Since the international financial crisis, how to effectively prevent the spillover risks of financial markets to the real economy and ensure the healthy and stable development of the macroeconomy has become a critical issue of global concern. Against the backdrop of increasing downward pressure on the world economy, growing instability, uncertainty, and unpredictability, and China’s goal of maintaining stable macroeconomic operations, this paper examines the impact of systemic risk on the distribution of shocks to China’s real economy through novel research perspectives, including the testing, simulation, and measurement of the “risk accelerator effect”. The results reveal significant asymmetry in the risk accelerator effect of systemic risk on the real economy. Specifically, an increase in systemic risk exhibits a pronounced risk accelerator effect on future real economic downturns, but no such effect is observed during real economic upswings. Second, this effect demonstrates persistence: a significant risk accelerator effect exists with a one-month lag period, exacerbating the downside risks to the real economy. However, the effect gradually diminishes over time and largely dissipates after 12 months. Third, the risk accelerator effect exhibits time-varying characteristics. By fitting the future distribution of the real economy using a skewed t-distribution and measuring it based on the concept of “entropy,” this study finds that the accelerator effect becomes more pronounced during periods of slowing macroeconomic growth. Furthermore, additional analysis highlights heterogeneity in the risk accelerator effects across different financial sectors, with the real estate and trust sectors identified as major contributors to this effect. Building on these insightful findings, this paper proposes policy recommendations for improving China’s macroprudential regulatory framework, thereby offering both academic value and practical significance.
Against the backdrop of the comprehensive deepening of capital market reform and the implementation of the high-quality development goals of the “New Nine Guidelines”, regulating the market value management of listed firms, curbing irrational speculation, and effectively protecting the rights and interests of small and medium-sized investors have become important components of financial regulation. In modern financial theory, the nominal stock price should not affect a firm’s intrinsic value. However, stock price stratification has, in practice, become an invisible barrier that shapes investor structure and investor behavior. Based on the catering theory of behavioral finance, this paper uses panel data of Chinese A-share listed firms from 2003 to 2024 and introduces dynamic heterogeneous trading behavior indicators, including changes in net buying by small orders and very large orders, to systematically examine how stock price stratification affects investor structure and managers’ catering behavior, thereby influencing firms’ long-term value. The results show that the low-price premium observed in the overall market and investors’ nominal price illusion constitute important mechanisms underlying managers’ catering behavior. Bonus share and capitalization policies generate heterogeneous effects in opposite directions over the short and long terms. In the short term, a decline in nominal price significantly increases net buying by small orders, whereas the response of very large orders is insignificant. In the long run, retail investor funds tend to flow out, while institutional investors exhibit a pattern of long-term increased holdings. In addition, although stock price management aimed at catering can improve short-term performance, it has negative effects on subsequent cumulative abnormal returns, Tobin’s
Climate risk has increasingly become an important factor affecting capital market volatility, and the stock prices of the materials sector, characterized by high energy consumption and high carbon emissions, are particularly sensitive to such risk. This paper constructs climate risk online concern predictors based on Baidu Search Index data, employs multiple machine learning models to examine their predictive value for the realized volatility of the CSI 300 Materials Index, and further introduces the expected utility framework and the SHAP interpretability analysis to evaluate the economic implications and underlying mechanism of the forecasts. The empirical results show that: i) The CROC predictors have pronounced predictive value for the realized volatility of the CSI 300 Materials Index, and their performance is particularly prominent within nonlinear machine learning frameworks, with XGBoost, GBRT, and SVR delivering the best forecasting results; ii) the CROC predictors contain independent and persistent incremental information beyond macroeconomic predictors, and incorporating them into portfolio optimization generates higher investment returns and greater risk-adjusted utility; iii) the CROC predictors generally produce a complementary effect in XGBoost and GBRT, whereas they play a dominant role in SVR, with “energy-saving technology” and “energy conservation” being the two core predictors with the largest contributions. This study not only reveals the predictive value of climate risk for stock market volatility in the materials sector, but also provides a useful theoretical basis for improving the identification and early warning of price volatility risk in high-carbon sectors under climate change shocks.
The monthly business report represents an exploratory initiative in the information disclosure system of China’s capital market. Its disclosure establishes an effective communication bridge between management and both external investors and internal shareholders. Based on a manually collected dataset of monthly business report disclosures by listed companies and using a sample of A-share listed companies from 2010 to 2023, this paper examines the impact of monthly business report disclosure on corporate investment efficiency. The findings indicate that the disclosure of monthly business reports by listed companies enhances investment efficiency, a conclusion that remains robust after a series of robustness tests. Mechanism tests reveal that the disclosure of monthly business reports primarily improves investment efficiency through three channels: alleviating financing constraints, enhancing managerial compensation incentives, and activating managerial learning mechanisms. Heterogeneity analysis shows that the positive effect of monthly business report disclosure is more pronounced in samples with higher disclosure quality, less idiosyncratic information, and greater difficulty in earnings forecasting. Moreover, both voluntary and mandatory disclosures significantly improve investment efficiency. Further research demonstrates that higher-frequency disclosure facilitates the timely correction of corporate underinvestment, improves investment performance, and enhances earnings persistence. To some extent, this study provides valuable empirical evidence for the development of China’s information disclosure system and offers insights for promoting the high-quality development of China’s capital market.
