Research background: In the context of green innovation as an important pathway to achieving global sustainable growth, developing green technologies is crucial to achieving China's sustainable development goals. Enterprises contribute to the low-carbon transformation of the economy through green innovation. However, the unpredictability of global economic policies has increased rapidly in recent years, raising decision-making risks for businesses and affecting their investment behavior. This, in turn, significantly impacts their green innovation activities. Purpose of the article: This study aims to analyze the impact of perceived economic policy uncertainty on corporate green innovation activities. This study focuses on exploring how perceived economic policy uncertainty affects green innovation. Methods: This study employs a panel Poisson model with panel data from Chinese A-share listed companies from 2010 to 2020. We verified the model's suitability and the robustness of the results through a likelihood ratio test and Poisson pseudo-maximum likelihood estimation. Findings & value-added: Perceived economic policy uncertainty significantly inhibits firms' green innovation activities. Mechanism tests indicate that perceived economic policy uncertainty suppresses green innovation activities by reducing environmental investment and financial asset allocation. Heterogeneity analysis shows that the detrimental effect of perceived economic policy uncertainty on green innovation activities is particularly noticeable among non-state-owned enterprises, pollution-intensive industries, and sectors with low environmental regulation. This study provides valuable information to promote green economy transformation in other economies.
Research background: Within the agri-food sector, listed companies classified in the Fishing and Farming category face increasing pressure to integrate sustainability practices into business operations, yet the relationship between environmental, social, and governance (ESG) factors and financial performance remains insufficiently explored. The growing relevance of Sustainable Development Goals (SDGs) in corporate strategy has intensified the need to understand how ESG commitments translate into measurable financial outcomes, especially in resource-intensive industries. Purpose of the article: This research examines how sustainability relates to the financial performance of listed fishing and farming companies that report SDG information, by analysing ESG practices alongside SDGs. Methods: Data were drawn from the London Stock Exchange Group database for listed companies classified under the Thomson Reuters Business Classification. Only firms disclosing SDG-related information were retained in the final sample. Financial performance is measured by means of return on common equity, return on assets, and income after tax margin. The methodology combines quantile regression, capturing differences across low, median, and high financial performance levels, with Bayesian network analysis to explore overall interlinkages, direct and indirect, between variables. Findings & value added: Results show that the relationship between sustainability and financial performance is strongly influenced by the firms’ performance position. While ESG practices can support financial performance, some may also generate short-term cost pressures. The study's novelty lies in its focused analysis of Fishing and Farming, its integration of ESG sub-scores with SDG contributions, and its examination of firms across the conditional distribution of financial performance, revealing important differences across quantiles. In the long term, this study contributes to sustainability theory by establishing that the relationship between financial performance and sustainability, whether measured through ESG factors or SDG contributions, is performance-conditional, urging future frameworks to incorporate firm-level heterogeneity when modelling sustainability outcomes in agricultural sectors.
Research background: Artificial intelligence (AI) has become a central element of firms’ digital transformation agendas, yet empirical evidence on how AI affects firm performance remains fragmented across theories, sectors, and performance dimensions, obscuring the mechanisms through which AI shapes organizational outcomes. Purpose of the article: The article aims to systematically map the scientific literature that explicitly examines the relationship between AI use and firm performance in the digital transformation era, and to clarify the main thematic areas, theoretical perspectives, and research gaps that structure this field. It addresses seven research questions covering the evolution of scientific production; key contributing countries, institutions, authors, and journals; dominant themes and conceptual clusters; the intellectual structure of the field; and the conceptualization of AI as a driver of business model and organizational transformation. Methods: The study uses a quantitative bibliometric design, drawing on 422 journal articles indexed in the Web of Science (WoS) Core Collection from 2000 to 2025 that explicitly address AI and firm performance. The Bibliometrix R package integrates performance indicators with science-mapping techniques, including co-word analysis, thematic mapping, and source co-citation networks. Findings & value added: The results show rapid post-2018 growth in research on AI and firm performance, strong internationalization, and a dual intellectual core that combines applied, technology-oriented journals with strategic management and organizational outlets. Conceptual analyses present AI as a strategic organizational meta-capability that affects performance through mechanisms of efficiency, agility, innovation, governance, and sustainability, while also identifying the field as fragmented and heavily reliant on cross-sectional designs and coarse measures of AI adoption. The article adds value by integrating dispersed contributions into a coherent bibliometric map, highlighting underexplored areas related to business model transformation, governance, sustainability issues, small businesses, and human-AI complementarity, and by outlining a future research agenda and implications for AI investment and governance.
