Time-Varying Global Financial Stress Contagion in a Decade of Trade Wars and Geopolitical Fractures

Abstract

The objective of this study is to explore the time-varying shock transmission mechanism between aggregated financial stress indices (FSIs) of developed economies (the U.S., the U.K., the European Union (EU) and Japan) and the emerging economy of China. We employ a novel Time-Varying Parameter Vector Auto-Regression (TVP-VAR)-based “connectedness approach” to capture dynamic shock spillovers without the limitations of arbitrarily chosen rolling windows, loss of observations, or excessive sensitivity to outliers, as it is grounded in a multivariate Kalman filter structure. The aggregated measures of the FSIs of China, the U.S., the U.K., the EU and Japan are incorporated from the Asian Development Bank’s data repository by using time-series observations from January 2010 to September 2023. The findings indicate that the FSI of China is influenced by financial stress shocks originating from Japan (18.35%) and the U.S. (16.86%) the most, whereas the U.K. (EU) contributes to only 8.42% (6.54%) of FSI shocks in China. This research article significantly captures China’s heightened vulnerability to external financial stress shocks from developed economic systems and underscores the critical importance of reinforcing financial resilience, strengthening macro-prudential regulations and early-warning systems, and expanding financial buffers during episodes of trade uncertainty like restrictions on China’s rare earth exports and solar panels, U.S. restrictions on industrial metal imports, Brexit, supply chain disruptions amid COVID-19, and geopolitical uncertainties like the Russia–Ukraine war. Overall, this study provides actionable guidance for mitigating the impact of global financial stresses, improving risk management, and safeguarding economic stability in an increasingly interconnected and volatile international environment.

1. Introduction

Because of the increase in globalization, global equity markets have become highly integrated, and volatility shocks in one economy are often transmitted across borders (Gkillas et al. 2019). Kasraoui et al. (2025) also found that macro-financial shock spillovers from the U.S. and global geopolitical risk are transmitted with higher intensity to emerging economies, whereas Yarovaya et al. (2016) focused on shock transmission in the equity market between emerging and developed economies and found that equity markets are highly susceptible to shocks originating within their own country or region compared with those transmitted from other regions. Therefore, existing studies have primarily focused on the transmission of shocks from U.S. stock market returns to China’s financial system (Chen et al. 2025), with limited attention given to the reverse—namely, the ability of the U.S. market to absorb disturbances originating from China (Zhang and Mao 2022). Some studies have examined the dynamic propagation of stock market shocks between China and other advanced markets (W. Li 2021). Others have extended the analysis to sectoral and thematic domains, including spillovers of financial stress to sustainable and green energy markets (Liu and Wang 2024), Sharia-compliant financial systems (Sheikh et al. 2024), global equity markets (Liang et al. 2023), and commodity sectors such as the metals industry (Chen et al. 2023a; Shahbaz et al. 2024). Despite these contributions, little effort has been directed toward analyzing how financial stress shocks originating in advanced economies—such as the U.S., the European Union (EU), the U.K., and Japan—propagate into China’s financial system. As W. Li (2021) also emphasized, advanced markets tend to serve as primary sources of systemic stress, while their economies are largely exposed to shocks within their own financial networks.

Furthermore, the equity market’s volatility shock transmission was intensified after the 2008 global financial crisis, increasing the scale and persistence of cross-border risk spillovers. In recent years, the COVID-19 crisis, coupled with persistent geopolitical tensions and trade policy shocks—such as China’s rare earth export restrictions (2012), U.S. curbs on Chinese solar panels (2012), the Russia–Ukraine agricultural trade dispute (2012), Brexit-induced trade uncertainty (2018–2019), and U.S. tariffs on Chinese goods and steel/aluminum imports (2018)—has intensified market volatility and undermined long-term financial stability (He et al. 2021a; H. Li 2020; Sheikh et al. 2025). Consequently, a growing scholarly focus has been placed on the nexus between trade and geopolitical risks and stock market performance (He et al. 2021a; Suwanprasert 2022; Zaremba et al. 2022). However, little attention has been directed toward examining the dynamic transmission of financial stress shocks between advanced economies such as the U.S., the U.K., the European Union (EU), and Japan and China’s financial system during prominent trade policy uncertainty and geopolitical risk events. For instance, Gkillas et al. (2019) only investigated the role of country-specific macroeconomic determinants in affecting the interconnectedness of equity market volatility. More recently, Sheikh and Suleman (2025) only investigated the influence of country-specific economic uncertainty and financial market contagion on non-equity market investors’ sentiments. Additionally, a few other studies have also explored direct shock transmission mechanisms from trade-related disruptions and risk evolving around geopolitical uncertainty on equity markets (Fiorillo et al. 2023; Li et al. 2022; Yu et al. 2026). Moreover, no effort has been made to explore time-varying dynamic shock transmission mechanisms between global aggregated financial stress indices of developed (U.S., EU, Japan and U.K.) and developing (China) economies during trade uncertainty and geopolitical risk events.

Research Objectives and Theoretical Contributions

The first objective is to explore the evolving mechanisms of shock transmission, focusing on how financial stress originating in Japan, the United States, the United Kingdom, and the European Union (EU) propagate into China’s financial system through the application of the advanced Time-Varying Parameter Vector Auto-Regression (TVP-VAR) connectedness framework by Antonakakis et al. (2020). Secondly, this article explores the dynamic, time-varying transmission of financial stress shocks from these developed economies (the U.S., the EU, the U.K., and Japan) to China’s emerging market context, particularly during episodes of trade disputes and geopolitical uncertainty. In doing so, this study aims to capture temporal shifts in both the transmission and absorption capacities of these economies under conditions of heightened geopolitical tension and trade restrictions.

This study advances the existing body of literature in the following distinct dimensions.

Firstly, this research article advances the existing literature on contagion and the financial market interdependence by exploring the dynamic time-varying shock transmission mechanism between financial stress indices (FSIs) of the developed and developing economic systems of the U.S., the U.K., Japan, the EU, and China. In the existing literature, studies are primarily focused on equity, forex, and commodity market volatility spillovers by employing the conventional generalized VAR-based connectedness approach (Brož and Teplý 2025; Guru and Yadav 2023; Iqbal et al. 2022). However, this research study shifts the analytical focus from market-specific volatility dynamics to the cross-border transmission of financial stress shocks measured through aggregate financial stress indices developed following the methodological approach by Park and Mercado (2014). These FSIs are the aggregated measure of financial fragility within an economy by integrating information from distinct segments of prominent financial architecture such as the banking and equity sectors, forex markets, and the sovereign bond market (Park and Mercado 2014; Xu et al. 2023). By incorporating aggregated financial stress indices for dynamic shock spillovers, this study conceptualizes contagion as the propagation of macro-financial fragility across developed and developing economies, rather than as a mere reflection of synchronized movements in financial market returns (Albrecht and Kočenda 2024; Sheikh et al. 2025). Furthermore, prior studies only explore the direct shock transmission from FSIs to the equity markets in the wake of extreme economic uncertainty events (see Hoque et al. 2024; Sheikh et al. 2024; Tabash et al. 2025c; Das et al. 2019). However, by exploring the time-varying shock spillovers between financial spillovers of developed and developing economies, this research article contributes by exploring whether these heterogeneous economic systems are interconnected through macro-financial architectures such as banking and equity sectors, forex and debt markets, not merely through synchronized equity market movements. For instance, Apostolakis et al. (2021) only explore the time-varying shock spillovers between the energy-related commodity market and the FSIs of developed economies. Therefore, this research article expands our understanding of the shock transmission mechanism between developing and developed economies’ financial architecture; therefore, financial market hedging strategies should incorporate systemic financial stress interdependence rather than relying exclusively on asset-level correlations.

Secondly, this research article also contributes to the existing literature by exploring time-varying financial stress shock transmission during prominent trade policy uncertainty and geopolitical risk events rather than focusing on the shock spillovers between financial and non-financial asset classes during the generalized 2008 financial crisis (Suleman et al. 2022) or health crises like COVID-19 (Al-Fayoumi et al. 2023; Chen et al. 2025; Liao et al. 2021; Yin et al. 2025). By focusing specifically on financial stress indices as that developed by Park and Mercado (2014), this study takes into account an aggregated measure of financial fragility within the broad spectrum of financial architecture, like sovereign bonds, forex, equity, and banking sectors, and captures how systemic fragility propagates between advanced and developing economies under time-varying geopolitical risk and trade uncertainty events. Therefore, by observing the time-varying aggregated measure of financial stress shock transmission during trade and geopolitical risk events, this research article contributes to the literature by exploring whether the financial stress contagion between developed and developing economies changes over time and by investigating the role of trade uncertainty and geopolitical risk events in altering the persistence and strength of financial stress shock spillovers. Unlike prior research that primarily focuses on static direct trade policy and geopolitical uncertainty shock transmission to equity markets (Li et al. 2022; Benguria et al. 2022; Zaremba et al. 2022), this study advances the literature by conceptualizing financial stress shock transmission as a dynamic and evolving process, sensitive to shifts in international trade relations, geopolitical policy disputes, and global crises. By embedding the investigation within major episodes of geopolitical tension and trade disturbances, this study illustrates how fragility in advanced financial systems becomes a pivotal conduit for the cross-border escalation and transmission of financial stress. From a theoretical standpoint, the analysis enriches contagion and vulnerability frameworks by revealing that the diffusion of financial stress is not a fixed phenomenon; rather, it evolves in response to geopolitical confrontations and shifts in trade or policy regimes, thereby providing deeper insights into the interplay between financial stress shocks originating in advanced markets and the inherent structural weaknesses of developing economies.

