May 2012

CEPAR Research Fellow, Rafal Chomik, blogs on the accuracy of the projections made in the Intergenerational Reports and concludes that Australian policy makers have a long way to go to prepare Australia for the challenges of population ageing.

Rafal Chomik, Senior Research Fellow

Birthdays are a good time for self reflection. This month's tenth birthday of the first Intergenerational Report (IGR) - an account of the long term fiscal sustainability of Commonwealth policies produced every 3-5 years - was no exception.

The occasion was marked with a conference in Canberra, bringing together leaders from the civil service, academia and industry to reflect on ten years of projections and policies coming out of the IGR, as well as some of the remaining problems. CEPAR Director, Professor John Piggott, spoke about pension policy and Deputy Director, Professor Peter McDonald, presented on population projections.

Of course, evaluating the past is easier than projecting the future, but instructive nonetheless [1]. The projections are for forty years and include the impact of population ageing on the retirement income regime.

The current picture looks positive. Australia's retirement system compares well internationally. Spending on age-related pensions, the aged dependency ratio, and old-age poverty (measured after housing costs) are low by international standards.  Super assets are approximately equal to GDP, one of the highest ratios in the world.

And the IGRs have shown successively less severe population ageing and less dramatic increases in related spending. The 2002 report projected that in 2041-42, the cost of pensions would be 4.6% of GDP. By the 2007 report, this was 0.3 percentage points lower. The 2010 projection was 0.6 percentage points lower still.

The first thing to note about these figures is that improvements in the projections within the three reports are not necessarily because of policy reforms, though there were many. Much of this is due to the underlying, long-term demographic and labour market assumptions. For example, the long-term fertility rate assumption increased from 1.6 to 1.7, and then to 1.9 births per woman; net migration from 90,000 to 110,000, and 180,000 people per year; and mature age participation saw dramatic upward revisions.

In most cases, Treasury was taking very recent experience and keeping it constant for 40 years. While IGR projection methodology is commendable on many counts, these examples suggest some assumption myopia. Increases to participation, for instance, have been variously predicted and documented[2]. Longevity assumptions also need careful analysis. Recent mid-life longevity increases took most people by surprise, particularly in Australia: between 1950 and 1980, life expectancy at birth and at 50 increased by 7% and 10%, respectively, but between 1980 and 2007 the increase was 12% and 26%.[3] CEPAR is looking at this by fitting a mortality model with data on reductions in smoking, a key explanation of the trends.

Other assumptions could be changing too (e.g. decumulation of Super) but the IGR does not make this clear. The reports already provide a lot of background information, but there is a case for still greater transparency, or even, as suggested by Adrian Pagan, Professor of Economics at the University of Sydney, that an independent body such as CEPAR be tasked to formally provide analysis for the IGR.

The second thing to note about the figures is that while they are less severe than in other countries and than was previously expected, they still show that demographic shift remains a challenge. IGR 2010 projects that, without policy change, ageing pressures will result in a fiscal gap of 2.75% of GDP by 2050. Policy measures implemented over the last decade addressed various objectives of a well-designed retirement income system to different degrees. But at times reforms were piecemeal and poorly balanced the competing objectives. The 2009 increases to Age Pension level (an adequacy measure) were balanced with a tighter means test and higher eligibility ages (to address sustainability), a sensible balance. By contrast, the 2012-13 budget savings sought from Super meant that the objectives of economic and administrative efficiency were undermined through greater complexity and uncertainty. As a consequence some issues remain unresolved, including those related to decumulation, taxation, early retirement, and choice.

A final point about the IGR process is that its methodology is seldom used to demonstrate specific policy impacts. This is a shame, because an important reason for predicting the future is in understanding how it can be shaped. It could answer questions about what changes will keep budgets balanced despite the demographic shift. One reading could be that if the IGR and the projected fiscal balances are tools to evaluate the fiscal sustainability of current policy then, by their own measure, successive Governments have failed to implement policies that will keep budgets sustainable.

The tenth birthday of the Intergenerational Report is also the twentieth birthday of the Superannuation Guarantee. These are a cause for celebration, but also a reminder that much remains to be done to shape policy structures for an older Australia.

READ THE WORKING PAPER > Long-term Fiscal Projections and the Australian Retirement Income System