Composite global indicators from survey data: the Global Economic Barometers

Abstract

This paper presents a coincident and a leading composite monthly indicator for the world business cycle—the Global Economic Barometers. Both target the world’s output growth rate and consist of economic tendency surveys results from many countries around the world. The calculation of these indicators comprises two main stages. The first consists of a variable selection procedure, in which a pre-set correlation threshold and the targeted leads to the reference series are used as selection criteria. In the second stage, the selected variables are combined and transformed into the respective composite indicators, computed as the first partial least squares factor with the reference series as response variable. We analyse the characteristics of the two new indicators in a pseudo real-time setting and demonstrate that both are useful additions to the small number of indicators for the global business cycle published so far. Finally, yet importantly, the Barometers were quick to plunge in the beginning of March 2020 and have since then given a reliable real-time reflection of the economic consequences of the Covid-19 pandemic.