Thirdly, on the methodological side, prior studies relied on the connectedness framework by Diebold and Yilmaz (2012), which is based on the generalized VAR (GVAR) model, in order to explore shock transmission between financial stress indicators and multiple financial asset classes (Yu et al. 2024). However, this traditional rolling-window approach has important drawbacks, including the arbitrary selection of window size, the loss of data at the sample edges, and heightened sensitivity to outliers (Asadi et al. 2023; Shahbaz et al. 2024). To address these issues, the present study employs the Time-Varying Parameter VAR (TVP-VAR) connectedness framework. The TVP-VAR model, built on a multivariate Kalman filter, offers several advantages: it allows shock spillovers to be estimated in a fully dynamic manner, preserves all available observations, and produces more robust results by avoiding distortions caused by rigid-window choices. The adoption of this approach is therefore theoretically significant, as it provides a more flexible and accurate representation of evolving spillover patterns in response to shifting economic and geopolitical conditions.

The remainder of the paper is organized as follows: Section 2 and Section 3 present the review of the relevant literature and outline the data set, together with descriptive statistics, respectively. Section 4 details the methodological framework, while Section 5 and Section 6 discuss the empirical findings, along with practical implications, and concluding remarks, respectively.

2. Literature Review and Conceptual Framework

One of the motivational factors for exploring financial stress shock transmission between developed financial systems (the U.S., the U.K., the EU, and Japan) is that in 2023, bilateral trade in goods and services between the U.S. and China was valued at roughly $643.2 billion, with American exports accounting for $195.5 billion and imports from China reaching $447.67 billion. In 2023, the percentage of U.S. FDI in Chinese manufacturing, wholesale trade, and financial as well as insurance services rose to $126.9 billion, reflecting a 3.8% increase from the preceding year. By contrast, Chinese FDI holdings in the United States amounted to $28 billion in 2023 (U.S. Mission China 2025). In the four quarters ending with the first quarter of 2025, trade flows between the United Kingdom (U.K.) and China amounted to approximately £99.7 billion, including both goods and services. During this period, British exports to China were valued at £28.8 billion, comprising £15.6 billion in goods (54.3%) and £13.2 billion in services (45.7%). In contrast, imports from China totaled £70.8 billion, the majority of which were goods at £67.3 billion (95.1%), while services accounted for £3.5 billion (4.9%) (Department for Business and Trade 2025). Moreover, China and the European Union (EU) maintained strong bilateral connections in foreign direct investment. In 2023, Chinese enterprises recorded outward greenfield investment flows of approximately €150 billion, while the European Union’s cumulative outward FDI stock in China reached nearly €230 billion during the same year (Durá and Vandermeeren 2024). The Literature Review Section is structured into two parts. The first part outlines the conceptual framework and economic rationale underlying the transmission of aggregated financial stress shocks between developed and developing economies in the context of trade and geopolitical uncertainty. The second part reviews the existing literature on volatility contagion across financial systems and identifies the significant research gaps that this study aims to address.

2.1. Conceptual Framework and Economic Rationality

Franch et al. (2024) reveal that the banking sector appears to be the main channel of financial shock transmission because such shocks propagate across different financial sectors and national boundaries due to the higher integration between banks and shadow banks. For instance, the progression of the global financial recession in 2007–2009 into the European debt crisis exemplifies that financial stress can be transmitted across integrated financial systems. Building on contagion theory (Masson 1999; Forbes and Rigobon 2002), disruptions in one financial market can propagate to others in ways that exceed what underlying economic fundamentals would predict. The deepening of global financial integration has substantially intensified cross-border linkages among national financial systems, strengthening their mutual dependence. As a result, disturbances that arise within a specific market can swiftly propagate across jurisdictions through multiple transmission mechanisms (Wu and Wang 2025). Such rapid cross-market spillovers amplify the potential for widespread instability and elevate the likelihood of systemic disruptions within the global financial architecture. In a highly integrated financial network, financial stress shocks are transmitted through both direct and indirect pathways. The direct channel for financial stress shock transmission includes formal contractual relationships among market participants. However, indirect propagation—such as fire sales, diffusion of imperfect information, and asymmetrical information flows—operates independently of such bilateral agreements and often emerges from collective behavioral responses of investors, such as broad-based changes in risk perception or investor sentiment (Mikropoulou and Vouldis 2025). Therefore, financial stress within developed financial systems (the U.S., the U.K., the EU, and Japan) can be transmitted to an emerging economy like China because of highly integrated banking systems, trade flows, capital flows and financial market integration (see Baele 2005).

Building upon the behavioral financial perspective, investors’ behaviors such as herd-driven investment patterns and uninformed investors’ trading (Sheikh and Suleman 2025), funding limitations, and the interdependence of asset holdings also contribute to the cross-border transmission of financial stress shocks. Another explanation for financial stress shock transmission from advanced economies to China is based upon the fact that asset commonality drives financial contagion in interbank markets through fire-sale feedback and mark-to-market accounting (Barucca et al. 2021). Under highly illiquid market conditions, the increase in substantial asset liquidation by distressed banks causes a decline in market prices, contributing to mark-to-market losses for other financial institutions holding similar asset classes. These financial institutions are engaged in further asset disposals, and this feedback mechanism intensifies financial stress contagion (Caiazzo and Zazzaro 2025). Therefore, financial stress from advanced economies can be transmitted to emerging economies because of higher trade linkages, capital flows and heightened integration of China with global financial systems (see Vuong et al. 2022). Periods of financial stress turbulence in advanced economies often trigger global investors to restructure their portfolios, commonly by withdrawing from risk-prone or non-core markets. As China has become increasingly interconnected with international trade and financial flows, these portfolio adjustments serve as conduits through which external financial stress shocks in the U.S., the EU, Japan and the U.K. penetrate its domestic financial environment. Consequently, financial stress shocks from developed financial systems like the U.S., the EU, the U.K. and Japan are transmitted to China with greater magnitude and intensity under global trade and geopolitical uncertainty. This is because financial stress shocks within advanced markets such as the United States, the United Kingdom, the EU, and Japan can induce capital flight, exchange rate instability, and liquidity constraints in emerging markets, thereby weakening their financial stability (see Shim and Shin 2021). In addition, contagion effects may hinder trade flows and investment activities, amplifying fragilities in countries with limited financial diversification (Park and Mercado 2014). For instance, Chen et al. (2025) documented the evolving process through which equity market disturbances originating in advanced economies such as the United States are conveyed to emerging markets like China. Their analysis highlighted the intermediary role of Hong Kong, which operates as a key transmission hub channeling U.S. equity market shocks into the Chinese system. Moreover, the study revealed that these spillover effects become more pronounced during periods of heightened geopolitical tension, the COVID-19 pandemic, and economic frictions—particularly the trade policy conflict between the two nations. However, existing studies only explore the shock spillovers between equity market returns and volatility of developed and developing economies (see Sheikh et al. 2025; Chen et al. 2025) and ignore time-varying financial stress shock transmission between advanced and emerging economies amid trade policy uncertainty (TPU) and geopolitical risk (GPR) events.

Furthermore, perspectives from globalization and financial integration theories emphasize that disturbances originating in advanced economies are rapidly transmitted to emerging markets through trade linkages (Guru and Yadav 2023), international capital flows (Ahmed and Huo 2019), and collective behavioral dynamics among investors (Sheikh and Suleman 2025). This supports the expectation of high dynamic financial stress shock transmission between developed (U.S., Japan and U.K.) and developing (China) economic systems. However, financial stress shocks can also be transmitted from China to developed systems, as the Evergrande crisis demonstrated by undermining global financial stability and intensifying spillover effects. For instance, Shahzad et al. (2025) explored the quantile domain equity market spillovers between China and developed financial systems, and the findings highlighted the higher shock propagation between global equity markets amid China’s Evergrande crisis. Therefore, these findings highlight the existence of a stronger contagion mechanism through which financial shocks from China can be transmitted to major international financial centers. This is because stress originating in emerging economies, particularly China, can also reverberate back to developed markets like the United States through global production networks and portfolio adjustments (Zhang and Mao 2022). Therefore, we can also infer that financial stress shocks from China are transmitted with higher magnitude and persistence to developed financial systems amid TPU and GPR. Hence, acknowledging the two-way nature of these spillovers is vital for policymakers to reinforce macro-prudential policies, improve crisis management tools, and safeguard international financial stability.

