This paper develops a novel intra-urban spatial structure model that extends traditional linear or circular city frameworks to accommodate cities of arbitrary shapes.For the first time,it examines how the number and location of urban sub-centers—key components of a polycentric urban layout—affect residents’ welfare by reshaping the spatial structure within cities.The theoretical model assumes that the number and locations of sub-centers are exogenous variables determined by urban planning as well as historical and geographical factors.Given the locations of these sub-centers,the spatial structure of the city is determined.This structure,in turn,influences commuting costs and agglomeration effects,leading to a redistribution of population and economic activities and,consequently,affecting residents’ welfare. The model reveals that increasing the number of sub-centers reduces commuting costs within the city.However,the dispersion of firms caused by a polycentric layout may weaken agglomeration economies.Urban planners must weigh these trade-offs and consider initial conditions such as natural geography and industrial bases to avoid coordination failures in sub-center development.Taking Chengdu as a case study,the paper calibrates the model and conducts counterfactual analyses to explore how different spatial planning strategies might reshape the distribution of population and economic activities,ultimately influencing residents’ welfare.
 The counterfactual analyses reveal that in monocentric cities,welfare levels are highly sensitive to changes in congestion factors,whereas polycentric cities are more resilient.Thus,under severe congestion,a polycentric layout offers significant potential to alleviate traffic-related challenges often faced by large cities.Moreover,Chengdu’s latest Land Spatial Master Plan (2020-2035),which integrates the existing western sub-center with the CBD area while prioritizing the development of eastern and southern sub-centers,appears well-positioned to enhance agglomeration effects,improve commuting efficiency,and elevate overall welfare.
 The findings provide valuable insights for urban policymakers in designing spatial plans.Furthermore,the proposed model and numerical simulation approach can be applied to other cities,offering a robust framework for evaluating and optimizing polycentric urban layouts—an increasingly critical policy instrument in the ongoing process of urbanization.

In recent years,China’s platform economy has developed rapidly,and its position and role in the overall economic and social development have become increasingly prominent.As a core element of the platform economy,the intrinsic relationship between Internet traffic and the platform economy needs to be examined.Internet Celebrity Economy/Influencer Economy refers to the phenomenon where creators/influencers on video-sharing and live-streaming platforms leverage their fame or talents to attract followers and then monetize the traffic through collaborations with businesses.The high loyalty,strong stickiness,and long retention time of the streamers’ fans significantly differentiate them from casual viewers.The direct competition among streamers underscores the importance of analyzing the oligopolistic nature of traffic entry points.

An interesting fact is that manufacturers of different categories often adopt different live-streaming e-commerce marketing models.Some product promotion links seem to be everywhere.Some merchants place advertisements on all platforms and multiple channels without distinguishing channel attributes.There even appears a contradictory situation where game zone anchors and knowledge zone anchors recommend the same product at the same time.Some merchants also place advertisements on multiple channels,but mainly in the form of pre-sales or flash sales,with a small amount of supply and no large-scale promotion.Other merchants focus on highlighting the “channel exclusive” feature in promotion,with limited release as the main selling point.

This paper argues that the differentiation of marketing models is related to product categories to a certain extent.Unlike registered users of traditional platforms,potential consumers in the field of live-streaming e-commerce often exist in the form of fans of anchors or brands.Fan attribute is the most distinct personal characteristic of the audience of live-streaming sales.The “die-hard” audience of anchors or brands stay in the live-streaming room frequently and for a long time,with high product exposure frequency,high trust in anchors or brands,and high possibility of accepting recommendations; The “passer-by” audience of anchors or brands do not have a strong awareness of the anchors or brands.They mostly enter the live-streaming entrance through the way of enterprise paid promotion of the live-streaming room,with a general stay time,and the possibility of accepting recommendations depends on the cost-effectiveness of the product; The “haters” of anchors or brands have a strong dislike for the anchors or brands,and it is almost impossible for them to become live-streaming audiences,let alone accept recommended marketing.The acquisition costs of these three types of users show significant horizontal differences.Hence,the recognition of commodity value by these three types of users,that is,the correlation between commodity preference and live-streaming audience,is the key to distinguish commodity categories.

