This paper examines how sovereign cash-bond relative value interacts with FX option-implied risk and whether predictive connectedness changes during market dislocations. Using daily data for a panel of currencies, the empirical analysis follows a two-stage design. It first asks whether sovereign Z-spread differentials add information to an option-only system and then models Z-spreads, at-the-money implied volatility, and risk reversals jointly to compare cross-market predictive relationships between tranquil and dislocation states. The analysis controls for macroeconomic news, interest-rate repricing, and spot exchange-rate movements.The results show that adding Z-spreads leaves connectedness within FX options virtually unchanged. In the joint system, however, predictive relationships run in both directions, although their contributions to forecast variation are modest. The most consistent finding is that the option-to-bond predictive relationship changes during market stress. This state variation remains when the common U.S. bond-market component is treated differently and when stress is identified using alternative bond-market thresholds or the VIX. Its strength and direction are not uniform across these choices, however, and become less conclusive when uncertainty over stress classification is allowed for. In the opposite direction, evidence that bonds predict options weakens once the common U.S. component is separated. A further finding is pronounced currency heterogeneity: the unconditional bond-to-risk-reversal relationship is driven largely by the Japanese yen. Overall, the findings show that FX options contain state-dependent predictive information about sovereign cash-bond relative value and that cross-market relationships vary across market conditions and currencies.
Supply chain shareholding means that suppliers or customers establish economic ties with firms through equity holdings. Its impact on corporate sustainable development capacity has gradually attracted widespread scholarly attention. Drawing on relational theory, supply chain shareholding may provide material support through resource collaboration, thereby improving operational efficiency and securing the material foundation for corporate ESG development. It may also strengthen corporate information quality and enhance process supervision through governance embedding, thereby guiding firms' ESG practices. Using a sample of Chinese A-share listed companies from 2009 to 2023, this study empirically examines the impact of supply chain shareholding on corporate ESG performance and the underlying mechanisms. The results show that supply chain shareholding significantly improves firms' ESG performance, and this finding remains robust after a series of robustness checks. Mechanism analysis indicates that supply chain shareholding enhances ESG performance through resource collaboration (increasing the acquisition of trade credit and improving working capital efficiency) and governance embedding (improving information disclosure quality and reducing the frequency of penalties). Heterogeneity analysis indicates that the positive effect of supply chain shareholding on ESG performance is more pronounced when firms have lower supply chain bargaining power and when they operate in the manufacturing sector. Furthermore, this study finds that, compared with supplier shareholding, customer shareholding has a stronger positive effect on ESG performance.
We use the sports betting market as a real-world market laboratory to study the home bias in individuals’ behavior. In contrast to traditional financial markets, where institutional frictions and the joint hypothesis problem confound explanations of the home bias, the sports betting market’s quasi-experimental features enable us to cleanly test whether preferential betting on home teams reflects informational advantages or behavioral forces. We find that individuals systematically favor local teams, domestic teams, and teams featuring players from their home country. This home bias does not yield superior performance but distorts individuals’ portfolios, generating welfare costs of similar magnitude to those in the stock market. Our findings help strengthen the foundation of behavioral explanations of the home bias in similar market environments.
This paper investigates the presence of psychological price barriers in crude oil prices and their interaction with geopolitical risk (gpr) and economic policy uncertainty (epu), using daily data for Brent and WTI crude oils from 1987 to 2025. While our analysis yields mixed evidence in support of psychological barriers, both Brent and WTI tens digit prices are characterised by results that are consistent with clustering away from psychological barriers. Our research provides novel insights in terms of uncovering an asymmetric relationship between geopolitical risk and the likelihood of prices encountering these barriers. To address a high-dimensional setting with numerous lags, the One Covariate at a Time Multiple Testing (OCMT) approach is employed for variable selection. We find that when gpr is heightened, there is more likelihood of crude oil prices encountering psychological price barriers. When gpr is subdued, decreases in geopolitical risk can be more impactful for downward breaches of such barriers. In contrast, epu does not exhibit any explanatory power regarding the likelihood of prices encountering psychological barriers.
