Business, Economics and Management

Abstract: Sustainable development of the agro-industrial complex (AIC) requires an integrated assessment of investment activity, production performance, financial returns, and resource provision at the district level. This study evaluates the production and economic performance of 13 districts of the North Kazakhstan Region using a composite scoring approach based on seven groups of indicators: investment activity, gross agricultural output, crop yields, livestock productivity, crop-production profitability, livestock-production profitability, and capital intensity. The information base includes official district-level statistics, materials of the regional agricultural authorities, data on fixed production assets for 2021–2023, and national land-use reporting. The results reveal substantial territorial differentiation. Kyzylzhar and Taiynsha districts form the high-performance group, Esil District is classified above average, whereas Ualikhanov District occupies the lowest position in the aggregate ranking. The analysis also identifies a scale-efficiency mismatch: large agricultural territories do not necessarily demonstrate high profitability or productivity, while several medium-sized districts achieve stronger returns on land and production resources. The findings provide a comparative basis for territorially differentiated agricultural policy, targeted investment support, technological modernization, and the development of processing infrastructure in the North Kazakhstan Region.

Abstract: Artificial intelligence is rapidly transforming education globally, yet the economic consequences of AI-enabled education remain insufficiently understood. Existing research has largely examined AI in relation to learning outcomes and instructional efficiency, with limited attention to its implications for human capital formation, productivity, and inclusive economic growth. This study develops and empirically examines a framework linking AI-enabled education with economic outcomes through the formation of human capital and the development of labour-market-relevant skills. Using OECD country-level panel data for 38 countries over the period 2015-2025 (N = 350 country-year observations), we construct a composite AI-Enabled Education Index (AIEI) using Principal Component Analysis. The index incorporates indicators of digital educational infrastructure, AI skills, educational technology adoption, teacher digital competence, and AI-related educational policies. We employ two-way fixed effects estimation with country and year fixed effects, mediation analysis, and robustness tests including system GMM and instrumental variables. The AIEI demonstrates a significant positive association with human capital formation (β = 0.28, p < 0.001). Human capital mediates the relationship between AI-enabled education and labour productivity, with the indirect effect accounting for 45.5% of the total effect (Sobel Z = 4.82, p < 0.001). AI-enabled education is also positively associated with inclusive economic growth (β = 0.28, p < 0.001). Heterogeneity analysis reveals that effects are larger in countries with higher digital infrastructure (difference = 0.20, p < 0.001) and stronger institutional quality (difference = 0.20, p < 0.001). AI-enabled education contributes to human capital formation, productivity, and inclusive economic growth, with human capital serving as a key mediating mechanism. Investments in digital infrastructure, teacher development, and institutional quality amplify these benefits. The findings provide evidence for policy interventions that leverage AI to enhance educational and economic outcomes.

Abstract: People with disabilities experience attitudinal, communication, environmental, and digital barriers when using financial services. This descriptive implementation case study examined a co-designed, three-day disability-inclusion training intervention delivered to 25 staff in a commercial bank in Bangladesh. Study-specific questionnaires assessed self-rated familiarity with disability types and inclusive-banking rules or policies, confidence identifying and addressing barriers, confidence supporting customers with disabilities, and frequency of considering disability inclusion before and immediately after training. Participant-level pre/post records could not be reliably linked, so group-level descriptive comparisons were used. The proportions selecting the two highest response categories were 4/25 (16%) before and 21/25 (84%) immediately after training for familiarity; 12/25 (48%) and 23/25 (92%) for confidence identifying and addressing barriers; and 8/25 (32%) and 20/25 (80%) for confidence supporting customers. The group mean for frequency of considering disability inclusion was 2.5 before and 4.4 immediately after training on a five-point scale. At approximately 9-10 weeks, participant-level follow-up data were available for 20 participants; 15/20 (75%) selected very or extremely confident when asked about initiating interactions with people with disabilities, while self-reported use of inclusive language and avoidance of assumptions varied. The follow-up questionnaire did not repeat the immediate measures and therefore could not assess sustained effects. Reported organizational responses included disability-inclusive wording in a proposed recruitment policy, video-call sign-language support and Braille materials reported as available, and digital accessibility work underway. Findings suggest disability-led training can be implemented in financial institutions, but the uncontrolled design, study-specific self-report measures, unpaired analysis, short follow-up, and lack of independent verification preclude causal conclusions.