2.2. Review of Existing Studies

The aggregated measure of financial stress index (FSI) developed by the Asian Regional Integration Center (ARIC) of the Asian Development Bank (ADB) following the methodological approach by Park and Mercado (2014) is an indicator designed to quantify the intensity of systemic strain within developed and emerging economies (Xu et al. 2023). These financial stress indices (FSIs) are the aggregated measure of financial fragility within an economy as they integrate information from distinct segments of prominent financial architecture such as the banking and equity sectors, forex markets and the sovereign bond market (Park and Mercado 2014; Xu et al. 2023). When observing the responses of U.S. equity market abnormal returns to the failure of a prominent financial institution (Silicon Valley Bank implosion), Tabash et al. (2024) utilized these disaggregated global financial stress indices in order to control for global economic fragility. Therefore, the information on financial fragility is aggregated from the vast financial architecture—namely, banking industry, foreign exchange market, equity market, and sovereign and corporate debt markets—to capture multidimensional vulnerabilities within the financial architecture. Elsayed and Yarovaya (2019) explored whether the global financial recession and the Arab spring intensified the time-varying regional financial stress indices of MENA economies by employing the generalized VAR-based connectedness approach by Diebold and Yilmaz (2012). Overall findings indicated the intensification of MENA economies’ financial stress shock spillovers during the economic recession of 2008 compared with the Arab spring. Sheikh et al. (2024) investigated whether a sustainable financial system in the developed economic context receives higher global financial stress shocks compared with Sharia-compliant and conventional equity sectors by employing the quantile domain vector auto-regression (QVAR) connectivity approach. Overall findings suggested the heterogeneous transmission of shocks across bearish, median and bullish quantiles and indicated that the sustainable financial system of Australia received the most global financial stress shocks in the short and long terms. Hoque et al. (2024) also explored non-homogenous global financial stress shock transmission across bearish and bullish quantiles of U.S. sectoral stock volatility by employing the QVAR-based connectedness approach. Overall findings suggested that the disaggregated measure of global financial stress indicators transmitted higher forecast error variances to U.S. sectoral equities across extreme higher and lower quantiles. In a similar manner, He et al. (2021b) explored the heterogeneous responses of sustainable clean energy firms’ stock returns to the financial stress shocks of developed economies like the U.S. and Europe by employing the QARDL approach. Overall findings suggested a persistent adverse effect of developed economies’ financial stress indices on sustainable energy equities, but this effect was more pronounced under bearish equity market conditions. Liang et al. (2023) conducted a comparative analysis on different global uncertainty factors in forecasting the realized volatility of emerging and developed equity markets by employing predictive regression models. Overall findings suggested that the global FSIs can forecast the realized volatility of global equity markets with higher precision compared with global economic, geopolitical and equity market volatility.

Therefore, existing studies primarily examine the role of aggregated global financial stress indices in explaining variations in equity market volatility, while fewer studies emphasize the importance of exploring the time-varying shock spillovers from these global financial stress indices to equity market returns and volatility dynamics (Tabash et al. 2024; Sheikh et al. 2024; Hoque et al. 2024; He et al. 2021b; Liang et al. 2023). Apart from these studies, Elsayed and Yarovaya (2019) only explored financial stress contagion among MENA economies. However, no effort has been made to explore whether the shocks within the aggregated global financial stress indices of developed financial systems (the U.S., the U.K., the EU and Japan) are transmitted with higher magnitude and persistence to the financial stress indices of a developing economy, e.g., China, and vice versa. This further motivates us to explore whether the shock spillovers between the financial stress indices of developed and developing economies are intensified during TPU and GPR adverse events.

Moreover, Cipollini and Mikaliunaite (2020) only explored the shock transmission mechanism between European macro-uncertainty indices and financial stress indices by employing the global vector auto-regressive approach. Overall findings suggested that time-varying shock spillovers between financial stress and global macro-uncertainty factors disintegrated after the EU sovereign debt crisis and have been continuously decaying since then. Apostolakis et al. (2021) investigated the dynamic shock propagation system between energy-related commodity market risk, the aggregated measure of financial stress indices (FSIs) and economic policy uncertainty (EPU) by employing the TVP-VAR-based connectedness approach. Furthermore, the authors also took into account the impulse response function and VAR-based GARCH-M approach in order to explore the impact of energy-related commodity market volatility on EPU and FSIs. Overall findings highlighted the intensified shock spillovers among financial stress, economic uncertainty, and oil uncertainty during the global financial recession compared with the recent health emergency (COVID-19). Ozcelebi et al. (2025) employed quantile-on-quantile and wavelet-based quantile regression models for examining the relationship between the aggregated measure of financial stress indices and equity market performance in prominent Asia-Pacific economies. The findings revealed that rising stress levels are associated with declining equity returns, with the negative impact being more pronounced in structurally fragile markets like the Philippines and export-centric countries like Thailand and Korea. Beyond studies linking financial stress to equity markets (Ahmed and Huo 2019; Fu et al. 2022), Chen et al. (2023a) investigated whether the precious metals can be characterized as potential safe-haven assets against financial stress across bearish, moderate and bullish quantiles by employing a quantile-on-quantile regression model. Their findings revealed that precious metals can serve as effective hedging instruments, although this capacity varies across quantiles and depends on the prevailing economic environment, particularly when comparing pre- and post-crisis regimes. Apart from these studies, Shahbaz et al. (2024) explored direct shock transmission between global disaggregated financial stress indices and industrial metals’ conditional variances by employing the TVP-VAR connectivity approach. The findings suggested that global disaggregated financial stress shocks are transmitted to industrial metals in greater intensity in the short term compared with the long term.

Xu et al. (2023) also utilized a similar metric of aggregated financial stress indices developed by Park and Mercado (2014) to examine the predictability of equity market returns using these financial stress indices. Employing predictive regression techniques, they found that aggregated financial stress within the Chinese economic context exerts a negative influence on equity market returns. The evidence further indicates that financial stress within the developing market context also serves as a stronger predictor of stock returns than conventional macroeconomic variables. However, Chen et al. (2023b) employed the Markov-switching VAR framework in order to observe the aggregated U.S. financial stress shock reception of implied volatility and returns of the energy-related commodity market. Overall findings also confirmed that an increase in these disaggregated financial stress indices contributes to higher volatility within the energy-related commodity market and that this effect is more prevalent and persistent under a high-volatility regime. In a related study, Sheikh et al. (2025) also took into account the disaggregated global financial stress indices for exploring whether the increase in global financial stress indices (FSIs) causes higher volatility contagion between global Sharia-compliant and non-Islamic sectoral equities by using a series of quantile regression models. Overall findings suggested that an increase in global uncertainty factors, including financial stress, drives volatility contagion between Islamic and non-Islamic sectoral equities. Similarly, Ferrer et al. (2018) assessed the dynamic spillover from the aggregated measure of financial stress indices to economic uncertainty in the U.S. context and confirmed the intensified adverse impact of financial stress during recessionary periods such as the economic recession of 2008. Yu et al. (2024) examined the interaction between capital flows and financial stress by applying the conventional GVAR-based connectedness framework introduced by Diebold and Yilmaz (2012). Their results indicated that the influence of stress shocks is highly time-dependent, with stronger effects being observed during severe downturns such as the global financial crisis.

Building upon the literature review of the above-mentioned studies, a few studies only explore shock transmission between macro-uncertainty and financial stress indices (Cipollini and Mikaliunaite 2020), as well as the association between financial stress and capital outflows across borders (Yu et al. 2024). Apostolakis et al. (2021) examine the time-varying shock spillovers among commodity market risk, economic uncertainty and financial stress. In a similar manner, Chen et al. (2023b) also explore shock transmission between financial stress and energy-related commodity markets. Some studies focus on shock transmission between the aggregated measure of financial stress indices and equity market dynamics (Ahmed and Huo 2019; Fu et al. 2022; Xu et al. 2023). Shahbaz et al. (2024) explore asymmetrical shock propagation between global financial stress indicators and industrial metals’ conditional variances across quantiles. In contrast to existing studies, by incorporating aggregated financial stress indices for dynamic shock spillovers, this study conceptualizes contagion as the propagation of macro-financial fragility across developed and developing economies, rather than as a mere reflection of synchronized movements in financial market returns (Albrecht and Kočenda 2024; Sheikh et al. 2025). Therefore, this research article contributes to the literature by exploring whether these heterogeneous economic systems are interconnected through macro-financial architectures such as banking and equity sectors, and the forex and debt markets, not merely through synchronized equity market movements. Furthermore, this research article expands our understanding about the shock transmission mechanism between developing and developed economies’ financial architecture and suggests that financial market hedging strategies should incorporate systemic financial stress interdependence rather than relying exclusively on asset-level correlations. Therefore, by observing the time-varying aggregated measure of financial stress shock transmission during trade and geopolitical risk events, this research article contributes to the literature by exploring whether the financial stress contagion between developed and developing economies changes over time and by investigating the role of trade uncertainty and geopolitical risk events in altering the persistence and strength of financial stress shock spillovers.

3. Data and Descriptive Statistics

3.1. Study Model

In order to explore the time-varying financial stress-based shock transmission between developed financial systems (the U.S., the U.K., Japan and the European Union (EU)) and developing economic context (China), we take into account the aggregated measure of financial stress indices for these economies.

Table A1. Study model.