This study constructs a comprehensive urban economic resilience index covering 281 prefecture-level cities in China from 2012 to 2022, encompassing three dimensions: recovery capacity, adaptability, and transformative capacity. Using this index, we exploit the pilot implementation of the “Delegation, Regulation, and Service” Reform in China's Tax System as a quasi-natural experiment to examine how optimizing the tax business environment relates to urban economic resilience. The findings indicate that the pilot reform is positively associated with urban economic resilience, and supplementary evidence is consistent with a cross-regional capital-flow channel. Furthermore, the reform's effects are particularly significant in southern cities, non-provincial capital cities, non-resource-based cities, and cities with higher levels of digital economy development. This study provides empirical evidence on the benefits of improving the tax business environment and offers insights for coordinated and resilient regional development.
This study examines how climate risk affects corporate borrowing using firm-level climate risk measures constructed from textual analysis of annual reports. Using a sample of Chinese A-share listed firms from 2010 to 2023, we find that climate risk is positively associated with corporate borrowing, especially long-term borrowing. This association is mainly driven by climate transition risk, whereas physical climate risk shows no significant average effect on loan size. The positive climate risk-borrowing association is more pronounced among state-owned enterprises. Moderating-effect tests show that information transparency and institutional investor monitoring strengthen this relationship. Further analyses show that firms exposed to higher climate risk adjust their loan maturity structure toward long-term borrowing. Quantile regression results show that the positive association between climate risk and loan size becomes weaker at higher quantiles of the loan distribution for both comprehensive and transition risks, suggesting a diminishing marginal effect. Overall, the evidence suggests that climate transition risk reshapes firms' borrowing behavior by increasing long-term financing needs during low-carbon transformation. This study extends the climate finance literature by documenting how different dimensions of climate risk affect corporate debt financing in an emerging market where borrowing remains central to external finance.
We ask whether weighting a stock's return path by its current shareholder base reveals information that price-only variables overlook. Using daily volume to reconstruct the current shareholder base, we estimate three intuitive variables from a stock's return path: a cumulative return that reflects the profit or loss of current holders, an average return that captures the typical daily experience of holding the stock, and a scale-invariant gain-loss ratio that records the net fraction of shareholders who are in profit. The shareholder-weighted variables forecast one, three, and six-month returns, subsume capital gains overhang, and complement classic momentum. Evidence from portfolio sorts and cross-sectional regressions remains strong after controlling for standard characteristics and across exchanges, price segments and sub-periods, including the post-2000 era. For practitioners, the new shareholder-weighted variables improve risk-adjusted returns relative to traditional momentum strategies and are easy to implement.
Does herding become more or less persistent as an episode lasts longer? We examine the duration dependence of herding persistence in cryptocurrency markets and investigate whether investor attention moderates this relationship. We identify daily herding episodes using five-minute cryptocurrency returns and estimate discrete-time hazard models of episode termination. Theoretical arguments yield competing predictions. Prolonged collective behavior may reinforce persistence through social confirmation and information cascades, whereas the accumulation and processing of new or conflicting information may encourage investors to reassess their decisions and terminate herding. The results show that longer episodes are associated with lower termination hazards when the moderating role of investor attention is not considered. However, once this moderating role is incorporated, the relationship between duration and persistence weakens and may even reverse. Additional analyses controlling for herding intensity and a range of robustness checks yield results that are generally qualitatively consistent with the main findings. Overall, investor attention acts as an important boundary condition shaping the relationship between episode duration and herding persistence.
