Maintaining investment efficiency has become increasingly important for firms in an environment characterized by elevated uncertainty. This study examines the relationship between firm-level uncertainty and investment inefficiency of Chinese listed firms, highlighting the crucial role of corporate internal control in preserving investment efficiency. Using newly constructed measures of firm-level uncertainty and an unbalanced quarterly firm-level panel data covering the period from 2009 to 2022, the results show that heightened firm-level uncertainty worsens investment inefficiency. However, the adverse effect is mitigated for firms with a robust internal control mechanism. Mechanism analysis shows that internal control offsets the effect of uncertainty on investment inefficiency by reducing corporate underinvestment. Moreover, analysis using disaggregated internal control indicators suggests that strategic planning, operating management, report reliability, legal compliance, and asset safety play a significant role in alleviating the effect of uncertainty on investment inefficiency . Heterogeneity analysis reveals that the moderating effect of internal control is significant for state-owned enterprises (SOEs), but insignificant for foreign-funded firms. Overall, this study highlights the importance of strengthening corporate internal control in preserving investment efficiency during periods of heightened uncertainty.

High-quality economic development (QUA) has become a central policy objective in emerging economies, yet achieving both sustained economic growth and environmental protection remains an unresolved challenge. Green technology innovation (INN) is increasingly viewed as a potential solution, but its effectiveness and and the institutional conditions that support it warrant systematic empirical examination. QUA remains a central goal of sustainable growth, yet balancing economic expansion with environmental protection remains a challenge. This study investigates the role of INN in advancing QUA, focusing on the moderating effects of environmental regulation (REG) and government intervention (GOV). Using panel data from 41 cities in the Yangtze River Delta of China (2010-2020), QUA is evaluated across five dimensions: innovation, coordination, openness, green, and shared development. Results from fixed-effects regression reveal that INN significantly enhances QUA, and this finding is robust to alternative specifications. REG strengthens the effect of INN, consistent with the Porter Hypothesis, whereas GOV weakens this relationship by crowding out private innovation and resource efficiency. Spatial econometric analysis further indicates that INN improves local upgrading but lacks significant spillover effects across neighbouring regions, reflecting barriers in technology diffusion and inter-regional cooperation. These results highlight the need for coordinated strategies that integrate innovation, effective regulation, and appropriately calibrated government involvement to foster sustainable and innovation-driven development. These findings advance the literature by showing how INN, under varying REG and GOV conditions, drives multidimensional QUA and offers actionable insights for policy design.

This study investigates the dynamic impact of climate change performance on extreme tail risk transmission across global financial markets. Based on the "Too Extreme to Fail" conceptual framework, we propose a cascading failure network model using QRNN-∆CoVaR and QRNN-∆CoES to quantify the domino effect of tail risk propagation. The model captures tail dependencies and reveals how variations in climate governance performance modulate the intensity and pathways of risk contagion. Our main analysis utilizes daily market data from 1998 to 2024, aligned with the Climate Change Performance Index data from 2007 through a matched time window approach. The findings demonstrate that climate-sensitive factors significantly amplify systemic vulnerabilities, whereas superior climate governance serves as a critical risk buffer during periods of extreme volatility. Empirical results reveal significant spatial and temporal heterogeneity in risk contribution, with certain regions exhibiting higher sensitivity and momentum during major financial crises. Backtesting results confirm that our proposed nonlinear framework provides superior accuracy in quantifying global systemic risks compared to traditional linear methods, offering a robust tool for climate-integrated financial stability monitoring.

This study examines machine learning methods for predicting S&P 500 returns using monthly data from November 1987 to February 2022 and employing 11 predictors derived via principal component analysis. This study contributes to the literature by proposing a practitioner-oriented, economically grounded evaluation framework for machine learning-based market timing that emphasizes long-term cumulative geometric total returns (including dividends), downside-sensitive risk measures, and benchmarking against the no-information rate. Additionally, this study provides the first systematic, investable assessment of state-of-the-art time-series classification frameworks, including the Diverse Representation Canonical Interval Forest (DrCIF), ARSENAL, and HIVE-COTE 2.0 (HC2), in a long-term financial forecasting setting. Model performance was evaluated using strict out-of-sample simulations in terms of predictive accuracy, risk-adjusted excess returns, and terminal portfolio values. The results indicate that DrCIF and Long Short-Term Memory (LSTM) models deliver the highest economic value relative to passive index investing, logistic regression, and commonly used machine-learning benchmarks. The findings suggest that industry practitioners may improve cumulative and risk-adjusted returns relative to the S&P 500 by implementing defensive market timing adjustments guided by machine-learning-based forecasts. The empirical results indicate that ensembling high-performing models does not necessarily improve economic outcomes, and correctly identifying a small number of extreme downside events can be sufficient to achieve long-run outperformance.

