AN ACTOR-NETWORK THEORY APPROACH IN INVESTIGATING THE INFORMATION SYSTEMS PERSPECTIVE OF ANTI-MONEY LAUNDERING COMPLIANCE THROUGH A CASE STUDY OF THE FOREIGN ACCOUNT TAX COMPLIANCE ACT (FATCA) IMPLEMENTATION IN A JORDANIAN LOCAL BANK

摘要

Implementation of the Foreign Account Tax Compliance Act (FATCA) goes beyond a technological modification to automate the identification of US clients and report their information to the IRS. FATCA implementation requires foreign financial institutions (FFIs) to learn the new requirements, to modify their organizational structures and their employees’ relationships and responsibilities, and to adjust the technology that helps the employees collect new FATCA-related information and to process that information so that it can be reported to the IRS in the correct format. In spite of that, research on FATCA implementation has focused on studying each constituent separately. However, according to the information systems (IS) body of research and from a systems thinking perspective, the whole (the bank that is complying with FATCA as a system) is more than the sum of its parts (the information, technology, and social structures that it includes). For this reason, this dissertation argues that in order to achieve an effective FATCA implementation and reduce tax evasion activity, FATCA implementation should be studied from an IS perspective. This will assist in appreciating the complexity of FATCA implementation and compliance and will help practitioners to better anticipate future uncertainties. The dissertation uses actor-network theory (ANT), as it is a socio-technical theory, to investigate the implementation of and compliance with FATCA in a Jordanian local bank. Our interpretation revealed a number of problems in the bank’s compliance initiative; among them were the issues of overlooking technology, information, and the bank’s customers as actors with interests of their own. Accordingly, we provide eight propositions that can enhance the effectiveness of FATCA compliance. Tax-evasion has been shown in the literature to be a predicate crime involving money laundering (ML), i.e., a crime that generates proceeds that need to be treated in secretive ways so that they can be falsely legitimized. We argue in this dissertation that the findings of our case study could provide lessons for the anti-money-laundering (AML) domain in relation to its structurally coupled domain of ML. Thus, we presented some lessons that can be tested in the ML/AML domains.