The Third Plenary Session of the 20th Central Committee of the Communist Party of China proposed “advancing the shift of the consumption tax collection to the retail stage and it's steady transfer it to local governments.” As the third-largest tax category in China's fiscal system, the consumption tax plays a significant role. Its revenue surged from 484.7 billion yuan in 1994 to 1611.781 billion yuan in 2023, with its share in total national tax revenue increasing from 4.74% to 8.90%. To strengthen local public finance systems and alleviate local fiscal pressures, the central authority advocates gradually transferring the consumption tax to local governments and shifting its collection point from the production stage to the consumption stage.
This paper examines how relocating the consumption stage affects prices. From a single-industry perspective, the tax base differs between the two stages: the ex-factory price at the production stage versus the final price excluding VAT at the consumption stage. Additional costs in the circulation phase lead to a larger tax base at the consumption stage, potentially raising final consumer prices. However, a general equilibrium perspective reveals three macro channels of influence: (1) mitigating the cost amplification effect of the tax along production chains, (2) directly altering the consumption tax base, (3) affecting tax bases of other levies (e.g. VAT and the urban maintenance and construction tax). Therefore, in the context of “low inflation accompanied by deflationary pressure,” assessing price changes solely from a single-industry perspective is insufficient. A comprehensive evaluation requires a macro perspective that incorporates inter-industry network linkages.
Using a 2020 Social Accounting Matrix(SAM) with disaggregated wholesale and retail sector, we develop input-output price models for consumption tax levied at both the production and consumption stages. These models simulate and quantify the impacts of the collection point shift on prices and welfare in various policy scenarios. Our marginal contributions are threefold: (1) We explore the impact of shifting the consumption tax collection price on prices and welfare, offering a relatively novel research perspective. Most existing studies on consumption tax reform focus on theoretical pathways, with limited quantitative research on the collection point shift. (2) Methodologically, we innovatively incorporate the consumption tax shifting mechanism into the input-output framework, enriching the application of this method. This paper is the first to use input-output analysis to examine the issue of the consumption tax collection point, extending its application in public finance. (3)We offer quantitative estimates of policy impacts across scenarios, providing data-driven insights for reform. Existing discussions on specific pathways for the shift often lack quantitative estimates. This paper calculates specific numerical effects on prices and welfare for different reform options, aiding the reform's implementation.
Our findings indicate: (1) After the shift, price in most industries show a downward trend due to reduced intermediate input costs for enterprises,with a pronounced effect of when taxing only final demand products. (2) Key industries like tobacco, alcohol, and automobiles face upward price pressure due to increased circulation costs, with the tobacco industry experiencing the most significant rise. (3) Taxing only final demand products reduces CPI by 0.55% and PPI by 1.65%, while taxing both intermediate and final products increases CPI by 0.23% and decreases PPI by 0.54%. (4) The reform enhances residents' real purchasing power by lowering industry prices, leading to overall welfare improvement. However, the welfare gains differ substantially between urban and rural residents, highlighting the need to address group inequality. (5) In a gradual reform, piloting the shift in the tobacco or petroleum industries would have a larger price impact, while piloting in the alcohol or automobile industries would have a smaller effect. (6) A higher pass-through rate of indirect taxes correlates with a more pronounced decline in industry prices and greater improvement in residents' welfare.
Based on the conclusions, we proposed three policy implications. First, the pace of the consumption tax reform should be carefully managed to guard against downside price risks. Given that China's CPI has remained below 1% for 27 consecutive months since February 2023, with an average inflation rate of only 0.12%, overly rapid implementation of the collection point shift could exacerbate price declines and harm economic stability. Therefore, the reform should proceed gradually and be implemented in phases. Initially, the tax could be applied to both intermediate inputs and final demand products, later transitioning to taxing only final demand products to avoid sharp price fluctuations. Second, consider launching pilot programs starting in the tobacco and alcohol industries, which have smaller spillover effects on other sectors.. Compared to the petroleum industry, tobacco and alcohol are positioned further downstream in the industrial chain, meaning that the reform would have a smaller price impact on other sectors. Pilots could begin in the alcohol industry, expanding to tobacco after tax procedures are standardized, and subsequently covering petroleum, automobiles, and other areas. Third, strengthen tax administration capacity and optimize tax rates. The dispersion of taxpayers at the consumption stage imposes higher requirements on tax supervision and administration, necessitating enhanced digital and intelligent tax administration to safeguard revenue. The statutory tax rates for key sectors such as high-energy-consumption and luxury goods could be increased initially, followed by gradual rate reductions and simplification of tax brackets. This study also has several limitations. Future research could incorporate differentiated treatments for various tax calculation methods and organizational structures of production at the micro-enterprise or industry level. Furthermore, extending the analysis to a general equilibrium framework could further explore the reform's impact on output and fiscal revenue.

“Implicit guarantees” are often used to study government bailout models for state-owned enterprises (SOEs). Their “implicit” nature carries two potential implications. First, implicit decision-making: governments do not provide guarantees based on publicly announced rules but instead act discretionarily according to circumstances and parameters unknown to the public. Second, implicit behavior: the government's guarantee or bailout actions themselves are sometimes not publicly known. This is particularly evident when bailouts are conducted either through fiscal funds or via internal transfers among affiliated entities, with the latter typically being more opaque. This paper refers to the implicit guarantee model that combines fiscal bailouts and internal transfers as the “dual implicit guarantee” model. It does not aim to propose a new model but to emphasize that implicit guarantee resources inherently originate from dual sources. Building on this, the paper employs theoretical modeling to gain a deeper understanding of the macro-level risks associated with implicit guarantee models from the following two perspectives.
