Housing Boom and Headline Inflation: Insights from Machine Learning
IMF Working Papers, July 28, 2022
Source details
- Canonical URL
- Housing Boom and Headline Inflation: Insights from Machine Learning
Other formats
Bibliographic details
- Authors: Yang Liu, Di Yang, Yunhui Zhao
- Published: July 28, 2022
- Series: IMF Working Papers
- DOI: https://doi.org/10.5089/9798400218095.001
Summary and context
- Title: Housing Boom and Headline Inflation: Insights from Machine Learning
- Authors: Yang Liu, Di Yang, Yunhui Zhao
- Date: July 28, 2022
- Publication: IMF Working Papers, Working Paper No. 2022/151
- Pages: 45
- Volume: 2022
- Issue: 151
- DOI: https://doi.org/10.5089/9798400218095.001
- ISBN: 9798400218095
- ISSN: 1018-5941
- Subject tags: Consumer price indexes, Economic forecasting, Housing, Housing prices, Inflation, National accounts, Prices
Stylized facts on housing and inflation
- Inflation rose during the pandemic against supply chain disruptions and a multi-year boom in global owner-occupied house prices.
- House prices are presented as a leading indicator of headline inflation in the U.S. and eight other major economies with fast-rising house prices.
- The analysis focuses on two housing components of headline inflation: rent and owner-occupied housing cost.
Forecasting approach and methodological insights
- The paper applies machine learning methods to forecast inflation in the two housing components (rent and owner-occupied housing cost) of headline inflation.
- For the vast majority of countries analyzed, machine-learning models outperform the VAR model.
- The methodological finding suggests potential value for incorporating machine-learning models into inflation forecasting.
Key empirical findings and inferences
- Results suggest that for most of the countries in the sample:
- The housing components could have a relatively large and sustained contribution to headline inflation.
- Inflation is just starting to reflect the higher house prices, implying further inflationary impact may materialize as housing cost measures respond.
Policy-relevant implications (as conveyed in the source)
- Monitoring owner-occupied house prices and housing cost components (rent and owner-occupied housing cost) is important for understanding near-term headline inflation dynamics in countries experiencing housing booms.
- Incorporating machine-learning forecasting models may improve inflation forecasts relative to VAR models for these housing-related inflation components.
IMF Working Paper No. 2022/151, July 28, 2022
Content in this bundle
- Working Paper