 2013.
29 June 2026 3
The English excerpt is extracted automatically from the cited source page and may contain layout or recognition errors. It is never batch translated.
Open report viewer