Resilience, crisis contagion, and vulnerability in Central Europe and the Baltics

The recent financial crisis had serious worldwide impacts. Initial resilience and good past performances led to the illusion that the Central and Eastern European (CEE) region was able to decouple from developments in advanced economies. This initial illusion was however immediately denied since the crisis spread to that region just with a lag. The CEE region was, in fact, suddenly placed at the epicenter of the emerging market crisis. Further, the consequences of the crisis were not uniform among countries of the CEE region. Strong cross-country disparities in the resistance and recovery capacities have been observed. Focusing on a CEE sub-region, the Central Europe and the Baltics (CEB), our research project aims to analyze and disentangle the resilience performance to the 2008 financial crisis within countries of this region according to their shock isolation and absorptive capacities. We develop a new methodology to investigate two important dimensions of resilience, namely recovery and resistance. The latter can be defined as the relative vulnerability or sensitivity of economies within CEB region to disturbances and disruptions, whereas the former is the speed and extent of recovery from such a disruption or recession. Our methodology is based on Bayesian estimation techniques for general equilibrium models. We build and estimate a DSGE model for a small-open economy, which features nominal wage and price rigidities, as well as financial frictions in the form of liquidity-constrained households and limited access to deposits for the bank system. Then we group our parameter estimates in two sets: structural parameters and stochastic structure. The former individuates the deep parameters affecting the economic recovery capacities after stochastic disturbances (innovations) occur; the latter governs the innovation distributions and their intrinsic persistence. Accordingly, we study the relative differences across CEB economies using Principal Component Analysis (PCA), obtaining synthetic orthogonal indexes of these differences in a parsimonious way. Finally, we use the two sets to compare the relative recovery (resistance) country performances of a single country to those of a hypothetical economy characterized by a CEB average structural (stochastic) set of estimated parameters. Precisely, considering estimated parameters as variables of a cross-sectional dataset organized by country, we first look at national differences considering as reference a hypothetical country, where there are no distortions and/or unaffected by disturbances; second we use, as reference, a hypothetical average country, built on the estimated parameter means.