Hence,this paper develops a duopoly model of platforms based on the Hotelling model,which allows firms to price traffic.It finds that when consumers’ preferences for goods are completely independent of their channel loyalty,firms always choose bilateral traffic diversion,and both partial and full market coverage are possible.When consumers’ preferences for goods are completely negatively correlated with their channel loyalty,the only possible scenario is bilateral traffic diversion by firms with partial market coverage.When consumers’ preferences for goods are completely positively correlated with their channel loyalty,two scenarios may occur:bilateral traffic diversion with full market coverage,or unilateral traffic diversion with partial market coverage.

Subsequently,this paper analyzes the impact of changes in platforms’ customer acquisition costs and market segmentation on social welfare.It provides targeted policy recommendations for various markets based on differences in consumer types:when consumers’ preferences for goods are completely independent of their channel loyalty,the degree of market coverage can be used as an intuitive indicator of welfare.Markets with higher coverage generally have better consumer welfare properties without losing too much total social surplus.When consumers’ preferences for goods are completely negatively correlated with their channel loyalty,central planners can restrict behaviors such as induced sharing and forwarding,and penalize false or exaggerated elements in promotional activities to reduce the willingness of undesired users to participate in marketing activities,thereby preventing firms from exiting the market.When consumers’ preferences for goods are completely positively correlated with their channel loyalty,central planners should control channels’ customer acquisition costs as much as possible to promote more transactions and should pay more attention to markets with lower segmentation,especially being vigilant against practices such as “big data price discrimination” that harm consumer rights.

This research not only has positive implications for guiding firms on how to choose appropriate traffic channel layouts but also fills a gap in the existing literature regarding the impact of the correlation between consumer preferences and traffic channel loyalty on market equilibrium.It provides a new perspective and explanatory framework for the theoretical study of platform economics.

Entrepreneurial activity across different cities in China exhibits significant regional disparities,with a distinct pattern of being “strong in the east and weak in the west.” Social organizations play a critical role in China’s socialist modernization efforts.As an important form of social organization,chambers of commerce serve as key bridges linking the government,enterprises,and markets.Chambers of commerce refer to legally registered social organizations composed of members engaged in similar economic activities,individuals,or economic entities within the same region,operating under principles of industry service and self-regulation.

In recent years,regional chambers of commerce in China have experienced rapid growth,playing an increasingly prominent role in supporting high-quality development.According to data from the China Social Organization Government Service Platform,as of December 2023,there were 24,970 registered regional chambers of commerce below the provincial level (excluding provincial-level chambers of commerce).However,the regional pattern of “strong in the east and weak in the west” remains prominent.This raises several critical questions:Can regional chambers of commerce enhance urban entrepreneurial activity? If a positive effect exists,what mechanisms drive it? What factors may strengthen or constrain the entrepreneurial impact of these chambers of commerce? Analyzing these questions not only contributes to understanding the theoretical and practical significance of regional chambers of commerce but also provides policy insights for optimizing the entrepreneurial environment and promoting employment.


Based on data manually compiled from the China Social Organization Government Service Platform on regional chambers of commerce below the provincial level,matched with the 2003-2021 China Industrial and Commercial Registration Database,this study empirically examines the impact of regional chambers of commerce on urban entrepreneurial activity.The findings reveal that regional chambers of commerce significantly enhance entrepreneurial activity in cities.Specifically,an increase of one standard deviation in the cumulative number of regional chambers of commerce per 100,000 people results in an increase of 0.419 newly registered enterprises per 100 people in a city.

The study conducted a series of robustness checks,including differentiating among types of chambers of commerce,replacing measures of urban entrepreneurial activity,modifying indicators of regional chambers of commerce development,and adjusting the research sample scope.All results consistently support the robustness of the main conclusions.To address potential endogeneity concerns,the study employs the number of chambers of commerce and river density in each prefecture-level city as of 1912 as instrumental variables for regional chambers of commerce.

Mechanism analysis indicates that regional chambers of commerce enhance urban entrepreneurial activity by improving social trust and optimizing the business environment.Further,the study examines the heterogeneous impacts of regional chambers of commerce from multiple dimensions,including ownership types,industry classifications,and city types.The results demonstrate that regional chambers of commerce primarily promote the entry of individual industrial and commercial households and private enterprises,have the most significant impact on labor-intensive industries,and exhibit stronger entrepreneurial effects in third-tier and smaller cities,non-urban agglomeration cities,and small to medium-sized cities.Additionally,the study finds that regional chambers of commerce generate synergistic effects with entrepreneurship-related policies,thereby boosting urban entrepreneurial activity and enhancing urban innovation levels.

