Portrayals of Chinese companies in American and British economic news tweets during China’s macroeconomic transitions 2007–2023

Abstract

This study investigates how Chinese companies are portrayed in American and British economic news tweets, as well as their relationship to Chinese economic fluctuations. The analysis included a corpus of 55,394 tweets (934,155 words) from well-known media outlets between 2007 and 2023. It also incorporated China’s quarterly Gross Domestic Product (GDP) and monthly Purchasing Managers’ Index (PMI) to contextualise the tweets in terms of their actual economic performance. Using van Dijk’s ingroup and outgroup ideologies, this project examined the ideological depiction through sentiments and emotions. RoBERTa-based transformer models were used to analyse sentiments and emotions, whereas Large Language Models (LLMs) were used for evaluative target annotation. Positive and negative sentiments were found to be significantly (p < 0.01) correlated with China’s macroeconomic indices. The representation of Chinese companies trended between ingroup and outgroup portrayals. Positive sentiment diminished as the economy transitioned from expansion to contraction, while negative sentiment increased. American news tweets were most positive during economic balance and most negative during downturns, while British news tweets were most positive during stability or early recovery and most negative when a downturn was predicted. As the economy shifted from growth to decline, positive sentiment emphasised corporate external opportunities alongside corporate strengths, with both evaluative targets reinforcing ingroup representation. Negative sentiment shifted from corporate weaknesses to contextual threats, both strengthening outgroup representation. Moreover, certain emotions had a significant (p < 0.0001) influence on sentiment swings and ideological transitions. These findings emphasise how economic changes and journalistic factors influence the ideological representation of Chinese firms in economic news on social media.