The European Central Bank has published a working paper, “Monetary policy without an anchor”, analysing monetary policy when households and firms are imperfectly informed about the central bank’s objectives and learn from its policy choices. The model generates a “reputation channel” in which policymakers tighten more aggressively after adverse supply shocks, accepting short-run output costs to secure more stable long-run inflation expectations. Using high-frequency identification around policy announcements, the authors estimate how long-horizon inflation expectations respond to interest-rate surprises across emerging and advanced economies and find especially large negative semi-elasticities for Brazil. A 1 percentage point surprise in the Selic rate is associated with a 0.48 percentage point decline in five-year, five-year forward market-based inflation compensation and a 0.18 percentage point decline in survey-based expectations four years ahead, while other economies show far smaller responses. Calibrating the model to match these estimates and Brazil’s macro dynamics, the paper quantifies sizable reputation-driven incentives to prioritise inflation stabilisation, with comparable inflation-output trade-offs in a setting without reputational motives requiring a Phillips curve slope about four times larger on average and about nine times larger in periods of high uncertainty.
European Central Bank 2026-04-13
European Central Bank working paper links reputation concerns to aggressive rate hikes and finds Brazil’s long-run inflation expectations drop 0.48 percentage points after a 1 percentage point surprise
The European Central Bank has published a working paper analysing monetary policy when households and firms are imperfectly informed about the central bank’s objectives and learn from its policy choices, generating a “reputation channel” that induces more aggressive tightening after adverse supply shocks. Using high-frequency identification, the paper finds especially strong sensitivity of long-horizon inflation expectations to interest-rate surprises in Brazil and, when calibrating the model to Brazilian data, shows that reputation-driven incentives can replicate observed inflation-output trade-offs that would otherwise require a much steeper Phillips curve.