News coverage of Confucius Institutes in the pre-Trump era. Issue 2 (3rd April 2022)

• Record Type: Journal Article
• Authors: Akhtaruzzaman, Muhammad; Berg, Nathan; Lien, Donald; Wu, Ying
• Is Part Of: Economic and political studies. Volume 10:Issue 2(2022)
• Page Start: 155
• Page End: 180
• Publication Date: 2022-04-03
• Subjects: Confucius Institute -- China -- English-speaking newspaper -- public diplomacy -- soft power; China -- Economic conditions -- Periodicals; China -- Politics and government -- Periodicals; Economic history; Politics and government; China; Electronic journals; Periodicals; 330
• DOI: 10.1080/20954816.2020.1870025
• Languages: English
• ISSNs: 2095-4816

Abstract

Abstract: A novel dataset is presented that reveals a general pattern of declining favourability in news stories about Confucius Institutes (CIs) from 2006 through 2015 in six English-speaking countries: the US, the UK, Canada, Australia, India and Kenya. There are significant differences in mean favourability and time trends across countries, however. Decreasing trends in favourability are observed in the US and Canada, in particular, whereas Kenya and India's time trends are nil or possibly increasing. We report empirical models of favourability conditional on GDP per capita, trade with China and the number of CIs across country-year observations. Consistent with Pew polling data, our data reveal an interesting pair of conflicting income effects. High-income countries tend to be more negative and less positive on CIs than low-income countries overall. Within-country positive shocks to real GDP per capita, however, are positively associated with favourability after cross-country differences in income levels are absorbed by country fixed effects. The number of CIs in a country has mixed effects on the favourability of news coverage about CIs across country-year observations. Despite billions of dollars spent on public diplomacy by the Chinese government, our findings suggest that its return on investment in public diplomacy in the form of CIs may not be as expected.