https://doi.org/10.52903/wp2024334
UNPACKING COMMODITY PRICE FLUCTUATIONS: READING THE NEWS TO UNDERSTAND INFLATION
Dimitris Malliaropulos
University of Piraeus and Bank of Greece
Evgenia Passari
Université Paris-Dauphine
Filippos Petroulakis
Bank of Greece
ABSTRACT
We show that text-based indicators of supply and demand disturbances in commodity markets provide distinct information about future inflation movements, relative to existing predictors and various measures of inflation expectations derived from financial prices or from surveys of professional forecasters and consumers. Specifically, we document that demand disturbances play a significantly larger role in prediction than supply disturbances because they typically lead to more correlated increases in quantities and prices of goods across the consumer basket, resulting in a clear and positive relationship between commodity prices and overall inflation. Local projections confirm that demand disturbances generate substantially larger inflation responses in the short- to medium-term. In terms of magnitudes, commodity-specific indicators reduce out-of-sample inflation forecast errors at the one-year horizon by up to 30 percent.
JEL-classifications: C19, E31, E37, Q02
Keywords: inflation, commodities, news, textual analysis, demand, supply
Disclaimer: : The views expressed in this paper are those of the authors and not necessarily those of the Bank of Greece or the European System of Central Banks. The authors thank George-Marios Angeletos, Marta Banbura, Gadi Barlevy, Francesco Bianchi, Ryan Charhour, Ludmila Fadejeva, Raffaella Giacomini, Yuriy Gorodnichenko, François Gourio, Refet Gürkaynak, Matteo Iacoviello, Michele Lenza, Andrei Levchenko, Francesca Monti, Dimitris Papanikolaou, Pascal Paul, Giorgio Primiceri, Ricardo Reis, Hélène Rey, Alexander Richter, and the participants of the Sciences Po - OFCE - Banque de France - CEPR Workshop on Empirical Monetary Economics 2023, the 2024 AEA session “New Developments in the Study of Inflation Origins and Inflation Dynamics”, the 31st CEPR European Summer Symposium in International Macroeconomics (ESSIM), the PSE Macro Days 2024, the ECB "New Methods for Sentiment Analysis with Artificial Intelligence in Central Banks" 2025 workshop, the 2nd LBS Alumni Conference, and the EC-CEPR-JIE Conference on "Global Shocks, Macroeconomic Spillovers and Geopolitical Risks" for comments and discussions for comments and discussions. Evgenia Passari gratefully acknowledges financial support from the Europlace Institute of Finance (EIF) & the Labex Louis Bachelier, and hospitality from the Kellogg School of Management, Northwestern University, the Federal Reserve Bank of Chicago and the Kiel Institute of the World Economy where this research was partly conducted.
Correspondence:
Filippos Petroulakis
Economic Analysis and Research Department
Bank of Greece
El.Venizelos 21, 10250 Athens, Greece
Tel.: +30-2103202416
email: fpetroulakis@bankofgreece.gr