Impact of COVID-19 on jump occurrence in capital markets

This research examines the impact of the COVID-19 pandemic on sudden and substantial price movements (jumps) in global capital markets. Jumps significantly impact financial markets, escalating volatility and signaling shifts in investor sentiment. The COVID-19 pandemic, with its unprecedented uncertainty and varying government responses, provides a unique setting to analyze jump dynamics. Economic intuition suggests that increased uncertainty surrounding the novel virus, its spread, fatality rates, and resultant government control measures (quarantine, lockdowns) would heighten market volatility and increase the likelihood of jumps. Daily news and announcements regarding pandemic progress significantly influence investor sentiment, leading to potential overreactions and price deviations from intrinsic value. The study aims to investigate the specific impact of pandemic-related information on the magnitude and frequency of jumps across six major global financial markets (China, France, Italy, Germany, UK, and US), considering the diverse approaches to pandemic management employed by each country. Existing literature focuses largely on market volatility and uncertainty, lacking a detailed investigation into the independent effects of sharp price movements during the pandemic, and most importantly the effects of information on these jumps.

The literature on COVID-19's impact on stock markets is extensive, covering areas such as market volatility and the effects of uncertainty. Studies have analyzed the impact of COVID-19 on market volatility in various countries, including the US and a broader selection of 12 countries. While acknowledging uncertainty as a driver of instability, these studies primarily focus on overall price fluctuations rather than independent jump occurrences. Recent research has explored the relationship between COVID-19 and market jumps, identifying the pandemic as a source of heightened risk and uncertainty leading to increased abrupt market movements. However, the impact of specific COVID-19 related information on jump occurrence in stock markets remains largely unexplored. Existing literature on the impact of COVID-19 information on stock returns largely focuses on the negative impact of daily growth in confirmed and death cases. Some studies have also touched upon the short-term effects of announcements related to control measures on stock returns. However, the relationship between COVID-19 information and jumps is not straightforward, as information leading to decreased index returns doesn't necessarily trigger a negative jump. The government's response also plays a crucial role, potentially moderating or exacerbating market reactions.

The study uses the autoregressive jump intensity (ARJI) model proposed by Chan and Maheu (2002) to analyze jump dynamics. This model filters out market jumps and allows for time-varying jump intensity, addressing the challenges of modeling infrequent, discrete jump events. The authors extend the ARJI model by incorporating external variables (daily confirmed cases and daily death cases from the WHO database) as proxies for COVID-19 into the jump intensity equation. This allows them to examine the influence of pandemic development monitoring indicators on the occurrence of jumps across the six selected markets. Four specifications of the extended ARJI model are developed, incorporating current and lagged values of the daily confirmed and death cases. Maximum likelihood estimation is employed to estimate the model parameters. A Monte Carlo simulation is conducted to evaluate the model's performance, comparing estimated parameters with true values. The study uses daily closing prices of stock indexes (from the Wind Database) from January 2, 2013 to December 31, 2023. The data is divided into pre-COVID (January 2, 2018 - December 30, 2019), COVID (January 3, 2020 - December 31, 2021), and post-COVID (January 3, 2022 - December 31, 2023) periods. The Oxford University stringency index is used to assess the stringency of COVID-19 measures in each country. A difference-in-differences (DID) analysis is conducted to assess the differential impact of COVID-19 on jump frequency between China (treatment group) and the other countries (control group). Additionally, a robustness test involving a regression analysis of weekly extreme events (defined as daily index returns exceeding twice the standard deviation) against pandemic tracking indicators is performed.

The study reveals several key findings. First, jump intensity was significantly higher during the COVID period than in the pre- and post-COVID periods across most countries. Second, the study identified three distinct patterns across the six countries. In China, daily confirmed cases significantly impacted jump intensity, more so than daily death cases. In the US, none of the pandemic proxies significantly affected jump intensity. In European countries and the UK, daily death cases significantly impacted jump intensity. This suggests that market reactions were driven by concerns about potential future control measures triggered by COVID-19 updates. The annual jump frequency analysis shows similar jump dynamics across European countries, with a peak during the pandemic's initial outbreak in 2020. China exhibited a unique pattern, with low jump frequency in 2021 and high frequency in 2022, reflecting its zero-COVID policy. The DID analysis confirms a significant negative impact of China’s effective COVID-19 control on jump frequency compared to other countries during the second wave. Parameter estimates from the extended ARJI models showed that daily confirmed cases significantly affected jump intensity in China, while daily death cases significantly impacted jump intensity in European countries and the UK. The robustness test using weekly extreme events confirmed these findings. Hierarchical cluster analysis grouped France, Germany, Italy, and the UK together based on similarities in confirmed cases, deaths, stringency index, and stringency-cases ratio, contrasting them with China (strictest measures) and the US (least stringent measures).

The findings address the research question by demonstrating a clear link between COVID-19, its associated information, and the occurrence of jumps in capital markets. The increased jump intensity during the COVID period, consistent across multiple analytical methods (ARJI model, jump frequency comparison, DID analysis, robustness test), supports the hypothesis of heightened market reactivity to pandemic-related uncertainty. The variations in the impact of pandemic proxies (confirmed cases vs. deaths) across countries highlight the crucial role of government policies and their effectiveness in shaping market sentiment. China's zero-tolerance policy focused on early containment of confirmed cases, while European countries prioritized healthcare capacity and thus focused on managing deaths. The US's decentralized approach resulted in less impact from confirmed cases or deaths. The results highlight the significance of policy choices and their indirect impact on market behavior through investor perception and anticipation of future actions. The implications extend beyond individual market analysis, highlighting the global interconnectedness of financial markets and the substantial influence of public health crises.

The study establishes a strong link between pandemic-related information and jump dynamics in capital markets, showing varied responses based on national COVID-19 strategies. The findings demonstrate the importance of considering the economic implications of pandemic control measures and highlight the need for transparent communication to mitigate market volatility. Future research could expand the sample of countries, incorporate other relevant variables (e.g., vaccine rollout, economic stimulus measures), and refine the models to better capture the complexity of market reactions to pandemic-related events.

The study's limitations include the limited number of markets analyzed (six major markets), and reliance on WHO data, particularly concerning the potential underreporting of COVID-19-related deaths in China. Future research should address these limitations by expanding the geographical scope to include a more diverse range of markets, particularly emerging ones, and employing alternative data sources or methodologies to validate findings.