Against the backdrop of global climate governance evolving from nationally determined contributions toward the coordination of cross-border rules, external carbon regulation is increasingly shaping firms’innovation decisions through trade channels. Unlike conventional environmental policies that rely on compliance costs at the implementation stage, the European Union’s carbon border adjustment mechanism (CBAM) — through advance legislative notice, a phased timetable, and a monitoring, reporting and verification (MRV) system — releases a clear and binding institutional signal to firms before any cost materializes. Drawing on signaling theory, this paper treats the 2021 CBAM legislative proposal as an exogenous institutional signal shock and, using data on Chinese A-share listed firms over 2015–2024, identifies policy exposure by matching listed firms with customs trade records according to whether a firm exported CBAM-covered products to the EU prior to the policy. A difference-in-differences approach is then employed to examine the effect of this forward-Looking carbon-regulation shock on firms’ green technology innovation. The results show that CBAM significantly promotes firms’ green technology innovation even before any actual payment obligation arises, and this finding remains robust across a series of robustness checks. The mechanism analysis indicates that the effect does not stem from a broad-based expansion in the overall volume of carbon information disclosure, but operates primarily through an optimization of firms’carbon-disclosure structure: firms structurally reorient their disclosure toward dimensions more directly relevant to cross-border carbon regulation — such as low-carbon transition strategy, value-chain and supply-chain carbon information, and verifiability-related information — thereby enhancing the identifiability, comparability, and verifiability of their green behavior. Heterogeneity analysis further reveals that this promoting effect is more pronounced among firms that are older, larger in pre-treatment size, and located in the eastern region. By adopting the perspective of a “signal-type institutional shock”, this paper extends research on how environmental regulation affects firms’ green innovation and provides new empirical evidence for understanding the micro-economic consequences of cross-border carbon regulation before compliance costs become explicit.
Bitcoin, as a decentralized cryptocurrency, is one of the important innovations in the field of digital currency. A large number of transactions has led to congestion in the Bitcoin system. The congested system not only affects the user’s trading experience, but also affects the user’s trading behavior. Therefore, this study investigates the impact of Bitcoin system congestion on users’ transaction behavior based on the analytical perspective of users’ replace by fee (RBF) strategy. In addition, this study examines the potential mechanisms by which congestion in the Bitcoin system affects users’ adoption of the replace by fee (RBF) strategy. The results show that Bitcoin system congestion can significantly increase the utilization of the replace by fee (RBF) strategy. The mechanism analysis suggests that Bitcoin congestion drives the adoption of replace by fee (RBF) through two channels:transaction cost effect and the risk aversion effect. This paper provides an important reference for understanding users’ trading behavior in the Bitcoin system.
Digital transformation is a systematic endeavor involving two core phases: Generating willingness and executing action. Each phase faces distinct constraints and challenges, necessitating differentiated key drivers. This paper constructs an analytical framework for influencing factors in corporate digital transformation across four dimensions: external environment, organizational resources, organizational capabilities, and executive team characteristics. Machine learning methods and accumulated local effects plots are employed to examine the predictive power, importance ranking, and predictive patterns of different characteristic variables on digital transformation willingness and actions. Findings reveal: 1) Among various machine learning methods, ensemble learning approaches represented by gradient boosting regression trees and random forests demonstrate the strongest interpretability and highest predictive accuracy. 2) Among multidimensional feature variables, peer effects and human capital are crucial factors in stimulating transformation willingness and driving transformation actions, with peer industry effects contributing most significantly to prediction. Concurrently, the amount of government subsidies, investment in digital technology, and absorptive capacity demonstrate strong predictive power for transformation willingness, while innovation capability, industry competitiveness, and firm size exhibit higher predictive contributions for transformation actions. 3) Accumulated local effects plots reveal distinct nonlinear relationships between these characteristic factors and digital transformation. 4) Heterogeneity analysis indicates that key determinants of digital transformation willingness and action vary across enterprises with different property rights structures and specialized expertise qualifications. The findings enrich the existing research on the drivers of digital transformation from a predictive perspective, offering theoretical guidance to help businesses overcome the practical challenges of “unwillingness, hesitation, and lack of capability”in digital transformation, as well as to assist the government in optimizing policies related to digital transformation.