This study investigates how monetary policy rate adjustments and inflation trends influence stock market performance in the United States (US), the European Union (EU), the United Kingdom (UK), and Japan, covering the Global Financial Crisis (GFC) and the COVID-19 pandemic. Using wavelet techniques, the analysis captures time- and frequency-specific relationships that move beyond static, single-economy approaches. Results show that interest rate cuts stabilised stock returns during the GFC, partically in the US and UK, while post-pandemic dynamics were dominated by inflation shocks that amplified volatility in all economies excpet for Japan, where inflation effects remained weak at medium-horizon. The EU's delayed tightening prolonged uncertainty and exhibited weaker short-horizon transmission, whereas Japan's ultra-loose stance muted the effects of rate hikes, and inflation remained a relatively mild and statistically weak influence even post-2021. The findings highlight a structural shift from interest rate dominace to inflation to inflation-driven volatility in all economies except Japan, with important implications for central banks and investors.

This study examines the impact of digital talent mobility on economic growth in the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) using the spatial Durbin model and spatial mediation model. Based on panel data analysis from 2003 to 2022, our findings show that, first, digital talent flow significantly enhances economic growth and generates positive spatial spillovers through knowledge and technology diffusion. Second, industrial structure upgrading positively mediates this relationship, but its spatial mediation negatively affects surrounding cities by causing talent drain. Lastly, future industrial development potential exhibits only a limited local effect while displaying negative spatial mediation. As a policy conclusion, this study proposes that within the complex institutional context of the GBA, dynamic equilibrium between the radiative effects of core cities and balanced regional development should be achieved through cross-regional industrial coordination and integrated talent governance mechanisms.

This study examines how intraday trading restrictions influence market quality through an agent-based computational finance (ACF) model calibrated to China's A-share market. The model simulates interactions among heterogeneous investors, under alternative trading rules (T+0 vs. T+1). Simulation experiments reveal that the T+0 trading rule significantly enhances price discovery efficiency and market liquidity while reducing volatility relative to T+1. These improvements arise from more rapid information incorporation and higher trading frequency. However, welfare effects are heterogeneous: informed investors benefit disproportionately, while noise traders experience minimal performance gains. Liquidity providers maintain stable profits under both regimes, suggesting that market-making remains resilient to settlement rule changes. The findings indicate that intraday trading flexibility can improve market efficiency without compromising stability when supported by appropriate regulatory safeguards. The study provides policy insights for regulators considering the reintroduction of T+0 trading rule in emerging markets, emphasizing the need to balance liquidity enhancement with investor protection and market stability. While the analysis focuses on the Chinese market, the results have broader implications for markets seeking to enhance liquidity and efficiency without compromising stability or investor protection.

This study examines the multi-factor models of the prospect theory (PT) in the Chinese stock markets. Based on all A- and B-share stocks from January 2000 to December 2022, it develops a new behavioural asset pricing framework that augments standard factor models with a PT-based component. Our results reveal the following: first, the PT value (PTV) factor has a significant negative correlation with stock returns in the A-share market, whereas it has a non-significant correlation with stock returns in the B-share market. Second, the effect of the PTV factor on stock returns remains significant in the A-share market even after extending the baseline specification by adding additional risk factors and lottery-type controls. Third, economic policy uncertainty is considered a non-traded macro-economic state variable and moderates the relationship between the PTV and stock returns. Finally, the extended behavioural asset pricing specification exhibits greater explanatory power than traditional asset pricing models. This study contributes to the extant literature by providing new evidence on Chinese A- and B-share markets.

In the context of global climate change and the growing emphasis on sustainable development, green total factor productivity (GTFP) is increasingly recognised as a critical indicator that integrates both economic efficiency and environmental performance. Meanwhile, environmental, social, and governance (ESG) ratings have become an important reference for investors and corporate decision-makers. However, there are notable inconsistencies in ESG scores across different rating agencies, due to divergent evaluation criteria. This paper explores the impact of ESG rating divergence on the GTFP of Chinese A-share listed firms for 2015 firms between 2013 and 2022. The empirical results reveal a significant negative relationship between ESG rating divergences and GTFP, indicating challenges to sustainable development. Transmission mechanism analysis suggests such divergences are conveyed through increased financing constraints. Nevertheless, market attention and a firm's innovation capacity can mitigate this negative impact. Furthermore, this relationship between ESG rating divergences and GTFP exhibits heterogeneity, as it is not observed for firms in primary industries, central, western regions, non-state-owned, non-high-technology sectors and polluting industries. These findings provide evidence for the necessity of standardised ESG rating disclosure to enhance the reliability of ESG assessments.

