Classification of the NATO Countries with Respect to Defence Spending Patterns: An Unsupervised Clustering Approach

Abstract

Although there has been growing interest in defence expenditure studies in recent years, no necessary attention has been paid in literature to empirical analysis of composition of defence expenditure. However, limited number of non-empirical studies state that understanding composition of military budget of a country reveals remarkable insights about its economic condition and defence strategy policy choices in the future. The purpose of this study has been to gain further understanding of basic differences in defence budget allocation between NATO ally counties. To fulfil this aim, K-Means clustering approach is used and three main groups are found with the dataset of NATO that formed by four main defence expenditure categories. Moreover, the reliability of the results has been proven by additional analysis such as hierarchical clustering and K-Medoids methods. Empirical findings state that NATO countries could be categorised into three groups which are named as equipment intense expenditure cluster, personnel intense expenditure cluster, and balanced expenditure cluster.