Variable Name	Identification of Variable	Definition of Variable	Data Ranges	Data Sources	Justification of Variable
U.S. Financial Stress Indices	Aggregated measure of financial stress indices for the U.S. economy (U.S. FSI)	The aggregated measure of financial stress indices (FSIs) as developed by the Asian Regional Integration Center (ARIC) of the Asian Development Bank (ADB) following the methodological approach by Park and Mercado (2014). These FSIs are indicators designed to quantify the intensity of systemic strain within developed (U.S., U.K., EU and Japan) and emerging (China) economies.	1 January 2010 to September 2023	https://aric.adb.org/database/fsi (accessed on 10 August 2025)	These FSIs represent an aggregate measure of financial fragility within an economy by integrating information from distinct segments of prominent financial architecture such as banking and equity sectors, forex markets, and sovereign bond markets (Park and Mercado 2014; Xu et al. 2023). In the existing literature, Xu et al. (2023) compared the predictability of equity market indices by using these aggregated measures of financial stress indices. The results reported that these stress indices outperform other macroeconomic determinants in predicting equity returns.
European Union (EU) Financial Stress Indices	Aggregated measure of financial stress indices for the EU economy (EU FSI)		1 January 2010 to September 2023	https://aric.adb.org/database/fsi (accessed on 10 August 2025)
United Kingdom (U.K.) Financial Stress Indices	Aggregated measure of financial stress indices for the U.K. economy (U.K. FSI)		1 January 2010 to September 2023	https://aric.adb.org/database/fsi (accessed on 10 August 2025)
Japan Financial Stress Indices (FSIs)	Aggregated measure of financial stress indices for the Japan economy (Japan FSI)		1 January 2010 to September 2023	https://aric.adb.org/database/fsi (accessed on 10 August 2025)
China Financial Stress Indices	Aggregated measure of financial stress indices for Chinese economic context (China FSI)		1 January 2010 to September 2023	https://aric.adb.org/database/fsi (accessed on 10 August 2025)

Note: This table explains the sources, identification, and data range for financial stress indices (FSIs). Furthermore, this table also explains the justification for using the aggregated measure of financial stress indices in order to explore the time-varying financial stress-based shock spillovers between advanced and emerging economies.

shows that these FSIs are sourced from the Asian Development Bank’s Asian Regional Integration Center (ARIC) database for the maximum time period spanning 1 January 2010 to September 2023 (https://aric.adb.org/database/fsi, accessed on 10 August 2025) and thus provides a robust empirical foundation for cross-market analysis. The aggregated measure of financial stress indices (FSIs) is developed by the Asian Regional Integration Center (ARIC) of the Asian Development Bank (ADB) by following the methodological approach by Park and Mercado (2014). These FSIs are indicators designed to quantify the intensity of systemic strain within developed (U.S., U.K., EU and Japan) and emerging (China) economies (see Table A1). One of the justifications of using FSIs is based upon the fact that they are the aggregated measure of financial fragility within an economy (developing and developed) as they integrate information from distinct segments of prominent financial architecture such as the banking and equity sectors, forex markets and the sovereign bond market (Park and Mercado 2014; Xu et al. 2023). Therefore, by incorporating aggregated financial stress indices for dynamic shock spillovers, this study conceptualizes contagion as the propagation of macro-financial fragility across developed and developing economies rather than as a mere reflection of synchronized movements in financial market returns (Albrecht and Kočenda 2024; Sheikh et al. 2025).

One justification for incorporating this specific aggregated measure of financial stress indices is that Xu et al. (2023) analyzed the predictability of equity market indices by using these FSIs and reported that these stress indices outperform other macroeconomic determinants in predicting equity returns. The incorporation of these financial stress indices (FSIs) also aligns with prior research examining the time-varying nature of shock transmission from the aggregated measure of financial stress indices to financial asset classes as well as non-financial asset classes, like energy-related and industrial commodity markets (Shahbaz et al. 2024; Sheikh et al. 2024; Sheikh et al. 2025). Within the scholarly discourse, Ferrer et al. (2018) incorporated measures of economic uncertainty alongside these financial stress indices to investigate their dynamic interconnections over time. Similarly, Elsayed and Yarovaya (2019) employed this aggregated measure of financial stress indices (FSIs) to analyze the transmission of stress-related shocks across the financial landscape of the MENA region. In a related contribution, Apostolakis et al. (2021) assessed the evolving spillover effects among financial stress indices, oil market volatility, and economic uncertainty, suggesting that the magnitude of such spillovers intensified significantly during the COVID-19 crisis. Park and Mercado (2014) also utilized aggregated financial stress indices (FSIs) and revealed that while internal financial disturbances remain the predominant source of volatility in national financial stress indices, regionally transmitted shocks exert a significant influence within emerging Asian economies. Therefore, FSIs represent a composite measure of systemic strain, encompassing pressures within the banking sector, fluctuations in foreign exchange markets, volatility in equity markets, and instability in debt instruments (see Table A1).

Another justification for incorporation of these financial stress indices (FSIs) developed by the ADB following the methodological approach by Park and Mercado (2014) is that these are aligned with this study’s objective of analyzing the cross-border transmission of systemic financial stress rather than volatility in individual markets. These FSIs can capture overall economic fragility by integrating financial stress information from a wider financial infrastructure such as banking, equity, foreign exchange, and sovereign bond markets. Therefore, these FSIs take into account an aggregated macro-financial fragility measure rather than relying on fragmented sources of shock spillovers (such as equity returns alone). The objective of this research article is to explain financial stress contagion as the transmission of macro-financial fragility rather than as simple return co-movement. Therefore, the incorporation of aggregated FSIs also simultaneously captures funding stress, balance-sheet risks, currency pressures, and sovereign vulnerabilities as prominent dimensions of overall systemic fragility within developed and developing economic contexts. However, while the FSI as incorporated in this study takes into account the fragility within the banking, equity, foreign exchange, and sovereign bond markets of emerging and advanced financial systems, using separate indicators (such as equity market risk or exchange rate risk) may also create redundancy, multicollinearity, and double-counting without improving explanatory power. The incorporation of these specific factors (other than the aggregated financial stress indices) may limit the ability to explore time-varying aggregated financial stress-based shock transmission. For the construction of these FSIs, the ADB applies a uniform methodology across economies following the methodological approach in (Park and Mercado 2014), ensuring harmonized weighting, normalization, and aggregation for estimating the financial stress indices. This standardization reduces measurement inconsistencies and strengthens the reliability of connectedness estimates.

Restricting the analysis to January 2010–September 2023 is justified by both data availability and analytical relevance. This period marks the longest horizon for which consistent financial stress index (FSI) data are obtainable from the Asian Development Bank’s Asian Regional Integration Center. Therefore, incorporating the long-horizon data set enhances statistical robustness, limits sampling bias, and allows for the examination of evolving cross-market financial stress shock transmission. The span also captures major external trade-related disruptions—including China’s rare earth export controls (2012), U.S. trade restrictions on solar panels (2012) and steel/aluminum (2018), the Russia–Ukraine agricultural dispute (2012), Brexit uncertainty (2018–2019), the COVID-19 pandemic (2020 onward), and geopolitical uncertainty such as the Russia–Ukraine war (2022). Situating these shocks within a unified framework enables the analysis of both cyclical fluctuations and structural disturbances, providing insights into China’s financial resilience under global uncertainty.

3.2. Descriptive Statistics

Table 1. Descriptive statistics and unit root characteristics of financial stress indices (FSIs) of emerging and developed economies.

	China	European Union (EU)	Japan	U.K.	U.S.
Mean	−0.46304	−0.60017	−1.28164	−0.48392	−0.26154
Median	−0.85735	−1.0022	−1.29025	−0.4861	−0.98495
Maximum	4.1928	4.1563	3.0373	3.6502	5.7319
Minimum	−3.2047	−3.2001	−3.9123	−2.9544	−3.1336
Std. Dev.	1.447167	1.375841	1.443699	1.377238	1.990602
Skewness	0.731484	0.981498	0.315023	0.665785	1.023174
Kurtosis	3.098571	3.759639	2.555719	3.544532	3.522269
Jarque-Bera	14.69163	30.27446	4.061345	14.1422	30.47873
Probability	0.000645	0	0.131247	0.000849	0
Sum	−75.9389	−98.4276	−210.188	−79.362	−42.892
Sum Sq. Dev.	341.3699	308.5492	339.7354	309.1757	645.887
Observations	164	164	164	164	164
Unit root test (at level)
ADF	−2.45 ***	−3.18 ***	−1.99 **	−3.02 ***	−3.47 ***
PP	−2.58 ***	−3.37 ***	−2.09 **	−3.06 ***	−3.22 ***

Note: The table presents the summary statistics of financial stress indices (FSIs) for advanced economies (U.S., U.K., European Union (EU), and Japan) alongside the emerging market of China. The stationarity properties of the FSIs are evaluated using the Augmented Dickey–Fuller (ADF) and Phillips–Perron (PP) unit root tests by Dickey and Fuller (1981) and Phillips and Perron (1988). The symbols *** and ** denote rejection of the null hypothesis at the 1% and 5% significance thresholds, respectively, confirming that the financial stress measures are level-stationary.

presents the descriptive statistics of financial stress indices for developed economies (the United States, the United Kingdom, Japan, and the EU) and a developing financial system (China). Among these, the United States record the highest standard deviation (1.99), followed by Japan (1.4437) and China (1.4471), indicating greater volatility in both upward and downward movements of financial stress within these economies. Moreover, the financial stress indices of the United States, the United Kingdom, and the EU report the highest kurtosis values of 3.52, 3.54, and 3.75, respectively. These elevated kurtosis estimates reflect a leptokurtic distribution, implying that financial stress in these markets deviates from the mean more frequently and sharply than under a normal distribution. In addition, results from the Augmented Dickey–Fuller (ADF) test by Dickey and Fuller (1981) and the Phillips–Perron (PP) test by Phillips and Perron (1988) confirm that the financial stress indices in both emerging and advanced economies are stationary at level. The rejection of the null hypothesis of non-stationarity is supported by p-values below the 1% and 5% levels of significance, and this also confirms that the mean and variance of financial stress within these developed and developing economies remain constant over time.

Figure 1 illustrates that the financial stress indices of the United States, China, the United Kingdom, and the EU exhibited notable upward fluctuations during the periods 2011–2012 and 2015–2016, with a pronounced surge observed amid the COVID-19 pandemic. In addition, both advanced economies (the United States, the United Kingdom, and the EU) and the emerging economy (China) experienced another marked escalation in financial stress throughout 2022 and 2023 (see Figure 1).

