| Author: | Yu, Jia |
| Title: | Risk management in buyer-guaranteed supplier finance : fraud, regulation, and interpretable AI |
| Advisors: | Ng, C. T. Daniel (LMS) Lo, K. Y. Chris (LMS) Cheng, T. C. Edwin (LMS) |
| Degree: | Ph.D. |
| Year: | 2026 |
| Subject: | Business logistics -- Risk management Artificial intelligence-- Industrial applications Corporations -- Finance Supply chain management Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Logistics and Maritime Studies |
| Pages: | xiii, 214 pages : color illustrations |
| Language: | English |
| Abstract: | Buyer-guaranteed supplier finance (BGSF) has become a key enabler of global trade by alleviating capital constraints for upstream suppliers. However, recent financial failures, such as the collapse of Greensill and Evergrande, have exposed significant vulnerabilities with complexity in decision making in BGSF systems, resulting in substantial losses for financial service providers (FSPs), supply chain participants, and broader society. In response, regulatory frameworks, like Basel III, have imposed strict risk management requirements on FSPs. On the other hand, artificial intelligence (AI) is emerging as an effective tool for real-time decision-making in complex environments. Despite this development, existing research predominantly focuses on supply chain actors, largely overlooking how AI can be applied to the regulatory obligations of FSPs and their strategic interactions with supply chain participants. Failing to bridge this divide, by not integrating FSPs regulatory obligations with real-world supply chain dynamics, leaves the entire system exposed to unmanaged risks, guaranteeing further large-scale financial crises and threatening the entire economic stability. To shed light on this underexplored area, this thesis develops integrated risk management frameworks that jointly consider the decisions of FSPs, suppliers, and retailers under Basel III with an actional decision-making tool based on interpretable AI. It aims to examine how regulatory-compliant risk management practices affect supply chain decision-making, financial stability, and broader economic outcomes. To achieve this objective, three important issues are studied in this thesis. In the first study, we investigate the risks inherent in buyer-backed purchase-order financing (BPOF) scheme caused by the fraud of the guarantor. This study proposes a decision-making framework, for both FSPs and supply chain parties, that incorporates risk management strategies when offering the BPOF scheme to unreliable supply chain participants with supplier's operational disruption in the presence of fraud on the retailer's bankruptcy risk. By employing expected shortfall (ES) and minimum capital reserves as risk control measures required by Basel III, we investigate the equilibrium decisions and outcomes for a bank and supply chain parties under three financing mechanisms based on these measures in the BPOF scheme: non-mechanism for guarantor fraud, screening mechanism, and checking mechanism. We also evaluate the performance of these mechanisms in controlling risks and optimizing social welfare. Our findings highlight the bank's risk controls for ensuring only authorized risks exist in the BPOF scheme, improving the reliability and robustness of the overall financing system including the FSP and the supply chain parties. Contrary to intuition, we show that the BPOF scheme can even improve the social outcomes in the presence of guarantor fraud under certain circumstances. We identify the conditions under which the BPOF scheme should be applied and determine which mechanism should be employed to handle risks caused by guarantor fraud and to enhance the overall social outcomes. In the second study, we address the joint decision-making problem of retailer's order allocation based on suppliers' wholesale pricing decisions and bank's loan portfolio pricing decision under risk control with expected shortfall as required by Basel III. Suppose that a retailer would like to choose and allocate purchasing orders to multiple heterogeneous suppliers, which are subject to capital constraints and potential operational disruptions. The suppliers' capital shortage problem can be solved by the prevalent BPOF scheme provided by a bank with loan guarantee from the retailer. All the parties involved in the BPOF scheme, including the suppliers, retailer, and bank, can make equilibrium decisions under competition based on our integrated supply chain financial and risk management framework. We also explore the impact of the BPOF scheme adoption on social welfare and the role of the government under this scheme. Our findings indicate that the bank's risk preference influences the performance of the entire supply chain and social welfare, which has not been revealed in the literature. We also find that the BPOF scheme not only can solve suppliers' capital shortage problem but also promote social and economic stability and prosperity at the expense of the suppliers' profits. Therefore, to foster sustainability and development in both social and economic areas, the government should monitor the implementation of the BPOF scheme and provide suppliers with subsidies for its adoption. We also identify special situations where the BPOF scheme should not be used, providing valuable managerial insights. In the third study, although the BGSF can alleviate the upstream suppliers' capital constraints, it can also amplify supply chain instability if poorly designed, evidenced by catastrophic failures like Evergrande. Basel III mandates risk-sensitive mechanisms, balancing regulatory compliance and resilience, and yet creates a critical tension: traditional banks exclude high-risk chains while non-bank lenders deploy opaque AI pricing. We resolve this by analyzing uniform financing (UF) versus Basel III-compliant dynamic financing (DF) under BGSF, examining how high-rate with low-guarantee versus low-rate with high-guarantee pricing models impact the performance. Key findings reveal: (i) DF enhances stability for low-risk supply chain; (ii) counterintuitively, BGSF adoption harms performance for high-risk supply chain; (iii) BGSF should be withheld from high-risk, low-return systems to prevent supply chain collapse. Our empirical verification underscores the critical importance of risk-differentiated pricing under the BGSF scheme. To enable real-time risk management and supervision for BGSF under Basel III, we develop an interpretable AI framework embedding game-theoretic axioms into machine learning and Large Language Model (LLM), generating risk-adjusted financing and supervision strategies tractable to our theoretical model. Our findings confirm that interpretable AI mitigates risk and creates value for supply chain system in BGSF by transforming regulatory constraints into stabilizing mechanisms, offering policymakers a transparent tool for resilient and regulatable financing. In conclusion, this research contributes to the literature on integrated supply chain finance (iSCF) and interpretable AI in the operations management field. The insights offer practical guidance and tools for FSPs, supply chain actors, and policymakers seeking to build resilient and socially inclusive financial systems. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/14553