Do Aggregation Methods Influence Efw Index Performance? an Examination Using Cluster Analysis
Abstract: Recent empirical studies provide evidence that economic freedom, as measured by the Economic Freedom of the World (EFW) Index of the Frazer Institute, is strongly related to economic growth. None the less, the nature of this effect is subject to debate. Criticism regarding the arbitrary composition of the index has been voiced, inspired by the unclear effect of its categories on GDP p.c. growth rates. This paper develops alternative categories for the 42 individual variables of the EFW Index, by using cluster analysis. Theory is very important in telling us which variables can be used to measure economic freedom, but it certainly tells us much less as to how these variables should be grouped together. Based on that premise, the new index categories will not be pre-determined in their composition. Using a cross-sectional data set, the performance of the recalculated index is then compared to that of the original EFW Index. It is shown how the results may help to solve some of the open issues regarding the effect of economic freedom on GDP growth. Particularly, multicollinearity between index categories is reduced and new conclusions are reached on what parts of economic freedom are responsible for causing elevated growth rates, depending on the use of the freedom level or increase.