Research background: As climate concerns intensify globally, organizations face increasing pressure to integrate sustainability into their core human resource strategies. Although green human resource management has received growing scholarly attention, the role of digital enablers, particularly gamification and AI-driven personalization, remains insufficiently examined. In particular, limited empirical evidence exists on how technologically enabled green HR interventions contribute to the development of employee environmental literacy through organizational learning mechanisms. Purpose of the article: This study investigates how technologically enabled green human resource management practices influence employee environmental literacy. It further examines the sequential mediating roles of sustainable learning engagement and green knowledge internalization, while also exploring the moderating influence of contextual and individual factors within organizations. Methods: Using survey data collected from HR managers working in multinational corporations, the study applies a dual analytical approach. Partial least squares structural equation modeling (PLS-SEM) is used to test the hypothesized structural relationships, while machine learning algorithms (XGBoost, LASSO, and Random Forest) are employed to evaluate predictive relevance and identify nonlinear relationships within the data. Findings & value added: The results indicate that gamified green human resource management practices and AI-driven green human resource personalization significantly enhance employee environmental literacy. The analysis also confirms the sequential mediating roles of sustainable learning engagement and green knowledge internalization, demonstrating that technology-enabled human resource interventions influence environmental literacy primarily through learning-based mechanisms. Machine learning results further support the predictive relevance of the proposed framework, with XGBoost achieving the strongest predictive performance, followed by Random Forest and LASSO regression. The partial least squares structural equation modeling analysis confirms sequential mediation (the gamified green human resource management path; the AI-driven green human resource personalization path), and green organizational climate moderation is significant. In addition, green organizational climate and environmental values strengthen key relationships in the model, highlighting the importance of contextual and individual factors. By integrating ability-motivation-opportunity theory, the knowledge-attitude-behavior model, and social cognitive theory within a unified framework, this study contributes to the green human resource management literature and provides practical insights for human resource leaders seeking to design technology-enabled sustainability learning systems.
Research background: The link between environmental, social, and governance (ESG) performance and bank financial stability is of high academic and regulatory interest, yet global evidence is mixed. Clarifying this relation is important for resilient banking systems under rising sustainability pressures. Purpose of the article: To examine the association between ESG performance and bank financial stability, assessing both composite scores and the individual pillars, and documenting heterogeneity across bank financial stability, ESG profiles, and economic conditions. Methods: The analysis draws on a global panel of 4,466 bank-year observations from 688 banks across 84 countries over 2013-2024. ESG data come from MSCI ESG Ratings, and bank financial stability is measured using the natural logarithm of the Z-score. Baseline estimates use fixed effects with bank, country, and year effects. To account for persistence and potential endogeneity in bank financial stability, we additionally estimate dynamic panel models using the two-step Arellano-Bond GMM estimator. Findings & value added: The composite ESG score is positively associated with bank financial stability, but the effect is economically and statistically negligible in fixed effects models. Pillars diverge: governance is positive and significant, social is negative, and environmental shows no clear link. Heterogeneity is pronounced, with ESG aligning with higher bank financial stability mainly among already stable banks, while fragile banks face adverse associations. During the COVID-19 period, the social pillar improves toward neutral or mildly beneficial, while the governance effect weakens. Dynamic GMM yields a stronger positive composite association and uniformly positive pillar effects, suggesting static models understate benefits due to endogeneity and persistence. The central contribution of this paper lies in its reconceptualization of the ESG-bank financial stability relationship as fundamentally state-and capacity-contingent. By demonstrating that ESG functions not as a universal remedy but as a conditional strategic asset that benefits financially robust institutions, and by revealing how pillar-specific effects exhibit distinct shifts during systemic crises, this paper provides a novel dynamic framework that advances both theoretical understanding and the practical design of risk management and prudential supervision in an era of escalating global uncertainty.
