The government announced in its recent budget that pensions (including the Disability Support Pension and Age Pension) will no longer be pegged to wages. Instead, they’ll only rise in line with the CPI.

In response, ACOSS said:

This would mean pensioners’ living standards would fall behind community living standards, as these rise with wages. After 10 years pensioners will be $80pw worse off.

David Crowe of The Australian described arguments based on this figure as  “exaggerated” and “a powerful scare”, saying the analysis:

ignores the actual increase in the pension twice a year.

The $80 figure is calculated by assuming one sort of growth in the pension under the government’s scenario but a slightly higher rate of growth under an alternative by 2027. Both calculations assume growth in the pension, so it keeps up with the cost of living. Stretched across more than a decade, though, the difference can seem enormous.

I don’t think ACOSS’ argument is just a scare. It’s true that pensions won’t fall below their current level, in inflation-adjusted terms. Pensioners won’t be worse off, in real terms[fn1], than they are today. But they will[fn2] be worse off than they otherwise would have been. The gap in living standards between workers and pensioners will also grow over time.

ACOSS’ numbers, I believe, assume that real wages will grow at about 1.5% or 1.6% a year. If you assume real wages grow at that pace, while the real pension is flat (because it’s indexed to the CPI), you get a gap of about $80 a week in 2027.

Single pension, with two different projections


Source: Total maximum periodic pension-related payment from Department of Social Services. CPI from ABS 6401. Wage indexation scenario assumes 1.6% real wages growth per year.

Here’s the difference between the level of the pension under the two scenarios:

Projected gap between wage-indexed pension and CPI-indexed pension


By 2030, under the CPI-indexation scenario, the pension will be a touch below 25% of male average weekly earnings. That will be the lowest ratio since 1989.


Source: Total maximum periodic pension-related payment from Department of Social Services. CPI from ABS 6401. Male total average weekly earnings from ABS 6302. Wage indexation scenario assumes 1.6% real wages growth per year.

Is it reasonable to assume real wages will grow at 1.6% per year? Well, that’s what the Intergenerational Report 2010 assumed. For long-run projections, it assumes that labour productivity growth will equal its 30-year average (1.6%), and that real wages will rise at the same pace. Treasury now uses a figure of 1.5% per year labour productivity and real wages growth. So the assumption that seems to have been made by ACOSS, and which I used to make the charts in this post, is consistent with the assumption used by the Treasury..

Is it reasonable to project  income support payment rates out to the future using these different indexation assumptions? Well, that’s what the Henry Review did, when it noted that the gap between Newstart and pensions would continue to widen over time, as the former is indexed to CPI while the latter has been indexed to wages. Here’s the chart from the Henry Review:

If you believe, as I do, that poverty and social exclusion should be measured in relative terms in a rich country like Australia, then it’s important that people reliant on benefits don’t fall too far behind typical community living standards. Otherwise, relative poverty will grow. In 2007-08, 27% of older people were in poverty in Australia (measured using a 50% of median income poverty line). That figure fell to under 12% in 2009-10, thanks to the pension increases in 2009. If pensions are indexed only to prices rather than wages, that poverty rate will start to rise again.

While the proposed changes to the pension won’t mean a cut in real terms relative to today’s payments, they will result in pensions being lower than they otherwise would’ve been, and they will result in a growing gap between pensioners and the rest of the community. Perhaps you can argue that indexing pensions to male average earnings resulted in a shrinking gap between typical community living standards and pensions[fn3], but I don’t think it’s disputable that CPI indexation will result in a growing gap. I don’t think the ACOSS figure is just a “scare.”

[fn1]By “real terms” I mean “after adjusting for changes in the consumer price index.” But the CPI may not be the best way to adjust pensions for the change in the cost of living. The ABS publishes living cost indexes for age pensioners and other transfer recipients. Since those series began in 1998, the CPI has risen at a compound annual growth rate of 2.9%, while the living cost indexes have risen by 3.1% and 3.2%, respectively. If the cost of living for pensioners continues to outpace the CPI then indexing pensions to the CPI will lead to a fall in pensioners’ real living standards.
[fn2]Pensioners will be worse off than they otherwise would have been if average wages rise faster than the CPI.
[fn3]This would be the case if male total average weekly earnings rose faster than whatever you consider to be the benchmark for community living standards.