First, the study investigates why implicit guarantees often fail to effectively mitigate the credit risk of SOEs. The capacity for internal transfers as an implicit guarantee offers two benefits to the government or SOE groups: one is the ability to liquidate corporate liquidity in advance (referred to as liquidity value), and the other is the capacity to retain some high-quality assets of a firm without bailing out the entire entity (referred to as industrial value). While this allows the government to preserve more assets without excessive fiscal expenditure, it exacerbates SOE credit risk in two ways. Initially, a weaker internal transfer capacity can alleviate SOE credit risk by easing fiscal distress. Conversely, a stronger internal transfer capacity may incentivize the government to abandon bailing out certain firms altogether, opting instead to transfer their funds and assets to conserve fiscal resources while retaining assets. Furthermore, internal transfer capacity amplifies the impact of fiscal conditions on SOE credit risk and weakens the mitigating effect of government guarantee willingness (i.e., government-firm affiliation). This occurs because the liquidity value offered by internal transfers becomes more attractive under poor fiscal conditions (governments facing fiscal strain are more inclined to transfer rather than bail out), while the industrial value is more appealing when guarantee willingness is high (governments with stronger affiliations can preserve more assets of defaulting firms through transfers).
The above conclusions partially explain the real-world phenomenon of SOE bond defaults despite the presence of implicit guarantees. By dissecting the liquidity and industrial values of internal transfers and their associated indirect risk effects, this study further advances the understanding within internal capital market theory regarding the impact of internal transfer behaviors.
Second, using a Bayesian model, the paper illustrates how implicit guarantees can also induce strong risk spillover effects; that is, the public default of a single firm may lead to a decline in the financing capabilities of numerous other SOEs. Primarily, since implicit guarantee decisions lack full transparency to the market, an unexpected public default leads investors to suspect causes such as insufficient government guarantee willingness or poor fiscal conditions. This information effect inherently affects the financing of other SOEs in the same region. Moreover, under the dual implicit guarantee model that accounts for internal transfer behavior, investors may further suspect that defaults are likely due to the proactive transfer of high-quality and liquid assets away from the firm. This further strengthens the risk spillover effect because the liquidation value of such defaulting firms is lower. Additionally, since both poor fiscal conditions and strong guarantee willingness incentivize internal transfers, trust in fiscal conditions can mitigate risk spillover, whereas trust in government-firm affiliation may instead exacerbate it.
These conclusions provide a more comprehensive explanation for why SOE bond defaults easily trigger risk spillover in reality and further analyze the potential macro-level risk implications of the tunneling behavior discussed in internal capital market theory.
Beyond deepening the understanding of the underlying logic of macro-level risks associated with implicit guarantees, this study offers direct policy insights in the following aspects. First, it demonstrates that higher asset liquidity (a key factor influencing transfer capacity), while helping firms better cope with forced default risk, may conversely increase the active default risk of group enterprises. Second, it shows that enhancing government-firm affiliation or market belief in it may not prevent and could even worsen macro-level risks. Furthermore, the modeling approach in this paper can be extended to analyze financing and guarantee financing issues for all group enterprises.

To strengthen the institutional development of the capital market, it is essential to enhance external supervision mechanisms. In light of persistent accounting irregularities, a thorough investigation into how accounting supervision can be deployed more effectively is warranted. While prior research has largely examined listed firms, focusing on how regulatory scrutiny influences inspected entities and generates corrective spillovers across the market, this study shifts attention to the bond market. Specifically, we explore the impact of the Ministry of Finance (MOF) random inspections of accounting information quality on both listed and non-listed bond-issuing firms.
Credit ratings provided by rating agencies constitute a critical information source in the bond market. Higher-quality accounting information generally leads to more favorable credit ratings, which, in turn, reduces the yield required by bond investors. The MOF random inspections of accounting information quality can thus be regarded as a significant form of external supervision in the bond market. A key question, therefore, is whether this supervision effectively complements the existing regulatory framework of bonds.
This study examines two channels through which MOF random inspections of accounting information quality may affect corporate credit ratings: an information effect and a catering effect. On one hand, the inspection process supplies credible new information to rating agencies, thereby improving assessment accuracy, particularly for firms with higher accounting transparency. On the other hand, rating agencies may engage in rating catering, strategically inflating ratings to accommodate inspected issuers, potentially undermining rating objectivity.
To estimate the effect of MOF random inspections of accounting information quality on bond issuers' credit ratings, we employ a Difference-in-Differences (DID) design, supplemented by a combined Propensity Score Matching and DID (PSM-DID) approach. Our sample covers both listed and non-listed bond-issuing firms from 2008 to 2023. Based on official MOF random inspections announcements of accounting information quality, we identify 80 bond issuers that underwent random inspections, including 46 listed and 34 non-listed firms. Firms selected for inspection constitute the treatment group, whereas those not inspected form the control group.
The results indicate that credit ratings improve significantly following the announcement of MOF random inspections. This effect is stronger among issuers with higher accounting transparency, while we find no statistically significant evidence of rating catering. Moreover, the effect is significant only in the subsamples of firms without tax violations, firms that have completed required rectifications, listed firms, and issuers experiencing financial distress. Furthermore, MOF random inspections of accounting information quality are associated with reduced rating divergence across agencies. Taken together, these findings suggest that MOF random inspections exert a meaningful governance effect on bond-issuing firms. This study offers practical insights for improving both credit rating practices in capital markets and the effectiveness of government accounting supervision.
