Measuring Macroeconomic Uncertainty: A Cross-Country Analysis

Andreas Dibiasi,Samad Sarferaz

doi:10.2139/ssrn.3628071

Abstract

This paper constructs internationally consistent measures of macroeconomic uncertainty. Our econometric framework extracts uncertainty from revisions in data obtained from standardized national accounts. Applying our model to quarterly post-WWII real-time data, we estimate macroeconomic uncertainty for 39 countries. The cross-country dimension of our uncertainty data allows us to identify the effects of uncertainty shocks on economic activity under different employment protection legislation. Our empirical findings suggest that the effects of uncertainty shocks are stronger and more persistent in countries with low employment protection compared to countries with high employment protection. These empirical findings are in line with a theoretical model under varying firing cost.

Highlights

In times of economic crisis measuring macroeconomic uncertainty and understanding its various effects on the economy are essential to an efficient and adequate response of policy makers
In this paper we have introduced new internationally comparable measures of macroeconomic uncertainty for a large set of countries using data revisions in aggregate variable that are bound to the system of national accounts
We have set up an econometric model and constructed a new real-time data set of real GDP for 39 countries that serves as the basis for our estimations

Summary

Introduction

In times of economic crisis measuring macroeconomic uncertainty and understanding its various effects on the economy are essential to an efficient and adequate response of policy makers. This paper constructs measures of macroeconomic uncertainty that are defined as the conditional volatility of an unpredictable forecast as in, e.g., Jurado et al (2015) and that are available for a large set of countries. To obtain this goal, we draw on the macroeconomic data revisions literature, thereby treating statistical agencies’ estimates of first releases of macroeconomic variables as forecasting exercises and their subsequent revisions as forecast errors.. In contrast to the bottom-up approach of, e.g., Jurado et al (2015), we follow a top-down approach, where we partly outsource the information acquisition to the statistical agency

Objectives

Methods

Findings

Conclusion