Business, Economics and Management

Abstract: Digitalisation is increasingly viewed as an important instrument for narrowing the VAT gap in the European Union, yet it remains unclear which forms of digitalisation are most closely associated with improved collection. This study distinguishes between two channels: the voluntary adoption of cloud accounting by firms and the mandatory digital reporting of transactions to tax administrations. The analysis uses an unbalanced panel for the EU-27 covering 2013–2024, with the main econometric estimations restricted to 2013–2023. Data from the European Commission, Eurostat and the World Bank are examined through sequential pooled OLS specifications, two-way fixed-effects models and robustness checks. The initially negative association between cloud accounting and the VAT gap disappears after income and government effectiveness are controlled for. Mandatory digital reporting, by contrast, is associated with a VAT gap approximately 3–4 percentage points lower across the main specifications. However, the limited number of adopting countries and the presence of a pre-adoption downward trend restrict causal interpretation. The findings suggest that digitalisation is more closely related to VAT collection when it provides tax administrations with timely, structured and verifiable transaction data. Voluntary business digitalisation may support compliance, but it does not substitute for mandatory administrative reporting.

Abstract: Tax administrations increasingly rely on data-driven risk models to prioritize enforcement and outreach resources, yet machine learning applications to personal income tax arrears remain scarce compared with corporate-tax and fraud-detection contexts. This study introduces TaxMind, a machine learning framework for predicting tax payment arrears, operationalized as escalation to mandatory (forced) collection, comparatively evaluating five classifiers and applying Shapley-value techniques to interpret the resulting model. A dataset of 16,010 administrative tax records was cleaned by removing 5,385 exact duplicate rows (33.6% of raw data), yielding 10,625 records across 9,264 taxpayers; a taxpayer-grouped train/test split prevented information leakage between related observations. Decision Tree, Random Forest, XGBoost, Support Vector Machine, and a Multi-Layer Perceptron were trained on eight demographic and fiscal attributes, with a signed-logarithm transformation correcting skew in the tax-amount field; interpretability was assessed using Shapley Additive Explanations (SHAP). XGBoost achieved the best discrimination (AUC = 0.779, accuracy = 70.2%, F1 = 0.699), outperforming the other models compared. SHAP identified tax amount, tax category, and taxpayer age, a socio-demographic attribute, as dominant predictors, diverging from gain-based rankings. These findings indicate that duplicate contamination and non-grouped validation can distort administrative model evaluations, and that debtor-related attributes carry under-exploited predictive value for tax segmentation.

Abstract: Artificial intelligence (AI) is entering audit workflows while sustainability reporting and assurance expand the volume, variety and uncertainty of information subject to professional evaluation. The policy question is whether firms disclose governance arrangements that keep AI-assisted work human-led, reviewable and accountable. This exploratory study analyses the complete cross-section of the 2024 UK transparency reports of Deloitte, EY, KPMG and PwC, coded against seven pre-specified dimensions of AI–judgment governance and aggregated into a transparent, replicable AI–Judgment Governance Disclosure Index (AI-JGDI). The analysis is triangulated with the UK Financial Reporting Council's 2024 inspection results and interpreted against the ISAs, the IESBA Code, the EU Artificial Intelligence Act, the NIST AI Risk Management Framework, CSRD/ESRS, IFRS S1 and S2, and ISSA 5000. All four firms dis-close deployed AI capabilities and explicitly retain human professional responsibility; disclosure is strongest for human oversight, governance ownership and training, and least consistent for AI-specific validation and for explanations that would allow an external reader to reconstruct how an AI output affected an audit judgment. AI-JGDI scores range from 71.4 to 100.0. The study contributes a public-document method, a disclosure index and a governance framework for accountable AI-assisted judgment, and identifies limited explicit integration between AI governance and sustainability-assurance methodology.