The policy implications of this study are threefold. First, continue advancing the core tasks of government accounting supervision to mitigate debt risks. Conduct targeted inspections of enterprises in financial distress. Establish threshold-based alert mechanisms within accounting supervision, calibrated according to industry-specific characteristics and credit ratings of bond-issuing firms. Second, a transparent platform should be developed to enhance the market's information environment. The platform would mandatorily disclose corporate remedial actions, with its follow-up mechanisms particularly focused on monitoring non-listed issuers. Third, a horizontally coordinated supervision framework should be established by enhancing information sharing between the MOF and financial intermediaries. This would provide robust support for the accurate identification of credit risks among bond issuers.
This paper highlights the beneficial effects of government accounting supervision on credit rating agencies and illustrates the governance efficacy of this supervision at the bond-issuer level. Future research could examine the synergistic effects and underlying mechanisms of accounting supervision in combination with other regulatory instruments, such as audit supervision.

A reasonable net interest margin (NIM) is a crucial guarantee for the stable operation of commercial banks, especially for rural financial institutions (RFIs). However, data show that the NIM of RFIs has narrowed to a historical low level, with the NIM of some rural commercial banks even falling below 1%, increasingly exposing potential risks to their safety. Furthermore, RFIs face limited channels for capital replenishment. Their core business consists of rural deposit and loan services, and are prohibited from cross-regional operations, making the deposit-loan spread their primary and single source of profit. Therefore, NIM compression may impact the operations of RFIs more significantly.
Based on this background, this paper first compares the typical differences in NIM compression between RFIs and large and medium-Sized banks (LMBs), discussing whether and through what inherent mechanisms NIM compression affects the risk of RFIs compared to LMBs. Secondly, it verifies effective pathways for RFIs to cope with the risks arising from NIM compression. This paper provides empirical evidence for the necessity of maintaining reasonable NIM to prevent financial risks in commercial banks and offers insights for preventing risks in RFIs.
The conclusions of this paper are as follows: NIM compression significantly increases the risk of RFIs. The mechanism lies in the fact that NIM compression increases pressure on the asset side and restricts the equity side, leading to a decline in the risk disposal capacity and risk resilience of RFIs. Further analysis finds that promoting diversified operations to increase non-interest income and controlling operational management costs are effective pathways for RFIs to alleviate the impact of NIM compression. Adjusting the loan structure based on customers does not help them effectively cope with the risk; instead, it may squeeze out personal loans and undermine their core mission of serving small entities.
The contributions of this paper are mainly reflected in the following aspects: First, it examines the risks of RFIs based on the typical fact of NIM compression, providing an important supplement to the existing literature on commercial bank risks. RFIs have distinct particularities, yet previous research has focused more on LMBs. Investigating the impact of NIM compression on the risk of RFIs adds an important perspective to current financial risk research. Second, it analyzes the inherent mechanism through which NIM compression increases the risk of RFIs from the perspectives of asset-side pressure, equity-side constraints, and liability-side adaptation. This paper argues that NIM compression leaves RFIs without sufficient provision coverage and loss reserves to guard against bad debt impacts, and also restricts their internal capital replenishment, significantly reducing their risk resilience. However, NIM compression does not drive RFIs to pursue high-risk investment returns. Third, it provides references for the operational strategies and pathways of RFIs in responding to NIM compression risks. This paper finds that diversified operations and reducing operational and management costs are effective pathways for RFIs to cope with NIM compression risks. However, adjusting the loan structure based on customers does not effectively address NIM compression; instead, it causes RFIs to deviate from their positioning of supporting small entities.
Based on the above empirical conclusions, to mitigate the risks associated with NIM compression in RFIs, this paper provides the following implications: First, guide and assist RFIs in maintaining reasonable NIM through structural monetary policy tools. The guiding role of structural monetary policy can be strengthened by flexibly using tools such as relending programs supporting agriculture and small businesses, and rediscounting to precisely reduce the liability costs of RFIs. Second, implement differentiated measures for preventing and resolving risks from NIM compression. RFIs can be appropriately encouraged to replenish capital through market-based methods in multiple channels, such as supporting the issuance of secondary capital bonds and special bonds, or conducting compliant mergers, acquisitions, and equity investments in accordance with the law. Third, strengthen business capacity and prevent RFIs from deviating from their focus on agriculture and small entities through enhanced supervision. Prudently guide RFIs to adjust their loan businesses and explore business growth points among long-tail customers; accelerate the transformation of intermediary businesses to broaden non-interest income channels.

According to the WIPO 2025 report, China continues to lead the world in patent volume, accounting for nearly 49.5% of global applications. However, international patent filings remain insufficient, and the relatively low proportion of invention patents highlights a need for improvement. Have innovation incentive policies acted as a catalyst for the phenomenon of “high quantity, low quality” in corporate innovation？ Addressing this question will not only help clarify the intrinsic logic behind corporate innovation behavior but also provide a micro-foundation for implementing national innovation strategies and offering practical insights for resolving innovation dilemmas. Adopting a dual perspective of innovation quantity and quality, this study examines whether and how R&D subsidies influence corporate innovation. Theoretically, information asymmetry between external investors and firms, compounded by the high-risk nature of corporate innovation, exacerbates the difficulty for external investors to assess corporate innovation performance. In this context, R&D subsidies signal strong corporate innovation capability and serve as a valuable reference point for external investor attention, thereby increasing the pressure of innovation expectations faced by management. Given that innovation quantity is more readily observable to outsiders, management prioritizes the expansion of innovation quantity over quality to sustain favorable stock market performance.