Abstract: Contemporary adults report declining social excitement, weakening conversational motivation and a sense of emotional flattening within peer groups, even with the sheer amount of experiential currency available around. This paper argues that this paradox arises not from individual habituation alone, but from what this paper proposes as Collective Hedonic Adaptation (CHA). Experiential homogenization (EH) across social groups systematically depletes the Interpersonal Information Utility (IIU) of shared narratives. This paper introduces the Avocado Paradox framework, integrating Hedonic Adaptation Prevention (HAP) theory, Optimal Distinctiveness Theory (ODT) and Socioemotional Selectivity Theory (SST) into a unified conceptual model linking Experiential Saturation (ES) to declining Conversational Novelty (CN), Social Engagement Quality (SEQ), Perceived Life Excitement (PLE) and Subjective Well-Being (SWB). Six propositions are advanced including Algorithmic Curation as a historically novel accelerant of narrative capital collapse. Implications for consumer experience design, brand differentiation strategy, and well-being research are discussed.

Abstract: Information Technology (IT), Information Systems (IS) and Computer Assisted Audit Techniques (CAAT) tools are common in auditing innovative environment. For instance, CAAT and associated auditing tools potentially enhance quality and enables auditors to achieve efficiency. However, audit firms still struggle to reduce resistance in the use of CAAT and other IT audit tools toward the plain compliance of the related standards. This study performs a comprehensive literature survey and analysis of frameworks of the main actors of the usage of IT audit tools. It toes the ISA 620 standard and the resistance theory. It further performs an in-dept analysis on the IT auditing tools in respect to application to auditing procedures. Findings show that the research concerning CAAT and IT auditing tools is increasing. So also, the networks of authors are going from the more traditional outlook spanning other frontiers. Findings also show that the application of IT tools generate more jobs and insights for the financial auditor while it spurs efficiency. We show that the adoption of the IT audit tools is strategically faced with conflicting priorities of the financial auditor inasmuch as the use of specialist is presently dominating the assurance services and it is hard to share from the acquired cake. The onus of application of innovative tools such as AI whose impact must be shared between the audit firm, and the client is yet under debate. We show that audit firms encourage compliance of ISA 620 where audit partners’ IT background strengthens the effective use of IT audit tools. We identified data from papers, frameworks and models that discusses the use of IT audit tools, in auditing. These constructed data were then categorised by technique, engagement phase, and attributes to enhance their effective usage. The analysis categorises into, IT auditing tools in lieu of dragged efficiency, ISA 620 adoption framework (620AF) resistance, overcoming regulatory and methodology hurdles, and identifying gaps for future research, and classifying topics mostly encountered in the literature. Overall, we contribute to the IT audit literature and practice by showing how engagement partners can be complaint with relevant standards and be less resistant and diffuse IT audit tools. The implications of this study to academia are apart from the traditional authors of IT/IS auditing and tools, other networks of authors need to be incentivised. Policy makers and assurance gatekeepers need to tighten their approaches to enhance compliance of ISA standards in the use of auditing tools in the auditing procedures.