Capital-structure models typically evaluate financing choices under a trusted return distribution. This paper develops a robust-solvency framework for settings in which firms are uncertain about that distribution itself. Distributional ambiguity enters a Roy–Telser safety-first constraint rather than the objective function, tightening the set of admissible financing policies. Under the stated monotonicity and boundary conditions, greater ambiguity reduces feasible debt capacity wherever the robust solvency constraint binds. The empirical implementation maps the mechanism into a firm-level levered-return model and recovers an ambiguity-equivalent wedge from observed leverage. For observations with an interior Gaussian benchmark, the wedge is exactly the debt-weight target gap scaled by the ratio of debt cost to asset volatility, making its economic content and measurement boundary explicit. In a panel of United States-listed non-financial firms, the wedge-by-interior differential remains negative across all failure-probability calibrations and under System GMM. The implied interior effect is likewise negative under both fixed effects and System GMM, although imprecisely estimated. Asset volatility predicts lower subsequent leverage, without a significantly stronger association in the interior region. The results establish directional coherence with relative debt-capacity contraction while showing that the accounting-based inversion does not separately identify an ambiguity increment beyond conventional risk and target-gap variation.
Global climate governance has increasingly exposed firms to climate transition risk (CTR), particularly as the depreciation of firms' high‑carbon assets conflicts with the low-carbon transition requirements. Developing a resilient strategy that accounts for environmental policy constraints, technological pathway choices, and market demand is a challenge for firms. Drawing on upper echelons theory and dynamic capability theory, this study uses a dataset of listed companies to examine whether firms with executives possessing environmental backgrounds (EEBs) exhibit different levels of CTR from the perspective of firm dynamic capabilities. Text mining techniques and fixed effects models are used to construct the index and estimate its effects. Results reveal that firms that appoint EEBs exhibit lower market-priced risk exposure. This relationship remains robust after endogeneity treatment and robustness tests. Additionally, when firm power concentration is low, external media attention is high, and the influence of Confucian culture is strong, EEBs are negatively associated with CTR. Mechanism analysis indicates that EEBs are better able to perceive CTR, leverage green development opportunities, and reconstruct green resources, consistent with the role of firm dynamic capabilities in explaining lower CTR. Further analysis demonstrates that a lower CTR is associated with enhanced environmental responsibility and economic performance, promoting sustainable development and increasing business value. This study documents a robust association between EEBs and CTR and advances our understanding of transition risk pricing in financial markets, providing practical insights for firms aiming to improve climate governance and promote an orderly low-carbon transition.
Under monetary policy uncertainty (MPU), accurate inflation forecasting is essential for macroeconomic policymaking and risk management. Yet existing studies have not systematically assessed the transmission mechanisms through which MPU affects inflation forecasts or quantified its marginal predictive value. This paper first introduces MPU shocks within a Dynamic Stochastic General Equilibrium (DSGE) framework to clarify potential channels through which MPU affects inflation and to guide the selection of forecasting variables and model structure. Building on this theoretical foundation, we propose the Dimension-Transition for Trend and Multiple-Attention for Residual Network (DTMR-Net), a deep learning forecasting model that decomposes inflation into a low-frequency trend component and a high-frequency residual component using moving averages and constructs theory-guided neural pathways for each. The trend path employs a multi-scale dimension-transition convolutional network to capture smooth and persistent movements driven by long-term expectations and policy effects. The residual path combines temporal convolutions, Long Short-Term Memory (LSTM), and multiple attention mechanisms to model short-run fluctuations driven by cyclical, seasonal, and policy shocks. Using U.S. data, the empirical results show that DTMR-Net achieves lower forecast errors than advanced models such as Gated Recurrent Unit (GRU), LSTM, Informer, TimesNet, and PatchTST in most settings, with greater stability during high-uncertainty periods such as the post-pandemic recovery and episodes of rising geopolitical risk. Interpretability analysis further indicates that MPU provides a limited but identifiable marginal contribution to short-horizon inflation forecasts, reducing errors by about 2% at the one-month horizon. This effect weakens as the forecast horizon extends and becomes less pronounced in stronger models. Overall, MPU is not a dominant driver of inflation forecasting performance; its value lies mainly in modestly improving short-term forecasts under high uncertainty.
