Comovements and heterogeneity in the euro area analyzed in a non‐stationary dynamic factor model

Abstract

This paper establishes stylized facts on comovements and heterogeneity of individual euro area countries' output and price developments in the past two decades. For this purpose, a non-stationary structural dynamic factor model is fitted to a large dataset of euro area macroeconomic variables. The main results are as follows. Both common factors and idiosyncratic components are important in explaining individual countries' output and price developments in the euro area and are also both very persistent. Idiosyncratic shocks and adjustments to these shocks are mainly responsible for cross-country heterogeneity. The asymmetric transmission of common shocks plays a minor role. Finally, there is no strong evidence that some common shocks lead to greater heterogeneity than others. . S. EICKMEIER1 Marcellino et al. (2000) do not identify the shocks behind the factors, but give the factors an economic meaning and therefore also do some structural analysis. 2 Slovenia, Malta and Cyprus are not included in the analysis.

Do Euro Area Countries Respond Asymmetrically to the Common Monetary Policy?

“…Notice that we estimate a strong effect of monetary policy shocks on the economy (in particular on both GDP and CPI). 13 However, our estimates are not far from those usually found by the literature (Monticelli and Tristani, 1999;Sala, 2003;Eickmeier, 2009). 14 13 The reason for these large magnitudes in the response of prices is due to the heavy transformations choice.…”

“…For Finland, Greece and Ireland we cannot find restrictions for the pre-euro sample, while for Portugal we cannot find them for the post-euro sample. Finally, it is worth noting that unconventional reactions of inflation or prices to the monetary policy shock are also found by Sala (2003) for Italy and Portugal, Eickmeier (2009) for Greece and Portugal, Boivin et al (2009) for Germany, and Peersman (2004) for Austria and Italy.…”

“…achieve identification, we impose economically meaningful restrictions as those in Forni et al (2009), Forni andGambetti (2010a) and Luciani (2013). 4 Finally, to account for estimation uncertainty, we build confidence intervals using a bootstrap algorithm as in Bernanke et al (2005) and Eickmeier (2009). At each iteration d, we bootstrap the estimated common shocksũ t d and we generate new common factors…”

“…We identify the monetary policy shock by means of sign restrictions (Faust, 1998;Canova and de Nicolò, 2002;Uhlig 2005), an identification strategy also used in the context of factor models by Eickmeier (2009) and Forni and Gambetti (2010b,c). Specifically, at each iteration we draw a vector of q(q − 1)/ 2 angles from a uniform distribution on [0, 2 ), which, by means of Givens transform, are used to construct an orthogonal matrix R( ) of dimension q × q.…”

Fundamentals and Exchange Rate Prediction Revisited

“…6 The finding of three estimated sources is consistent with Bai and Ng (2007) who point out three primitive shocks in 132 monthly U.S. variables. It also agrees with Eickmeier (2009) pointing out that two to six latent common factors are sufficient to explain variations in most macroeconomic variables, and it agrees with Greenaway-McGrevy et al (2014) indicating three factors in 27 monthly exchange rates. Since no article discusses the criterion of selecting the number of independent sources, to the best of our knowledge, we assume that it is the same as the number of PCs determined by IC p2 .…”