Abstract: To promote the sustainable development of New Energy Vehicles（NEVs）, multilevel governments are formulating an increasing number of policy mixes. A policy mix combines different policy tools from various governance levels, where interactions may occur. Using data from China’s NEV industry, this study creatively divides policies into strategic, supply-side, demand-side and environment-side policies, then further divides the policies into central policies and local policies. Then we investigates how multi-type policies at both the central and local levels affect innovation output and further explores their synergistic effects. The findings show that at the single-level government level, the demand-side subsidy phase-out policies have an inverted U-shaped impact on the innovation output of NEV. And the other multi-type policies exert a positive influence on NEV innovation output. However, when all policies are implemented simultaneously, the involvement of supply-side policy and demand-side policy at the local government level have a policy superposition effect, the effects of local environmental-side policy is crowded out. Synergy is present both within same-type policies and across multi-type policies in multilevel governments. The multilevel governments should pay attention to the coordination and optimization of different types and levels of policies in the NEV industry.

Abstract: Based on Information Processing Theory (IPT), this study examines the impact of suppliers’ digital transformation (SDT) on focal firms’ green innovation (FGI) through supply chain information processing. From an information processing perspective, it identifies three boundary conditions: geographical distance, environmental regulation intensity, and analyst attention, and further analyzes how they moderate the relationship between SDT and FGI. A “focal firm–year–supplier” dataset of Chinese listed firms from 2008 to 2022 was established through data matching. A series of robustness checks and endogeneity treatments were conducted to ensure the reliability of the findings. The results indicate that SDT can significantly enhance FGI by facilitating information transmission and integration across the supply chain. Moreover, this positive effect becomes more pronounced as geographic distance decreases, environmental regulation intensity increases, and analyst attention rises; these findings provide indirect evidence supporting the information-processing in supply chain. By systematically analyzing how SDT promotes innovation performance beyond organizational boundaries through information transmission and integration, this paper reveals a novel type of interaction between suppliers and focal firms, thereby contributing to the literature on the determinants of green innovation and information spillovers.

Abstract: Many economic time series are compositions: educational attainment shares, waste treatment routes, energy mixes, sectoral employment. Forecasters usually transform such series to log-ratio coordinates and apply generic univariate or multivariate methods, imposing no economic restriction on how the parts move. Evolutionary game theory offers an obvious candidate restriction, since replicator dynamics makes the growth rate of a share proportional to its relative payoff advantage. We take that restriction seriously as a forecasting device. A discrete-time replicator-mutator map with linear payoffs and a common congestion term is estimated under three pooling regimes, country-specific, fully pooled, and shrunk between the two, and raced against five log-ratio benchmarks in an expanding-window rolling-origin design on three Eurostat panels. The central finding is that the pooling regime governs accuracy more strongly than the choice between structural and statistical models. Moving from country-specific to pooled parameters reduces the mean absolute scaled error by 27.8% on educational attainment and by 15.3% to 19.8% on municipal waste routes, whereas the best structural specification differs from the best benchmark by only 1.3% on educational attainment. Structural specifications enter the model confidence set on the attainment panel but not on the waste panels, where no-change forecasts dominate. Win rates diverge sharply from mean losses, which suggests combination rather than selection.

Abstract: This study examines entrepreneurs’ awareness and preventive activities regarding the environment and employees as key elements of the Quality 5.0 concept. Quantitative research was conducted among manufacturing companies in Greater Poland that had not implemented or certified a quality management system. Data were collected using CATI interviews based on an original questionnaire assessing the importance and implementation of 24 preventive management factors, with detailed analysis of 10 factors. A quota-random sample of 380 enterprises, stratified by company size, was used. The findings indicate that maintaining good relationships with the environment, ensuring occupational safety and ergonomics, careful supplier selection and continuous evaluation, and improving employees’ qualifications and skills are perceived as the most important preventive management factors. Environmental relations and workplace safety also achieved the highest implementation levels, suggesting their strongest contribution to a preventive management approach. In contrast, participatory management received relatively low importance and implementation ratings, highlighting the need for further investigation. The results emphasize the role of preventive management practices in supporting the Quality 5.0 concept and provide practical guidance for organizations seeking to strengthen sustainable, employee-oriented, and proactive management approaches.

