Submission to: Treasury

17 April 2015

Submission on Income Streams Review

This submission considers the two Income Streams Review Consultation Papers concurrently. It is largely based on the findings of two Workshops on retirement incomes policy conducted by the Academy of Social Sciences in Australia (ASSA). 

Download this submission as a PDF file

Submission in full


This submission considers the two Income Streams Review Consultation Papers concurrently. It is largely based on the findings of two Workshops on retirement incomes policy conducted by the Academy of Social Sciences in Australia (ASSA). The most recent was held in November 2014 and comprised academic experts, industry representatives and representatives from Treasury, the Department of Social Services and the Australian Government Actuary. The Workshop canvassed many of the topics raised in the Consultation Papers but did not discuss them in detail.

Tax support for superannuation is substantial and will provide financial challenges for governments. This becomes most apparent in the de-accumulation phase. The Workshop strongly supports the Government trying to address these challenges. There is a need to make changes to the current arrangements.

We also supported the objective, as stated in the Consultation Papers, that the primary objective of the superannuation scheme should be to provide income for retirement. Furthermore, the Workshop agreed that “the objective of the current regulatory regime in the retirement phase is to ensure that the capital underpinning a retirement product is drawn down over time in order to:

  1. provide income over the course of a person’s retirement,
  2. prevent large amounts of concessionally taxed superannuation savings being passed on as bequests, and
  3. facilitate the provision of a ‘smooth’ level of income.”

Specifically, the majority of Workshop participants (but not all) agreed that one of the undesirable outcomes of the current arrangements is that large amounts of tax subsidised savings can be left in bequests.


The Workshop agreed that life time annuities have a pivotal role in a retirement incomes stream especially through the management of longevity risk. Their take up is very low at present. They are seen as poor investments and probably are in practice. In part, this reflects a lack of understanding that life-time annuities represent an insurance product and not just an investment product to be compared to other investment products.  Such insurance can only be efficiently covered through pooling, and in a form that limits the risk of adverse selection in order to reduce costs.  Arguably it also requires government intervention to facilitate market trade in the risk, such as through the provision of longevity bonds or other long-term indexed bonds that might reduce the impact of current regulatory provisions, designed to protect institutions’ financial sustainability, which in practice presents a barrier to affordable lifetime annuities.

The Workshop considered some submissions from the Centre of Excellence in Population Ageing Research (CEPAR). The submissions propose both the introduction of longevity bonds or similar financial instruments, and the mandating or defaulting of deferred annuities to be purchased when de-accumulation begins and to commence payment at, say, age 85.  Both proposals received strong (but not universal) support from the Workshop but we did not discuss the proportion of accumulated savings that should be directed into a deferred annuity. That requires further research. However, it was suggested that the compulsion only apply when total savings exceed some threshold.

The alternative approach favoured by other participants was similar to that proposed by the Financial System Inquiry. This is to mandate funds to offer comprehensive products that effectively include a deferred annuity but not make take up compulsory. We would regard this as a minimum. Given the tax support involved in accumulating superannuation savings, there is a strong case for making some form of annuity or deferred annuity compulsory especially if there is not widespread changes in behaviour that results in greater compliance with objectives A, B and C.

These arrangements were also seen as a way to reduce reliance on the age pension as insurance against longevity by those with the ability not to rely on taxpayers, and also to counter overly conservative behaviour (that seems to be widespread) of not drawing down accumulated savings but allowing them to increase and consequently to be left in estates for the next generation even when that is not the intended purpose.  Requiring the purchase of a deferred annuity would also address the problem of adverse selection and thereby reduce the cost of annuities. The encouragement or mandating of deferred annuities might also require adjustment to age pension means test arrangements to ensure there is no inappropriate penalty involved.


The Workshop did not specifically discuss the establishment of principles for Retirement Income Products. However, from the tone of the discussion, we are confident the majority of Workshop participants would support their establishment. We think it would be possible to develop principles especially given the specified objectives of the regulatory schemes on retirement income products. If it was thought desirable to involve the research community, ASSA would be prepared to host and organise a Workshop on principles.

These principles could be used to provide a framework to determine whether individual retirement income products should be approved or not. Product specific regulation may be undesirable but it is better than having the current restrictions on what are clearly appropriate products. Delegated legislation could be used to reduce the cost of ‘legislating’ the approval of individual products.

Retirement Income Products

A key issue for Australia’s superannuation scheme is not being able to manage longevity risk. This risk is generally not managed in accounts based pensions, for example. This encourages conservative behaviour that builds up savings beyond a level that is really necessary often resulting in significant residuals on death that are left to estates.

For these reasons, we support the availability of both Category A and Category B retirement income products. There should be scope for innovation so both categories should be allowed as long as they conform with the principles discussed above.

A key requirement is to remove regulatory restrictions that limit the provision of worthwhile longevity products. Furthermore, the government might consider increasing the availability of instruments (eg CPI indexed bonds) to support the development of longevity products.

We would support proposals which allow the purchase of a single annuity product during the accumulation phase with multiple premiums. The regulations restricting this seem to be unnecessarily restrictive.

Superannuation is extremely complex and not all advisors act in the best interest of clients or may not fully understand how to best manage longevity risk except by using the age pension as a safety net. Defaults may play an important role in ensuring people’s retirement savings are used prudently. Case studies of people in different circumstances might also be useful to help good de-accumulation decisions.

Budgetary restrictions were mentioned several times in the Consultation Paper. They have to be a consideration, of course, if the superannuation arrangements are to be sustainable. However, they should be looked at from a medium to long term perspective. We are concerned that too much emphasis might be given to short term considerations impacting on what might be sensible policy decisions when looked at from a longer term perspective.

Limiting Bequests

The Workshop also discussed whether there should be any restriction on the use of tax-supported superannuation savings for bequests.  It recognised that, on the one hand, to the extent that the superannuation system is simply a mechanism for spreading consumption over lifetimes, and that bequests are just another form of voluntary consumption, there would be little reason to constrain their use for this purpose.  On the other hand, a focus on the core objectives of the retirement income system, and evidence that bequests are often unplanned and are the result of poor management of accumulated superannuation, suggests a more restrictive approach such as through mandating deferred annuities.  Concern was also expressed that rising levels of bequests funded from taxpayer-supported savings might accentuate wealth inequality into the future.  However, the Workshop also noted that the main element of bequests were owner-occupied housing that is also a tax-favoured form of savings.

The Financial System Inquiry concluded that the superannuation system should accommodate bequests, but not do so to the detriment of retirement incomes. Our Workshop will agree with that principle. In other words, the de-accumulation scheme should try to discourage people from using savings for retirement income purposes for estate planning purposes. However, any residuals at death should be able to be commuted at death.

 Dennis Trewin, AO FASSA, on behalf of the Policy and Advocacy Committee

 APRIL 2015

For more information, please contact:
Mr Murray Radcliffe
Deputy Director
murray.radcliffe [at]
+61 .2 62491788