Research background: Environmental disclosure has emerged as both a financial signal and an ethical commitment in sustainable finance. However, in emerging economies with weak institutional enforcement, it remains unclear whether environmental transparency improves debt financing outcomes. Purpose of the article: This study examines whether environmental information disclosure enhances firms' debt capacity and debt maturity in pollution-intensive industries, where environmental risk and information asymmetry are high. Methods: The analysis uses an unbalanced panel of 212 pollution-intensive firms listed on the Dhaka Stock Exchange during 2010-2023. Firm fixed-effects regressions with year controls and firm-clustered standard errors address unobserved heterogeneity and serial correlation. To mitigate potential endogeneity, the study incorporates lagged disclosure measures and dynamic specifications as robustness checks, thereby strengthening identification and reducing reverse-causality concerns. The framework draws on Information Asymmetry Theory, Stakeholder Theory, and the Natural Resource-Based View, with weak institutional enforcement modeled as a contextual boundary condition. Findings & value added: Higher environmental disclosure significantly increases debt capacity but has no unconditional effect on debt maturity. Financial performance strengthens the positive association between disclosure and debt capacity, indicating that profitability enhances disclosure credibility. The limited maturity effect reflects conservative lending practices in weak enforcement environments, where short-term financing structures persist despite improved transparency. By analyzing environmental disclosure under weak institutional enforcement, this study contributes to the sustainable finance literature beyond a single-country case. Methodologically, it combines fixed-effects estimation with lag-based and dynamic approaches to reduce bidirectional bias. The findings show that disclosure reduces information risk and improves debt access, but its effectiveness depends on institutional quality and firm-level financial strength. The results generalize to bank-dominated emerging markets, where disclosure credibility is constrained by monitoring capacity and enforcement limitations. These insights inform regulators and financial institutions seeking to strengthen disclosure frameworks and integrate environmental risk into lending decisions.
Research background: The social capital of failed entrepreneurs affects resource integration during re-entrepreneurship and determines the quality and competitiveness of subsequent entrepreneurial activities. However, existing studies have not yet clarified the possible paths of influence and boundary conditions linking the social capital of failed entrepreneurs and re-entrepreneurship quality. Purpose of the article: This study investigates whether the increased social capital possessed by failed entrepreneurs can enhance re-entrepreneurship quality and competitiveness, as well as the mediating effect of entrepreneurial self-efficacy in the above relationship, and further explores how business dynamics moderates these effects. Methods: Drawing on social capital theory and the resource-based view, this study utilizes a large-scale cross-national sample of 6,805 failed entrepreneurs from 79 countries in the Global Entrepreneurship Monitor database from 2010 to 2020. A multiple regression model was employed to examine the U-shaped relationship between social capital and re-entrepreneurship quality, while mediation and moderation analyses were applied to test the roles of entrepreneurial self-efficacy and business dynamism. Findings & value added: Results demonstrate significant U-shaped relationships between social capital and both re-entrepreneurship quality and entrepreneurial self-efficacy. Specifically, social capital exerts a negative effect before reaching a critical threshold, beyond which its influence becomes positive. Entrepreneurial self-efficacy partially mediates this U-shaped relationship, while business dynamism strengthens the positive U-shaped effect of social capital on re-entrepreneurship quality and renders the U-shaped curve steeper. The primary value added of this study lies in revealing these nonlinear mechanisms and context-dependent conditions, thereby providing deeper theoretical insights into how failed entrepreneurs can effectively leverage social capital to enhance re-entrepreneurship quality and competitiveness.
Research background: The global competition for talent has made talent management (TM) a critical organizational challenge. TM systematically attracts, identifies, develops, retains, and engages talent. Organizations are increasingly integrating environmental sustainability into TM and focusing on green talent management (GTM) to hire, develop, and retain the right talent for green workplace initiatives. Purpose of the article: The study examines whether GTM contributes to employee retention through employees' competency development and top management support. Methods: A quantitative, cross-sectional survey was conducted among 500 employees (faculty and administrative staff) working in higher education institutions (HEIs) in Oman. SmartPLS (4.0) software was used to test the study's proposed hypotheses. Findings & value added: The results supported three direct hypotheses. Top management support had a significant moderating effect on the relationship between GTM and green competency development. Additionally, green competency development mediates the relationship between GTM and employee retention. This study contributes to the emerging field of green human resource management (GHRM) by highlighting its role in retaining talented, environmentally conscious employees. GHRM is an organizational function that is planned and aligned with its environmental strategies. The findings imply that organizations that hire, train, and develop environmentally conscious talent with the support of top management enhance employee commitment and long-term retention.