Compared to existing research,this study offers three main contributions:

(1) Existing empirical studies on chambers of commerce primarily focus on provincial-level nonlocal chambers of commerce in China,with limited discussion on local chambers of commerce and inadequate systematic analysis of regional chambers of commerce below the provincial level due to data constraints.This study is the first to extend the research perspective to regional chambers of commerce below the provincial level nationwide,encompassing both local and nonlocal chambers of commerce.It refines the spatial scale to include municipal-level,county-level,township-level,and neighborhood associations.


(2) Previous literature has explored the impact of provincial-level nonlocal chambers of commerce on interregional enterprise development,cross-regional trade,and interprovincial labor mobility.This study enriches the empirical literature on the economic effects of chambers of commerce by demonstrating that regional chambers of commerce significantly enhance urban entrepreneurial activity.

(3) The study investigates the specific mechanisms through which regional chambers of commerce influence urban entrepreneurial activity,showing that these chambers of commerce enhance social trust and optimize the business environment.These findings deepen the understanding of regional chambers of commerce as a vital form of social organization and enrich research on factors influencing urban entrepreneurial activity.The findings of this study provide valuable insights for fully leveraging the positive role of regional chambers of commerce,optimizing urban entrepreneurial environments,and promoting high-quality economic development.

This paper examines the impact of China’s two-child policy on household risky asset allocation.Since the 1970s,China’s fertility rate has declined significantly,raising concerns about economic growth,such as an aging population and labor shortages.In response,the government relaxed birth restrictions,introducing the two-child policy to address these issues.Beyond influencing fertility intentions,the policy change may affect household financial decisions.Families anticipating higher child-rearing costs might seek to invest in riskier assets to improve returns,while others may reduce risky investments due to financial constraints.Understanding how fertility policy affects household investment choices is crucial for both policy design and the long-term sustainability of economic growth.Using the selective two-child policy in 2014 as a quasi-natural experiment,this study analyzes the China Household Finance Survey (CHFS) data to examine the causal relationship between the policy and household allocation of risky assets.


The findings indicate that the implementation of the two-child policy has led to a notable 7.3-percentage-point increase in the proportion of household risky assets,particularly in urban families.The study posits that families with stronger fertility preferences and higher income,who expect an increase in future child-rearing expenses,tend to invest in risky assets to increase their asset return rate.The research reveals that the increase in risky asset allocation is not due to new participation in risky asset investment,but rather an increase in risky asset allocation by households that already held risky assets and a decrease in the risk aversion of middle- and high-income households in urban areas.To test the robustness of the benchmark results,we used a series of robustness tests,including a placebo test,propensity score matching (PSM),exclusion of childless households,inclusion of rural households in the control group,addition of fixed effects,and control variables.All results are consistent with the baseline results.


In conclusion,this study finds that the implementation of the two-child policy has led to a significant increase in risky investment for eligible families.The rise in the risky investment rate is closely tied to family fertility preferences and income levels,establishing a logic chain of “relaxation of birth control-increase in fertility desire-increase in liquidity demand-increase in risky investment”.The research suggests that the increase in risky investment is due to residents’ anticipated increase in child-rearing expenses,leading to a preference for risky assets to increase asset returns.


The research contributes to the literature by linking fertility policy to household asset allocation and being the first to identify a causal effect of having an additional child on risky asset investment.It offers a new explanation for the low level of risky asset investment among Chinese households.The findings have important policy implications,highlighting the need for financial market reforms that take into account both the impact of population policies and the dynamic changes in household asset allocation.Policymakers should consider balancing the costs of child-rearing with opportunities for financial investment.Additionally,the government should strengthen financial market regulation to protect the rights of family investors,especially those with multiple children,and ensure greater fairness and transparency in the market.

