Bache: Norges Bank refines models for policy decisions
Ida Wolden Bache, Governor of Norges Bank, discussed the central bank's use of macroeconomic models in monetary policy decision-making. She also outlined the ongoing project to renew Norges Bank's modelling system at a seminar in Oslo.
Modeling the Norwegian economy
Norges Bank relies on macroeconomic models to guide its monetary policy, aiming for low and stable inflation near 2% and high employment.
For short-term analysis and forecasts, the central bank employs data-driven empirical models, such as autoregressive, factor, and vector autoregressive (VAR) models.
These are part of a system that updates forecasts with new data, weighting models based on performance.
Key information sources include Norges Bank's Regional Network, conducting 400 quarterly interviews with 1600 firms, and the Expectations Survey, monitoring inflation expectations.
For medium-term forecasts (three to four years), Norges Bank uses NEMO, a dynamic, stochastic, general equilibrium model adapted for Norway's small open economy, including its petroleum sector.
NEMO links current conditions with long-term trends, providing a crucial input for policy rate discussions.
Evolving models for new realities
Recent global shocks have challenged Norges Bank's models, leading to an underestimation of inflation impulses.
This experience underscores the need for models that better capture abrupt economic shifts and international supply-side impulses affecting Norwegian wage and price formation.
Norges Bank is renewing its modelling system, exploring artificial intelligence and machine learning to enhance forecasting and model development.
The bank is also incorporating more heterogeneity and microdata, especially on households' finances.
Research using this microdata reveals that households with higher debt-to-income ratios significantly reduce consumption after policy rate increases, highlighting the cash-flow channel's importance.