As the digital economy becomes increasingly embedded in public governance and everyday life, digital government development (DG) is transforming public service delivery and potentially reshaping household consumption through changes in information access, transaction costs, credit conditions, and policy expectations. Using household-level panel data from the China Family Panel Studies (CFPS) for 2018–2022, this study employs a two-way fixed-effects model to identify the effect of DG on household consumption expenditure, mediation models to examine underlying mechanisms, quantile regressions to capture distributional heterogeneity, and a moderating-effects model to assess the role of digital finance. The results show that DG significantly increases household consumption expenditure. Mechanism analyses indicate that this effect operates primarily through alleviating household credit constraints, enhancing social interaction, and improving expectations about future economic conditions. The consumption-enhancing effect is stronger among female-headed, urban, and high-income households and households in eastern China, while more developed digital finance further amplifies the effect of DG. The impact also varies across consumption categories, with the magnitude of the estimated effects ranking as enjoyment-oriented consumption, development-oriented consumption, and subsistence consumption. Quantile regression estimates reveal that the positive effect of DG increases along the household consumption distribution, indicating substantial asymmetric effects. These findings highlight the role of digital government in stimulating household consumption and promoting consumption upgrading, while underscoring the importance of coordinated development between digital government and digital finance for more targeted and sustainable consumption policies in China.
Using firm-level data from 2008 to 2022, this paper measures the innovation resilience of Chinese A-share listed firms and, on this basis, employs multiple linear regression models and a double machine learning framework to examine the relationships between multiple dimensions of corporate culture (integrity, innovation, cooperation, and competition), supply chain finance (SCF), and innovation resilience. The results show that integrity culture, innovation culture, cooperative culture, and competitive culture are all significantly positively associated with firms' innovation resilience. Mechanism analysis based on the main fixed-effects specification indicates that the four cultural dimensions are positively related to SCF, and SCF is positively related to innovation resilience. Further heterogeneity analysis reveals that under low market competition, innovation culture, cooperative culture, and competitive culture have positive effects. Integrity culture is insignificant. Under high market competition, integrity culture, innovation culture, and cooperative culture have positive effects. Competitive culture is insignificant. From the perspective of the corporate life cycle, integrity culture, innovation culture, and cooperative culture are significant during the introduction stage. Innovation culture and competitive culture are significant during the growth stage. Only integrity culture remains significant in the maturity and decline stages. In the exit stage, none of the four cultural dimensions is significant. Among firms certified under ISO 9001, only innovation culture has a significant positive effect. Among non-certified firms, all four types of culture exhibit positive effects. The study further finds that extreme events induce fluctuations in innovation resilience, and that the 2008 global financial crisis, the 2015 A-share market crash, and the COVID-19 pandemic are associated with different patterns of changes in innovation resilience. Overall, the findings enrich the typological analysis of the impact of corporate culture on firms' innovation resilience and provide more targeted recommendations for cultivating innovation resilience at the firm level.
The Environmental Kuznets Curve (EKC) proposes an inversed U-shaped relationship between countries' stages of economic development and environmental degradation, indicating that countries in early stages of economic development may experience significant environmental degradation. Green finance forms a relatively new form of finance encompassing various financial instruments and policies which aim to incentivise and reward reductions in environmentally-damaging economic activity. This study examines whether green finance can “flatten” the EKC in developing countries by lowering the peak of environmental degradation and advancing the turning point where growth aligns with environmental improvement. To address this question, we conduct a systematic review of the literature on green finance in developing countries using Web of Science and Scopus, covering studies published from 1 January 2010 to 31 December 2025. Across the evidence base (largely concentrated in China) we find qualified support for the proposition that green finance can contribute to EKC flattening. Green finance is consistently associated with strengthening the technique effect through cleaner technologies, energy efficiency, and green innovation, and more selectively with compositional change via a shift in energy inputs toward renewables. Direct links from green finance to measured environmental outcomes remain comparatively scarce and mixed, reflecting measurement and attribution challenges. Based on the findings, we propose several policy recommendations to help developing countries advance green economic transformation and reduce environmental degradation.