The issuance of the CSI A500 ETF has shifted market attention towards passive investment strategies.ETFs,as a well-established passive investment tool,have garnered increasing attention due to their unique and innovative trading attributes.This paper investigates whether passive index investment tools enhance market monitoring functions from the perspective of ETF ownership.The findings reveal that a higher proportion of ETF ownership in individual stocks significantly reduces the prevalence of shareholder equity pledging.An exploration of the underlying mechanisms suggests that higher ETF ownership mitigates information delay and offsets existing noise trading,thereby improving stock price information efficiency.This enhanced efficiency generates a monitoring effect,ultimately reducing equity pledging by shareholders.In terms of ownership characteristics,the monitoring effect is more significant in firms where the largest shareholder holds a higher ownership stake or in non-SOE enterprises.Further,the improvement effects brought about by ETFs are more pronounced in firms with lower levels of information disclosure quality.The results of the study remain robust after undergoing various tests,including instrumental variable analysis.The conclusions of this study provide important insights for improving market efficiency,strengthening market monitoring functions,and fostering a virtuous governance cycle.

We primarily examine the enhancement of market monitoring functions from the perspective of ETF ownership,using equity pledge as a measure of the effectiveness of market supervision.The interaction between institutional investors and corporate governance generally operates through three mechanisms:active governance,“voting with their feet”,and market-based information mechanisms for indirect supervision.Given the diverse investor base and dispersed composition of ETFs,we argue that the third mechanism predominantly underpins their market monitoring role.We investigate whether an increase in the proportion of ETF holdings can reduce equity pledging,thereby curbing corporate tunneling behaviors and enhancing effective monitoring.We identify improvements in stock price information efficiency as the primary channel through which ETFs exert their supervisory function.Key factors influencing stock price information efficiency include the speed of information transmission,liquidity depth,and the proportion of noise trading.Our empirical analysis demonstrates that the introduction of ETFs accelerates information transmission,offsets existing noise trading,and enhances the efficiency of stock price information.This,in turn,creates capital market pressure that imposes supervisory constraints on equity pledging by shareholders,although we find that the liquidity channel appears less significant.Further analysis reveals that the supervisory effect of increased ETF ownership is more evident in firms where the largest shareholder holds a higher proportion of shares or in non-state-owned enterprises,as these firms often exhibit stronger incentives for tunneling.We also find that the effects of ETFs are more pronounced and significant in firms with weaker information disclosure practices.To address potential endogeneity concerns,we leverage the exogenous increase in passive ownership resulting from index reconstitutions,referring to changes in the bottom constituents of the Russell 1000 Index and the top constituents of the Russell 2000 Index.Similarly,in the Chinese market,we use changes in the CSI 300 and CSI 500 indices,and construct an instrumental variable based on newly added top-tier constituents in the CSI 500 Index.Our results remain robust,thereby ensuring the reliability of our findings.


The primary contributions of this paper are as follows:1)Expanding the understanding of the impact of passive index investment tools from the perspective of ETFs:While existing literature on ETFs largely focuses on their influence on the characteristics of capital markets,there is a lack of in-depth exploration of their effects on corporate governance.Our findings provide a robust interpretation of the role ETFs play in corporate governance,demonstrating how ETF ownership influences equity pledging through its effects on the informational environment.This contribution offers new insights and evidence regarding how financial instruments can address corporate governance issues via information mechanisms.2)Augmenting research on the factors influencing equity pledging:Prior studies on equity pledging have predominantly focused on its consequences or subsequent abnormal behaviors,leaving the determinants of equity pledging relatively underexplored.By adopting the unique perspective of ETFs,our study identifies how different informational environments affect equity pledging decisions,filling an important gap in the literature.3)Exploring the channels through which ETFs exert influence:We show that ETFs play a role in corporate governance by reducing information transmission lags and mitigating noise trading.This enriches the understanding of the mechanisms through which passive index investment tools operate.Given the growing scale of ETF investments,our findings highlight the practical importance of examining how ETFs shape the ecological dynamics of capital markets.Overall,the results of this study have significant implications for future research on passive index investment tools.They contribute to a deeper understanding of how such tools enhance market efficiency and foster greater market depth,offering valuable guidance for both academic inquiry and practical application.