Abstract: This conceptual article examines how artificial intelligence can support sustainability assurance in the transition from voluntary ESG disclosure to regulated, assurance-oriented sustainability reporting. The study is situated in the context of the Corporate Sustainability Reporting Directive, the European Sustainability Reporting Standards, ISSA 5000 and the EU Artificial Intelligence Act. It argues that AI can strengthen ESG verification by supporting disclosure identification, ESRS mapping, anomaly detection, consistency checks, greenwashing risk screening, external data triangulation and working-paper documentation. At the same time, AI introduces specific assurance risks, including data quality risk, reliability and hallucination risk, explainability risk, bias risk, overreliance risk, documentation risk, confidentiality risk, boundary and materiality risk, and accountability risk. The article develops a Responsible AI-Assisted Sustainability Assurance Framework that integrates ESG data inputs, AI analytical procedures, assurance risk assessment, human professional judgement, validation controls and documented assurance outputs. The central conclusion is that AI should be used as an analytical support layer within a human-in-the-loop assurance process, not as an autonomous source of assurance conclusions.

Abstract: This study responds to the call to apply the Impression Management Narrative Reporting (IMNR) Index to a larger sample, extending its application to the top 40 JSE-listed companies in South Africa. While prior literature has established the presence of impression management (IM) in narrative reports of JSE-listed companies, no study has quantitatively measured its level using multiple IM tactics combined into a single metric. Using content analysis, secondary data were collected from CEO letters to shareholders in annual or integrated reports. A purposive sample of 26 companies was analysed following the IMNR Index methodology. The findings reveal a median IMNR score of 5.20 (out of 8), indicating high levels of IM. Tone manipulation emerged as the dominant tactic (median 0.90), followed by rhetoric (0.83) and readability (0.81), while comparison was the least used tactic (0.38). These results confirm that CEOs of JSE-listed companies engage in IM strategies, consistent with Agency Theory and Signalling Theory. The study demonstrates the practical relevance of the IMNR Index in emerging market contexts and contributes to the literature by providing quantitative evidence of IM prevalence in CEO communications. Investors should approach CEO letters with scepticism, as high IM levels and tone manipulation may obscure underlying performance. Regulators should consider enhanced disclosure guidelines and potential assurance requirements for narrative reports. Audit committees must exercise active formal oversight to ensure integrity, balance, and faithful representation in these disclosures.

Abstract: Purpose. This study proposes a two-stage methodology for using drone-captured inspection imagery as an audit-quality signal in detecting earnings management. The study integrates a computer-vision pipeline that verifies physical inventory with a financial-statement analysis grounded in the discretionary accruals (DACC) literature. Design. Stage 1 applies a pretrained YOLOv8 object-detection model to oblique low-altitude drone imagery of vehicle storage facilities to produce an Inventory Verification Discrepancy Score (IVDS). Stage 2 estimates Modified-Jones discretionary accruals on a panel of 281 firms (1,048 firm-year observations, 2011–2022) and tests how IVDS relates to DACC, with inventory intensity as a theoretical moderator. Robustness checks include firm and year fixed effects and the exclusion of financial firms. Findings. YOLOv8n achieves 95.5% mean vehicle-detection accuracy on oblique drone imagery without domain fine-tuning, establishing the feasibility of off-the-shelf models for inventory verification in audit contexts. In the financial panel, DACC is significantly higher in inventory-intensive firms (t = 6.02, p < 0.001) and in non-Big4-audited firms (t = −3.01, p = 0.003) at the univariate level. Multivariate regressions reveal that, after controlling for inventory intensity, the IVDS × HighInv interaction is positive and marginally significant (p < 0.10), providing tentative support for the moderating role of inventory intensity. Firm- and year-fixed-effects specifications confirm a significant Big4 effect (β = −0.008, p < 0.05) that is otherwise absorbed by industry and size controls. Contribution. To our knowledge this is the first study to specify a complete pipeline linking drone-based asset verification to accrual-based earnings management. We provide a methodological foundation for integrating unstructured visual audit evidence with traditional financial-statement analysis, together with empirical evidence on the feasibility of off-the-shelf computer-vision models for low-altitude oblique drone deployment scenarios.