One of the justifications for the increase in financial stress in China, the U.S., the EU and Japan is that between 2011 and 2012, China introduced anti-dumping and countervailing tariffs, reaching nearly 22%, on imports of large automobiles and sports utility vehicles from the United States (Bradsher 2011). During the same period, the United States, the European Union (EU), and Japan filed disputes against China regarding its restrictions on the export of rare earth elements essential to advanced technological production (Gillispie and Pfeiffer 2012). In addition, the United Kingdom’s referendum in June 2016 to withdraw from the European Union created profound ambiguity over future trade arrangements and potential regulatory divergence. Similarly, the negotiations leading up to the Trans-Pacific Partnership (TPP), officially concluded in October 2015 (Gleeson et al. 2017), introduced substantial ambiguity for global trade during 2015–2016, as businesses confronted uncertainty over prospective tariff frameworks, particularly concerning U.S. participation. In addition, the year 2018 marked a period of intensified financial strain in the United States, the United Kingdom, Japan, and China, largely coinciding with Washington’s decision to levy global tariffs of 25% on steel and 10% on aluminum imports (Sachdev and Rao 2025). These developments suggest the need to examine how shocks arising from significant trade policy disruptions propagate financial stress between advanced and emerging economies.

Furthermore, the pronounced escalation of financial stress during the COVID-19 pandemic can be linked to China’s adherence to stringent zero-COVID-19 containment policies, which persisted until late 2022 and severely constrained industrial production, supply chains, and cross-border commerce. The eventual relaxation of these restrictions in November 2022 triggered abrupt disturbances (Tabash et al. 2025b), as both pent-up demand and disrupted supply networks produced heightened volatility (Tabash et al. 2025a). Furthermore, the intensification of trade-related uncertainty in 2022 was amplified by Russia’s invasion of Ukraine in February of that year. The subsequent sanctions imposed by Western economies—targeting energy, food, and trade flows—along with retaliatory measures, disrupted established trade corridors and supply chains while simultaneously generating policy instability regarding export prohibitions, import costs, and overall market accessibility. Therefore, this motivates exploring the time-varying dynamic financial stress shock transmission mechanism between developed and developing financial systems.

4. Methodology

To investigate spillover dynamics across financial markets, Diebold and Yilmaz (2012) integrated a VAR framework with Cholesky decomposition to estimate aggregate spillover magnitudes. This approach is vulnerable to ordering and lag-length choices and cannot identify directional spillovers (Deng and Xu 2024). To address these weaknesses, Diebold and Yilmaz (2012) introduced Generalized Forecast Error Variance Decomposition (GFEVD), which alleviates order dependence. Nevertheless, it does not accommodate the heteroskedastic features typical of financial series. Additionally, reliance on rolling-window estimation entails information loss and sensitivity to window size and remains highly affected by extreme observations. To overcome these constraints, Antonakakis et al. (2020) integrated the Time-Varying Parameter Vector Auto-Regression (TVP-VAR) framework with Diebold and Yilmaz’s (2012) spillover index, thereby formulating the TVP-VAR model for capturing both static and time-varying connectedness indices. Furthermore, the lag length is selected according to the minimum values of the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) and restricted to p = 1. For exploring time-varying financial stress shock spillovers, the rolling-window size of 10 is taken into account. This rolling-window size is appropriate for aggregated FSIs because of their higher persistence and the need to avoid over-parameterization in relatively short samples. The H-step-ahead forecasting horizon is set to 3 (H = 3). This is consistent with short-term spillover analysis in the financial stress literature. Furthermore, the selection of a particular lag order and a shorter H-step-ahead forecasting horizon is consistent with the approach in (Gabauer and Gupta 2018; Shahbaz et al. 2024).

One justification for selecting the H-step-ahead forecasting horizon is that financial stress-based shock spillovers are derived from GFEVD; therefore, the selection of a shorter H-step-ahead forecasting horizon is sufficiently important to estimate systemic transmission effects while restricting noise from long-horizon projections. Parameter evolution follows a Bayesian TVP-VAR specification with shrinkage priors (prior = “BayesPrior”) (see Antonakakis et al. 2020). The factors (κ₁ = 0.99, κ₂ = 0.96) control the degree of time variation in the state and covariance equations. The selection of κ₁, κ₂ and rolling-window size is closely matched with that in (Shahzad et al. 2025). Specifically, κ₁ governs the smooth evolution of coefficients, implying gradual parameter drift. Meanwhile, κ₂ allows for moderate stochastic volatility in the error covariance matrix. Estimation is conducted using a Bayesian state-space representation with Kalman filtering, which recursively updates time-varying coefficients and variance–covariance matrices. This approach is preferable to rolling-window VAR because it avoids arbitrary window truncation and yields smoother and more efficient parameter estimates.

According to Deng and Xu (2024), construction begins with the specification of the TVP-VAR(p) process as follows:

$$y_t=A_t{z}_{t-1}+{\epsilon }_{t,}{\epsilon }_t|{\mathsf{\Omega }}_{t-1}~N(0,{\sum }_t)$$

(1)

$$\mathrm{v}\mathrm{e}\mathrm{c}\left(A_t\right)=vec\left(A_{t-1}\right)+{\xi }_{t,}{\xi }_t|{\mathsf{\Omega }}_{t-1}~(0,{\Xi }_t)$$

(2)

with

$$z_{t-1}=\left[\begin{array}{c}{y}_{t-1}\\ y_{t-2}\\ .\\ .\\ .\\ y_{t-p}\end{array}\right]A_t=[A_{1t}{A}_{2t}\dots \dots ..A_{pt}]$$

(3)

Here, $y_t$ denotes an N × 1 vector, while $A_t$ represents an N × N parameter matrix. The matrices ${\Sigma }_t$ and ${\Xi }_t$correspond to time-varying variance–covariance structures, and vec($A_t)$ indicates the vectorization of $A_t$. The core idea of the spillover index approach is to evaluate the generalized forecast error variance decomposition (GFEVD) at an H-step horizon using the TVP-VAR(p) framework, expressed as follows:

$$y_t=A_t{z}_{t-1}+{\epsilon }_t={\displaystyle {\sum }_{j=0}^{\infty }}{B}_{jt}{\epsilon }_{t-j}$$

(4)

Building upon the GFEVD, the shock transmission from a variable $j$ towards a variable $k$ can be rewritten as

$${\varphi }_{jk,t}\left(H\right)={\displaystyle \frac{{\displaystyle {\sum }_{kk,t}^{-1}}{{\displaystyle {\sum }_{h=0}^{H-1}}\left({\displaystyle {\theta }_j^{\prime }}{B}_{h,t}{\sum }_t{\theta }_k\right)}^2}{{\displaystyle {\sum }_{h=0}^{H-1}}\left({\theta }^{\prime }{B}_{h,t}{\sum }_t{\theta }_j\right)}}$$

(5)

In this setting, ${\sum }_h$ denotes the covariance matrix of the error terms that evolves over time, while ${\sum }_{kk,t}$ reflects the time-dependent standard deviation associated with the $k^{th}$ disturbance. The vector ${\theta }_j$ serves as a selector, taking the value of one at the $i^{th}$ position and zero elsewhere. To guarantee that each row of the variance decomposition matrix sums to unity, a normalization procedure is employed.

$${\tilde{\varphi }}_{jk,t}\left(H\right)={\displaystyle \frac{{\varphi }_{jk,t}\left(H\right)}{{\displaystyle {\sum }_{j=1,k=1}^N}{\varphi }_{jk,t}\left(H\right)}}\ast 100$$

(6)

Whereas, ${\displaystyle {\sum }_{k=1}^N}{\tilde{\varphi }}_{jk,t}\left(H\right)=1,{\displaystyle {\sum }_{j=1,k=1}^N}{\tilde{\varphi }}_{jk,t}\left(H\right)=N$.

The Total Spillover Index (TSI) is derived from the normalized variance decomposition matrix and serves as a measure of the aggregate intensity of spillover effects across all variables in the system:

$${TSI}_t\left(H\right)={\displaystyle \frac{{\displaystyle {\sum }_{k=1,j\ne k}^N}{\tilde{\varphi }}_{jk,t}\left(H\right)}{{\displaystyle {\sum }_{j=1,k=1}^N}{\tilde{\varphi }}_{jk,t}\left(H\right)}}$$

(7)

In addition, detailed directional spillover measures, denoted “TO” and “FROM”, can be computed for variable $j$. These indicators capture both the extent to which variable $j$ transmits shocks to others and the degree to which it is affected by shocks originating from the rest of the system. Therefore, the directional TO, FROM and NET spillover of shocks between the financial stress indices can be written as

$${TO}_{J\to *,t}\left(H\right)={\displaystyle \frac{{\displaystyle {\sum }_{k=1,k\ne j}^N}{\tilde{\varphi }}_{kj,t}\left(H\right)}{{\displaystyle {\sum }_{j=1,k=1}^N}{\tilde{\varphi }}_{jk,t}\left(H\right)}}\ast 100$$

(8)

$${FROM}_{J\leftarrow *,t}\left(H\right)={\displaystyle \frac{{\displaystyle {\sum }_{k=1,j\ne k}^N}{\tilde{\varphi }}_{kj,t}\left(H\right)}{{\displaystyle {\sum }_{j=1,k=1}^N}{\tilde{\varphi }}_{jk,t}\left(H\right)}}\ast 100$$

(9)

By taking a differencing between the TO and FROM directional spillover measures, the Net Spillover Index (NET) for variable $j$ is derived, indicating its overall net transmission of shocks to the remaining variables in the system:

$${NET}_{j,t}\left(H\right)={TO}_{J\to *,t}\left(H\right)-{FROM}_{J\leftarrow *,t}\left(H\right)$$

(10)

According to Deng and Xu (2024) the time-varying net pairwise dynamic connectedness (NPDC) indices can be computed as

$${NPDC}_{jk,t}\left(H\right)=\left({\displaystyle \frac{{\tilde{\varphi }}_{kj,t}\left(H\right)}{{\displaystyle {\sum }_{j=1,k=1}^N}{\tilde{\varphi }}_{jk}\left(H\right)}}-{\displaystyle \frac{{\tilde{\varphi }}_{jk,t}\left(H\right)}{{\displaystyle {\sum }_{j=1,k=1}^N}{\tilde{\varphi }}_{jk}\left(H\right)}}\right)$$

(11)

5. Results and Discussion with Practical Implications and Policy Guidelines

Table 2. Financial stress shock transmission from the EU, Japan, the U.S. and the U.K. to China.

