Volume 10 (2026)
Explore peer-reviewed cutting-edge academic research published in Shareholding journal for 2026. Featuring advanced empirical studies on digital assets, generative AI in corporate governance, global tax architecture, and sustainable financial markets.
Generative Artificial Intelligence on Corporate Boards: Fiduciary Duty, Algorithmic Decision-Making, and Shareholder Accountability
Prof. Jonathan Sterling-Vance1, Dr. Klára Nováková2
1Department of Corporate Law and Governance, Oxford Saïd Business School, UK
2Institute of Cybernetics and Economic Policy, Charles University, Prague, Czech Republic
Abstract
The rapid integration of advanced generative artificial intelligence systems into executive decision-making and board-level advisory committees has precipitated unprecedented legal and governance dilemmas across global capital markets. Traditional corporate jurisprudence rests upon the premise of human agency, cognitive discretion, and personal fiduciary responsibility. As multinational corporations increasingly deploy autonomous algorithms to simulate strategic mergers, forecast capital allocation scenarios, and draft governance policies, courts and regulatory bodies face a normative vacuum regarding legal accountability. This paper investigates the intersection of generative AI and fiduciary duty under corporate statutory frameworks in the European Union and Anglo-American jurisdictions as of 2026.
Utilizing a comparative legal analysis combined with case studies of artificial intelligence sandbox implementations across 150 publicly traded corporations, we examine how delegation of strategic deliberation to neural networks alters the duty of care and loyalty. Our findings indicate that while generative models substantially reduce strategic planning latency and enhance data synthesis, they concurrently introduce severe opacity risks, liability shields for negligent management, and potential algorithmic collusion. Furthermore, we propose a novel regulatory taxonomy—the Algorithmic Fiduciary Oversight Model (AFOM)—designed to preserve human ultimate control while capitalizing on machine intelligence efficiencies.
Ultimately, this study contributes to corporate law literature by establishing clear liability boundaries for directors utilizing synthetic intelligence, ensuring that shareholder protection mechanisms evolve in tandem with technological disruption.
Tokenized Equities and Distributed Ledger Settlement: Eliminating Friction in Post-Trade Shareholder Clearing
Dr. Marcus Aurelius Thorne1, Prof. Meiling Zhang2
1Department of Financial Engineering, National University of Singapore, Singapore
2School of Banking and Finance, Frankfurt School of Finance, Germany
Abstract
Traditional securities clearing and settlement infrastructure has long been encumbered by multi-day settlement cycles (such as T+1 and T+2), counterparty credit risks, and complex multi-tiered custodian hierarchies. The maturation of permissioned distributed ledger technology (DLT) by 2026 has enabled institutional tokenization of corporate equity, facilitating atomic delivery-versus-payment (DvP) and instantaneous settlement. This study evaluates the microstructural impact and liquidity implications of tokenized share issuance across experimental public-private market sandboxes in major international financial centers.
Utilizing high-frequency transaction data and queuing theory models, we analyze the reduction in collateral requirements and operational costs achieved when equity shares are represented as smart contract tokens on enterprise blockchains. Our empirical results demonstrate that instantaneous settlement eliminates overnight counterparty risk, reduces required clearing capital by over 65%, and enables real-time capitalization table transparency for corporate issuers. However, the study also highlights significant regulatory hurdles, including cross-border legal enforceability of tokenized shareholder rights, privacy compliance under data protection statutes, and the risk of smart contract exploits.
The paper concludes with a strategic roadmap for stock exchanges and central securities depositories (CSDs) transitioning toward hybrid DLT settlement architectures.
The OECD Pillar Two Global Minimum Tax: Corporate Restructuring, Cross-Border Investment, and Shareholder Yields
Dr. Henrik Van Der Bilt1, Prof. Amara O’Connor2
1Department of Public Economics and Taxation, University of Vienna, Austria
2School of Economics and Political Science, Trinity College Dublin, Ireland
Abstract
The full implementation of the OECD/G20 Pillar Two framework establishing a 15% global minimum corporate tax has fundamentally transformed international tax planning and multinational enterprise (MNE) capital structures as of 2026. For decades, multinational corporations leveraged low-tax jurisdictions and intellectual property holding structures to optimize effective tax rates, boosting post-tax cash flows available for shareholder dividends and buybacks. This empirical study investigates how the enforcement of Pillar Two rules across over 40 implementing countries has impacted corporate effective tax rates (ETRs), cross-border foreign direct investment (FDI) flows, and shareholder payout yields.
