Measuring competition between the great powers across Africa and Asia using a measure of relative dispersion in media coverage bias

The prevailing narrative in the US, echoed by media outlets and think tanks, frames Great Power Competition (GPC) as a decline in American influence relative to China and Russia. This perception is reflected in US governmental strategies since 2015, with successive administrations prioritizing GPC in their military and economic policies. This study aims to empirically assess the existence and nature of GPC in Africa and Asia, regions crucial to understanding the competition due to their resources and the significant presence of initiatives like China's Belt and Road Initiative (BRI). Competition is defined as multiple state actors pursuing similar or interconnected national interests within the same region. The study proposes a novel country-level measure of competitiveness using the coefficient of variance (CoV) of media coverage bias as a proxy for the relative power of each great power. This CoV-based measure, termed GPC-CoV, quantifies the dispersion of media attention given to the five great powers within each country. The research period spans from March 2015 to March 2021, utilizing the GDELT database for data collection. To validate the GPC-CoV, a separate subregional measure (GPC-Narrative) is developed through a comprehensive review of literature from government reports, think tanks, and academic sources. This dual approach allows for the validation of the GPC-CoV measure against established narratives of great power actions and interests. The study utilizes these metrics to understand not just dominance, but the opportunity for great powers to pursue their interests in a given country, even with the presence of other great powers.

The paper draws upon existing literature on Great Power Competition, primarily focusing on the perceived decline of US influence and the rise of China and Russia. It leverages frameworks from international relations theory, particularly focusing on the "race" for economic privilege and the creation of economic zones by great powers. The competitive exclusion principle from ecology provides a theoretical lens to understand competition, framing it as a dynamic struggle for resources and influence. This principle, however, is contextualized to reflect the nuance of great power interaction. Furthermore, the study incorporates existing analyses of great power actions in specific regions of Africa and Asia to develop a narrative-based measure of competition. This review involves sources from government, academic, and think tank reports which helps in establishing the context for the study's primary findings. The existing literature is used to inform the creation of a qualitative, subregional measure of competition, which then allows for cross-validation with the primary quantitative metric.

The study's core methodology relies on a two-pronged approach for measuring great power competition: a quantitative method using media coverage bias and a qualitative method using expert narratives. The quantitative approach leverages the Global Database of Events, Languages, and Times (GDELT) dataset, a large-scale repository of global media coverage. The dataset covers news articles from March 2015 to March 2021. Data filtering focuses on events occurring within African and Asian countries and involving the five great powers (US, China, Russia, India, and France). Articles are classified as ‘local’ based on the news agency’s location and a careful manual check of the reporter's address. The key quantitative measure, GPC-CoV, is constructed using the coefficient of variance (CoV) applied to the media ratio of each great power within each country. The media ratio represents the proportion of articles about a specific great power relative to the total articles on all five powers for that country. The CoV is preferred over the standard deviation because it allows for comparison across countries with varying average media coverage. A higher CoV indicates greater dispersion of media attention and, hence, greater competition. The qualitative approach involves an extensive literature review of government reports, think tank publications, and academic articles focusing on great power interests in various sub-regions of Africa and Asia. This literature is synthesized to create a subregional measure of competition (GPC-Narrative), focusing on eight key interest areas: arms sales, security agreements, use of force, energy expansion, fossil fuel extraction, maritime control, infrastructure projects, and trade/FDI. For each interest area, a competition score is calculated based on the number of great powers actively pursuing that interest. The resulting subregional competition scores serve as a qualitative benchmark against which the aggregated subregional GPC-CoV scores are validated. The correlation between the GPC-CoV (aggregated to the subregional level) and GPC-Narrative is analyzed to assess the robustness of the primary measure.

The study ranks 88 African and Asian countries based on their GPC-CoV scores, revealing a spectrum of competitive environments. Countries like Egypt, Seychelles, Oman, Uzbekistan, Mauritius, and Angola rank as the most competitive, indicating a relatively balanced presence of great power influence based on media coverage. In contrast, countries such as Israel, Kenya, Swaziland, Liberia, Somalia, and the Central African Republic show significantly lower competitiveness, with a clear dominance by one great power (usually the US). The GPC-CoV metric explicitly quantifies the opportunity for a great power to pursue its interests in a target country, even with the presence of other great powers. This is illustrated using the comparative analysis of Nigeria and Angola: Angola, despite its lower economic standing, scored higher in competitiveness than Nigeria due to a more diversified presence of great powers. The GPC-Narrative measure, derived from qualitative analysis of the literature, shows significant overlap with the GPC-CoV measure, demonstrating a weak negative correlation. The inconsistency in correlation direction between the two measures is explained by the differences in their construction. The study further uses the GPC-CoV to examine the relationship between competition and local press sentiment towards Chinese activities. The results indicate that the distance from China, and the level of great power competition, significantly influences press sentiment. Countries experiencing high competition and being moderately distant from China showed a more positive sentiment towards China. In contrast, countries with little competition or those closer to China showed largely negative sentiment. Interestingly, the analysis of press sentiment regarding Chinese activities reveals a strong positive correlation with sentiment towards American activities regardless of the competition level, highlighting potential alignment in how the media perceives both powers. However, the correlation between press sentiment toward Russian and Chinese activities differs significantly based on competition levels. In less competitive countries, press sentiment towards both powers is unrelated, while in competitive countries, a strong positive correlation emerges.

The findings demonstrate the value of the GPC-CoV metric in quantifying the complexity of great power competition. The study successfully created and validated a novel measure of GPC using media coverage bias. This approach offers a distinct perspective, focusing on local perceptions rather than solely on the resources or capabilities of the great powers. The validation using a qualitative, narrative-based measure lends confidence to the GPC-CoV's capacity to capture the dynamics of GPC. The analysis of press sentiment towards Chinese activities provides further insight into the interplay between geographical proximity, competition intensity, and public perception. The results underscore the importance of contextual factors in shaping local reactions to great power presence. This multi-faceted approach significantly contributes to a more nuanced understanding of GPC compared to existing metrics that may overlook local dynamics or focus primarily on aggregated metrics of power. This research has clear implications for policymakers, as it highlights countries where competition is most intense and where opportunities for influencing local dynamics may be most promising. This complements existing research focusing on the capabilities of great powers, providing insights based on local media and their perception of various actors’ activities.

This research presents a novel, empirically validated measure (GPC-CoV) of great power competition based on media coverage bias in Africa and Asia. The study uses this measure to reveal the nuanced nature of GPC across various countries, distinguishing between simple dominance and genuine competition. The findings suggest a complex relationship between geographical distance, competition intensity, and public perception, particularly in the context of China’s influence. This study’s contribution is the development of an empirically-grounded metric which can inform strategic decision-making. Future research could explore the temporal dynamics of GPC using longitudinal data, further refine the GPC-CoV metric, and extend the analysis to include additional great powers or regions.

The study acknowledges certain limitations. The reliance on GDELT data introduces potential biases inherent in the data collection and processing algorithms. The manual verification of local press agencies, while crucial for accuracy, is time-consuming and prone to human error. The focus on five great powers may not fully capture the influences of other significant actors in the region. While the correlation between GPC-CoV and GPC-Narrative offers validation, the weak correlation highlights the complexities of measuring GPC and the limitations of capturing multifaceted dynamics in a single metric.