	China FSI	EU FSI	Japan FSI	U.K. FSI	U.S.A. FSI	FROM
China FSI	49.83	6.54	18.35	8.42	16.86	50.17
EU FSI	4.73	34.45	14.88	28.13	17.81	65.55
Japan FSI	12.17	13.81	31.86	16.64	25.52	68.14
U.K. FSI	5.99	26.58	16.92	32.17	18.34	67.83
U.S.A. FSI	10.7	16.17	24.17	17.01	31.94	68.06
TO	33.59	63.1	74.32	70.2	78.54	319.74
Inc.Own	83.42	97.55	106.18	102.37	110.48	TCI
NET	−16.58	−2.45	6.18	2.37	10.48	79.93%

Note: This table reports the transmission of financial stress shocks between advanced and emerging economies using the TVP-VAR connectedness framework by Antonakakis et al. (2020). The Total Connectedness Index (TCI) captures the aggregated share of forecast error variance explained by spillovers of financial stress shocks across the entire TVP-VAR system. “TO”, “FROM” and “NET” explain the directional financial stress spillovers. The “TO” measure indicates the proportion of shocks transmitted from variable “i” to all other variables, i.e., “j”, while the “FROM” measure reflects the share of shocks received by variable “i” from all other variables. The “NET” measure represents the difference between “TO” and “FROM,” with positive values signifying that an economy acts primarily as a transmitter of financial stress shocks, whereas negative values suggest its role as a shock absorber.

indicates that shocks within the U.S. and Japanese financial stress indices account for relatively large error variance contributions—16.86% and 18.35%, respectively—in forecasting China’s financial stress. Similarly, financial stress shocks originating in the U.K. also contribute 8.42% of shocks in China’s financial stress index. Moreover, the table demonstrates that financial stress in the European Union (EU), the U.S., the U.K., and Japan transmits substantial proportions of financial stress shocks to other financial systems, amounting to 63.1%, 78.85%, 70.2%, and 74.32%, respectively, which are considerably higher than those transmitted by China. By contrast, the Chinese financial system contributes the lowest spillover effect, transmitting only 33.59% of its stress shocks to other advanced economies. More specifically, Table 2 reveals that financial stress shocks originating in China contribute the smallest proportions to the EU, Japan, the U.K., and the U.S., compared with the levels of stress that China receives from these advanced economies. For example, financial stress shocks from China account for only 4.7%, 12.17%, 5.99%, and 10.7% of the financial stress transmitted to the EU, Japan, the U.K., and the U.S., respectively.

The TVP-VAR estimations in Table 2 reveal that approximately 80% of the total forecast error variance can be associated with the transmission of financial stress shocks across the system, underscoring the strong interconnectedness of financial fragility between China’s financial system and the developed economic markets of the U.S., EU, Japan, and the U.K. The higher percentage of overall shock spillovers implies greater contributions of forecast error variance transmission to the Chinese FSI, due to the shocks in the FSI of the developed financial system (see Table 2). Furthermore, a higher TCI also implies that financial stress within China is also highly co-integrated with the developed economies’ FSIs, which increases the risk of transmitting financial fragility from China to developed economic systems (Zhang and Mao 2022). These TCIs are the aggregated values of the forecast error variances due to the spillover of overall financial stress shocks within the entire TVP-VAR system (Shahbaz et al. 2024). Furthermore, a higher TCI also implies that financial stress is largely externally driven, confirming that financial markets (developed and developing) are deeply integrated (see Bouri et al. 2021). Therefore, domestic financial stability in developing economies is highly vulnerable to shocks originating in advanced markets. During periods of high spillover of financial stress shocks between developed and developing economies, domestic macro-prudential or monetary policies may be insufficient during global turmoil. A higher TCI (80%) also implies that financial stress shocks cannot be treated as a purely domestic phenomenon; it is structurally embedded in global transmission networks. Furthermore, this study provides strong empirical evidence that cross-border financial stress transmission is quantitatively dominant (80%) and that the Chinese economy is also more systematically exposed to liquidity shocks, funding and credit risk, macroeconomic imbalances, and sovereign default risk in the developed financial system.

These outcomes hold significant technical and policy-oriented implications. As most developed and developing financial systems are not immune to financial fraud and cybercrime (see Preciado Martínez et al. 2025), exploring the developed financial systems’ overall vulnerability to financial stress emanating within the emerging financial market system provides practical implications for regulators.

Firstly, from a regulatory perspective, the magnitude of the 80% spillover effect points to systemic risk externalities that extend beyond the capacity of domestic instruments to manage effectively. Consequently, regulators should integrate financial stress index (FSI)-based early-warning systems into supervisory frameworks to facilitate the real-time monitoring of cross-market vulnerabilities. In addition, the findings call for enhanced macro-prudential cooperation, including the development of cross-border surveillance platforms and the alignment of capital adequacy standards, which would help curb regulatory arbitrage and contain the amplification of stress shocks. Moreover, Table 2 further illustrates that advanced economies transmit substantial stress to one another, reinforcing the need for coordinated policy instruments such as joint liquidity facilities and central bank swap arrangements. These mechanisms are vital to dampening destabilizing contagion channels and preserving financial stability during episodes of elevated market stress.

Secondly, the substantial variance in contributions originating from advanced economies—including the United States, the EU, and the United Kingdom—indicate that China’s financial system exhibits pronounced exposure to external shocks. Given the strong transmission of financial stress from mature markets, Chinese authorities should strengthen engagement in international regulatory platforms to promote consistent standards regarding leverage management, derivative positions, and capital adequacy. Such coordination would diminish susceptibility to cross-border spillovers and help curb systemic vulnerabilities. Domestically, Chinese policymakers are advised to implement flexible countercyclical capital buffers alongside dynamic provisioning practices, thereby enhancing the resilience of financial institutions against disturbances imported from the U.S., the EU and the U.K. Moreover, stress-testing frameworks for China’s financial sector should explicitly integrate shock scenarios linked to the United States, the United Kingdom, and Japan, in light of their outsized impact on China’s financial stress dynamics. The findings further indicate the importance of establishing regional liquidity facilities and expanding bilateral or multilateral currency swap arrangements with major central banks. These mechanisms would serve as critical safeguards against liquidity strains induced by foreign shocks, while simultaneously curbing contagion risks. The results reveal that China’s financial system absorbs comparatively larger stress shocks from advanced economies—including the United States, the United Kingdom, Japan, and the EU—while exerting only limited spillover effects in return (see Table 2). This asymmetry indicates China’s dependence on global financial conditions and emphasizes its ongoing vulnerability to instability in major advanced financial centers. Moreover, the finding that advanced economies transmit substantial shocks not only to China but also among themselves illustrates the deeply interconnected nature of systemic stress within mature markets.

5.1. Time-Varying Financial Stress Shock Spillover Between Advanced and Emerging Financial Systems During Episodes of Heightened Trade Policy and Geopolitical Uncertainty

Figure 2 graphically depicts the time-varying aggregate measure of financial stress shocks among advanced and emerging economies, represented by the TCI. In contrast, Figure 3 and Figure 4 illustrate the directional spillovers of financial stress shocks, specifically “TO” and “FROM” transmission across financial systems, respectively. Figure A1 shows the difference between outward and inward spillovers of financial stress shocks (“TO”–“FROM”). Specifically, values plotted below the horizontal axis reflect an economy’s capacity to absorb external financial stress shocks, while values positioned above the axis represent its role as a transmitter of such shocks to stress in other financial systems.

Figure 2 illustrates the dynamic evolution of shock spillovers, indicating that the aggregated spillover effects of financial stress between developed and emerging financial systems intensified in 2010, between 2011 and 2012, and again in the 2014–2015 period. Moreover, the figure suggests pronounced surges in the overall connectedness indices of financial stress during 2018, the COVID-19 crisis, and the subsequent years of 2022 and 2023. Figure 3 and Figure 4 explain the directional “TO” and “FROM” shock spillovers. Figure 3 graphically shows the time-varying transmission of shocks from the financial stress indices of major advanced economies (the U.S., the EU, Japan, and the U.K.) to China’s financial stress indices. Meanwhile, Figure 4 shows the overall reception of these shocks by China’s financial stress indices originating from the financial stress in the same advanced economies. Therefore, Figure 3 captures the time-varying directional spillovers of financial stress transmitted from advanced financial systems (the U.S., the EU, Japan, and the U.K.) to China. These spillovers exhibit notable upward trends in 2012, 2015, and 2016 and intensified markedly during 2018, the pandemic period, and the years 2022–2023. The findings suggest that developed economies not only transmitted substantial stress to China but also reinforced one another’s systemic vulnerabilities in episodes of heightened global uncertainty. Similarly, Figure 4 documents the financial stress absorbed by China as a recipient of shocks originating from advanced markets. The results reveal that China’s financial system bore relatively higher spillover effects in 2010, 2012, and 2014–2015, while its vulnerability deepened further during 2018, the COVID-19 era, and 2022–2023. The evidence indicates the susceptibility of China’s financial markets to shocks originating in advanced economies, revealing China’s evolving but still subordinate position in transmitting global financial stress. Such insights are critical for regulators and policymakers, as they point to the need for stronger macro-prudential frameworks and international coordination to mitigate systemic contagion risks during periods of global financial instability.

