Better member outcomes in superannuation through data driven financial literacy prediction
- Publication Type:
- Thesis
- Issue Date:
- 2021
Open Access
Copyright Clearance Process
- Recently Added
- In Progress
- Open Access
This item is open access.
Through a combination of product innovation and government reform the retirement savings system in Australia has become increasingly complex. Everyday Australians are now required to possess sophisticated financial decision making skills in order to safely navigate the many risks and opportunity costs associated with their superannuation. The decisions that savers make regarding their investments will have a significant impact on their retirement wellbeing. As a result, it is paramount for any responsible financial institution to actively measure, monitor and elevate the financial literacy of its members. To date, there is no research which proposes a suitable context specific construct for financial literacy in superannuation which is predictive of financial outcomes and utilises passive administrative data to enable ongoing measurement. To address these challenges this research first proposes a measurement construct for financial literacy in superannuation informed by the results of a financial literacy survey enriched with administrative member data. Next, it proposes a novel solution for the prediction of superannuation literacy using a vast dataset of demographic and behavioural features. The prediction framework addresses the issue of non-response bias while maximising predictive performance. Finally, the prediction framework is validated against a real world business problem, customer churn. The findings of this research indicate that the measurement construct for superannuation literacy significantly outperforms the conventional measure against a number of financial outcomes. Superannuation literacy outperforms common financial literacy by a multiple of 7.1, 11.2 and 8.9 for account balance, portfolio return and portfolio volatility respectively. The prediction framework for superannuation literacy outperforms a number of state of the art algorithms for prediction. The aggregate measure for superannuation literacy achieves an R square of 84.6% and is highly correlated to positive financial outcomes in super. Validation against customer churn provides an insight into the complex relationship between financial sophistication and decision making. This research provides the framework and tools to monitor and engage superannuation members based on their sophistication and intervene where they are determined to be at-risk, requiring additional support to manage their retirement savings, or to maximise member satisfaction and engagement.
Please use this identifier to cite or link to this item: