Research on the dynamic spillover of stock markets under COVID-19—Taking the stock markets of China, Japan, and South Korea as an example

Baicheng Zhou, Qingshu Yin, Shu Wang, Tianye LiFrontiers in Public Health 2022 · DOI

TL;DR

This paper investigates the dynamic spillover effects between the stock markets of China, Japan, and South Korea, particularly in the context of the COVID-19 pandemic. Using a novel time-varying parameter vector autoregression (TVP-VAR) model with a large set of economic variables, the study reveals that China's influence on its East Asian neighbors' stock markets has been growing, especially since its financial market liberalization.

What Problem Does This Paper Address?

The increasing interconnectedness of global financial markets means that shocks in one market can quickly propagate to others. While previous research has explored these "spillover effects," many studies have relied on traditional econometric models that struggle to capture the dynamic, non-linear, and time-varying nature of these relationships. Furthermore, understanding how major global events, like the COVID-19 pandemic, alter these interdependencies is crucial for economic stability and policy-making. This paper specifically addresses the gap in understanding the evolving spillover dynamics between the key East Asian economies of China, Japan, and South Korea, especially under the unprecedented shock of COVID-19, and how their financial integration has progressed.

Methods

The authors employ a State-Space Model with Time-Varying Parameters and Factor Augmentation (SV-TVP-FAVAR). This advanced econometric technique is designed to:

  • Capture Time-Varying Relationships: Unlike traditional VAR models with constant coefficients, SV-TVP-FAVAR allows the relationships between variables to change over time, reflecting evolving market dynamics.
  • Incorporate a Large Number of Variables: The model can handle a high-dimensional information set (239 economic variables) by extracting common latent factors, thus avoiding the "curse of dimensionality" that plagues standard VARs. This allows for a more comprehensive understanding of the underlying economic drivers.
  • Handle Non-Linearity: The framework is better equipped to capture the complex, non-linear interactions that characterize financial markets, especially during crises.
  • Integrate Mixed-Frequency Data: The model can accommodate data with different frequencies (e.g., daily stock market data and quarterly macroeconomic data), which is crucial for leveraging high-frequency financial information.

The authors also utilize Augmented Dickey-Fuller (ADF) tests to ensure data stationarity and Bayesian estimation (Markov Chain Monte Carlo - MCMC) for parameter estimation.

Datasets

The study uses daily stock market data for:

  • China (Shanghai Securities Composite Index - SSEC, and Shenzhen Component Index - SZI)
  • Hong Kong (Hang Seng Index - HSI)
  • Japan (Nikkei 225 Stock Index - N225)
  • South Korea (Korea Composite Stock Price Index - KS11)

Additionally, they incorporate external financial market data, including:

  • Commodity Price Index (RJ/CRB)
  • U.S. 10-year Treasury yield (USG10Y)
  • Real Effective Exchange Rate Index of the US dollar (REER)
  • Brent Crude Oil Futures Settlement Price (CFD(OIL))

A broad set of 239 economic variables were used to extract common factors, including industrial added value, GDP, consumer price index, industrial producer price index, purchasing power parity, money supply, interest rates, and exchange rates. Data was sourced from the Wind database. The sample period spans from the first quarter of 1995 to the second quarter of 2022.

Key Insights and Ideas

  • China's Growing Influence: The study highlights a significant and increasing spillover effect from China's stock market to Japan and South Korea, particularly since China's financial market liberalization (e.g., the QFII system implementation in 2002). This suggests a shift from China being a "passive receiver" to an "active guide" in regional financial markets.
  • Hong Kong's Evolving Role: While Hong Kong historically served as a crucial "window" for China's price information exchange with Japan and South Korea, China's direct influence has grown to a point where its spillover effect is now comparable to, and in some aspects surpassing, that of Hong Kong.
  • Crisis Amplifies Linkages: Both the Asian Financial Crisis (1997) and the Global Financial Crisis (2008) led to a strengthening of co-movements and spillover effects between these stock markets, consistent with the "herding effect" observed during turbulent times.
  • COVID-19's Unique Impact: The COVID-19 pandemic's impact on spillover dynamics shows distinct characteristics compared to previous crises, underscoring its unprecedented nature and the increased uncertainty it introduced.
  • External Markets Matter: Commodity markets and crude oil prices tend to have positive interactions with the stock markets of these countries, while foreign exchange and bond markets show negative interactions. Hong Kong's stock market appears particularly sensitive to external financial market shocks.

Key Results

  • The return spillover effect of China's stock market on Japan and South Korea has gradually increased since 2003.
  • While China's direct influence is growing, the spillover effect between Japan and South Korea remains stronger than China's direct spillover to either of them.
  • During crises, the normal transmission channels between stock markets can break down, leading to increased interactions as investors exhibit "herding behavior."
  • The impulse response functions indicate that the price information transmission between stock markets caused by return rate changes has a primarily short-term effect, typically returning to zero within eight periods.

Limitations and Future Work

  • Data Limitations: While the study uses a large number of variables, some missing data were supplemented by interpolation, and non-stationary data were transformed, which can introduce some limitations.
  • Model Complexity: The SV-TVP-FAVAR model, while powerful, is complex, and the interpretation of its results requires careful consideration.
  • Future Directions: The authors suggest that further research is needed to explore the similarities and differences among the stock markets of China, Japan, and South Korea from both internal and external market perspectives. They emphasize the need to strengthen regional monetary synergism and enhance communication regarding market rules and institutions for better financial and economic stability in East Asia, especially in the post-pandemic era. The development of more secure and diversified financial products is also highlighted as crucial for internationalizing a country's currency.

Summary updated

Abstract

Examining stock market interactions between China (mainland China and Hong Kong), Japan, and South Korea, this study employs a framework that includes 239 economic variables to identify the spillover effects among these three countries, and empirically simulates the dynamic time-varying non-linear relationship between the stock markets of different countries. The findings are that in recent decades, China's stock market relied on Hong Kong's as a window to the exchange of price information with Japan and South Korea. More recently, the China stock market's spillover effect on East Asia has expanded. The spread of the crisis has strengthened co-movement between the stock markets of China, Japan, and South Korea.
BibTeX
@inproceedings{zhou2022research,
  author = {Baicheng Zhou and Qingshu Yin and Shu Wang and Tianye Li},
  title = {{Research on the dynamic spillover of stock markets under COVID-19—Taking the stock markets of China, Japan, and South Korea as an example}},
  booktitle = {Frontiers in Public Health},
  year = {2022},
  doi = {10.3389/fpubh.2022.1008348},
}
Figure