In recent years, frequent international emergencies have significantly increased the volatility of crude oil futures prices, profoundly affecting the stability of global financial markets. Against the backdrop of the launch of China’s crude oil futures, this paper examines the time-varying dynamic relationships among China’s crude oil futures, stock, foreign exchange, and bond markets, aiming to provide insights for maintaining financial market stability and mitigating systemic risk contagion. The findings reveal only short-term interactive effects between China’s crude oil futures market and the financial markets. Specifically, during bull markets, oil market shocks exhibit positive impacts on the stock market, with the opposite holding true during bear markets. Oil market shocks impose negative impulse impacts on the foreign exchange market, while their impacts on the bond market alternate between positive and negative. Furthermore, exchange rate shocks show short-term positive effects on the oil market, but their long-term influence is limited. Stock market shocks exert negative impacts on the oil market during downturns, whereas the bond market has the weakest effect, presenting a negative influence. Finally, within the risk spillover transmission network of “oil market–stock market–foreign exchange market–bond market”, the oil and stock markets act as sources of risk spillovers from return fluctuations, while the foreign exchange and bond markets are net recipients. Notably, during periods of unexpected events, oil market price fluctuations are the most pronounced, constituting a core factor affecting financial market stability. These conclusions remain robust after being tested using advanced methods such as the elastic net shrinkage technique. Based on these findings and considering the development of China’s financial market, this paper offers relevant insights and policy recommendations to mitigate the impact of crude oil futures price volatility and sustain financial market stability.
Improving firms’ environmental performance is central to achieving energy conservation, emission reduction, and green development. Taking China’s emissions trading scheme as a quasi-natural experiment, this paper examines the impact of market-incentive environmental regulation on Chinese manufacturing firms’ environmental performance and its underlying mechanisms. Departing from the existing literature that mainly tests the mechanisms underlying the Porter Hypothesis, this study uses firm-level big data from China to investigate the micro-level mechanism of firms’ emission reduction and efficiency improvement from the perspective of digital product imports. The results show that the emissions trading scheme significantly reduces the pollutant emission intensity of Chinese manufacturing firms. The effect remains robust after a battery of robustness checks and after addressing endogeneity using an instrumental-variable strategy. Mechanism analysis shows that the emissions trading scheme induces firms to increase imports of digital products. The emission-reducing effect and efficiency-enhancing effect of digital products constitute two important channels through which firms’ environmental performance is improved, with digital capital investment serving as the main pathway for emission reduction and efficiency gains. Further analysis shows that the beneficial effect of the emissions trading scheme on firms’ environmental performance is more pronounced among firms facing weaker financing constraints, firms in heavily polluting industries, and firms located in non-resource-based cities or cities with higher levels of science and education. This study provides empirical evidence and policy implications for promoting market-based environmental regulation, releasing institutional benefits, and supporting firms’ digital transformation and green sustainable development.
The accumulation of large-scale medical data and rapid advances in artificial intelligence have created new opportunities for developing intelligent models for early lung cancer diagnosis. However, black-box models often struggle to gain clinical trust, while post hoc interpretability methods frequently suffer from limited fidelity and poor stability. In addition, the severe class imbalance commonly observed in lung cancer data substantially weakens model performance in identifying minority-class samples. To address these issues, this study proposes a cost-sensitive optimal classification tree with hyperplanes (COCT-H) model, which integrates cost-sensitive learning with hyperplane-based partitioning. Built on a mixed-integer optimization framework, the proposed model employs linear hyperplanes for node splitting and directly generates auditable if-then decision rules. It further incorporates an improved big-
This paper develops a group interactive fixed effects model for policy evaluation that captures the time-varying structure of policy impacts as well as regional heterogeneity in responses to common shocks. We establish the consistency and large-sample properties of the proposed estimators and provide practical criteria for selecting the appropriate group level and the number of latent factors. Compared with the traditional two-way fixed effects model, the group interactive fixed effects model effectively controls for time-varying confounders and substantially reduces estimation bias; it also outperforms the standard interactive fixed effects model when the time dimension of the panel is relatively short or the error terms exist serial and cross-sectional correlation. Monte Carlo simulations and empirical evidence show that two-way fixed effects model may misidentify policy effects due to its inability to account for region-specific time-varying structures, whereas the group interactive fixed effects model yields more accurate policy effect estimates. Overall, this framework offers a more interpretable and reliable econometric tool for conducting robust policy evaluation in complex and dynamically evolving environments.
During the 15th Five-Year Plan period, advancing people-centered new urbanization requires ensuring high-quality employment and income growth for rural and migrant populations while eliminating economic barriers to their integration into urban citizenship. This serves as a critical foundation for achieving Chinese-style modernization and common prosperity. Focusing on the dynamic characteristics of land use and population changes in peri-urban areas, this paper analyzes how productive land transfer behaviors, influenced by varying local government development preferences, affect labor markets in peripheral counties. It further examines the inverted U-shaped relationship between the proportion of land allocated for productive use and the wage levels of non-urban populations. Using land market transaction data, this study collects information on the type, location, and area of construction land transfers in peri-urban areas, matching it with CHIP micro-survey data at the county level and regional economic and social data for empirical analysis. The results indicate that this inverted U-shaped relationship— Where wages initially rise and then decline — Is significant among rural residents and migrant workers. The underlying mechanism driving this nonlinear effect includes wage constraints in the productive sector during the low-proportion phase and intensified labor market competition in the high-proportion phase. An excessive preference for industrialization or an insufficient push for industrial development in local governments’ land transfer strategies can lead to lower wage levels for rural and migrant populations. The findings suggest that local governments should adjust future land transfer structures based on regional conditions, aligning urban-rural land planning, population mobility trends, and labor market development to promote long-term urban-rural integration.