International organizations such as the G20 and OECD have been attempting to promote the adoption of the best corporate governance mechanisms across countries.As a result,many scholars have predicted that the corporate governance of companies in different countries would increasingly become similar.However,significant differences still exist in corporate governance among firms from different countries.Over the past three decades,scholars have attempted to examine the origins of these differences through large-sample cross-country comparative studies,which is defined as comparative corporate governance.However,the challenges of conducting solid comparative corporate governance research are substantial.Limited consensus exists over what factors best explain the diversity of corporate governance across countries.Moreover,few articles provide adequate attention to cross-country comparability testing,raising the concern that the reported findings may be spurious.In light of these opportunities and challenges,we systematically review comparative corporate governance studies published in top journals both domestically and internationally from 1990 to 2022.In this review,we not only draw on the available literature from international business (IB) and management but also borrow from comparative research in neighboring disciplines such as political science and sociology.The objectives of this review are three-fold:① take stock of the growing literature on comparative corporate governance by adapting Kohn’s typology of models of comparative research; ② provide a critical assessment of the literature and identify the gaps and problems in the extant studies; ③ set an agenda for future comparative corporate governance research.Drawing upon the cross-national comparative research framework proposed by sociologist Kohn,we categorize comparative corporate governance research into four domains:nation as the object of study,nation as the context of study,nation as the unit of analysis,and nation as part of a larger system.Firstly,13 studies hold a nation as the object of study.In this type of comparative corporate governance research,the authors’ primary interest lies in corporate governance in the particular countries studied.It is often descriptive,seeking to describe the similarities and/or differences in corporate governance in different nations.Secondly,18 studies have treated a nation as the context of the study.In such studies,the authors are primarily interested in examining whether and how national context is related to  corporate governance or  the relationship between corporate governance and its antecedents/consequences.Thirdly,research treating a nation as the unit of analysis has flourished in the comparative corporate governance field.In this type of research,the pivotal distinguishing national characteristics become variables in the analysis.Research that treats nations as units of analysis is typically labeled as “large-N comparative analysis”.Scholars pursuing this line of research are less interested in the unique context of the nations under study and more interested in the abstract relationships among quantifiable national characteristics.Finally,only four studies have treated nations as part of a larger international system.In these studies,scholars interpret a nation’s corporate governance as influenced by transnational systems or processes.

Based on our review of a sample of 115 papers,we identified significant  gaps and areas of concern that limit the impact and rigor of this line of research.Firstly,a deep understanding of the national context of corporate governance has been limited,which deserves further exploration.Secondly,scholars have also not paid enough attention to concept equivalence.Few comparative studies,particularly those treating a nation as the unit of analysis,have offered sufficient discussion of concept equivalence.Thirdly,the current literature has primarily focused on developed countries for comparative analysis,while relatively few studies have focused on emerging economies such as China.In addition,prior studies have adopted different methods and samples,making it difficult to determine whether the differences in corporate governance across countries are caused by variations in national institutions or by methods.Finally,although a number of scholars have effectively demonstrated key shortcomings in the law and finance approach,most scholars have predominantly relied on this approach to conduct comparative corporate governance research.

This paper has also proposed five avenues for future research based on the knowledge gaps identified in our critical assessment of the literature.① Scholars should conduct more systematic analyses of national context before translating “nations” into “variables.” Only with a deeper understanding of the national context can the country-level variables extracted from research become more explanatory.② Future research should take advantage of qualitative comparative analysis (QCA) and hierarchical linear modeling (HLM) in conducting comparative corporate governance research.③ Additional effort needs to be made to achieve comparability of concepts to enhance the reliability and validity of research.④ Future studies should present a justification for country selection.Scholars may only select countries to which they have access,which results in an over-representation of developed economies with better access to data on corporate governance.However,any similarities or differences revealed by a comparative corporate governance study may be no more than an artifact of the choice of countries.As a result,country selection must be theoretically justified.Given the significant differences in institutional environments and cultural contexts between emerging markets and developed countries,future research should incorporate emerging markets into the sample.⑤ Comparative corporate governance research can benefit from the  integration of theories from different disciplines.Therefore,future research should draw upon insights from other fields such as sociology and political science.

In the midst of digital revolution,it is both critical and urgent to promote the digital transformation of Chinese firms,as they face numerous obstacles in this process.These include,for example,a lack of clear strategic goals and practical pathways,a scarcity of digital talent,and insufficient financial resources.Emerging market governments,as a crucial force in market operations,play a significant role in coordinating and allocating market resources,while providing policy support and guidance.However,we know little regarding how local politician continuity,a systemic setting and political phenomenon in China,affects firm digital transformation.

