Understanding Superannuation Contribution Decisions: Theory and Evidence

JUN FENG, MONASH UNIVERSITY, PAUL GERRANS, THE UNIVERSITY OF WESTERN AUSTRALIA, GORDON CLARK, MONASH UNIVERSITY/OXFORD UNIVERSITY

28 February 2014

This paper reviews individual retirement savings decisions focusing on the contributions or savings behaviours. It examines lessons learned from international retirement saving systems and the limited number of studies examining the Australian superannuation system, to provide a better picture of who and why people make (or don’t make) voluntary contributions. Using the Mercer database, preliminary analysis of member contribution behaviours in Australia indicates consistently declining voluntary post-tax contributions across all age groups over the 2002−12 period.*

Australia’s compulsory employment-based superannuation system complements the means‐tested government age pension and the voluntary savings of individuals. This paper focuses on the latter third ‘pillar’ of the retirement income system, i.e. the voluntary savings undertaken and managed within the superannuation system.

Despite the fact that total superannuation assets exceed annual GDP, concerns remain as to the adequacy of retirement savings. In addition to the mandatory component of superannuation, various mechanisms are in place to promote voluntary retirement savings via superannuation, with voluntary contributions having been less successful. This pattern is not unique internationally.

Participation in voluntary contributions requires people to have a certain level of involvement in retirement planning, knowledge of the system, and interaction with their employer, superannuation fund or both. Surveys by Roy Morgan Research in recent years have suggested fund member financial knowledge is not high with only one‐third reportedly having considered how much they needed to save for retirement.

The direction of recent reviews, most recently the Super System Review and subsequent policy changes (MySuper), and previously Simpler Super and the Review of Australia’s Future Tax System, reflect a direct challenge to the assumption of an informed and engaged membership of funds.

Conclusion

The modern phase of Australia’s superannuation system now extends into its third decade. While this is now close to an individual’s expected working lifespan, the history of the system is marked by frequent rule changes impacting contribution rates, contribution limits and taxation.

Recent policy changes have entrenched the ‘libertarian paternalistic’ design of the system. The system is based on compulsion which requires individuals to save for their retirement though this contribution is made by the employer on the employee’s behalf.

A belief that the compulsory rate is inadequate has seen the compulsory rate increased alongside policies to encourage more voluntary retirement savings through tax incentives and provision of options within superannuation funds. Despite these efforts, participation in voluntary retirement savings appears to be declining.

This paper presents an overview of international literature relating to the determinants of voluntary retirement saving participation, with a focus on available Australian evidence. From this review a number of key socio‐economic and demographic factors, as well as job characteristics are identified to be important predictors of participation. Other factors such as retirement planning, financial literacy and status quo bias can also explain the lack of voluntary contributions.

A preliminary analysis of a new, large database made available by Mercer Australia provides a valuable addition to this literature by allowing a view from within a superannuation fund instead of the population surveys currently available. The longitudinal nature of the database allows a better understanding of trends, impact of external events, and persistence in actions. Future analysis of this database will continue to add to our understanding of member behaviour in relation to voluntary contributions.

In view of the preliminary evidence suggesting a decline in the number of those making voluntary contributions, future work will examine to whom this applies. For example, is this due to those who were previously contributing ceasing to do so, but otherwise continuing to receive employer contributions, or is it because those who were previously contributing are leaving the fund (retiring from their employer) and new younger members are not choosing to contribute? Also, what has happened to the amount that has been contributed? How has this changed over time and how significant have rule changes been in changing the amount contributed? We will address these issues in future work.

* We gratefully acknowledge Mercer Australia for the provision of data for the empirical analysis and Dileepa Diyagama in aiding this. Thanks also to Rohan Fletcher and Jacqui Whale for assistance with preliminary data management.