Research background: The rapid adoption of digital transformation technologies has reshaped corporate operational and governance practices, with important implications for ESG performance, particularly in China. While digital transformation can enhance transparency, efficiency, and sustainability outcomes, its effectiveness largely depends on companies' innovative capabilities. Green innovation serves as a key moderating mechanism that enables companies to leverage digital technologies to achieve superior ESG performance by supporting environmentally sustainable processes and responsible governance practices. Purpose of the article: This study aims to investigate the moderating role of green innovation in the relationship between digital transformation and ESG performance from the perspec-tives of stakeholder theory and dynamic capabilities theory in the context of China. Methods: The study utilised data comprising 33,411 observations from non-financial A-share companies listed on the Shenzhen and Shanghai stock exchanges for the period 2013-2023. Findings & value added: The study contributes new insights by examining the moderating role of green innovation in the relationship between digital transformation and ESG perfor-mance from the perspectives of stakeholder and dynamic capabilities theories. The results show a positive moderating effect of green innovation on this relationship. This enables a deeper understanding of how companies can leverage green innovation to enhance corpo-rate sustainability and meet stakeholder expectations. Although this study examines firms in China, its findings have significant implications for corporate decision-makers, researchers, and policymakers, both within China and internationally, demonstrating how organisations across different institutional contexts can strategically align digital transformation initiatives with sustainability objectives.
Research background: The rapid diffusion of generative artificial intelligence (GenAI) is transforming how employees perform and redesign their work. However, limited research has examined the psychological and contextual mechanisms through which perceived technological support translates into proactive job adaptation. This study draws on the conservation of resources (COR) theory, and social cognitive theory (SCT), to investigate how perceived GenAI support fosters job crafting. It focuses on self-efficacy toward GenAI as a mediating psychological resource and GenAI usage as a boundary condition that shapes this process. Purpose of the article: This study examines the mediating role of self-efficacy toward GenAI in the relationship between perceived GenAI support and job crafting, as well as the moderating effect of GenAI usage on this relationship. Methods: A two-wave, time-lagged survey was conducted with South Korean employees who had experience using GenAI at work. Path modeling with bootstrapping was used to test the mediation, moderation, and moderated mediation effects while controlling for demographic variables, industry, and affectivity. Findings & value added: Perceived GenAI support was positively associated with job crafting through self-efficacy toward GenAI. Furthermore, GenAI usage strengthened the effect of perceived GenAI support on self-efficacy and the indirect effect of perceived support on job crafting through self-efficacy. This study advances the literature by clarifying how external technological support is internalized into psychological resources and activated through usage to promote proactive work redesign. The findings provide a refined theoretical explanation for adaptive employee behavior in AI-enabled workplaces by extending COR theory and SCT to human-AI collaboration contexts and highlighting the central role of domain-specific self-efficacy.
