new evidence based on textual analysis of Chinese media big data
Trade tensions between China and the United States have created volatility in the global stock market in recent years, but their effects are difficult to quantify. News related to tariffs and other factors, including trade barriers between the two countries, have made analyzing these effects particularly difficult. For this reason, the literature has so far mainly analyzed case studies that refer to specific episodes with a time horizon of only a few days.
We contribute to literature in three ways. First, we introduce a new Trade Sentiment Index (TSI) that captures, on a more continuous basis, the tone regarding trade in Chinese media, and examine its ability to explain the behavior of global stock markets. Second, we analyze the effects of Sino-U.S. business sentiment on stock prices at the country, sector, and firm level over the period January 2018-June 2019. Third, we disentangle the stock effects of business sentiment arising from social media (i.e. news articles on the web, forums and the versatile social media platform WeChat) from those of traditional media (i.e. newspapers and magazines ).
No stock market is benefiting from deteriorating Sino-US trade sentiment, and Asian markets tend to be more negatively affected. In particular, we find that sectors most affected by tariffs – such as those related to information technology – are particularly sensitive to the tone of trade tensions. The TSI accounts for about 10% of the model’s ability to explain stock price movements in countries with high exposure to the Sino-US value chain, with social media accounting for the majority of sentiment (9%) and traditional media only modestly (1%).
Trade tensions between China and the United States have played a significant role in swinging global stock markets, but the effects are difficult to quantify. We are developing a new Trade Sentiment Index (TSI) based on textual analysis and machine learning applied to a big data pool that assesses the positive or negative tone of Chinese media coverage and assesses its ability to explain the behavior of 60 global stock markets. We find that the TSI contributes about 10% of the model’s ability to explain stock price variability from January 2018 to June 2019 in countries most exposed to the Sino-US value chain. Most of the contribution is made by the tone extracted from social media (9%), while that obtained from traditional media explains only a modest part of the variability in stock prices (1%). No stock market is benefiting from the Sino-US trade war, and Asian markets tend to be more negatively affected. In particular, we find that sectors most affected by tariffs, such as those related to information technology, are particularly sensitive to the tone of trade tensions.
JEL codes: F13, F14, G15, D80, C45, C55
Keywords: stock returns, trade, sentiment, big data, neural network, machine learning