Abstract: This study investigates whether gender diversity on the client-governance and external-audit sides is related to the way key audit matters (KAMs) are communicated. The sample comprises 9,808 firm-year observations for Japanese listed companies during 2021–2023. KAM headings are manually extracted from statutory audit reports and separated into account-level and entity-level matters. Neither female board representation nor the presence of a female signing audit partner is individually associated with a lower KAM count. Their interaction, however, is negatively related to both the number of KAMs and the logged length of KAM narratives. The two stand-alone gender variables are positively related to entity-level KAM disclosure, whereas female board representation is negatively related to account-level KAM disclosure. These patterns indicate that gender diversity is associated with the allocation of disclosure attention rather than with a uniform expansion or contraction of reporting. Joint female involvement coincides with a more concise KAM section, while firm-wide risks remain more likely to receive explicit attention. The findings extend research on audit reporting, corporate governance, and risk communication by showing that gender diversity may influence which audit risks are emphasized and how extensively they are described in a low-KAM institutional setting.

Abstract: The Markets in Crypto-Assets Regulation (MiCA) harmonizes market rules across the European Union, but it does not by itself determine how entities should classify, measure, document, and tax crypto-asset transactions. This study examines whether recent Slovak implementation measures have translated formal harmonization into operational accounting clarity. An anonymous online survey of 34 accounting, tax, finance, and business professionals was analyzed using descriptive statistics, Wilson confidence intervals, a Fisher–Freeman–Halton exact test, Cramér’s V, and thematic coding of open responses. Although 52.9% of respondents viewed legislative development positively, 72.7% of valid respondents considered current valuation rules inadequate, 60.6% reported that reforms had not increased accounting clarity, and 60.6% perceived greater uncertainty. Tax obligations (50.0%) and record-keeping and documentation (35.3%) were affected more often than bookkeeping mechanics (14.7%). Practical experience was strongly associated with monitoring legislative change (exact p < 0.001; Cramér’s V = 0.62). The findings reveal a regulatory–operational clarity gap: legal taxonomy and market supervision have advanced faster than implementable valuation and documentation guidance. A valuation hierarchy, standardized audit-trail requirements, transaction-specific examples, coordinated accounting–tax guidance, and proportionate support for smaller entities are recommended.

Abstract: The European Union has emerged as the world’s leading region in green bond issuance, yet market development across Member States remains markedly uneven. This article investigates the factors underlying these differences by distinguishing between two dimensions: entry into the sovereign green bond market and the depth of the green debt market as a whole. Using a balanced panel of the EU-27 for 2021–2025, constructed from ECB Securities Issues Statistics (CSEC) and Eurostat data, market entry is modelled using a Probit specification, while market depth is examined using a Tobit specification that accounts for countries remaining outside the market. The results show that the two dimensions are driven by different factors. Sovereign market entry is associated primarily with economic scale, whereas the fiscal position has no detectable effect on participation. Market depth, by contrast, is associated not only with economic scale but also with a stronger budget balance, suggesting that green debt markets deepen where fiscal credibility already exists rather than serving as a substitute for it. No statistically significant negative relationship is identified between environmental taxation and green debt market depth, providing no evidence of substitution and remaining consistent with the use of the two instruments as distinct and potentially parallel policy tools. The principal policy implication is that efforts should focus on reducing market-entry costs and strengthening institutional capacity in smaller Member States.

Abstract: This study examines the relationship between environmental, social, and governance (ESG) disclosure and corporate tax avoidance among Vietnamese non-financial listed firms (2020-2024). Using panel data regression, we find the composite ESG index and its individual environmental (E), social (S), and governance (G) dimensions are negatively associated with tax avoidance. Crucially, we identify two distinct moderating effects. As expected, financial constraints weaken the mitigating impact of ESG. More intriguingly, we uncover a highly surprising finding: state ownership also significantly attenuates this effect, highlighting complex institutional nuances in emerging economies. Furthermore, by developing machine learning models to forecast tax avoidance, we demonstrate that incorporating ESG variables substantially improves predictive accuracy compared to baseline models. This research contributes novel evidence from an evolving market, offering practical implications for policymakers, investors, and firms regarding the interplay of sustainable governance, concentrated ownership, and tax transparency.