Drawing on resource dependence theory,this study investigates the impact of local politician continuity on firm digital transformation.Politician continuity in a local area lowers uncertainty in public policymaking and resource allocation,thereby facilitating local firms’ digital transformation.Furthermore,firm specific characteristics and institutional factors determine the extent of firms’ resource dependence on the local government,thus moderating the aforementioned effect.Specifically,we focus on how the proportion of overseas revenue,research and development (R&D) investment,firm ownership,and local digital economy policies moderate the relationship between local politician continuity and firm digital transformation.

The empirical setting is Chinese listed firms from 2010 to 2022.Firm-level data came from the CSMAR database,Wind database,and China Research Data Service Platform (CNRDS).Information of local politicians were from the personal resumes of local officials published on websites,such as People’s Daily Online,Xinhua Net,Zecheng Net,and Baidu Encyclopedia,and are supplemented by news reports and government websites to ensure the comprehensiveness and authenticity of the data.The names,birth dates,genders,educational backgrounds,places of origin,time of joining the CPC,first working time,time of taking office,time of leaving office,reasons for leaving office,and other information are manually collected.In addition,other relevant data at the city level,such as the city’s gross domestic product and local digital economy policy orientation,was sourced from the China Statistical Yearbook and the China Research Data Service Platform.The final sample covers a total of 29967 observations for 4499 companies,involving a total of 249 prefecture-level cities.To validate the hypotheses,this paper constructs a fixed-effect Tobit model.The core explanatory variables and control variables were lagged by one year to promote the construction of causal relationships.


The study demonstrates that local politician continuity promotes digital transformation of firms within their jurisdictions.The proportion of overseas revenue and R&D intensity weaken the positive impact of local politician continuity while in state-owned firms,its impact gets more pronounced.Further,local digital economy policies weaken the effect of local politician continuity on firm digital transformation.


Additional analyses show that local politician continuity has a greater positive impact on firm digital transformation in central affiliated state-owned firms than local affiliated ones; and there is no difference between officials’ internal and external promotion.


Our study enriches the literature on politician continuity in two ways.①Based on resource dependence theory,we explicate how local politician continuity affects the digital transformation of firms within their jurisdictions.②We incorporate “central strategic objectives/local policy responses” into the theoretical framework of local politician continuity.We add to the emerging research on the antecedents of firm digitization transformation by providing a new perspective for understanding firms’ motivation for engaging in digital transformation.③By exploring the moderating effects of firm-level factors (i.e.,the proportion of overseas revenue and R&D investment) and institutional factors (i.e.,ownership and local digital economy policies),we demonstrate the importance of market strategies and institutional factors in managing non-market dependent relationships.


The findings of this study provide valuable management and policy insights.For managers,it is essential to actively manage the potential risks related to local politicians.In response to the negative impact of frequent local personnel changes on digital transformation,they can adopt market-oriented strategies such as increasing investment in overseas markets and R&D.For state-owned enterprises and firms located in areas with less digital economy policies,more attention should be paid to potential changes in local officials and sufficient preparations should be made to mitigate the negative impact on digital transformation.For governments:Firstly,it is necessary to enhance the continuity and stability of local policies during the transition period to provide a stable and predictable businessenvironment.Secondly,the study found that compared to state-owned enterprises,private enterprises still face significant disadvantages in accessing government resources.Although reducing dependence on the government mitigates the positive impact of local official continuity on their digital transformation,local governments should fully consider the actual needs of private enterprises,allocate resources reasonably,and improve resource utilization efficiency.Thirdly, it is crucial to strengthen policy guidance for local development of the digital economy.By enhancing policy supply in areas such as promoting industrial digitization,digital infrastructure construction,digital resource integration,digital talent cultivation,and network and information security system construction,the impact of local officials on firm digital transformation can be mitigated.

