The study explores the long-term effects of major crises on the crude oil market and global economy. While short-term impacts are readily apparent, long-term consequences are more difficult to identify and are crucial for understanding market dynamics and equilibrium. The paper highlights the difficulty in defining 'short-term' and 'long-term' periods, but adopts a one-year delineation in the context of financial markets and OPEC+ production quota adjustments. Existing literature predominantly focuses on short-term impacts or individual events, lacking a comprehensive framework for analyzing long-term, cross-event effects. This study addresses this gap by developing a novel approach to quantify and analyze the long-term consequences of major crises across various categories.

Existing research on major crisis events and their impact on the oil market and economy employs several methods with limitations. Studies using 0-1 dummy variables offer rough estimates, failing to capture intensity or time-varying impacts. Event analysis methods are suitable for short-term impacts but are ineffective for long-term analysis due to the influence of other factors. Methods comparing pre- and post-event performance are event-specific and may mix effects from other events. Finally, indices like the Geopolitical Risk Index are limited to specific event types. This paper addresses these limitations by using a Proxy-SVAR model and a novel proxy variable.

The study employs a Proxy-Structural Vector Autoregressive (Proxy-SVAR) model to analyze the impact of major crisis events. The model uses monthly data from January 1990 to December 2022, incorporating 50 events classified into four categories: geopolitical conflicts, natural disasters, economic and financial crises, and public health emergencies. A key innovation is the use of the percentage change in WTI crude oil futures prices as a proxy variable for each event, addressing the challenge of quantifying events. This proxy variable is supported by its F-value and historical decomposition results. The Proxy-SVAR model addresses the endogeneity of the events by employing changes in oil futures prices as instrumental variables. The study examines the impact of these events on various economic indicators, including crude oil prices, global oil production, the S&P 500 index, industrial production indices of major countries (OECD+6, US, China), and CPI indices for the US and China. Unit root tests (ADF and PP) are used to ensure stationarity of the time-series data before applying the Proxy-SVAR model.

The study's findings reveal that major crisis events generally lead to increased crude oil prices and a decline in global crude oil production initially, followed by a gradual recovery. The response of different economic indicators varies. The S&P 500 index initially shows a positive response, then weakens and becomes negative before eventually approaching zero. The world and US industrial production indices generally respond positively, while China's industrial production index initially responds negatively before showing a positive response and then reverting to negative before converging to zero. The CPI in both the US and China increases following events, with China experiencing a more prolonged and severe impact. Geopolitical conflicts and natural disasters lead to higher oil prices, whereas financial crises and public health emergencies lead to lower prices. The impacts' duration differs across event types, with financial crises having the most prolonged impact (over four years on the economy) and natural disasters the shortest (about one and a half years). Historical decomposition analysis shows major crisis events' significant influence on actual crude oil price fluctuations, particularly sudden changes. Further analysis by event type reveals nuanced impacts: Wars and geopolitical conflicts initially increase oil prices and decrease oil production; the S&P 500 shows a V-shaped response, and the CPI increases substantially in both the US and China. Natural disasters result in similar initial responses as wars, but the impact is short-lived. Economic and financial crises decrease oil prices and increase oil production, with more volatility in China's industrial production index. Public health emergencies initially lead to a slight increase and then a significant decrease in oil prices. The US CPI shows an initial positive response followed by a decline, while China's CPI fluctuates around zero.

The findings highlight the complex and multifaceted impacts of major crisis events on the crude oil market and economy. The varying responses of different economic indicators across event types underscore the importance of considering event-specific characteristics when assessing their impact. The study's consistent finding of prolonged impact on the Chinese economy, compared to the US economy, indicates potential vulnerabilities in emerging markets. The findings support the development of targeted and dynamic policy responses to better manage these risks. The use of the Proxy-SVAR model and the novel proxy variable offers a robust approach for future research exploring the long-term impacts of major crises. The findings have significant policy implications, particularly given the increasing interconnectedness of global markets and the potential for cascading effects from major crisis events.

This study provides a comprehensive analysis of the long-term impact of major crisis events on the crude oil market and the global economy. The key findings highlight the differential impacts of various event types and their diverse implications for policymakers. The research emphasizes the need for proactive risk management strategies and the development of a more robust and resilient global economy. Future research could extend this study by incorporating additional indicators, exploring specific regional variations, and examining the effectiveness of different policy interventions.

The study's limitations include the use of a proxy variable for major crisis events, which might not fully capture the complexity of each event. Additionally, the model relies on monthly data, which may mask some short-term fluctuations. Finally, the study focuses on the effects of major crises on a global level, limiting the analysis of specific country-level responses.