Empirical results based on data from non-financial listed companies from 2007 to 2023 indicate that R&D subsidies significantly increase the quantity of corporate innovation without improving its quality, thereby exacerbating the “high quantity, low quality” dilemma. Specifically, this effect is contingent upon whether a firm receives such subsidies, which validates the external investor attention hypothesis. Subsequent heterogeneity tests reveal that the characteristics of government R&D projects and alternative information channels exert differentiated impacts on the innovation effect of subsidies. Collectively, these results demonstrate that the observed effect weakens when external investor attention regarding subsidies is dispersed. Further mechanism analysis finds that once a firm receives an R&D subsidy, the positive impact of innovation quantity on excess stock returns is significantly attenuated; that is, obtaining subsidies raises external investors' expectations regarding corporate innovation capabilities.
Based on the theoretical analysis and empirical findings, this study proposes the following policy recommendations: First, cultivate a favorable and systematic environment for innovation policy. The findings indicate that the external investor attention effect induced by government R&D subsidies exerts a detrimental impact on the quality of corporate innovation. This implies that while industrial policies aim to incentivize an increase in the quantity of innovation, they must prioritize supporting the quality of innovation and encourage firms to adhere to long-termism in their innovative endeavors. Second, foster an equitable and stable external innovation environment. R&D subsidies concentrated on specific market participants exacerbate the negative impact of investor attention. A lack of equity distorts corporate behavior, disrupts technology diffusion and competition, and amplifies uncertainty within the innovation environment. Consequently, subsidies must prioritize equity and enhance the predictability of their acquisition. Third, information asymmetry between the capital market and firms may distort the strategic direction of corporate innovation. Therefore, it is essential to improve capital market institutions and enhance the quality of information disclosure. It would help ensure that innovation quality is fully incorporated into external investors' decision-making.
The contributions of this study are in the following aspects: First, it analyzes the innovation effects of government R&D subsidies through the lens of external investor attention. While existing literature has primarily focused on financing constraint effects and strategic innovation behaviors aimed at securing subsidies, this paper attempts to further elucidate the association between R&D subsidies and the “high quantity, low quality” phenomenon in corporate innovation. Second, this study extends the application of external investor attention within the context of corporate investment and financing decisions. Existing corporate finance literature regarding external investor attention has largely focused on examining market reactions. Third, this study uncovers the interactive mechanisms among government R&D subsidies, external investor attention, and corporate innovation, thereby enriching the literature on the relationship between government subsidies and external investors. By establishing the logical chain of “R&D Subsidies-External Investor Attention-Expectation Pressure Effect-Corporate Innovation”,this paper demonstrates how R&D subsidies influence corporate innovation decisions through the channel of external investor attention.

In the context of national efforts to promote talent development and modernize tax governance, tax authorities, highly specialized public sector agencies with broad service coverage, have long prioritized the development of strong tax audit teams. Through initiatives such as talent pool development, professional competitions, and certification programs, these authorities have sought to enhance team competence, particularly by recruiting certified tax agents and accountants to bolster audit capacity. However, the impact of such talent investments on tax governance outcomes and their underlying mechanisms remains insufficiently explored. Existing literature identifies gaps in the analysis of the economic value of human capital in the public sector. Using national tax survey data from 2009 to 2016, this paper systematically investigates the impact of personnel investment in tax audit departments on the tax decision-making behavior of micro-enterprises, with the goal of addressing gaps in research regarding the economic effects of public sector human capital and the factors influencing corporate tax avoidance. The findings contribute significant theoretical insights and practical recommendations for advancing the modernization of tax governance.
The study demonstrates that investment in specialized tax audit personnel significantly enhances corporate tax compliance. This effect results from enhancing the effectiveness of tax audits, particularly by increasing audit volume, improving case selection accuracy, and raising case resolution rates. These improvements in audit processes strengthen deterrence and enhance both efficiency and precision, thereby raising the cost of tax avoidance for enterprises. As a result, companies are more likely to comply with tax regulations, weighing the marginal benefits of tax avoidance against penalty costs, which ultimately leads to greater tax compliance. This baseline result holds even after a series of robustness tests, ensuring the reliability of the results. Additionally, by addressing endogeneity, the study mitigates potential issues with causal identification, further strengthening the validity of the results.
Heterogeneity analysis uncovers significant variations in the effectiveness of this governance strategy across different regions and firm types. At the regional level, the positive effects of talent investment are more pronounced in areas with greater public awareness of tax auditors and higher levels of digitalization in tax governance. The former amplifies deterrence signals by enhancing firms' perceptions of audit capacity, while the latter generates a multiplier effect through the integration of “talent and technology”. At the enterprise level, the impact is more significant for larger firms or those with a more diverse range of tax obligations. These companies, due to their higher tax complexity and more concealed tax avoidance behaviors, are more likely to benefit from the constraints and guidance provided by specialized audits.