Figure A1 shows that China experienced heightened financial stress spillovers from advanced financial systems during 2010 and again between 2012 and 2014. A key factor underlying the 2010–2011 surge was the dispute at the WTO concerning China’s export restrictions on raw materials, which were challenged by the U.S. (Office of the United States Trade Representative USTR 2011). In the same period (2010–2011), the European Union (EU) launched its “Trade, Growth and World Affairs” strategy, signaling a shift in trade priorities (Dee and Mortensen 2014), and this may have generated uncertainty for trading partners. Similarly, the U.S. 2010 Trade Policy Agenda emphasized stronger enforcement measures—particularly targeting China—which contributed to heightened bilateral trade tensions and uncertainty. The intensification of financial stress spillovers from all other advanced economies to China in 2012–2014 can be linked to several prominent trade policy disputes. In May 2012, the U.S. imposed anti-dumping duties on Chinese solar panels (Hufbauer and Vieiro 2012), provoking strong retaliation threats and adversely affecting the bilateral trade linkages between the two. During this period, China’s restrictions on rare earth exports also became the subject of WTO disputes (2012–2014), with rulings against China’s quotas and export duties (United States Trade Representative USTR 2014) may have created uncertainty for global manufacturers reliant on these inputs. Additionally, Japan’s entry into the Trans-Pacific Partnership (TPP) negotiations in 2013 reshaped expectations regarding the Asia-Pacific trade architecture (Schott and Solís 2013), while the launch of U.S.–EU TTIP negotiations in July 2013 opened a new chapter of trade policy uncertainty, with multiple negotiation rounds continuing until 2014.

Figure A1 shows that China’s financial system received higher financial stress spillovers during 2015–2016, largely associated with major trade policy shocks. For instance, the Brexit referendum of 2016 introduced significant uncertainty for the EU and the U.K., with global financial stress spillover effects. Kim et al. (2024) also found that Brexit led to higher uncertainty shocks and caused lower liquidity in financial markets. Moreover, the U.S. presidential election in November 2016, which brought Donald Trump to power, raised uncertainty regarding NAFTA, U.S. commitments to the WTO, and the future of the Trans-Pacific Partnership (TPP), thereby amplifying stress transmission across financial systems. Baker et al. (2016) also mentioned the increase in economic uncertainty amid U.S. presidential elections. Figure A1 shows that China’s financial system was subjected to intensified spillover effects during 2018 and the COVID-19 period. The escalation in 2018 can largely be linked to heightened trade policy uncertainty following President Trump’s proclamations that introduced 25% tariffs on steel imports and 10% tariffs on aluminum imports (Tabash et al. 2024). In retaliation, China imposed countermeasures the same year by levying tariffs on most of the U.S. goods, including products such as frozen pork, wine, and nuts, thereby amplifying bilateral trade tensions and contributing to elevated financial stress transmission. According to Li et al. (2021), the U.S.–China trade conflict substantially altered the interlinkages between economic policy uncertainty (EPU) and financial networks, whereas the emergence of the COVID-19 pandemic further intensified these linkages by increasing the overall density of the integrated network. Furthermore, COVID-19 has also caused trade disruptions because of the restricted supply chain distribution channels (Tabash et al. 2025a, 2025b).

Figure A1 shows that the Chinese financial system absorbed a comparatively larger share of stress spillovers from other advanced economies during 2022 and 2023. Complementing this, Figure A2 presents the time-varying net pairwise dynamic connectedness (NPDC) indices, capturing bilateral stress transmission between China and major developed economies. The results indicate that both the United States and Japan acted as dominant sources of financial stress shocks to China over this period. In addition, the NPDC linkages between China and the United Kingdom, as well as between China and the EU, reveal that China consistently functioned as a net recipient of financial stress from these economies throughout 2022 and 2023. Figure 5 further confirms that China functions primarily as a recipient of financial stress shocks, while the United States, the United Kingdom, and Japan act as dominant transmitters, causing greater stress spillovers toward China than they receive from other advanced financial systems and China.

Following Russia’s large-scale incursion into Ukraine in February 2022, the United States, the European Union (EU), the United Kingdom, and allied states enacted sweeping punitive measures against Moscow, encompassing restrictions on finance, trade, and energy exports (Boubaker et al. 2022; Sampedro et al. 2024). These actions, coupled with disruptions to global supply chains—particularly in energy and critical raw materials—generated spillover effects (Alberini et al. 2023) that reverberated through international financial markets. Consequently, heightened risk premia and elevated market volatility intensified cross-border financial interconnectedness and the transmission of financial stress shocks between China and developed financial systems. In parallel, U.S.–China relations between 2022 and 2023 became increasingly adversarial in the spheres of trade and technology policy. Key developments included the imposition of export controls on semiconductors and dual-use technologies (Council on Foreign Relations n.d.), restrictions on cross-border investment, and intensified political discourse centered on strategic dependencies. Such measures amplified uncertainty for firms and investors, particularly concerning regulatory exposure, vulnerabilities in supply chains, and the stability of investment returns. Additionally, in mid-2022, geopolitical developments between China and Taiwan also heightened geopolitical tensions (Reals 2025). These events elevated perceptions of geopolitical risk, with implications for shipping lanes, regional trade flows, and supply chain stability. The associated uncertainty translated into increased financial stress through channels such as higher risk premia, rising insurance costs, and abrupt market adjustments in response to geopolitical shocks.

5.2. Robustness Analysis

In prior scholarship, Suleman et al. (2023) investigated how shocks propagate across financial asset classes by applying Diebold and Yilmaz’s (2012) connectedness framework while assessing robustness by recalculating spillover effects over multiple H-step-ahead forecasting horizons. Therefore, consistent time-varying shock spillover between energy and sustainable financial markets across different H-step-ahead forecasting horizons characterize the stability of the model. The re-estimation of the TVP-VAR connectedness model across multiple forecast horizons is crucial to testing the robustness of the transmission dynamics of financial stress shocks between developed and developing financial systems. By applying 1-, 2-, and 3-month-ahead H-step-ahead forecasting horizons, the analysis allows for the verification of whether the identified spillover patterns are persistent across different time frames rather than being an artifact of a specific horizon (see Figure 6). The similarity of the financial stress domain TCIs across varied H-step-ahead forecasting horizons, as illustrated in Figure 6, reinforces the reliability of the estimated connectedness structure. Furthermore, Figure 7 also shows the reliability of TVP-VAR-based shock spillovers between financial stress indices of developed and developing financial systems, as the overall connectedness between variables remains similar across multiple lag orders (lag orders of 1 and 2).

Moreover, Figure 6 shows that the TCI, as a time-varying and aggregated measure of forecast error variance contributions, remains consistent across multiple H-step-ahead forecasting horizons and intensifies during well-known episodes of global and regional uncertainty. Furthermore, Figure 7 also shows that overall shock spillovers between financial stress indices of developed and developing financial systems also remain consistent across different lag orders and intensify during prominent TPU and GPR events. These include the 2010 sovereign debt tensions, 2012 EU financial stress, 2014–2015 market volatility, the 2018 trade policy disputes, the COVID-19 crisis, and the geopolitical and trade tensions of 2022–2023. This indicates the consistent responsiveness of financial stress spillovers across different H-step-ahead forecasting horizons (Figure 6) and multiple lag orders (Figure 7) to systemic shocks. This suggests that interconnectedness is not static but amplifies during periods of heightened uncertainty, underlining the importance of dynamic modeling. The convergence of TCI patterns across different forecasting horizons (Figure 6) and lag orders (Figure 7) also implies that the spillover effects are not merely short-lived or horizon-specific; rather, they represent a persistent, consistent and coherent feature of global financial architecture. This strengthens the argument that TVP-VAR connectedness provides a robust framework for capturing time-varying systemic linkages and offers policymakers and regulators valuable insights into the evolution of cross-market vulnerabilities.

5.3. Discussion

The findings regarding the higher time-varying financial stress shock transmission from developed financial systems to the financial stress of a developing economy (China) extend the literature on contagion and financial market interdependence. In the existing literature, studies are primarily focused on equity, forex and commodity market volatility spillovers by employing the conventional generalized VAR-based connectedness approach (Brož and Teplý 2025; Guru and Yadav 2023; Iqbal et al. 2022). However, the findings regarding the higher financial stress-related connectedness between developed and developing financial systems shift the analytical focus from market-specific volatility dynamics to the cross-border transmission of financial stress shocks measured by incorporating aggregated financial stress indices within the TVP-VAR framework. These FSIs are developed by the ADB by following the methodological approach by Park and Mercado (2014) and are classified as an aggregated measure of financial fragility within an economy that integrates information from distinct segments of prominent financial architecture such as banking and equity sectors, forex markets and sovereign bond markets (Park and Mercado 2014; Xu et al. 2023). By incorporating aggregated financial stress indices for dynamic shock spillovers, this study conceptualizes contagion as the propagation of macro-financial fragility across developed and developing economies, rather than as a mere reflection of synchronized movements in financial market returns (Albrecht and Kočenda 2024; Sheikh et al. 2025).

