PensionReforms' summary and comments

Australia's mix of private and public retirement income/savings arrangements conform with the World Bank's 1994 vision of the three pillar model and has been publicly applauded by the Bank's advisers. In the meantime the World Bank seems to have moved on to a five pillar version (see here for more).

"This [2001] paper has sought to explain the Australian approach to retirement income provision.  Over the past 100 years, retirement income provision in Australia has evolved into a multi pillar arrangement comprising the Age Pension (a publicly provided safety net), the Superannuation Guarantee (private mandatory retirement saving) and voluntary retirement saving (including voluntary superannuation and private pensions)."

The piece of the Australian story that attracts the most attention is the compulsory, workplace arrangement that requires employers to contribute, now, 9% of pay.  Dubbed the "Superannuation Guarantee", it derived from national, union-based awards dating from 1986.  It became compulsory for all in 1992 because, apparently, "award superannuation proved to be difficult and costly to enforce."  After 1992,  "despite some early implementation problems, this is now operating quite smoothly: superannuation coverage is around 92% of employees and compliance is high."

The tax-free "Age Pension" is, unusually, both income and asset-tested and has been since it started in 1909 (the family home was dropped from the asset test in 1912).  Despite the quite stringent tests, the Age Pension is still an important part of retirees' incomes.  In 2001 about 80% of the retired received at least some with about half of all retirees' receiving the full pension.  In relative terms, the pension is round 25% of average male earnings for a single person and 20% each for a married couple. Allowing for the tax-free status coverts the post-tax equivalent of 37% of the average male wage.  The paper explains the quite complex rules that determine how much is payable.

"In terms of the individual and economy-wide criteria for assessment, the Australian arrangements rate favourably. In particular, the Superannuation Guarantee does well in the accumulations phase, because the mandatory contributions ensure full fundedness and the private basis of the policy helps provide political insulation. However, it performs less well in the decumulation (benefits) phase, because retirement income streams are not mandatory. Given the historical right to take superannuation benefits as lump sums in Australia, mandating retirement income streams is politically difficult. In the long term, however, the success of Australia's current suite of retirement income policies will depend upon the introduction of such a policy."

PensionReforms notes that the compulsory Superannuation Guarantee has at least one design curiosity - despite its near universal coverage of employees, superannuation enjoys quite generous tax breaks (recently improved).  It seems a little illogical to increase everyone's taxes to pay for an "incentive" to save for retirement when there is no choice about joining.

Needless to say, the tax-favoured environment, coupled with compulsion and the Age Pension's income and asset tests induces some rational responses.  Although what is called the "preservation age" (the earliest age from which the compulsory superannuation savings can be withdrawn) will increase to age 60 by 2025, PensionReforms notes that the Australian combination of public policy has seemingly induced a lowering of retirement ages.  That's partly because, there is no compulsory annuitisation of the Superannuation Guarantee savings.  The paper notes this gap and PensionReforms expects it to be closed by further regulation in due course.

So, is the Australian version of the World Bank's model a "success"?  Australians seem to think so (especially Australian financial service providers), despite the relatively high delivery cost of compulsion.  Australia's Reserve Bank wasn't sure at first whether it increased 'saving' (see here) but thinks now that it might have (see here).  However, PensionReforms suggests that the jury is still out on the overall question despite, now, 15 years of full compulsion (and another six years of a lesser version).

PensionReforms suggests that, if there is a concern about the future costs of state provision, the asset and income tests of the Age Pension are the logical concomitant of tax-advantaged, compulsory private provision.  Without those means tests, the state cannot achieve fiscal savings from its intervention in private decisions about saving for retirement. The Australian scheme is poorly integrated in this regard as the means test can be readily avoided.  If tax incentives are involved, the state's costs in effect may rise not fall. It's possible however that the intention of the compulsory scheme is to facilitate a higher replacement rate overall to higher income people.  If that's the case, the intention is poorly articulated and a policy to require mandatory annuitisation of the compulsory scheme's benefit seems called for at the very least.  The debate about the true objectives of overall retirement income policies (and associated issues) has yet to be held in Australia.  (file size 137KB) 112