How organizations adapt to rapidly changing and dynamic environments through continuous learning is considered central to organizational learning theory.Organizational learning has been driven by human experience and social interaction,facilitating knowledge creation,transfer,and application.However,the introduction of artificial intelligence (AI) is redefining the premises and mechanisms of learning,influencing traditional theoretical frameworks,and introducing complexities in multi-level interactions.These changes are primarily reflected in three areas:the expansion of learning actors,the restructuring of learning mechanisms,and the governance of learning outcomes.Although human-machine interactive learning is reshaping the paradigm of organizational learning,there is still a lack of a systematic understanding of the evolution of organizational learning research under the impact of AI.Thus,this study aims to comprehensively review the evolution of organizational learning research in the context of AI.


Specifically,this study conducts a systematic review of 74 key articles on organizational learning in the context of AI.Using the “individual-group-organization” framework of organizational learning as its foundation,the study conducts a review across four key themes:① individual-level learning in the AI context; ②group-level learning in the AI context; ③organizational-level learning in the AI context; ④ multi-level interactive learning in the AI context.Individual-level learning focuses on how a single individual modifies their cognitive structures and behavioral patterns through intuition and interpretation,with its core centered on knowledge construction at the individual level.Group-level learning refers to the process in which multiple individuals engage in collaborative learning within a group,emphasizing the interpretation and integration of individual knowledge within the team.Organizational-level learning concerns the integration and institutionalization processes across teams,focusing on how knowledge extends from the team level to the entire organization.Multi-level interactive learning examines the flow and interaction of knowledge across individual,group,and organizational levels,emphasizing how feed-forward and feedback processes shape learning at different levels.

The findings reveal several critical points.Specifically,at the individual level,AI enhances intuition,supports interpretive reasoning,and promotes action reflection,improving decision-making and adaptability in complex environments.However,it may also lead to negative consequences such as over-reliance on technology and a decline in autonomous learning capabilities.At the group level,AI facilitates the building of cognitive consensus,enhances collaboration efficiency,and improves decision-making effectiveness,but it may also reduce the diversity of knowledge within the group.At the organizational level,AI restructures digital memory mechanisms,optimizes knowledge management,and drives cross-boundary learning through data,enhancing organizational adaptability and stakeholder engagement,while also introducing potential ethical risks.In terms of multi-level interactive learning,AI accelerates the speed and breadth of knowledge flow both internally and externally,breaking down the knowledge barriers between organizations and ecosystems,and making multi-level interactive learning within organizations more dynamic,complex,and competitive.Building on this,this study constructs a research framework for organizational learning in the context of artificial intelligence,revealing the systematic evolution of organizational learning research under the impact of AI.First,AI is identified as a catalyst for changes in learning contexts,driving individuals,groups,and organizations to adopt integration pathways.Second,the integration of AI is observed to reshape adaptive actions in human-AI collaboration,with these actions influenced by technological,organizational,and environmental factors.Finally,the outcomes of these adaptive actions reinforce organizational systems through feedback mechanisms,enabling coevolution between AI and organizational systems,driving firm’s adaptive innovation and growth,and  forming a self-reinforcing cycle of coordinated evolution.

This study identifies research gaps and proposes four future research directions:①analyzing the multidimensional interactions of individual learning in the AI context,focusing on complex tasks,cross-boundary integration,and long-term effects.For example,the cognitive reshaping mechanism of AI in key managerial learning,the potential exploration of AI in individual cross-domain knowledge integration,and the negative effects of long-term interaction with AI on individual learning.②Deconstructing the psychological processes of group learning in the AI context,including trust building,role perception,and collaboration structures.For example,the trust mechanism of shared cognition in teams under AI contexts,AI-driven role cognition and dynamic task allocation models in teams,and the impact of team structure optimization on fostering diverse perspectives and collective intelligence.③Uncovering the dynamic adjustments of organizational learning in the AI context,such as memory updates,knowledge integration,and contextual adaptation.For example,the systematic management of organizational digital memory in AI contexts,the integration mechanism of AI knowledge and human knowledge within organizations,and the multi-context adaptation model of data-driven learning.④Constructing the co-evolution mechanism between AI capabilities and multi-level learning,emphasizing influence pathways,capability emergence,and governance frameworks.For example,the influence pathways of multilevel organizational learning on AI capabilities,the emergent capabilities and strategic consequences of AI in interaction with multilevel learning,and the critical role of AI governance in the co-evolution process.


