Macroeconomic models and monetary policy in the U.S.: the last fifty years
Fisher et al. in this paper present a comprehensive overview of how macroeconomic theory used by monetary policymakers has evolved over the past 50 years in the U.S. The authors analyzed the frequency with which specific topics were discussed by FOMC participants during their scheduled meetings. They focused on six key topics, namely, monetarism, inflation expectations, credibility, forward guidance, quantitative easing, and balance of risks.
Among the key findings are:
- Shift from monetarism: Until the mid-1980s, monetarism was the dominant paradigm for policymakers, but its influence waned by the early 2000s.
- Rise of New Keynesian (NK) model: The paper traces the development of the NK model, which has become central to modern macroeconomic policy discussions.
- Importance of expectations: The role of inflation expectations in policymaking grew significantly since the 1990s, becoming the second-most mentioned topic among those tracked.
- Ongoing evolution: While the NK model has been crucial for policy discussions, recent events have revealed its limitations, leading to ongoing research and model development to better capture complex economic dynamics
The Evolving Core of Usable Macroeconomics for Policymakers
Authors: Jonas D. M. Fisher , Bart Hobijn , Alessandro Villa
From: Federal Reserve Bank of Chicago
The rise of ABMs in central banking
Borsos et al. , in this paper, provide a comprehensive review of agent-based modeling (ABM) research conducted at twenty central banks and five related policy institutions over the past decade.
Key findings:
- ABMs have emerged as effective complementary tools for central banks, especially after their mandates expanded following the 2007-09 global financial crisis.
- ABM research has evolved from early work on payment systems to a current focus on financial stability, with growing applications in areas like price stability and climate change.
- ABMs offer several advantages for policy analysis, including:
– Ability to model heterogeneity and complex agent interactions
– Generation of nonlinear dynamics like boom-bust cycles
– Flexibility to assess policy impacts on multiple measures
– Capability to capture country-specific details and adapt to changing circumstances
While acknowledging room for improvement remains, the authors argue that integrating ABMs into central banks’ analytical frameworks can support more effective policy responses to both existing and emerging economic challenges.
Agent-based modeling at central banks: recent developments and new challenges
Authors: András Borsos, Adrian Carro, Aldo Glielmo, Marc Hinterschweiger, Jagoda Kaszowska-Mojsa, Arzu Uluc
From: Magyar Nemzeti Bank, Banco de España, Banca d’Italia, Bank of England, Narodowy Bank Polski
The Bank of Canada’s agent-based model
In this paper, Cars Hommes et al. discuss the ABM called CANVAS at the Bank of Canada. It should be noted that the Bank of Canada has been a pioneer in incorporating ABMs into its general toolkit for macroeconomic projections and policy analysis. The CANVAS (Canadian behavioral Agent-based model for moNetary policy AnAlySis) model introduces several key innovations in macroeconomic modeling:
- Heterogeneous agents: CANVAS incorporates detailed household and firm differences at the individual level, using rich micro-level data on demographics, household balance sheets, and firm characteristics.
- Realistic behavior: The model moves beyond rational expectations by incorporating adaptive learning and heuristics to simulate more realistic behaviors of individuals and businesses.
- Production network modeling: CANVAS explicitly models the Canadian production network, capturing complex interactions between different sectors of the economy.
- Data integration: The model harnesses large-scale Canadian micro- and macroeconomic datasets, including financial flows, national balance sheet accounts, input-output tables, government finance statistics, and Labor Force Survey data.
- Forecasting capability: CANVAS demonstrates competitive out-of-sample forecasting performance compared to benchmark vector auto-regressive (VAR) and dynamic stochastic general equilibrium (DSGE) models.
- Sectoral analysis: The detailed structure of CANVAS allows for forecasting medium-run macroeconomic effects at the sector level, enabling assessment of events like the COVID-19 pandemic’s impact on different industries.
- Inflation dynamics: By modeling strategic price-setting behavior of individual firms based on lab and survey evidence, CANVAS captures inflation dynamics through factors such as demand, supply, and expectations.
CANVAS: A Canadian behavioral agent-based model for monetary policy
Authors: Cars Hommes, Mario He, Sebastian Poledna, Melissa Siqueira, Yang Zhang
From: Bank of Canada, University of Amsterdam, Tinbergen Institute, International Institute for Applied Systems Analysis