Research background: The trajectory of virtual shopping has evolved beyond twodimensional interfaces toward highly immersive, three-dimensional environments. As the retail sector pivots into the metaverse, delineating the fundamental shifts in consumer behav-Copyright (c) Instytut Bada & nacute; Gospodarczych / Institute of Economic Research (Poland)
Research background: With the rapid development of new-generation digital technologies, including big data, blockchain, and artificial intelligence (AI), the deep integration of AI into traditional industries has induced unprecedented economic changes. The development of AI technologies requires complementarity between capital and highly skilled employees, and the optimization of enterprise human capital structures warrants investigation; therefore, AI has significant potential to improve employees' human capital and enhance enterprise productivity. This study reveals the economic consequences of AI, examines the internal logic of optimizing enterprises' labour structures amid AI's rise, and explores how the application of AI affects enterprise productivity at the enterprise level, as well as the role of upgrading human capital in this process. Purpose of the article: This study aims to explore how AI application impacts enterprise productivity, as well as the mechanism by which such an impact is achieved through the enhancement of human capital. To begin with, we integrate advanced academic research to explore the theoretical relationships among AI applications, human capital upgrading, and enterprise productivity, thereby clarifying the inherent drivers and practical methods for boosting enterprise productivity. Subsequently, the study employs a dataset of 5,167 listed companies from the Shanghai and Shenzhen A-share markets to investigate how the adoption of AI affects corporate productivity through the lens of human capital enhancement. Furthermore, we offer strategic recommendations to adjust the workforce configuration and boost corporate efficiency by improving human capital quality and promoting AI applications. Methods: Using panel data from 3,646 Chinese A-share listed companies for the period 2011-2024, the study applies machine learning methods to generate an AI dictionary and investigates the relationships among enterprise AI application, human capital upgrading, and productivity. Findings & value added: Theoretical exploration indicates that AI applications will increase demand for high-skilled labour and crowd out some low-skilled labour to optimize the human capital structure, thereby improving enterprise productivity. A mechanism test reveals that AI applications enhance enterprise productivity by upgrading human capital. Heterogeneity analysis suggests that the impact of AI application on productivity is more significant for non-state-owned, small-and medium-sized, and non-technology-intensive enterprises. This research applies machine learning methods to generate an AI dictionary from a text-analysis perspective and constructs AI application indicators at the micro-enterprise level. At the same time, from the perspective of human capital upgrading, the study analyzes the impact of AI applications on enterprise productivity and provides a more reasonable explanation for the improvement of enterprise production efficiency in the AI era.
Research background: Many studies on the effects of economic activities on the environment indicate that our world has exceeded many critical thresholds, and that the biosphere's future is at risk. In addition, the ability of our planet to renew its natural resources and heal itself keeps our hopes for our future alive. Purpose of the article: The impacts of economic activities on environmental quality are typically measured according to the Environmental Kuznets Curve Hypothesis. However, a new variable called the load capacity factor (LCF) is used to comprehensively assess these impacts. Therefore, this study investigates the impact of economic activities on environmental sustain-ability through the load capacity factor (LCF), a comprehensive indicator calculated as the ratio of biocapacity to ecological footprint for the Bucharest Nine (B-9) countries (Bulgaria, Czechia, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, and Slovakia), using annual data for 2001-2022. Methods: The research employed feasible generalized least squares (FGLS) and panel-corrected standard errors (PCSE) to reveal the long-run relationship between variables such as economic growth, energy efficiency, human development, and renewable energy (REN) consumption. Findings & value added: Our empirical findings reveal that economic growth and energy efficiency reduce the LCF, indicating increased environmental degradation. Conversely, human development and renewable energy consumption have positive impacts on LCF. Additionally, our results confirm the load capacity curve (LCC) hypothesis, suggesting a U-shaped relationship between income level and LCF. Our results suggest that policymakers and administrators in the B-9 countries must take concrete steps to support human development, ensure inclusive economic growth, and improve renewable energy policies. Since energy efficiency alone cannot provide the necessary savings, it is recommended that efficiency improvement policies be supported by energy-saving practices.
Research background: The European Union's transition towards a green and sustainable-growth paradigm has been accelerated by flagship initiatives such as the European Green Deal, the NextGenerationEU Recovery Plan, and the Just Transition Mechanism. While the literature increasingly addresses renewable energy, energy efficiency, and circular economy dynamics, empirical evidence on how climate-related investments, public environmental expenditure, technological innovation, and structural disparities interact to shape competitiveness remains limited. Existing studies often analyze these factors in isolation, overlooking their systemic interdependencies and the heterogeneity among EU member states. Purpose of the article: This study examines how green transition investments shape economic competitiveness and circular value-added in the European Union. It evaluates the effects of climate mitigation spending, environmental protection, energy efficiency, renewable integration, and R&D on sustainable performance while considering national heterogeneity. By capturing these interdependencies, the article advances the theoretical and empirical understanding of the green transition. Methods: A quantitative econometric framework based on panel data for the 27 EU member states over the period 2010-2023 was employed. The analysis relies on official Eurostat indicators capturing climate investment, public environmental expenditure, energy intensity, renewable energy share, gross R&D expenditure, and aggregate green capital. Findings & value added: The analysis demonstrates that targeted green investments and R&D foster circular economic growth, whereas energy inefficiency and poorly integrated renewables constrain performance. Cross-country disparities highlight uneven institutional capacities, pointing to the need for adaptive yet coordinated EU strategies to ensure climate neutrality and sustainable competitiveness. The study also offers a generalizable theoretical perspective by illustrating how green investment, innovation capacity, and energy efficiency jointly shape circular economic performance, providing long-term relevance for sustainability transitions across diverse economic contexts.