Abstract: The quantification of structural resilience and sub-percentile tail risk represents a major challenge across both corporate financial engineering and modern industrial logistics. Traditional aggregation architectures—principally linear risk matrices and parametric extreme-value copulas—harbor critical, systemic blind spots. This paper highlights a numerical limitation of the Gumbel extreme-value copula in deep-tail regions (F≥0.999). Analytical results indicate that the logarithmic structure of the tail generator produces progressively higher sensitivity near the distribution boundary, yielding an empirical condition number greater than 1,220 at the regulatory 99.9% Value-at-Risk (VaR) threshold. This numerical conditioning issue increases sensitivity to sample noise and data scarcity, resulting in a 36.5% underestimation of systemic tail damage. The proposed model formalizes risk scenarios by mapping multi-node threats as normalized directional unit vectors within a compact 3D vector space. Interactions are then calculated algebraically using geometric invariants—the Dot Product for root-cause convergence and the Cross Product Norm for dynamic, second-order risk resonance—effectively contracting high-dimensional combinations into a stable framework. Rather than treating risks as frame-dependent scalar probabilities, this generalized High-Dimensional Geometric Invariant Operational Risk Framework replaces legacy structures with domain-agnostic invariants capturing dynamic risk resonance and multi-trigger cascades. Empirical results across rugged operational environments, acute data scarcity (Ntrain = 100), and high-dimensional scaling (50 risk factors) demonstrate that the proposed model outperforms standard alternatives by a factor of approximately 13 in out-of-sample predictive accuracy (MSE = 0.08193) while maintaining absolute parametric stability. Furthermore, a Taylor-series tensor contraction successfully collapses 1,275 second-order interactions into just 2 free parameters. This framework bypasses iterative Maximum Likelihood Estimation (MLE) bottlenecks, unlocking real-time, low-latency Monte Carlo stress testing for systemic banking compliance and global supply chain risk governance.

Abstract: Sustainable investing has moved environmental, social, and governance (ESG) criteria toward the center of cross-border capital decisions, yet the country-level evidence on whether these criteria attract foreign direct investment (FDI) rests almost entirely on static models and rarely accounts for the quality of a country’s accounting and reporting environment. This study estimates a dynamic model of FDI for 265 economies observed from 2006 to 2020, combining the three ESG pillars with four accounting and tax variables: mandatory adoption of International Financial Reporting Standards (IFRS), the strength of auditing standards, the extent of business disclosure, and the corporate tax burden. A fixed-effects estimator and a two-step system generalized method of moments (GMM) estimator address persistence and the reverse causality between FDI and national conditions, and a panel smooth transition regression tests whether the relationship is nonlinear. Accounting transparency attracts FDI: audit quality and disclosure carry large positive and significant coefficients under fixed effects and across income groups, while the aggregate governance index enters negatively. The corporate tax burden deters FDI with no evidence of an optimal-tax turning point. The relationship is nonlinear in governance rather than income: a panel smooth transition regression locates a governance threshold near 0.97 on the standardized scale, above which the negative association between IFRS and FDI disappears and the effects of disclosure and sustainability reporting strengthen. The results reframe the ESG–FDI question around the reporting and assurance environment through which investors read a country’s ESG credentials, and around the governance quality that makes that environment credible.

Abstract: This research considers the acceptance of Artificial Intelligence (AI), technological and organizational readiness, willingness to change, and concerns related to AI adoption among employees of local government or-ganizations in Greece. Drawing on the Unified Theory of Acceptance and Use of Technology (UTAUT), extended featuring Organizational Readiness for Change theory and AI-specific individual and organizational determinants, the study puts forward an integrated conceptual framework for examining employees' behavioral intention to adopt AI. The research adopted a quantitative research methodology using a structured online questionnaire dis-tributed through Google Forms to municipalities and public organizations across Greece, yielding 239 valid responses from local government employees. The outcome of the primary research indicates that there is a positive attitude towards AI and a moderately strong intention to use AI-based applications in the workplace, with participants recognizing AI's potential to improve ef-ficiency, service quality, and decision making. Nonetheless, still the technological and organizational readiness remains at a transitional stage, with considerable challenges in digital skills, data governance, strategic planning, and infrastructure. Employees demon-strate preparedness to engage with the organizational changes AI adoption requires, however, concerns continue about job transformation, new skill requirements, and the wider impact of AI on public sector employment.