Abstract: This study examines whether ESRS-based ESG disclosures capture a life cycle perspective and whether they provide sufficiently risk-relevant information for sustainable finance decision-making. Drawing on Life Cycle Sustainability Assessment (LCSA), the paper analyzes the extent to which sustainability reports of major European non-financial enterprises reflect upstream, operational, downstream, and end-of-life impacts. The study applies qualitative comparative content analysis to the 2024 sustainability disclosures of Enel, Unilever, and Siemens AG. The findings indicate that ESG reporting remains predominantly entity-centered, indicator-based, and dimensionally segmented. Although value chain impacts and circular economy initiatives are increasingly disclosed, comprehensive cradle-to-grave integration remains limited. Environmental life cycle elements are more visible than social and economic dimensions, while cross-dimensional integration across life cycle stages is weak. These limitations may reduce the ability of ESG disclosures to capture systemic sustainability risks, including transition risks, supply chain vulnerabilities, and long-term value chain externalities. In response, the study proposes an LCSA–ESRS Operationalization Model based on boundary reconfiguration, stage-based indicator mapping, dimensional harmonization, and an accounting translation layer. The paper contributes to sustainability accounting and sustainable finance by showing how life cycle thinking can enhance the decision-usefulness, risk transparency, and systemic coherence of ESRS-based reporting.

Abstract: There are lots of perennial species that enable multiple harvests over years from one planting in nature. These crops require no repeated tillage and can promote root accumulation, thus leaving rural landowners with time for reproduction and further production, but this model is difficult for complex knowledge and operational difficulty. Focusing on the supplementation of distinctive species in rural household agriculture, this paper sorts out existing problems and compiles a biological resource list including perennial crops and self-reproducing animals. Combined with methods such as using bamboo trellises and other climbing structures to block light for non-crops, a household-based perennial agricultural scheme of "one-time work, continuous harvest" is constructed to ease reproductive pressure and accelerate civilizational development. Studies show that perennial, self-propagating, storable crops allow people to run a food company, avoid repetitive labor, and gain stable family food dividends; some resilient perennial species can gain competitive advantages with simple artificial tools, and combining the innate advantages of plants with the acquired strengths of tools can resist various risks; A diversified species lifespan table helps people plan investment according to species longevity and their own needs, allowing some species to form a cycle where longer lifespan is accompanied by larger root tubers and higher fruit yields.

Abstract: This research applies the principles of human resource accounting (HRA) and intangible asset valuation frameworks under IAS 38/IFRS to examine the relationship between the quality of player performance metrics, human capital metrics, and the quality of their financial reporting on the market valuation of football players and the financial performance of the leading football clubs in Europe. Based on a dual-level database, composed by 20 leading European clubs (club-level) and by 120 players (player-level) in the season 2023/24, the study constructs a performance-adjusted valuation model for estimating the interconnection between on-field statistics (goals, assists, expected goals, defensive actions, and performance indices imposed on a composite measure) and accounting or financial number (transfer fees, amortization charges, intangible asset values, book values) and financial results (ROA, club market valuation). The Outcome of multiple OLS Regression Models using Robust Standard Errors shows that Performance Index is the most important predictor of player market value (max 0.497, p < 0.01) whereas club revenue is the most important predictor of club market valuation (max 0.009, p < 0.01, R2 = 0.879). The market to book ratio analysis shows systematic difference between economic value and accounting book value based on player age, duration of contract signed, and performance indicators (Adj. R² = 0.363). The Moderated regression shows existence of positive moderating relationship between IFRS compliance and Big4 audit quality with on-field performance and financial outcomes. The findings add to the intersection of sports finance, accounting and human capital theory, stressing the inadequacy of current IAS 38 provisions in capturing the true economic value of football players as human capital assets.

Abstract: Background: The advent of agentic artificial intelligence (AI) is a paradigm shift in auditing practices, from the traditional rule-based automation towards autonomous, goal-oriented machines that have the capacity to make decisions without human interaction. Agentic AI systems use large language models (LLMs) as well as multiple agents to perform complex audit procedures with very little human interaction. Objective: Within a literature review in a systematic way, an analysis of the current situation for the application of agentic AI in auditing will be accomplished, the relevant frameworks and architectures, advantages and challenges will be assessed and some recommendations for research and practice will be introduced. Methods: Using the criteria recommended by PRISMA 2020, we performed a systematic search of academic databases (Web of Science, Scopus, IEEE Xplore and ACM Digital Library) and grey literature from the period January 2020 to February 2026. Studies were evaluated using formative inclusion and exclusion rules and finally 47 studies were included in a synthesis. Results: The applications uncovered during the review were among five major areas: (1) financial statement audit (2) internal audit and control testing (3) compliance and regulatory audit (4) fraud detection and risk assessment (5) audit planning and resource assignment. The most important agentic AI frameworks include CrewAI, LangGraph, AutoGen, and audit-focused systems. Major benefits are increased efficiency (40-60% reduction of time), increased accuracy (15-25% reduction of error), continuous monitoring capability and scalability. Critical challenges include limitations of transparency and explainability, regulatory uncertainty, data security, data ownership and accountability, model drift, and data integration. Conclusions: Agentic AI shows transformative potential for auditing in terms of facilitating change from periodic to continuous auditing and reactive to proactive risk management. Successful adoption requires governance frameworks, transparent audit trails, specialized auditor know-how, and regulatory standards.