Further analysis demonstrates that investment in tax audit professionals provides additional governance benefits. On the one hand, it promotes the effective utilization of available tax incentives by firms, ensuring proper policy implementation and preventing firms from missing benefits due to misunderstandings or insufficient reporting. On the other hand, it enhances tax burden fairness among firms, fostering a more equitable tax environment and mitigating market imbalances caused by unequal tax burdens, thereby maintaining market stability. Notably, empirical results show no significant long-term negative impact on corporate performance following a reduction in tax avoidance due to specialized audits. In contrast, these audits incentivize firms to increase innovation investment and enhance internal control, thereby indirectly boosting their performance. This suggests that talent investment primarily results in efficiency gains rather than merely increasing the tax burden on businesses.
The academic contributions of this paper are threefold. First, it broadens the traditional framework for analyzing corporate human capital by focusing on public sector human capital, emphasizing its roles in public goods provision, externalities, and institutional execution. Second, it fills a gap in the existing literature concerning the impact of the qualifications of tax policy implementers (i.e., tax auditors) on corporate tax compliance, expanding the study of factors influencing corporate tax avoidance and clarifying the core role of human capital in tax administration. Third, it clearly elucidates the mechanisms and boundary conditions of tax audit talent investment, and provides practical insights.

As a critical component of the innovation system, collaborative innovation serves as a common approach to improving innovation efficiency and driving economic growth. However, this topic has received relatively limited attention in the academic literature. This study develops a two-sided random search-and-matching model with firm heterogeneity to examine the determinants of innovation-output quality and partner selection in collaborative R&D, focusing on search frictions and equity linkages. Based on the model, three hypotheses are proposed. First, due to the presence of search frictions, the “productivity threshold” required for an innovating firm to collaborate with a non-equity-affiliated firm is higher than that for collaborating with an equity-affiliated partner. This selection mechanism implies that although innovating firms engage in a larger number of collaborative R&D projects with equity-affiliated firms, the innovation outcomes generated from collaborations with non-equity-affiliated partners tend to be of higher quality. Second, regardless of whether the partner is equity-affiliated, innovating firms consistently select collaborators with similar innovation capabilities. Third, innovating firms prefer non-equity-affiliated partners of comparable firm size, whereas firm size does not influence partner choice when collaboration occurs within equity-affiliated relationships.
Empirically, the study tests these hypotheses using collaborative patent data from Chinese industrial enterprises. Equity affiliations are identified via the “relationship mapping” function of Qichacha (a Chinese enterprise information platform); firms' innovation capabilities are measured by their patent application counts in the collaboration field; and firm size is proxied by the percentile rank of their main business revenue within the industry. The empirical results strongly support the theoretical predictions concerning the role of search frictions in shaping partner selection and collaborative outcomes.
Furthermore, the analysis extends the research framework by examining the economic consequences of collaborative innovation from the perspective of firms' operational decisions. The findings reveal that collaborative R&D generates substantial knowledge spillovers, guiding firms' subsequent innovation trajectories toward the technological fields associated with their joint patents. At the same time, collaborative innovation enhances firms' innovation performance, production efficiency, and operating outcomes. After participating in collaborative projects, firms exhibit higher total patent application counts, increased total factor productivity, and larger main business revenue compared with independent-innovation firms in the control group. Finally, the paper investigates the sources of post-collaboration growth in firms' patenting activities. The results show that the improvement in innovation performance is primarily driven by an increase in jointly filed patents, while the number of independently filed patents declines. This pattern is particularly pronounced in collaborations with equity-affiliated firms.
The potential contributions of this study are as follows. From a research perspective, this paper is the first to introduce the concept of equity affiliation into the analysis of inter-firm patent collaboration. By integrating a theoretical model with empirical evidence, it offers an in-depth examination of search-matching patterns in collaborative innovation, the quality of collaborative outcomes, and the extent to which such collaborations achieve the social optimum. Regarding the theoretical model, this paper embeds equity affiliation into a random search-and-matching framework and demonstrates how such affiliations generate heterogeneous sorting patterns across firms with respect to technological accumulation and firm size. The model further shows that these observed collaboration patterns are endogenously driven by firms' bargaining power and search costs, thereby offering a novel theoretical explanation for the differential probabilities of collaboration and heterogeneous quality of innovation outcomes between equity-affiliated and non-affiliated firms. Empirically, this study is the first to investigate, using Chinese micro-level enterprise data, how participation in collaborative innovation projects influences firms' subsequent innovation trajectories and overall performance. It also innovatively measures knowledge spillovers arising from collaboration by analyzing patent abstract texts. This approach provides stronger and more credible empirical evidence on whether collaborative innovation truly generates knowledge spillovers that enhance firms' innovative capabilities, thereby deepening our understanding of the strategic importance of collaboration in corporate innovation and business operations.

In the context of intensifying climate risks and China's continuous efforts toward its “dual-carbon” goals (carbon peaking and carbon neutrality), improving firms' accuracy in assessing climate risk and strengthening climate governance outcomes have become a central question in climate finance and corporate governance research. Prior work largely examines downstream outcomes, such as environmental disclosure, carbon performance, or green investment, to understand how governance structures and climate policies shape corporate behavior, but it has paid comparatively less attention to a more fundamental element of climate governance: whether decision-makers' subjective perceptions of climate risk align with the firm's objective risk exposure. When firms systematically misperceive climate risk, strengthened external regulation and improved internal governance may still fail to translate into effective climate action. Accordingly, identifying the drivers of Climate Risk Perception Bias and evaluating feasible corrective channels is important both theoretically and practically for improving corporate climate-governance effectiveness and enhancing the stability of the climate-finance system.
