Synchronization patterns in the European Union

Abstract

We propose a novel approach to investigate the synchronization of business cycles and we apply it to a Eurostat database of manufacturing industrial production time-series in the European Union (EU) over the 2000-2017 period. Our approach exploits Random Matrix Theory and extracts the latent information contained in a balanced panel data by cleaning it from possible spurious correlation. We employ this method to study the synchronization among different countries over time. Our empirical exercise tracks the evolution of the European synchronization patterns and identifies the emergence of synchronization clusters among different EU economies. We find that synchronization in the Euro Area increased during the first decade of the century and that it reached a peak during the Great Recession period. It then decreased in the aftermath of the crisis, reverting to the levels observable at the beginning of the 21st century. Second, we show that the asynchronous business cycle dynamics at the beginning of the century was structured along a East-West axis, with eastern European countries having a diverging business cycle dynamics with respect to their western partners. The recession brought about a structural transformation of business cycles co-movements in Europe. Nowadays the divide can be identified along the North vs. South axis. This recent surge in asynchronization might be harmful for the European Unio because it implies countries' heterogeneous responses to common policies.

The anatomy of government bond yields synchronization in the Eurozone

“…The theorem stated in Marčenko and Pastur (1967)-Marčenko-Pastur theorem henceforthplays a central role in RMT. This theorem states that the probability density function of the eigenvalues of a random correlation matrix obtained from normal independent and identically distributed series distributed with variance σ 2 is distributed according to the Marčenko-Pastur distribution [see Marčenko and Pastur (1967); Laloux et al (1999); Guerini et al (2022)]. More formally Theorem-Marchenko-Pastur law.…”

“…It can be observed, accordingly, that portfolios weighted by principal components do not outperform alternative portfolios strategies, but that principal components strategies are comparable and sometimes preferable in terms of Sharpe ratio maximization to alternative portfolio strategies. 5 PCA has also been applied to the analysis of systemic risk in financial markets and business cycle synchronization [see Guerini et al (2022)]. In particular, Billio et al (2012) include principal components extracted from financial series as measures of systemic risk and early warnings indicators.…”

Business Cycles Synchronization: Literature Review

“…More recently, Guerini et al (2023) studied the synchronization patterns in the European Union based on monthly industrial production from 2000 up to 2017. They concluded that i) after the introduction of the common currency and before the Great Recession, the synchronization has increased, but ii) after the Great Recession two separated clusters of countries have appeared, the northern economies against the southern ones.…”

E Pluribus, Quaedam: Gross Domestic Product out of a Dashboard of Indicators

“…This technique has also been recently adopted in the business cycle and financial economics literature. (see respectivelyGuerini et al, 2019;Barbieri et al, 2021). The main advantage of RMT is that it provides more precise and accurate information about a panel of time-series compared to basic PCA analysis, which does not allow one to distinguish between factors reflecting spurious correlations obtainable with a finite number of observations and those that instead contain relevant information about the similarity of the series.…”