Money management difficulty, customer protection and the detection of early-stage dementia

Abstract

Financial institutions need to protect their customers in vulnerable circumstances. We test if older persons' money management difficulty, e.g., paying bills and making pension decisions, can help identify vulnerable individuals with early-stage dementia. We show that clinically informed lead indicators of dementia, extended to include money management difficulty, can inform a high-performance AI model to predict a clinical diagnosis. Invariant to extensive tests, after age, money management difficulty is consistently ranked the most important lead indicator of dementia. Financial institutions can enhance their protection of customers with dementia and inform the inter-generational transfer of financial control to a reliable agent.

Note:

Funding Information: Cal Muckley is grateful for support provided by the VAR consortium, Science Foundation Ireland and an Operational Risk industry consorium comprising: Bank of Ireland, Citibank Europe Plc, Deloitte Ireland and Institute of Banking. Shivam Agarwal also gratefully acknowledges support of a generous UCD Business School Doctoral Scholarship. Cal Muckley would like to acknowledge the financial support of Science Foundation Ireland under Grant Numbers 16/SPP/3347 and 17/SP/5447 and funding from the ADAPT Centre for Digital Content Technology, funded under the Science Foundation Ireland Research Centres Programme (Grant 13/RC/2106 P2), and co-funded by the European Regional Development Fund.

Conflict of Interests: We have no competing interests to declare.

Keywords: Financial services, Household finance, Aging, Cognitive decline, Money management difficulty, Vulnerable customers, Machine learning, Explainable AI

JEL Classification: G51, G53, I18, J14, J26