Bank of Canada Overhauls Forecasting Models After Failing to Predict Persistent Inflation
-
The Bank of Canada is overhauling its economic forecasting models after failing to predict the extent and persistence of recent inflation. The models underestimated supply chain issues and changes in consumer behavior.
-
The new models will move away from the assumption that inflation expectations are always anchored at the 2% target. They will better incorporate real world complexity around how businesses set prices.
-
There will be a core forecasting model, plus a range of "satellite" models with differing assumptions to challenge the main model and specialty models to examine outlier events.
-
The models aim to better capture uncertainty and understand what happens when the economy enters "dark corners" like financial crises, allowing central bankers to hedge risks.
-
Even the best models have limitations, so decision-making will still require judgment. But better models can help identify hidden risks and interactions.