Benchmarking the competitiveness of nations: non-uniform weighting and non-economic dimensions

This paper addresses two methodological issues concerning the construction of a composite indicator of national competitiveness. The first is the choice of weights to be used to aggregate the underlying primitive data. Most composite indicators use predetermined fixed weight values that are then applied uniformly to all countries. However, such uniform weighting fails to recognize that countries may have different policy priorities with respect to the different primitive dimensions used to form a given index. To the extent a country’s mix of priorities differs from that represented by the imposed weight pattern, the use of uniform weighting may bias measurement, and hence inferences, of relative competitiveness. We propose in this paper a procedure that allows weights to be endogenously determined and that are applied individually to each country. These country specific weights explicitly take account of a country’s own choices and performance across primitive dimensions. We illustrate our procedure by applying it to examine the widely cited Growth Competitiveness Index developed by the World Economic Forum. The second issue we address is the conceptual scope of national competitiveness indicators. Such indicators often consider only the “economic” domain when assessing relative performance. This singular focus may also bias inferences about relative performance since it ignores other national goals or priorities. We therefore broaden the scope for evaluating relative performance by combining a traditional index of competitiveness with two additional dimensions that may impinge on a country’s performance: environmental sustainability and governance. The resulting Composite Inclusive Index is formed using weights that allow that countries may choose different combinations of the three dimensions but still achieve the same level of overall performance. A performance ranking across countries is then presented based on values of the Composite Inclusive Index.