Current progress toward the United Nations Sustainable Development Goals is frequently constrained by coordination gaps, greenwashing, and overreliance on fiscal subsidies. To address these challenges, this study develops a four-party evolutionary game model involving local governments, financial institutions, enterprises, and residents to investigate how digital governance shapes green transformation. Within this framework, digital governance improves regulatory monitoring and green credit screening, while residents provide market validation through green consumption choices and trust-based participation. By endogenising digital regulation, green finance adoption, enterprise green transformation, and resident green consumption, the model identifies multiple evolutionary regimes, ranging from traditional lock-in to Pareto-optimal coordination. Stability analysis shows that reaching the optimal equilibrium requires digital cost reduction, market premiums, and resident green utility to collectively outweigh transformation and financing costs. Numerical simulations further suggest that the system can evolve toward a coordinated state in which digital regulation, green finance, enterprise transformation, and resident participation reinforce one another, indicating that partial transformation does not persist as a stable interior outcome. The findings suggest that sustainable green transformation cannot rely on unilateral policy intervention or short-term subsidies alone. Hence, policymakers should use subsidies as earlystage catalysts while prioritising investments in digital traceability and consumer trust mechanisms to foster a self-reinforcing green ecosystem.

Understanding how monetary conditions influence safe-haven assets is increasingly important during periods of financial instability and large-scale monetary policy intervention. This study examines the dynamic relationships between monetary liquidity, inflation, interest rates, and gold returns in the United States from January 2005 to August 2025, encompassing the global financial crisis and the COVID-19 pandemic. An integrated framework that combines wavelet analysis and wavelet-conditional structural Granger causality is employed to capture time-frequency co-movements and horizon-specific predictive relationships across short-, medium-, and long-term horizons. The results indicate that the predictive information associated with monetary liquidity, the underlying inflation proxy, and interest rate changes varies across reconstructed frequency bands. Statistically significant predictive relationships are observed primarily over the medium- and long-term horizons, whereas predictive relationships are relatively weak over the short-term horizon. The findings indicate that the relationship between monetary liquidity and gold returns is frequency-dependent after accounting for the underlying inflation proxy and interest rate changes. These findings provide useful evidence for understanding how the predictive relationship between monetary liquidity and gold returns varies across investment horizons.

This study examines machine learning methods for predicting S&P 500 returns using monthly data from November 1987 to February 2022 and employing 11 predictors derived via principal component analysis. This study contributes to the literature by proposing a practitioner-oriented, economically grounded evaluation framework for machine learning-based market timing that emphasizes long-term cumulative geometric total returns (including dividends), downside-sensitive risk measures, and benchmarking against the no-information rate. Additionally, this study provides the first systematic, investable assessment of state-of-the-art time-series classification frameworks, including the Diverse Representation Canonical Interval Forest (DrCIF), ARSENAL, and HIVE-COTE 2.0 (HC2), in a long-term financial forecasting setting. Model performance was evaluated using strict out-of-sample simulations in terms of predictive accuracy, risk-adjusted excess returns, and terminal portfolio values. The results indicate that DrCIF and Long Short-Term Memory (LSTM) models deliver the highest economic value relative to passive index investing, logistic regression, and commonly used machine-learning benchmarks. The findings suggest that industry practitioners may improve cumulative and risk-adjusted returns relative to the S&P 500 by implementing defensive market timing adjustments guided by machine-learning-based forecasts. The empirical results indicate that ensembling high-performing models does not necessarily improve economic outcomes, and correctly identifying a small number of extreme downside events can be sufficient to achieve long-run outperformance.

This study examines changes in China’s fiscal policy before and after the US–China trade war during President Trump’s first term using the Granger causality test with monthly data from January 2015 to July 2021. The findings are as follows: Before the trade war, taxation played a key role in supporting exports, controlling imports and promoting consumption, while government expenditure had a limited impact on exports and domestic consumption. After the trade war, exchange rate volatility increasingly influenced government expenditure and tax revenues. Government spending significantly impacted exports, imports, taxes and consumption, while taxes continued to affect exports and imports. The trade war created uncertainty in the global trade environment, challenging China’s fiscal policy. However, the government’s policy adjustments before and after the trade war showcased its flexibility and resilience, highlighting the fiscal policy’s role in adapting to new economic conditions, stabilizing the economy and sustaining international trade. This study provides insights into the potential effects of US–China trade tensions during Trump’s second term and offers a basis for understanding future fiscal policy adaptations. It is recommended to enhance export competitiveness, maintain stable tax policies, adopt proactive fiscal measures, manage exchange rates and boost domestic consumption.