Abstract: Given the growing global concern about sustainability, green energy has emerged as a key pathway to protecting the environment and ensuring economic stability. The study examined the mediating role of green procurement policy in the relationship between green energy projects and environmental value for Independent Power Producers (IPPs) in Ghana. The study was explanatory and quantitative in nature and underpinned by Stakeholder Theory and Resource-Based View. The study included 153 respondents selected from various IPPs and data collected from them using structured questionnaires. Structural Equation Modelling (SEM) with AMOS was used to analyse data. Results showed that green finance and technology adoption have a significant impact on environmental value, while green procurement policy has a strong positive influence on it. Moreover, green procurement policy was found to partially mediate the relationship between green finance and technology adoption and green value. The study finds that implementation of green procurement systems facilitates the utilization of financial resources and technology to support environmental sustainability in the Ghanaian energy sector. It highlights the urgent need for policies and industry leaders to implement sustainable financing, technology innovation and green procurement as a way of paving the way for Ghana's green energy transition.

Abstract: Background: The Middle Corridor has become a strategically important Eurasian transport route, yet limited research has examined the configuration of intermodal networks for hazardous materials transportation under its unique operational and regulatory conditions. This study develops a decision-support framework for optimizing hazardous materials transport along this corridor. Methods: A multi-objective mixed-integer linear programming (MOMILP) model was formulated to optimize transportation cost, risk, and transit time simultaneously. The model incorporates modal selection, intermodal terminal transfers, infrastructure capacity constraints, border delays, and regulatory compatibility under ADR, RID, and IMDG requirements. A simplified Middle Corridor case study was used to evaluate Pareto-optimal network configurations and sensitivity to safety preferences and infrastructure capacity. Results: The numerical analysis shows that rail-dominant intermodal configurations consistently achieve the lowest transportation risk while requiring only modest increases in transit time. Sensitivity analysis indicates that assigning moderate weight to safety leads to stable rail-oriented solutions, whereas railway capacity limitations force shifts toward higher-risk road transport under increasing demand. Conclusions: The proposed framework provides a transparent and mathematically rigorous approach for balancing safety, efficiency, and cost in hazardous materials transportation. It offers practical decision support for infrastructure planning, intermodal terminal development, and policy evaluation along the Middle Corridor.

Abstract: The COVID-19 pandemic and subsequent economic crises uncovered deep vulnerabilities in social, economic, and institutional systems across the world. In this context, the corporate social responsibility (CSR) and sustainability strategies emerged as strategic tools for resilience and collaboration overcoming the crisis. This study examines the role of CSR and sustainability-oriented approaches in strengthening partnerships between for-profit and non-profit organizations in Slovenia during periods of crisis. Focusing on the various cross-sector collaborations formed or adapted during crises in the context of COVID and other events, the research explores how organizations mobilized resources, shared knowledge, and co-created social value to respond to urgent community needs. Using a qualitative method based on interviews the study identifies the various CSR and Sustainability strategies that companies and NPO/NGOs in Slovenia have adopted to combat the impact of pandemic and subsequent crises in society. Particular attention is given to how sustainability-driven innovative strategies enable longer-term, trust-based partnerships for achieving better organizational outcomes and provide recommendations for companies and NPO/NGOs moving forward. Findings indicate that organizations with embedded CSR and sustainability practices were better positioned to pivot from transactional relationships to strategic alliances with non-profits. These partnerships enhanced crisis response capacity and contributed to economic and social resilience to various crisis situations. Moreover, the crises acted as a catalyst for redefining value creation, shifting the focus from compliance management toward shared impact and collaboration.

