Pill: Should Act Based On 2-3Y Inflation Outlook, Not Just Incoming Data

Q: If we don't rely on the models so much, does that mean we have to rely on each data point more?

• A: Would push back on that quite strongly. Instrument is Bank Rate and there are uncertainties about how that is transmitted, how the economy behaves and what new shocks hit the economy.
• Always address that by looking at the data. Whether that's a formal model or informal model. What does that tell us about the inflation outlook in 2-3 years? And how should we steer the path of inflation so it remains at target in 2-3 years time. That's what we have always done and what we have continued to do, even before inflation targeting. That framework is "timeless".
• Always a question of how much we can rely on forecasts and models. Should not be just responding to today's data, but to what that data does to inflation in 2-3 years. Some things - e.g. gas prices in response to Ukraine invasion - pass through much faster than monpol can react.
• Given shocks we have uncertainties, our ability to filter those using models developed when those shocks were absent, views are up for grabs. We have to approach by saying, what is plausible, and what conclusions can we draw from that?
• Another key challenge is how to communicate what is being done. There is a balance between simplicity and too much complexity for people to understand.