This study examines the spatio-temporal heterogeneity impacts of digital talent mobility on economic growth across five major Chinese urban agglomerations. Based on panel data analysis using the Geographically and Temporally Neural Network Weighted Regression (GTNNWR) model, our findings show first, overall digital talent flow has sustained positive effects in advanced regions. Second, intra-regional flows generate more stable growth than inter-regional flows. Third, while inflows promote growth, outflows reduce innovation in developed areas but ease labor imbalances in periphery cities. As a policy implication, this study advocates region-specific talent strategies to optimize mobility and support balanced development.

This study analyzes the Exchange Rate Pass-Through (ERPT) on export and import prices in Indonesia’s manufacturing trade with ASEAN Plus-three countries (China, Japan, and South Korea) using monthly data from 2016 to 2022. Employing the Nonlinear (NARDL) models, we examine the degree and asymmetry of ERPT across five key manufacturing sectors for each partner. The NARDL model confirms asymmetric price responses to currency appreciation and depreciation in several sectors. The findings reveal significant sectoral and country-specific variations. Some export sectors experience complete pass-through in the short term but shift to negative complete passthrough in the long run. Meanwhile, imports from China and South Korea generally show a negative complete pass-through, while Japanese imports exhibit no significant ERPT. The results highlight the importance of exchange rate policies tailored to sector characteristics and partner countries, along with risk mitigation strategies, to strengthen Indonesia’s export and import competitiveness.

As sustainable development pressures escalate, stakeholders increasingly prioritise sustained corporate innovation resilience. However, these pressures also incentivise firms to adopt symbolic greenwashing tactics to mislead stakeholders, while the existing literature has paid limited attention to how such deceptive practices erode firms’ innovation capabilities. This study examines the impact of corporate greenwashing on innovation resilience using a panel dataset of 5,183 Chinese listed firms from 2015 to 2024. We find a significant negative relationship between corporate greenwashing and innovation resilience, suggesting that symbolic environmental management does not generate the technological depth required to withstand external shocks. Mechanism analysis indicates that this adverse effect operates through increased negative media coverage and reduced research and development expenditure, which jointly undermine corporate reputation and heighten structural rigidity. Further evidence shows that external institutional investors mitigate this negative relationship by strengthening monitoring and information verification, whereas government subsidies exacerbate it by creating soft budget constraints and fostering organisational inertia. The findings extend the literature by showing that innovation resilience depends not only on substantive technological investment and capability building, but also on limiting symbolic environmental communication that misallocates resources and weakens external support.

The environmental, social, and governance (ESG) performance of firms in emerging economies is critical for achieving global sustainable development. However, prior research has not sufficiently examined how global buyers shape these firms' ESG performance. Using hand-collected data on the major foreign customers of Chinese A-share listed companies from 2009 to 2024, this study shows that: First, major foreign customers significantly improve firms' ESG performance, but the effects are uneven across dimensions, with stronger effects on the environmental and governance dimensions than on the social dimension. Second, the impact of major foreign customers is greater when they have stronger ESG preferences, higher regulatory quality, and closer geographic proximity. Third, ESG improvements driven by major foreign customers further improve firms’ productivity, economic value added, and regional economic development. Overall, the findings suggest that global buyers are not only channels of demand and production relocation but also carriers of ESG-related governance pressure.

Based on stakeholder theory, when analyzing the influence of customers on firms' environmental, social and governance (ESG) behaviors, it is necessary to consider the customer ESG preferences. In this study, we propose that customer ESG preferences drive firms' ESG performance. In addition, we further propose that customer power, customer urgency and customer legitimacy are important relational attributes for customers to affect firms' ESG performance. Our argument is supported by empirical tests of Chinese A-share listed companies between 2009 and 2020. Our findings identify two impact channels through which customers influence firms' ESG behaviors - the customer governance and support channels. Finally, heterogeneity analysis shows that the influence of customer ESG preferences is more pronounced for state-owned enterprises and samples with low initial ESG values. This study enriches the ESG literature and stakeholder theory.

The advancement of computational modeling, data systems, and digital infrastructure has enabled the rise of agent-based computational finance (ACF). This study models interactions among heterogeneous investors. By embedding behavioral logics such as environmental, social, and governance (ESG) preferences and volatility thresholds, the model captures microstructural dynamics under different trading rules. Using ACF, the authors compare transaction plus 0 day (T+0) to transaction plus 1 day (T+1). Results show that T+0 improves price discovery, deepens liquidity, and reduces transaction costs. From a computational perspective, this research contributes to ACF by showing how policy logic and investor heterogeneity can be encoded and tested in a replicable simulation environment. The simulation method offers a scalable approach for regulators to assess sustainability-aligned reforms under diverse institutional settings. The study bridges computational finance and digital governance, highlighting the potential of integrating advanced IT into ACF to support adaptive, climate-conscious market design.