Utilizing a difference-in-differences econometric model on a panel of 1,100 multinational firms, we analyze the reallocation of earnings and operational substance following tax harmonization. Our findings indicate that while average ETRs for large MNEs have risen by 4.2 percentage points, the anticipated contraction in dividend distributions has been partially mitigated through operational efficiency gains and supply chain reshoring. Furthermore, firms previously relying heavily on tax havens experienced negative abnormal stock returns upon legislative enactment, whereas enterprises with robust domestic operating margins exhibited valuation resilience.
The study contributes to public finance and corporate finance literature by quantifying the structural shift in corporate tax burdens and offering guidance for investors re-evaluating multinational equity valuations in a post-tax-arbitrage global economy.
Natural Capital Accounting and Biodiversity Metrics in Institutional Portfolio Risk Assessment
Dr. Elena Rostova-Dumont1, Dr. Santiago Morales2
1Department of Environmental Economics, Sorbonne University, Paris, France
2Center for Sustainable Finance, Autonomous University of Barcelona, Spain
Abstract
While carbon accounting and greenhouse gas emissions reporting have become standard components of institutional portfolio risk management, biodiversity loss and natural capital degradation represent equally critical yet historically unquantified systemic risks for global financial markets. Following the institutional adoption of the Taskforce on Nature-related Financial Disclosures (TNFD) standards in 2026, institutional investors are increasingly required to evaluate corporate dependency and impact on natural ecosystems. This paper examines the integration of biodiversity metrics and natural capital accounting into portfolio asset allocation models across European institutional funds.
Utilizing spatial geographic data combined with corporate supply chain mapping for 800 publicly listed agricultural, mining, and consumer goods enterprises, we construct quantitative biodiversity footprint indexes. Our empirical findings indicate that firms operating in high-risk biomes without proactive natural capital preservation strategies face mounting regulatory penalties, supply chain disruptions, and cost-of-capital premiums. Conversely, corporations implementing regenerative resource management exhibit superior resilience against environmental regulatory shocks and maintain higher long-term Tobin’s Q valuations.
The study provides vital quantitative tools for asset managers seeking to hedge against nature-related financial risks while fulfilling fiduciary sustainability mandates.
Private Credit Expansion and Bank Disintermediation: Implications for Corporate Borrowing Costs and Shareholder Returns
Dr. Liam M. O’Malley1, Prof. Beatrix von Stauffenberg2
1Department of Banking and Financial Markets, London School of Economics, UK
2Institute for Advanced Financial Research, Goethe University Frankfurt, Germany
Abstract
The extraordinary post-2022 expansion of private credit markets and non-bank financial intermediation has reshaped the corporate debt landscape as of 2026, supplanting traditional commercial bank syndicated lending for middle-market and leveraged enterprises. While private credit funds offer flexible capital structures and rapid execution speed, their opaque loan covenants, floating-rate structures, and limited regulatory transparency introduce novel macroprudential risks. This paper investigates the empirical consequences of private credit reliance on corporate borrowing costs, financial leverage sustainability, and ultimate equity holder returns across North American and European corporate sectors.
Utilizing a proprietary dataset of 1,500 private debt transactions, we evaluate how borrowers dependent on private credit navigate high interest rate regimes compared to public bond issuers. Our econometric models reveal that while private credit facilitates continued capital formation during periods of traditional banking contraction, it carries significantly higher all-in borrowing costs (spreads plus PIK toggles). Furthermore, highly leveraged firms financed through private credit exhibit elevated restructuring vulnerability when cash flow volatility spikes.
The findings provide crucial insights for corporate treasurers optimizing capital structure financing mixes and regulatory authorities monitoring systemic shadow banking exposures.
