The Diffusion Speed of Good vs. Bad News in Geopolitics

Abstract

This article investigates the dissipation speed of positive and negative news in a geopolitical context. We perform a sentiment analysis of geopolitical news and measure the gamma of the corresponding sentiment scores per time unit in order to compare the travel speed of news with positive sentiment scores with news having negative sentiment scores. While prospect theory suggests that bad news is perceived as more impactful than good news, we show that this does not necessarily hold for the travel speed of news. On the contrary, we find that good news linked to keywords, which have usually a negative association, travel faster than bad news, and vice versa; a seeming repudiation of folk wisdom. Since our use cases were geopolitical crises, we associate phrases connected with conflict or the potential for conflict to have a broadly negative association. The implications of our insights suggest that the dissipation speed of news can be improved by framing and releasing positive news about events or entities with a negative association. Keywords: sentiment analysis, sentiment score, sext mining, geo-politics, news flow