Research background: Enterprise generative, multimodal, and agentic artificial intelligence (AI) technologies facilitate transformative productivity and workforce adaptation gains in innovative organizations, redesigns autonomous team and talent management for workforce and job rotation planning, skill development, and career paths, handle context-specific collaborative business processes, workflows, and decision-making, and augment multi-agent system scaling for labor productivity and operational efficiency, redefining agile and adaptive organizational performance in dynamic business environments, driving interoperable big employee data and strategic decision management, and creating strategic fluidity and synchronized digital labor for sustainable business value. Connected and interoperable agentic AI systems can carry out multistep tasks autonomously, reduce operational costs and unemployment rates, and manage big data-based organizational workflows and management pipelines, driving business value creation and productivity gains, reallocating digital labor, and redefining employee experiences and labor markets in terms of job loss and creation by upskilling and retraining. AI labor impacts predictions are based on multimodal data and labor force productivity modeling in relation to how job and skill creation can affect economic conditions and workforce development, while driving business model transformation. Purpose of the article: We aim to clarify whether enterprise generative, multimodal, and agentic AI-based task automation and machine performance complements technology-driven employment changes and algorithmic efficiency, resulting in workforce reduction and competitive pressures due to economic incentives in terms of how i) deep reinforcement learning algorithms can build digital agentic workflows for autonomous Internet of Things (IoT) sensor-based industrial robotic machines, leading to employment relation, personnel retention and recruitment, work reorganization, and labor productivity optimization, engaged productive staff flexibility and autonomy, and job performance and satisfaction, ii) how task automation and augmentation disrupt labor markets and reshape workforce for either more layoffs or more new hires, predicting both increased or lower wages, high or decreased unemployment, and job creation or elimination, and iii) how computer vision-based task automation and augmentation technologies redesign business-critical workflows and workforce upskilling processes across collaborative enterprise IoT and sluggish hiring environments for task automation and augmentation, streamlining personalized human resource support, resource efficiency, and enterprise productivity, driving economic growth. Methods: A quantitative literature review of ProQuest, Scopus, and the Web of Science databases was carried out and the most relevant research published between 2024 and 2025 was identified and analyzed. The Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) and the web-based Shiny app were harnessed for search results and screening. Dimensions (for bibliometric mapping) and VOSviewer (for layout algorithms) were the deployed data visualization tools. Evidence synthesis screening software and reference and review management tools leveraged included AMSTAR, CADIMA, DistillerSR, JBI SUMARI, MMAT, Nested Knowledge, PICO Portal, and SRDR+. Findings & value added: The main value added derived from the systematic literature review is that enterprise generative, multimodal, and agentic AI system applicability correlates with occupational task operation completion, wage, employment prospects, and education, driving business choices and transformation, labor markets, and economic growth. The benefits for theory and current state of the art are that enterprise generative, multimodal, and agentic AI-based flexible work arrangements and increased employee tracking for organizational and workforce performance can improve job quality while reducing pay inequity, staff absenteeism, job turnover, and widespread unemployment, affecting labor markets and resulting in long-term business values and outcomes. Occupational AI and computer vision technologies impact predictions with regard to work activity automation and augmentation in terms of job loss, labor productivity, and wage raising or lowering. Policy implications reveal that employee productivity and performance tools entail job displacement and creation, requiring emerging workforce reskilling or upskilling for talent attraction, retention, progression, and promotion across structural labor market transformation.
