Quantifying globalization

Abstract

Globalization is much debated, but is it possible to make reliable ranks of which countries are the most integrated internationally? Traditionally resort is taken to trade measures, but even considering only economic integration this measure disregards a number of aspects. This paper proposes a single measure or index of globalization based on several indicators of economic integration combined by use of the multivariate technique of factor analysis. The index is calculated for 23 OECD countries, and among the findings are that Ireland is ranked as the most globalized country during the 1990s, while the UK was at the top during the 1980s. Some of the most notable changes in the rankings are the decline of the USA, Canada, and to a lesser extent Japan and Norway. There are notable improvements in the ranking for Finland, Italy, Portugal, Spain and Sweden. For Portugal and Spain the changes seem to follow EU membership in the mid-1980s.

Acknowledgement

We would like the thank Asta Herdis Hall and Marias H. Gestsson for research assistance and Dani Rodrik for providing us with some of the data used in the analysis.

Notes

Usually defined as 0.5 (imports =exports) as a share of GDP.

It has been suggested the loadings should at least be greater than 0.60, but many researchers have used cut-off values as low as 0.40; cf. Sharma (1996).

The Bartlett test of spherity has also been suggested. However, this measure is dependent on sample size and thus rarely used (Sharma, 1996).

It is suggested that the KMO measure should be greater than 0.80. However, a measure above 0.60 is tolerable (Sharma, 1996).

When there is more than one factor extracted from the dataset an infinite number of solutions to the factor analysis problem exists. The method we use here aims at obtaining high factor loading on only one factor.