What we don’t know about retirement

Despite decades of study into our superannuation system, surprisingly little is known about the level of financial assets Australians have when they retire, particularly for households.

Now new research by researchers from Monash Business School and CSIRO, working with industry partners Challenger and Accurium, has painted a more complex picture of how retirees structure their retirement than first thought, pointing to a need for a greater diversity of income-stream producing financial tools to be developed.

For the first time, the means-testable assets of older Australians outside of superannuation was able to be studied at a household level, not just as individuals, thanks to a previously inaccessible database of aged pensioners at retirement.

As part of an ARC linkage grant project, researchers used a de-identified database from the Department Human Services to study the wealth of single and partnered households, including superannuation assets and non-superannuation means-testable assets.

Superannuation data is usually described in individual terms because it is accumulated that way. However, it ignores the reality that many people are part of a household.

This is recognised by the age pension, which is calculated on household earnings.

For those who do not own their own home, ensuring a secure income stream would be particularly valuable to meet future rent.

Initial research findings included that partnered households held twice as much superannuation as single households and were more likely to own a home.The research showed 87 per cent of partnered households owned a home at retirement, compared to 59 per cent of single households. It found 3.6 per cent of households held a mortgage loan for an investment or commercial property at the time at retirement.

Other findings include:

  • Single households with no super are much more likely to have means-testable assets of less than $100,000 than partnered households
  • Partnered households with no super are much more likely to have testable assets of more than $100,000 than single households
  • Most households with super assets have testable assets of less than $200,000
  • Single households with super assets are more likely to have testable assets of less than $100,000
  • Partnered households with no super are much more likely to have total assets exceeding $200,000
  • Partnered households are much more likely to be home owners than single households.

“This is the first time that data on this scale has been analysed to paint a picture of the actual circumstances of people in retirement,” says ARC project’s lead investigator, Monash Business School’s Professor Colin O’Hare.


Developing a framework

The research comes as the Federal Government develops a framework for Comprehensive Income Products for Retirement (CIPRs) – now called the MyRetirement framework – which emerged from David Murray’s Financial System Inquiry.

The team, made up of CSIRO researchers Alec Stephenson, Zili Zhu, Peter Toscas and Andrew Reeson, with Monash Business School’s Professor Colin O’Hare and Professor Bonsoo Koo and Challenger’s David Cox, analysed transactional DHS records used to assess eligibility for social security payments.

The database covers 11 years of data from between July 1, 2004 to July 1, 2015 and covers individuals aged over 60 years who are eligible for social security or pension concessions along with their partners.

It is estimated this covers around 75-80 per cent of the population in this age demographic. The database can be divided into categories such as Indigenous, rural or city area, by asset value, and by household.

The research found while a relatively small number of retirees in this dataset had superannuation after retirement, the balances tended to be larger. Of the 14,864 dual households in which both partners appear to have retired in 2013-14, only 6361 (43 per cent) had superannuation after retirement.

However, the median balance was $354,744 among these households, and 25 per cent had over $554,855.

“This is likely to reflect the fact that households with smaller balances will elect to withdraw it as a lump sum as the tax benefits of retaining wealth in superannuation are less relevant to them,” the report says.

Younger retirees were likely to have more assets in superannuation and less in held directly in business assets and savings.

But the majority of retirees with superannuation kept their money in an account-based pension, which they typically drew down at the minimum allowable rate.

The case for annuities

The report argues there is a case for annuities. “This may be particularly useful for the thousands of single people retiring each year with superannuation. For those who do not own their own home, ensuring a secure income stream would be particularly valuable to meet future rent; they are also less likely to face one-off expenses such as home repairs, for which the flexibility of an account-based pension is well suited.”

Professor O’Hare said further research needed to be done to better understand life expectancy to reduce the cost of products such as annuities. But he said there were limited options and advice to help people manage their retirement funds.

“Often people move this money from their super fund to one of the big banks because the branding is better – they feel like they have a stronger relationship with their bank than their super fund from which they may just receive a statement once a year. There is a need for more post-retirement products and advice in this stage of life,” he said.

Over the next two years, the project will continue to look at the wealth of retiree households with no superannuation or those who were not employees when working and so not eligible for the superannuation guarantee. It also looks at the differences in wealth of homeowners and renters.

Published on 30 Aug 2017