Abstract: The Solvency II regulatory framework (Directive 2009/138/EC) defines the Solvency Capital Requirement (Solvency Capital Requirement, SCR) using a Value-at-Risk (VaR) approach with a confidence level of 99.5% and a one-year time horizon. This structure makes regulatory capital extremely sensitive to the characteristics of the right tail of the loss distribution and, consequently, to the effectiveness of risk transfer mechanisms. This study analyzes the impact of the five main types of reinsurance contracts quota share, surplus, quota-surplus, excess-of-loss (XoL), and stop-loss on the transformation of the net loss distribution and the resulting dynamics of the SCR. The empirical analysis is based on a computational experiment using real data from a ten-year insurance portfolio covering the period 2016–2025. The results show that the use of reinsurance leads to a reduction in the total capital requirement in the range of 18.4–23.4% on an annual basis, with the effect exhibiting an approximately linear relationship with the size of the cession quota. The stratified comparative analysis conducted identifies significant differences in the effectiveness of individual contract structures with regard to the reduction of tail risk. In particular, XoL contracts demonstrate the strongest effect on the extreme quantiles of the loss distribution, with a reduction reaching −52.5% at the 99.5% VaR level and −68.6% at the 99.9% VaR level. In contrast, quota-share contracts result in a practically proportional scaling of risk, characterized by a symmetric reduction of approximately −40% across all confidence levels. The results further show that multi-tiered reinsurance programs combining quota share, excess, catastrophe XoL, and stop-loss components, provide the highest degree of capital relief, reaching 48.8%, which indicates the presence of significant nonlinear diversification and complementarity effects among the individual risk transfer mechanisms. A waterfall decomposition was applied to identify the main factors determining the difference between the standard formula and the internal model. The analysis finds that the dominant drivers of the observed capital relief are the effect of precise risk calibration (on average −8.3%) and the effect of diversification (−4.7%). These results underscore the importance of adequately modeling the interdependencies among risk modules and the limitations of standardized regulatory parameterizations. In addition, a “wrong-way risk” stress scenario was developed, involving the simultaneous occurrence of a catastrophic risk and the insolvency of two key reinsurers. Under this scenario, the effectiveness of risk transfer is reduced to −27.3%, and the solvency ratio falls below the minimum capital requirement (12.5%). This result empirically confirms the cautious regulatory stance of the European Insurance and Occupational Pensions Authority regarding the limited recognition of capital reliefs that do not demonstrate resilience under extreme stress conditions. The study provides a quantitatively grounded framework for optimizing reinsurance programs under the Solvency II regime. The main conclusion is that capital efficiency is not a function of a single “optimal” reinsurance contract, but rather results from the strategic combination of various reinsurance mechanisms capable of simultaneously reducing tail risk, improving diversification, and limiting vulnerability to systemic stress events.

Abstract: The transition to a low/non-carbon energy system requires both (i) further electrifying energy services and (ii) increasing the generation of that electricity from nonfossil fuels. We contribute to the nascent but growing literature on how energy prices impact the electrification of energy services by considering OECD country macro panel data and the electrification rates (share of energy consumption from electricity) of the residential and industrial sectors. For both sectors, we find a negative, significant, but relatively small (around -0.1 to -0.07) response for electricity prices and a positive, significant, and small (around 0.1 to 0.05) response for an index of direct use fossil fuel (e.g., coal, peat, natural gas, and oil products) prices. So, the combination of carbon taxes and encouraging renewables in electricity generation could harness both price incentives to in-crease electrification.