However, in the existing literature, studies mainly take into account the role of aggregated global financial stress indices in explaining the variations within the volatility of equity market returns, whereas few other studies accentuate the importance of exploring time-varying shock spillovers from these global financial stress indices to equity market returns and volatility dynamics (Tabash et al. 2024; Sheikh et al. 2024; Hoque et al. 2024; He et al. 2021b; Liang et al. 2023). Apart from these studies, Elsayed and Yarovaya (2019) only explored financial stress contagion among MENA economies. However, no effort has been made to explore whether the shocks within the aggregated global financial stress indices of developed financial systems (the U.S., the U.K., the EU and Japan) are transmitted with higher magnitude and persistence to the financial stress indices of a developing economy, e.g., China, and vice versa.

The findings also suggest that time-varying shock spillovers between financial stress indices of developed and developing economic contexts are intensified amid TPU and GPR adverse events. Furthermore, existing studies have only emphasized the influence of uncertainty-related factors—such as economic policy uncertainty, geopolitical tensions, oil price volatility, or global volatility indices—on financial stress (Li et al. 2025; Das et al. 2019). Others have examined the implications of stress shocks for asset return predictability (Xu et al. 2023; Ozcelebi et al. 2025), international capital flows (Bathia et al. 2023; Wang et al. 2022; Yu et al. 2024), macroeconomic dynamics, and commodity markets (Chen et al. 2023b). However, no effort has been made to explicitly observe whether the financial stress-related shocks generated within advanced financial markets are transmitted systematically into emerging economies’ financial stress during trade uncertainty and geopolitical risk events. Therefore, by observing the time-varying aggregated measure of financial stress shock transmission during trade and geopolitical risk events, this research article contributes by exploring whether financial stress contagion between developed and developing economies changes over time and the role of trade uncertainty and geopolitical risk events in altering the persistence and strength of financial stress shock spillovers. Unlike prior research that primarily focuses on static direct trade policy and geopolitical uncertainty shock transmission to equity markets (Li et al. 2022; Benguria et al. 2022; Zaremba et al. 2022), this study advances the literature by conceptualizing financial stress shock transmission as a dynamic and evolving process, sensitive to shifts in international trade relations, geopolitical policy disputes, and global crises.

FSI domain shock spillovers between developed and developing financial systems can capture overall economic fragility, as financial stress information is integrated from a wider financial infrastructure, including fragility within the banking, equity, foreign exchange, and sovereign bond markets of developed and developing economies. Therefore, these FSIs take into account an aggregated macro-financial fragility measure rather than relying on fragmented sources of shock spillovers (like only equity returns). Table 2 also confirms the existence of financial stress contagion as the transmission of macro-financial fragility rather than as simple return co-movement. Therefore, the incorporation of aggregated FSIs also simultaneously captures the funding stress, balance-sheet risks, currency pressures, and sovereign vulnerabilities as prominent dimensions of overall systemic fragility within developed and developing economic contexts. However, prior studies only highlight shock transmission between the aggregated measure of financial stress indices and equity market dynamics (Ahmed and Huo 2019; Fu et al. 2022; Xu et al. 2023). Shahbaz et al. (2024) explore asymmetrical shock propagation between global financial stress indicators and industrial metals’ conditional variances across quantiles. In contrast to existing studies, by incorporating aggregated financial stress indices for dynamic shock spillovers, this study conceptualizes contagion as the propagation of macro-financial fragility across developed and developing economies, rather than as a mere reflection of synchronized movements in financial market returns (Albrecht and Kočenda 2024; Sheikh et al. 2025).

The results also indicate a stronger transmission of financial stress shocks from advanced economies—including the United States, the EU, the United Kingdom, and Japan—to China’s emerging financial system, which aligns with earlier research while extending its implications. For instance, Dovern and van Roye (2014) suggested the detrimental consequences of financial stress in developed economies on global macroeconomic performance. Similarly, Dufrénot and Keddad (2014) documented that turbulence in advanced financial markets during the subprime crisis generated sustained volatility across emerging economies. More recently, Li et al. (2025) reinforced this perspective by demonstrating that financial stress is closely intertwined with global uncertainty factors, with shocks from advanced financial centers exerting disproportionately strong and widespread effects on other markets. In parallel, Fink and Schüler (2015) argued that systemic financial stress in developed markets represents a central determinant of cyclical patterns and instability in emerging economies. Collectively, these findings converge on the conclusion that stress episodes in advanced financial systems are transmitted with considerable force to emerging markets, shaping both their financial stability and broader economic trajectories. By conceptualizing the propagation of financial stress as a contagion-like process—operating through trade linkages, capital movements, portfolio reallocations, and equity market interdependence—this study extends contagion theories that traditionally focus on return or volatility co-movements (Albrecht and Kočenda 2024; Sheikh et al. 2024; Sheikh et al. 2025).

6. Conclusions with Future Research Directions and Limitations

In prior research, the emphasis has largely been on the transmission of financial stress shocks to equity or commodity markets, while limited attention has been given to contagion dynamics between aggregated measures of financial stress indices of advanced and emerging financial systems within a TVP-VAR connectedness framework. This study contributes to filling the gap by being the first to examine the evolving transmission of financial stress shocks from major developed economies—the U.S., the U.K., the EU, and Japan—to China, particularly during periods marked by significant trade and geopolitical uncertainty. Therefore, this study contemplates the use of aggregated financial stress indices for dynamic shock spillovers. This study conceptualizes contagion as the propagation of macro-financial fragility across developed and developing economies by employing the TVP-VAR approach, rather than as a mere reflection of synchronized movements in financial market returns.

The results suggest that disturbances in the U.S. and Japanese financial stress indices exert comparatively larger influences on the forecast error variance of China’s financial stress, contributing 16.86% and 18.35%, respectively. In parallel, stress originating from the U.K. explains 8.42% of the fluctuations in China’s financial stress index. Furthermore, the findings illustrate that the European Union, the U.S., the U.K., and Japan function as dominant transmitters of financial stress, with spillover magnitudes of 63.1%, 78.85%, 70.2%, and 74.32%, respectively—substantially surpassing those generated by China. In contrast, the Chinese financial market exhibits the weakest transmission capacity, disseminating only 33.59% of its stress shocks to other advanced economies. More precisely, Table 2 shows that China’s spillovers to the EU, Japan, the U.K., and the U.S. are minimal—4.7%, 12.17%, 5.99%, and 10.7%, respectively—when compared with the larger stress inflows it receives from these developed financial systems. Moreover, the evolving dynamics of financial stress transmission indicate that cross-market spillovers tend to escalate during periods characterized by heightened geopolitical tensions and trade-related uncertainties. The findings reveal an asymmetric contagion structure in which advanced economies consistently operate as the main sources of financial stress, whereas China predominantly absorbs shocks. The variation in contributions (U.S., 16.86%; Japan, 18.35%; U.K., 8.42%) to China’s stress levels indicates the heterogeneity of cross-border transmission channels. This suggests that contagion intensity is not uniform but shaped by differences in market depth, financial integration, and systemic relevance. Importantly, China’s relatively weak outward influence, despite its large economic scale, challenges the notion that economic size directly translates into systemic transmission capacity.

This research article explains the time-varying financial stress-based shock transmission mechanism between developing and developed financial systems amid TPU and GPR shocks. Future studies should take into account the role of developed and developing economies’ financial stress in transmitting shocks to the macroeconomic policy uncertainty of oil-exporting and oil-importing economies. Furthermore, future studies should also take into account the quantile domain VAR approach in order to observe the shock transmission mechanism across bearish, bullish, and moderate quantiles.

Study Limitations

One of the limitations of this study is connected with the data availability constraints, as the analysis on financial stress shock transmission incorporates a limited set of developed and emerging economies, i.e., China, the European Union (EU), Japan, the United States and the United Kingdom. Therefore, this restricts the broader generalizability of the results across the full spectrum of global financial systems because we did not include the major oil-importing emerging economies, i.e., GCC region. This limitation is generally due to the unavailability of global aggregate financial stress indices for the majority of the emerging and developed financial systems rather than a conceptual shortcoming of the research design. However, our selection of emerging and developed economies for observing financial stress domain shock transmission is also consistent with the study design in (Chen et al. 2025; Vespignani 2015), as these studies emphasize equity market volatility shocks and monetary policy shock transmission among U.S., China, Japan and EU. Furthermore, this research article only observes financial stress domain shock transmission for the period from January 2010 to September 2023. This is because aggregated financial stress indices for the selected developed and developing financial systems are only available for this specific timeframe. One of the benefits of exploring the time-varying financial stress-based shock spillovers during this time is to capture the fluctuations in spillovers amid trade and geopolitical risk events such as the U.S. trade restrictions on Chinese solar panels, Brexit, the heightened trade uncertainty following the U.S. presidential elections, COVID-19, the Russia–Ukraine war and the SVB implosion. However, this timeframe (January 2010 to September 2023) does not cover earlier crisis episodes such as the 2007–2008 global economic turmoil and the 1998–2000 Asian financial crisis. Nevertheless, the selected sample period remains sufficiently extensive to provide robust inference regarding contemporary interconnectedness patterns and evolving cross-border spillovers in the post-crisis era. Overall, these limitations are largely data-driven rather than methodological weaknesses.