Abstract: Most audit fee studies treat the relationship between fees and client risk as symmetric. A unit increase and a unit decrease in client risk are assumed to produce equal but opposite fee responses. We examine whether that assumption holds in the U.S. audit market using 4,090 firm-year observations of U.S. listed companies from 2010 to 2022 and a first-difference specification with firm and year fixed effects. The data show that audit fees rise by about 1.06 percent for each one-unit increase in the Audit Analytics Risky Client Score (p < 0.001). The response of fees to risk decreases is not statistically different from zero (coefficient = 0.001, p = 0.708). The implied stickiness differential is 0.0093 (p = 0.058). The stickiness ratio is approximately 0.13. Fees adjust downward at about 13 percent of the rate at which they adjust upward following an equivalent risk movement in the opposite direction. The pattern is robust to a strict definition of risk decreases, holds in both early (2010–2016) and late (2017–2022) sub-samples, and is corroborated by an alternative risk proxy based on loss-status transitions, where fees rise 4.3 percent on entry to loss status and do not adjust on exit. The result has implications for audit pricing models, audit committee oversight, and the way fee dynamics are interpreted by users of audit fee data.

Abstract: Accurate product-to-catalog invoice matching is a foundational internal control critical to financial oversight and audit quality, yet it is often bottlenecked by inconsistent vendor descriptions. Traditional rule-based matching fails to address this "long tail" of supplier heterogeneity, leading to costly manual reconciliation. This study presents an end-to-end system for automated invoice reconciliation. We introduce a novel “augment-both-sides” strategy: catalog entries are proactively enriched with LLM-generated keywords and synonyms before vectorization, while incoming invoice line items undergo query expansion to bridge the semantic gap between vendor terminology and master data. A final LLM-based reranker applies context-aware judgment to produce highly accurate Top-3 match candidates. We evaluate this system using three diverse entity resolution benchmark datasets, Abt-Buy, Amazon-Google and Walmart-Amazon, structured to simulate real-world ERP environments. The system achieves a Top-3 Recall of 93.14% to 97.96% across all domains, effectively narrowing the search space for accounting and auditing professionals from thousands of SKUs to a precise set of candidates. These results demonstrate that the architecture functions as a highly reliable intelligent decision aid, standardizing complex reconciliations, and structuring the reconciliation task for subsequent human verification.

Abstract: Following the global financial crisis, the transition to IFRS 9’s forward-looking Ex-pected Credit Loss (ECL) model has introduced significant implementation complexity, particularly in emerging markets facing data limitations. This study investigates the heterogeneous ECL compliance strategies adopted within the Cambodian banking sector during a period of heightened credit stress, marked by a system-wide non-performing loan ratio of 8.6%. Utilizing a multiple-case study design and replication logic, a quali-tative content analysis was conducted on the 2024 audited financial statements of 13 representative institutions, ranging from market leaders to international subsidiaries. The findings reveal a pronounced technical divide: market leaders utilize advanced internal statistical methods, such as cohort analysis, while international subsidiaries rely on top-down parent-group proxy models to bridge local data gaps. A “macro-correlation paradox” was identified, where certain institutions prioritize faithful representation by excluding macroeconomic variables when statistical links to historical defaults remain weak. Furthermore, a significant transparency gap exists, where granular disclosures are leveraged as strategic communication tools to signal institutional safety. These results suggest that ECL compliance in data-limited environments is a strategic management choice rather than a standardized technical exercise, highlighting the need for regulatory standardization of modeling assumptions to improve inter-bank comparability.

