The Scottish Government has defined inclusive growth as ‘growth that combines increased prosperity with greater equity; that creates opportunities for all; and distributes the dividends of increased prosperity fairly’.  This is an attractive concept because it aims to ensure that economic growth and increased equality can be promoted in tandem rather than being at odds with each other.  

Previously, collaborators at the James Hutton Institute, BioSS, and Highlands and Islands Enterprise used exploratory factor analysis to define seven socio-economic indicators that summarise key characteristics of inclusivity.  These factors are related to the concepts of economic strength and large employers, community support, private sector, quality of life, provision of rural services, small diverse business activity and deprivation in small areas.  A hierarchical clustering of areas in Scotland, using these factors, identified five geographical clusters.  

More recent analysis has extended this work by assessing the differences between these clusters.  The magnitudes of the differences between the five clusters were assessed using hypothesis tests and graph theory.  Firstly, hypothesis tests were conducted for each pair of clusters to determine if the mean values of each factor were statistically significantly different.  To visualise which characteristics are common among the clusters, an undirected graph was used, with multi-edges colour-coded by factor.  In this graph, the absence of an edge, which represents a factor, between two clusters means that there is no characteristic that is similar across the two clusters.  This visualisation allows us to identify which factors are most often similar in pairs of clusters, which, in turn, is inversely related to our ability to distinguish between clusters.  A statistic for cluster discrimination can then be defined to identify which factors most often support discrimination of clusters.  For each factor this was defined as the proportion of cluster comparisons where the null hypothesis (of equal values between clusters) was rejected.  This discrimination ratio was highest for provision of rural services, which can be interpreted as the most salient characteristic to take account of when seeking to identify a cluster.  At the opposite extreme, community support was found to have the lowest relevance when distinguishing clusters.  The information obtained through this analysis can help inform the targeting of government policies to promote inclusive growth.

This work was done in collaboration with Jon Hopkins at the James Hutton Institute and was funded under the Scottish Government's Strategic Research Programme for environment, agriculture and food.