This study has three contributions:First,it reveals the transformations in organizational learning under the AI context,encompassing changes in learning contexts,processes,and outcomes.These include the disruptive impact of AI on learning contexts,the expansion of feedforward and feedback mechanisms,and the adaptive evolution of human-AI collaboration,thereby extending the applicability and developmental scope of organizational learning theory in the AI era.Second,it systematically reviews and synthesizes the core themes and limitations of existing research on organizational learning in the AI context,constructing a comprehensive research foundation and theoretical framework.Third,it addresses the major gaps in current studies by proposing future research directions,offering clear pathways and guidance for advancing organizational learning research in the AI era.

The tourism industry has experienced sustained growth in recent years.However,the COVID-19 pandemic led to a sharp decline in global tourism in 2020.As the impact of the virus wanes and epidemic management becomes more standardized,the tourism sector is gradually rebounding.China,the world’s largest outbound tourism market,has significantly contributed to the global recovery of tourism through its policy approach to COVID-19 normalization.Forecasting tourist numbers enables more strategic allocation and adjustment of tourism resources,enhancing service quality.The pandemic has influenced tourism demand,making it essential to analyze these shifting patterns for the sustainable growth of the industry.


Research in tourism forecasting has yielded several insights:①Tourism demand forecasts benefit from incorporating relevant external factors,such as online search indices,to boost accuracy and interpretability.②Combining forecasting methods can enhance,or at least maintain,the accuracy of single forecasts. ③Unforeseen events like COVID-19 outbreaks can compromise the accuracy of traditional methods,necessitating specialized forecasting approaches during crises.However,existing methods often overlook the varying impact of exogenous factors on tourism demand over different periods,assuming a stable relationship between explanatory and target variables even in crisis conditions.In reality,the pandemic altered tourists’ risk perceptions,which,in turn,influenced their demand and consumption preferences,impacting their travel behavior.This indicates that an explanatory variable’s effect on tourism demand may differ across pre-pandemic,pandemic,and post-pandemic periods,meaning that variables from distinct periods may need to be treated as independent new variables.

This research primarily employed a Vector Autoregression model with exogenous variables (VARX),an adaptation of the Vector Autoregression (VAR) model that allows for the simultaneous analysis of endogenous and exogenous variables.The study began by gathering data on the number of Chinese outbound tourists,with a particular focus on trips to Japan,South Korea,and Singapore.Using web scraping techniques,55 Baidu search terms related to visa applications,trip planning,dining,accommodation,transportation,and shopping were collected; following correlation analysis,52 variables were retained.The classic factor model (FM) and dynamic factor model (DFM) were then used to combine the Baidu search indices into a composite indicator that retained dynamic relationships among the original multiple indicators.To account for the variable effects of exogenous factors over time,the Baidu search index was segmented into three phases—peak,trough,and recovery—based on tourists' search behaviors.Each segment served as an exogenous variable in the VARX model to forecast tourist numbers under different influences.Subsequently,a Stacking approach was applied in machine learning to combine various predictions,evaluated using RMSE,MAPE,and MASE.Finally,out-of-sample forecasts were produced to inform projections for Chinese outbound tourism.

The findings indicate that VARX models effectively predict Chinese outbound tourist numbers and that accounting for the time-specific effects of exogenous variables can enhance accuracy.Key insights include:①Aggregating new indicators from numerous web search indices is effective for high-dimensional time series forecasting,with the dynamic factor model proving superior in high-dimensional contexts,reducing prediction error and improving average accuracy by 12.25% over the classic factor model.Practically,managers can compile various search indices to create a comprehensive indicator that reflects market trends and use it to forecast tourism demand.② Segmenting variables based on significant time-based changes is practical,as the segmented combination model improves accuracy over traditional models and is valuable for tourism demand forecasting under crisis impacts.As tourism normalizes,people’s behavior and choices may carry over from previous patterns with some adjustments.③ The study not only evaluates the model’s validity but also projects an increase in Chinese outbound travel in 2024,expecting it to recover to at least 60% of pre-pandemic levels.


This study introduces a segmented combination forecasting approach for predicting Chinese outbound tourism.Utilizing a dynamic factor model to manage exogenous variables,it captures the original data’s multiple external factors while preserving their dynamic relationships.This method is not only relevant to tourism demand forecasting in special circumstances like pandemics but is also applicable in other contexts where data may be limited.