Aggregate and distributional analysis of Australian superannuation using the RIMGROUP Model

Date
Author
George Rothman
Publication type

Downloads

Conference Paper 96/3

Paper presented at the Fourth Colloquium of Superannuation Researchers, University of Melbourne.

RIMGROUP is a comprehensive cohort based lifetime accumulation model which makes projections on both the amounts and distribution of superannuation, other savings, retirement incomes, social security payments and tax concessions. This paper uses the RIMGROUP model to provide projections of superannuation aggregates and flows over the period to 2020 together with some representative distributional data by account type, gender and income.

Superannuation asset levels are projected to reach $370b in June 2000, $930b in 2010 and $1810b in 2020. Putting a reasonable bound of uncertainty around these figures suggests a range of $350-390b for June 2000, with higher percentage error bars as we go further out. The principal distributional feature by account type is the dramatic rise of Superannuation Guarantee (SG) accounts which increase from 12% of total superannuation assets now to 32% in 2020.

Average superannuation assets rise rapidly for all gender age and income groups over the period with some significant levelling of relative assets by gender and income.

A shorter term analysis to June 2005 is also presented which examines the impacts of recently announced superannuation initiatives of the Coalition Government, namely, the facility for lower paid workers to opt out of compulsory superannuation, spouse contributions and the introduction of Retirement Savings Accounts (RSAs). The individual impacts of the policies are examined together with their overall impact. Overall, these initiatives are estimated to reduce superannuation assets by only about $2.5b in 2004 (over a base of $600b) and national savings by about 0.1% of GDP. These estimates are much lower than those provided in the recent AMP paper (Issue 2 of 'SuperTrends', May 1996) but necessarily rely on judgements, particularly on take up rates, as well as analysis of available data.