Can soft power and nation brands predict economic performance of nations? A machine learning approach

Abstract

This research examines the predictive capabilities of soft power and nation brand dimensions on nations’ economic performance. It aims to elucidate the complex relationships between these constructs and economic outcomes. The study employs machine learning techniques, specifically random forest regressions, to analyse data from the Anholt-Ipsos Nation Brands Index (NBI) and the Global Soft Power Index (GSPI). By integrating various dimensions of soft power and national branding as predictors, the research models economic performance, focusing on GDP and GDP per capita for the period between 2010 and 2023. The findings indicate that the Investment and Immigration and Culture dimensions of the NBI, along with the Reputation dimension of the GSPI, are significant in forecasting GDP per capita. For predicting actual GDP, the Export and Culture dimensions of the NBI and the Influence and Familiarity dimensions of the GSPI are crucial. This study pioneers the application of machine learning to assess the relationships between specific dimensions of soft power, nation branding, and economic performance. It uncovers non-linear effects, interactions, and inflection points in economic performance predictions. Given the nascent nature of the GSPI dataset, this study takes an exploratory approach, acknowledging the evolving methodologies and scope of soft power measurement. As such, the findings provide initial insights into how emerging soft power indices can inform economic performance modelling while highlighting the need for further refinement in the future research.