Building on this premise, we examine how corporate governance quality shapes firms' Climate Risk Perception Bias. Drawing on polycentric collaborative governance, upper echelons theory, and the dynamic capabilities view, we develop a “government-firm dual-correction” framework that integrates internal governance mechanisms with the external policy environment. The core premise is that Climate Risk Perception Bias is not merely an information-deficit problem. Rather, it emerges from the joint influence of governance complexity, managers' cognitive capacity, and external institutional signals. Therefore, correcting this bias requires firms to recalibrate internal cognitive anchors and strengthen resource-reconfiguration (adaptation) capabilities, while governments provide stable and credible external constraints and incentives through sustained policy attention and institutional investment.
Methodologically, we move beyond the conventional single-data-source measurement paradigm and integrate machine learning, textual analysis, and panel econometric techniques to systematically quantify firms' Climate Risk Perception Bias. First, to measure objective climate risk, we use disaster-monitoring records from the National Meteorological Information Center and incorporate regional economic resilience and population vulnerability to construct a dynamic, comparable regional climate-risk index via a LightGBM model. Second, to capture subjective risk perception, we develop a climate-risk dictionary tailored to China's institutional context and apply it to firms' annual reports to quantify decision-makers' attention to climate risk. We then define Climate Risk Perception Bias as the systematic deviation between a firm's perceived climate risk and its objectively measured exposure. In our empirical design, we estimate two-way fixed-effects models and test for nonlinearity by including a quadratic term of governance quality. We further address potential endogeneity using instrumental variable estimation and system GMM.
We examine Chinese A-share listed firms on the Shanghai and Shenzhen exchanges from 2007 to 2022. We merge firm-level governance, financial, and macroeconomic data from CSMAR and WIND with annual-report texts and meteorological disaster records to construct an integrated dataset that links micro-level corporate behavior to the broader institutional environment. We construct a composite measure of corporate governance quality using principal component analysis (PCA) on multidimensional proxies capturing managerial incentives, board monitoring, and ownership structure, thereby providing a comprehensive assessment of firms' overall governance quality.
Our empirical results indicate a statistically significant inverted U-shaped relationship between corporate governance quality and Climate Risk Perception Bias. Specifically, in the early stage of governance improvement, the rapid accumulation of governance instruments and the rising complexity of information processing can increase managers' cognitive load, thereby amplifying firms' misperceptions of climate risk. Once governance quality surpasses a critical threshold, firms enhance their ability to identify and respond to climate risk through organizational learning and strategic allocation of slack resources, which significantly attenuates Climate Risk Perception Bias. This evidence challenges the prevailing linear premise in the literature that better governance monotonically improves risk management, and instead highlights the stage-dependent and nonlinear role of corporate governance in shaping climate-risk cognition.
Mechanism analyses further suggest that disclosure quality and commercial credit constitute two key channels through which corporate governance shapes Climate Risk Perception Bias. In the early phase of governance improvement, strategically ambiguous disclosure and contractions in commercial credit may reinforce managerial short-termism, thereby exacerbating misperceptions of climate risk. By contrast, at mature governance stages, high-quality and forward-looking climate disclosures, together with improvements in the supply-chain credit environment, enhance cognitive transparency and financial flexibility, effectively recalibrating managers' subjective assessments of climate risk. We also find that government-firm collaborative governance exerts a pronounced amplifying effect in correcting climate-risk misperceptions. Coordinated increases in local governments' policy focus on climate issues and ecological investment can substantially boost the efficiency of corporate governance mechanisms in mitigating Climate Risk Perception Bias, through enhanced credibility of regulatory signals and lower transition costs.
Based on these findings, we derive three policy implications. First, regulators should explicitly target the correction of Climate Risk Perception Bias by further strengthening climate-risk disclosure requirements and improving the comparability, forward-looking content, and verifiability of disclosures. Second, differentiated yet stable local green-regulation arrangements can help reduce the noise introduced by institutional complexity and policy uncertainty in firms' climate-risk cognition. Third, firms should strengthen climate-risk-oriented executive selection and incentive schemes within corporate governance to reduce the likelihood of climate-risk misperceptions at their cognitive source.
This study makes three primary contributions. First, adopting a cognition-based perspective, we position Climate Risk Perception Bias as a key mediating mechanism linking corporate governance to climate-governance performance, thereby extending the analytical frontier at the intersection of climate finance and corporate governance. Second, we integrate meteorological disaster data with text-mining techniques to develop a replicable firm-level measurement framework for Climate Risk Perception Bias. Third, through the lens of government-firm collaborative governance, we systematically document the complementarities between regulatory instruments and corporate governance mechanisms in correcting climate-risk cognition.
Future research can extend this work in three directions. First, a dynamic perspective could model how the governance threshold endogenously adjusts as the policy environment evolves. Second, incorporating industry heterogeneity and cross-country samples would allow comparative tests of how Climate Risk Perception Bias arises and is corrected across institutional contexts. Third, future studies could unpack governance components to examine the differential roles of institutional investor types, board structures, and incentive schemes in shaping climate-risk cognition.

