Menu Close

Explainer: the policy challenge of indexing welfare payments

The way pensions are indexed can have a significant impact on the end payment. Alan Porritt/AAP

Changes to the way Australia’s Age Pension is indexed, flagged in treasurer Joe Hockey’s first budget, are not due to commence until September 2017. But amid widespread disquiet in the community, Social Services Minister Scott Morrison has said he is open to working through “alternative scenarios” to make the pension sustainable.

In attempting to balance the budget welfare is an obvious area for reform. Welfare spending in 2014-15 is predicted to be A$146 billion of which aged pensions, disability pensions and unemployment benefits make up over A$70 billion.

It is common in many countries, including Australia, for social security payments such as pensions to be indexed in order to maintain their real value over time which would otherwise be eroded by inflation.

Automatic indexing of pensions to the Consumer Price Index (CPI) (the commonly used measure of the cost of living of the average household) began in 1976, but there have been several variations to the form of indexation and differences in treatment of different types of payments. Currently all payments are increased twice a year according to the method of indexation. Unemployment income support payments, Newstart, are adjusted according to the CPI.

Pensions, on the other hand, are increased six-monthly by a more complex procedure. Because pensioners have somewhat different spending patterns to the average household another cost of living index which is intended to better reflect pensioner spending has been constructed - the Pensioner and Beneficiary Living Cost Index (PBLCI). Indexing by the CPI or the PBLCI is intended to maintain the real value of payments, that is, pensioners will be still able to buy as much as usual when prices of goods and services rise. Note that maintaining the real value of payments is not the same as maintaining the standard of living relative to those in work. This is done by “benchmarking” benefits to Male Total Average Weekly Earnings (MTAWE).

The couple pension benchmark rate is equal to 41.76% of MTAWE; while the single pension rate is 66.33% of the couple rate. The method of setting the pension is to calculate the increase that would occur under indexation to CPI and the increase that would occur under indexation to PBLCI. If the highest rate of pension under indexation is lower than the benchmark percentages of 41.76% and 66.33%, respectively, then the rates are increased to the appropriate benchmark level. Parenting Payment (Single) was previously adjusted in the same way as other pensions, but from 2009 has been indexed to CPI and benchmarked to 25% of MTAWE.

In 1997 the Howard government introduced automatic benchmarking of the single pension rate to MTAWE. In 2009, the Rudd government increased the single rate of pension, and introduced the current indexation method which reflects movements in the CPI, PBLCI and MTAWE.

Other income support payments such as Newstart are only indexed to movements in the CPI. Since MTAWE has increased over time at a greater rate than CPI, which we would expect as real wages and standard of living generally increase over time, the value of unemployment benefits relative to pensions has fallen significantly over time. This has resulted in important distortions.

The figure below shows the number of people receiving certain benefit and pension categories, and how sole parent, disability and unemployment benefits have changed over time. There are some interesting features of these data in terms of the number of recipients at any point in time and with respect to trends over time.

Social security recipients, ’000s, 1980-2014

Department of Social Services (various years), Income support recipients

The number of people receiving unemployment benefits tracks roughly the ABS unemployment estimates. The number of people receiving single parent pensions, however, almost tripled over the period 1980 to 2005 but fell in the following years mainly due to the tightening of eligibility to the sole parent benefit.

Perhaps most interesting is the rise in people on disability pensions, which is in inverse relation to those on unemployment benefit. There appears to have been a movement from people on unemployment benefits to pensions which, while reducing the figures for those on unemployment benefits, is costly for the government since, as discussed above, pensions are indexed to average weekly earnings while unemployment benefits are indexed to the CPI.

Since the rate of increase in income received from a pension has been increasing faster than the rate of increase in the income received from the unemployment benefit, it is more attractive to be on a pension than on unemployment benefits. Also, not only are the payments greater, but the requirements to look for work are less for people on pensions than on Newstart.

This illustrates a major problem when designing a system of welfare benefits and arises largely because welfare has several, often conflicting, objectives. The most obvious of these objectives is to provide a minimum standard of living for those unable to provide sufficiently for themselves. But what this minimum should be and how it should be adjusted over time is not obvious – it is a policy decision for government.

Another common objective of welfare is to provide incentives for those not in work to look for work and, more recently to get retirees back to work. Income support needs to be low enough in order to create incentives for people to work, while fulfilling the social objective of ensuring people do not live in poverty. Add another objective - to reduce government expenditure, deficits and debt - and the problems inherent in a welfare system are multiplied.

Want to write?

Write an article and join a growing community of more than 181,800 academics and researchers from 4,938 institutions.

Register now