Algorithmic Momentum Feedback Loops and Retail Sentiment Cascades in Flash Volatility Events
Dr. Kenji Takahashi1, Dr. Chantal Moreau-Lalande2
1School of Finance and Quantitative Analytics, Waseda University, Tokyo, Japan
2Department of Financial Econometrics, ESSEC Business School, Paris, France
Abstract
The convergence of high-frequency algorithmic momentum trading models and real-time social media retail sentiment aggregation has fundamentally altered intraday market microstructures in 2026. When coordinated retail sentiment spikes coincide with automated trend-following execution algorithms, vicious feedback loops can be triggered, resulting in severe intra-day flash crashes or irrational price surges detached from fundamental asset values. This paper investigates the mechanics of sentiment-induced feedback loops by analyzing tick-level order book data and natural language processing sentiment streams across major global equity exchanges.
Utilizing vector autoregression (VAR) models and Granger causality tests, we demonstrate that algorithmic market makers increasingly incorporate real-time sentiment metrics into their inventory pricing models, inadvertently amplifying retail sentiment cascades. Our empirical analysis identifies specific structural triggers—such as options gamma squeezes and fragmented liquidity pools—that accelerate price dislocations during high-volatility trading sessions. Furthermore, we evaluate the effectiveness of exchange circuit breakers and dynamic collar mechanisms in dampening artificial volatility spikes.
The study contributes to behavioral market microstructure literature by providing quantitative insights into how machine-human interaction shapes modern equity pricing volatility.
Dual-Class Share Structures and Minority Shareholder Expropriation: Empirical Evidence from Tech Unicorn IPOs
Dr. Sarah L. Jenkins1, Prof. Nikolas V. Demetriou2
1Department of Corporate Finance, Manchester Business School, UK
2Department of Financial Management, Athens University of Economics and Business, Greece
Abstract
The prevalence of dual-class share structures featuring super-voting rights for founders and insiders has been a defining characteristic of technology initial public offerings (IPOs) over the past decade. Proponents argue that multi-class voting structures insulate visionary founders from short-term market pressures, enabling long-term capital allocation. Critics, however, contend that entrenched voting asymmetry severely exacerbates agency conflicts, disempowers institutional shareholders, and creates avenues for minority shareholder expropriation. This study provides an updated empirical reassessment of dual-class governance across international stock exchanges from 2020 through 2026.
Utilizing matched-sample panel regressions across 300 technology IPOs, we examine the long-term stock performance, capital expenditure efficiency, and related-party transaction frequency of dual-class versus single-class firms. Our findings reveal a pronounced governance discount that intensifies five years post-IPO, as insider entrenchment correlates with lower return on invested capital (ROIC) and sub-optimal strategic pivoting. Furthermore, we document that sunset provisions—when present—successfully restore market discipline upon expiration.
The paper offers crucial governance recommendations for institutional investors and index providers evaluating voting rights structures in public equity listings.
Machine Learning Forecasting of Systemic Banking Crises Using Macro-Financial Network Spillovers
Dr. Florian G. Becker1, Dr. Mei-Ling Zhou2
1Department of Financial Econometrics, University of Zurich, Switzerland
2Center for Financial Stability and Risk Analysis, National University of Singapore
Abstract
Anticipating systemic banking crises and macro-financial distress remains one of the most vital yet challenging mandates for central bank economists and institutional risk managers. Traditional early warning econometric models often suffer from poor out-of-sample predictive accuracy and failure to capture non-linear contagion pathways across interconnected financial networks. This research paper develops an advanced machine learning early warning system (EWS) that integrates macro-financial network spillover indices with gradient-boosted decision trees and recurrent neural networks (RNNs) as of 2026.
Utilizing a global panel dataset spanning 50 banking systems over four decades, our hybrid models analyze time-varying interbank contagion channels, asset price misalignments, and credit-to-GDP gaps. The empirical results demonstrate that our machine learning network architecture achieves an out-of-sample crisis prediction accuracy rate of 89.4%, significantly outperforming traditional logit models while reducing false-positive rates. Furthermore, SHAP (Shapley Additive exPlanations) value attribution reveals that cross-border wholesale funding freezes and real estate price velocities serve as the primary leading indicators of imminent systemic stress.
These findings offer institutional risk officers and regulatory bodies powerful quantitative tools for proactive macroprudential capital buffer calibration and crisis mitigation.