Abstract: Household plastic waste management is increasingly shaped by two policy pressures that are not fully captured by conventional closed-loop supply chain design models: the need to internalize carbon emissions and the need to distribute local waste-management burdens fairly across regions. This study extends the decentralized closed-loop supply chain framework to incorporate carbon taxation and regional cost-equity standards into a manufacturer-public-sector bilevel model. The manufacturer chooses virgin-resource purchases, recycled plastic waste-bale purchases, and product shipments, while local public sectors manage collection, transfer, disposal, and interregional waste flows. Carbon costs are assigned separately to manufacturer activities and public-sector activities to avoid double counting. Regional equity is modeled as a proportional bound on deviations from the average disposal burden, converting the local follower system into a regulated decentralized equilibrium. We reformulate the public-sector optimality conditions as an updated mathematical program with equilibrium constraints and use a two-stage computational design: a small-scale continuous MPEC/KKT validation solved with open-source SciPy routines, followed by reproducible normalized policy simulations calibrated to the operational ranges reported in the source study. The results show that carbon taxation increases the real recycling rate from 25.02% to 28.09% before the recycled-bale supply capacity becomes binding, reducing system emissions by approximately 2.73%. At a carbon tax of 30 USD/tCO2, tightening the regional equity threshold from 25% to 5% reduces the standard deviation of regional waste-management costs by approximately 26.1% and slightly increases the real recycling rate. The findings suggest that carbon taxation and equity standards are complementary: carbon pricing strengthens demand for recycled bales, while equity rules redirect recycling efforts toward high-burden regions. The study contributes a sustainability-oriented extension of decentralized plastic waste supply chain design and offers policy guidance for balancing recycling, emissions reduction, and regional fairness.

Abstract: The aim of the article is to analyze the capacity of the Polish third sector to act as a strategic actor in the process of national sustainable transformation. The article examines the central paradox of a numerically vast and dynamic non-governmental organization (NGO) sector, whose potential is systemically limited by structural, financial, and institutional barriers. The research employs a qualitative, descriptive-analytical methodology based on a critical synthesis of secondary data. Methodological triangulation was achieved by integrating diverse sources: peer-reviewed scientific literature, legal acts, national statistical data (GUS), industry reports (Klon/Jawor Association), and organizational case studies. The analysis confirms the existence of the "large but small" sector paradox: although numerically vast, it is highly fragmented and chronically underfunded. The dominant reliance on short-term public grants ("grantoholism") generates deep financial instability, hindering strategic development and independent advocacy. Despite these limitations, NGOs make a key contribution to the implementation of specific Sustainable Development Goals (SDGs), especially in the area of social services (SDG 1, 3, 10), environmental protection (SDG 14, 15), and climate action (SDG 13). The growing importance of the ESG framework presents both a significant opportunity for strategic partnerships and a serious challenge due to the sector's competence and technological deficits.

Abstract: Enhancing the resilience of agricultural industry chains is essential for safeguarding food security and advancing agricultural modernization. Using panel data for 280 Chinese prefecture-level and above cities from 2011 to 2023, this study treats the staggered rollout of the “Broadband Village” Pilot Program and the Universal Telecommunications Service Pilot Program as quasi-natural experiments in rural digital infrastructure and estimates their effects with a staggered difference-in-differences design. The results show that rural digital infrastructure significantly strengthens agricultural industry-chain resilience. This finding is robust to parallel-trends tests, placebo tests, winsorization, alternative sample periods, and estimators that are robust to heterogeneous treatment effects. Dynamic estimates indicate that the effect is delayed and cumulative rather than immediate. Mechanism tests show that the policy significantly promotes agricultural agglomeration and has a weakly significant positive effect on industrial structure upgrading, providing supportive evidence for the agglomeration and structural-upgrading channels. Subsample estimates suggest that the effects are concentrated in eastern China, major grain-producing regions, and major grain-consuming regions. Overall, the resilience gains from digital infrastructure depend on the joint presence of an adequate industrial base, digital adoption capacity, and efficient circulation and logistics systems.