In the capital markets, information intermediaries (such as securities analysts and financial media) play a critical role in disseminating, interpreting, and verifying information, which significantly influences market information efficiency and capital allocation efficiency. However, with the rapid increase in listed companies and the explosive growth of information, the attention resources of information intermediaries have become increasingly scarce. This has led to some companies being gradually “marginalized” by the market due to a lack of attention. This paper defines such companies as “market-marginalized companies”, referring to those that are neither tracked by analysts nor covered by mainstream media. Statistics indicate that approximately 15.29% of observed samples fall into a state of information intermediary absence.
The existence of market-marginalized companies significantly undermines the effectiveness of the capital market's pricing mechanism, causing the potential investment value of these companies to remain insufficiently reflected. In the long term, the continued distortion of the pricing mechanism not only weakens investors' ability to price risk, but also erodes their trust in the market, ultimately harming the core function of the capital market in supporting economic growth. Particularly in the context of the registration-based IPO system reform, when the market expansion rate exceeds the service capacity of information intermediaries, the new generation of growth-oriented enterprises may fall into pricing errors due to their inability to break through the attention threshold, thus falling into a vicious cycle of “listing-neglect-delisting”, which ultimately hinders the healthy metabolism of the capital market. Therefore, systematically examining the stock price behavior and governance pathways of marginalized companies in the market holds significant theoretical and practical implications for enhancing the overall efficiency of capital markets, protecting investor interests, and promoting high-quality financial development.
Using a sample of Chinese A-share listed companies from 2005 to 2021, this paper innovatively focuses on “marginalized companies” that are neglected by information intermediaries. It explores the impact of information intermediary absence on post-earnings announcement drift (PEAD) and further investigates how internal and external governance mechanisms can effectively mitigate this effect. The empirical tests reveal that, compared to companies with comprehensive information intermediary coverage, market-marginalized companies exhibit more serious post-earnings announcement drift, strongly corroborating the important function of information intermediaries in price discovery. Governance effect tests indicate that a high-quality information environment and effective corporate governance mechanisms significantly mitigate the magnitude of PEAD. These findings suggest that enhancing information transparency and optimizing corporate governance structures can partially offset the adverse effects of information intermediary gaps, reducing irrational stock price volatility caused by information asymmetry. This provides actionable insights for firms seeking to address the challenges posed by attention scarcity from information intermediaries.
This study makes three primary contributions: First, this study systematically identifies and reveals the existence and market characteristics of the long-neglected group of “market-marginalized companies”, broadening the perspective of capital market research. Unlike prior studies that focus predominantly on prominent key enterprises, this paper delves into the potential risks and behavioral traits of these “marginalized companies”. This not only assists investors in identifying market blind spots but also offers new insights for regulators to optimize policy frameworks. Second, this paper innovatively introduces the “market-marginalized companies” perspective to expand the study of PEAD. Unlike existing literature that primarily examines PEAD phenomena at the aggregate market level, this study emphasizes corporate heterogeneity within external information environments. It analyzes the phenomenon based on the practical constraint of attention faced by market information intermediaries, offering a novel theoretical explanation for the formation mechanism of PEAD anomalies. Third, this paper proposes actionable governance strategies to address the issue of information intermediary absence, offering feasible pathways for market optimization and policy formulation.
The policy implications are as follows: First, enhance the role of information intermediaries by incentivizing broader and deeper coverage of neglected firms and supporting their use of big data and AI to improve information processing and price discovery. Second, actively promote the standardization and development of interactive online platforms to establish real-time, open, and trustworthy communication channels. This will compensate for the coverage gaps of traditional intermediaries, thereby improving market transparency and pricing efficiency. Third, guide companies to refine their internal governance structures. Relevant institutions should promote best governance practices and reinforce external oversight to ensure sound governance, fostering mutual prosperity between the real economy and capital markets.

Environmental, Social, and Governance (ESG) investment is increasingly favored in China's public mutual fund industry. Prior to 2008, this field was still in a nascent or even blank state, but by the end of 2023, China's green loan balance had reached as high as 30.08 trillion RMB, marking significant progress in China's financial market in the area of sustainable investment. This growth aligns with national priorities like the carbon peaking and carbon neutrality goals and sustainable development, yet it faces challenges such as inconsistent information disclosure, non-uniform standards, and greenwashing risks. Currently, academic and practical interest have surged, with studies highlighting ESG's role in enhancing fund performance, attracting inflows, and mitigating risks during market uncertainties (e.g., Lins et al., 2017; Hartzmark and Sussman, 2019). However, while overall ESG impacts are well-documented, the independent role of the governance (G) dimension remains underexplored, particularly in how it influences investor behavior. Good governance is foundational to corporate sustainability, which can boost firm value, performance, and improve long-term stability (Gompers et al., 2003; John et al., 2008). Investors, driven by social preferences and non-financial motives, may favor funds emphasizing governance to align with ethical norms, even at potential financial return trade-offs (Riedl and Smeets, 2017; Pastor et al., 2021). Drawing on a comprehensive dataset of MSCI ESG and China's mutual funds from 2007 to 2023, this paper addresses this gap by constructing a novel fund-level governance score, which is based on MSCI governance pillar scores weighted by fund holdings, and investigates its link to fund flows amid rising ESG adoption and investor sophistication. We have the following main results.