Abstract: This study set out to explore the role that microfinance institutions (MFIs) could play in promoting sustainable environmental practices within Zimbabwe’s arti-sanal gold mining sector. Despite the global economic benefits of the gold mining sector, it poses significant environmental challenges for Zimbabwe, including deforestation, water contamination, and soil erosion. The growing global focus on sustainability has prompted many questions about the role MFIs could play in fostering environmental responsiveness among enterprises in Zimbabwe. This study adopted an interpretivist research approach and a qualitative research design, conducting in-depth interviews with senior personnel from MFIs and artisanal gold miners in the critical gold mining areas of Shurugwi, Gwanda, Masvingo, and Kwekwe. Ten informed participants were purposively selected from these mining areas, and their opinion regarding the role that MFIs could play in promoting environmental responsiveness was examined. The study found that while MFIs' primary role and function is to focus on financial access and poverty reduction, they can also play an essential role in advancing sustainability initi-atives among community members in gold-mining areas. MFIs were found to have the potential to use monetary rewards to encourage environmentally friendly behaviours and to conduct ecological education among members of the gold mining community. However, challenges such as a weak regulatory framework and resource constraints were identified as major obstacles to MFIs' ability to promote environmental respon-siveness among members of gold-mining communities in Zimbabwe. The findings of this study underscore an essential, though underappreciated, role that MFIs could play in promoting environmental responsiveness in Zimbabwe's gold mining communities by incorporating environmental responsiveness into their mission statements. Accom-plishing this goal requires greater stakeholder cooperation and greater governmental oversight. The study advocates that the Zimbabwean government enact policies that integrate environmental management into MFIs' operations, offer incentives for micro-finance products, and create strategic partnerships to deliver technical support and environmental management education to the artisanal gold mining sector to curb growing environmental degradation in Zimbabwe. In addition, it is recommended that enhancing environmental regulatory supervision and creating reliable monitoring mechanisms are essential for preserving the environment for future generations.

Abstract: This study examines how digitalization can reduce the cost and complexity of ESG and circular economy reporting for women-led SMEs within the evolving EU sustainability reporting framework, with a focus on the Danube Region. Using a conceptual accounting approach grounded in EU regulatory documents, academic literature, and prior bibliometric research, it identifies four key challenge do-mains: measurement, valuation, disclosure, and professional judgment. The analysis is complemented by an exploratory empirical extension based on publicly available documents and illustrative cases of women-led SMEs from the Danube Region. It develops an accounting-oriented problem matrix linking these challenges to digital enablers such as data platforms, automation tools, and traceability technologies. The findings suggest that digitalization improves not only efficiency, but also the reliability, auditability, comparability, and scalability of ESG reporting. A conceptual framework is proposed, connecting regulatory drivers, digital accounting capabilities, and reporting outcomes, including improved assurance readiness and access to finance. The paper also provides practical recommendations, including minimum viable ESG datasets and a staged digital adoption approach, alongside policy implications related to harmonized data requests and targeted capacity-building for SMEs. The study contributes by integrating ESG reporting, circular economy, digitalization, and gender-sensitive SME constraints into an explicitly accounting-centered analytical framework.

Abstract: Small and medium-sized enterprises are prone to errors or evasion in areas such as medical insurance reimbursement, pension contributions, and unemployment insurance, which not only harm workers' rights but also impact the sustainability of federal and state public funds. Therefore, an explainable intelligent tax compliance screening system is urgently needed. This paper uses multi-task learning as the backbone, while modeling the shared risk representations of sub-tasks such as wages, pensions, unemployment insurance, and medical insurance contributions. Subsequently, our model combines multi-task learning with time-series anomaly detection to capture overlooked scenarios and structural avoidance in long-term trends. On heterogeneous time-series graphs, SHapley Additive exPlanations (SHAP) contributions are propagated and decomposed according to 'feature-module-time-task,' and combined with gated dependencies and consistency constraints to generate auditable risk paths, achieving an upgrade from anomaly scoring to evidence-chain visualization. Experiments show that this method improves both identification performance and interpretable localization efficiency.