Dissecting the purchasing managers’ index: Are all relevant components included? Are all included components relevant?

Abstract

In this paper we scrutinise the composition of one of the most renowned economic indicators that is regularly released for more than 30 countries and regions. The composite Purchasing Managers’ Index (PMI) is constructed by pooling several survey based sub-components with certain fixed weights. Its characteristic feature is that its computation is based on the standardised methodology by that was developed for the PMI in the US more than thirty years ago. Though the uniform methodology makes the international comparison of national PMIs an easy and transparent task, it is not immediately clear whether the current fixed weighting scheme of the PMI components is supported by the data for other countries than US. We address this question using Switzerland as an example and our approach, based on Boriss Siliverstovs (2017), can be easily extended to other national PMIs. We find that the relative weights of the PMI components are generally supported by the data, except the fact that one component, found very informative for explaining GDP growth, is currently omitted from the PMI composition.

Keywords

• PMI
• MIDAS
• LASSO
• real-time data
• Switzerland