First, this paper reveals a significant “corporate governance premium” flow phenomenon in the China's fund market, and this phenomenon is more reasonably explained by investors' social preference theory. Specifically, investors are more inclined to allocate capital to funds within their portfolios that actively practice good governance principles. This indicates that, driven by social preferences and sustainable development concepts, Chinese fund investors incorporate the governance level of portfolio companies into their decision-making framework when selecting funds, thereby forming capital flow characteristics based on governance quality. Second, investors' positive responses to corporate governance exhibit significant asymmetric characteristics, primarily manifested in buying rather than selling. Further analysis shows governance-driven fund flows differ by fund size and star ratings. Smaller funds and highly rated funds experience more significant effects. Third, this paper further strengthens the reliability of these conclusions through various robustness tests, including instrumental variable (IV) analysis and propensity score matching (PSM).
This study contributes threefold to the literature. First, this paper extends the study of the information sets utilized by equity fund investors in their selection processes (Berk and Van Binsbergen, 2016; Barber et al., 2016). Second, it advances understanding of mutual fund dynamics by highlighting buy-sell asymmetries in governance sensitivity, which contributes to behavioral finance in decision contexts. Third, it illuminates non-financial motives of fund investors in China's market, where investors trade off returns for sustainability, which aligns with global evidence (Hartzmark and Sussman, 2019; Pastor et al., 2021).
In practice, our findings suggest fund managers may focus on governance transparency to get more societal preference-driven inflows, especially for smaller or higher-rated products. Regulators may tackle greenwashing by encouraging standardized ratings and ESG disclosure. The result will assist with the ESG market grow towards “dual carbon” goals.
In conclusion, this paper offers insights into fund investor behavior that are centered on governance. It illustrates how, in China's evolving ESG landscape, social and sustainability preferences impact the mutual fund market. It provides useful recommendations for enhancing fund designs and regulatory procedures by illustrating how governance drives asymmetric, heterogeneous flows, which support China's sustainable financial development.

Passive investing has expanded rapidly across global markets, and China has undergone a similar transition. By 2023, the size of domestic index funds had grown nearly eightfold since 2010, with the number of funds rising from fewer than one hundred to more than one thousand. In the third quarter of 2024, equity holdings of index funds exceeded those of active funds for the first time. In this context, understanding whether index funds amplify or stabilize stock-price fluctuations in China has become a central question for both regulators and market participants.
Theoretically, index funds may influence price stability through competing channels. On the one hand, inclusion in major indices increases visibility and reduces the sensitivity of prices to short-term noise. On the other hand, as index-fund scale expands, large passive flows may reduce free-floating shares and create mechanical trading pressure during subscription and redemption cycles, potentially amplifying short-term volatility. Determining which effect dominates requires credible empirical identification.
This study exploits the institutional features of the CSI 300 and CSI 500 indices to construct a quasi-natural experiment. Index constituents are ranked by average total market capitalization, with semiannual adjustments that shift firms across a well-defined threshold. Near this cutoff, firms in the lower end of the CSI 300 and firms in the upper end of the CSI 500 share similar fundamentals, yet their index weights and index-fund ownership differ sharply due to index-tracking rules. This discontinuity provides a clean source of variation that allows us to identify the causal impact of index-fund ownership on price stability.
Using data from 2013 to 2020, we implement a fuzzy regression discontinuity design (RDD). The first-stage results reveal a sizable jump in index-fund ownership at the cutoff, with CSI 300 tail firms receiving substantially higher passive holdings. This “tail-head” pattern reflects China's market characteristics, including the dominance of large-cap index products and the smaller overall market scale. The second-stage estimates show that higher index-fund ownership significantly enhances price stability, reducing volatility, idiosyncratic volatility, return amplitude, downside tail risk, and limit-down events, while leaving limit-up events unchanged. These patterns indicate an asymmetric stabilizing effect that mainly suppresses extreme negative movements.
To uncover the mechanisms behind these patterns, we examine firms' information disclosure, managerial tone, earnings quality, and equity-pledging behavior. Firms with higher index-fund ownership exhibit more cautious narrative tone in annual reports, higher reporting quality, and lower reliance on equity pledging, suggesting that index funds improve internal governance and discipline managerial behavior. These improvements reduce firms' sensitivity to short-term sentiment shocks and strengthen the anchoring role of fundamentals in price formation. Additional analyses rule out alternative channels such as liquidity effects and active-fund rebalancing.
Further results show that the stabilizing effect persists across different market conditions. Index funds reduce price swings in both bull and bear markets, although the channels differ. In bull markets, the effect is more apparent in limiting excessive short-term movements, whereas in bear markets, it is more closely associated with mitigating crash risk and preventing panic-driven sell-offs.
This study makes several contributions. First, it provides the first systematic evidence on the unique “tail-head” ownership structure in China, revealing how the configuration of passive capital differs from that in developed markets. Second, moving beyond exchange-traded funds, it evaluates the comprehensive influence of index funds, on price stability, showing that passive ownership has long-term governance implications. Third, it identifies the governance channel as the primary mechanism through which index funds stabilize prices, supported by evidence on disclosure behavior, managerial tone, and earnings quality. Fourth, it offers new insights for policy design, particularly regarding the development of patient capital, the optimization of index-fund product structures, and the long-term stability of China's capital market.
The findings yield several policy recommendations. Strengthening long-horizon assessment frameworks, improving product-design and index-construction rules, and enhancing standards for disclosure, internal control, and risk management will help reinforce index funds' stabilizing function. At a broader level, increasing the share of long-term funds and improving investor structure can enhance market resilience and reduce the influence of short-term speculative trading. These implications align with current efforts to support the high-quality development of the public-fund industry and to promote the long-term healthy functioning of China's capital market.