RDP 9511: Superannuation and Saving 6. Conclusion
December 1995
- Download the Paper 103KB
Our results suggest that, in the past, there has been a significant degree of substitution between superannuation and other forms of saving, although the offsets have not been complete. Our estimates are not very precise, but they suggest that, over the past 35 years, about three-quarters of the variation in superannuation saving was offset by changes in other forms of saving. At a more disaggregated level it appears that there has been a high degree of substitution between net superannuation contributions and non-superannuation saving, but little, if any, substitution between superannuation earnings and other saving. These estimates are generally consistent with observations by Edey and Britten-Jones (1990) and Goode (1994) as well as a number of recent overseas studies of saving through retirement income schemes.
It is too early to tell how big an effect the new superannuation arrangements will have on aggregate household saving in the long term. However, there are a number of aspects which might be expected to reduce the degree of offset between superannuation and other forms of saving. The compulsory elements of the new arrangements will force some households, who would otherwise save very little (particularly those in the initial stages of the life cycle for whom retirement is a long way off, or those on low incomes for whom it is optimal to rely on the government pension to fund retirement), to save for retirement. Liquidity constraints will limit the extent to which some of these households can offset increases in superannuation contributions by borrowing, or by reducing other forms of saving.
There is a danger, however, that some households will offset some of the increase by reducing other forms of saving and that, even those households who are currently liquidity constrained, may compensate for the increase in superannuation saving by reducing saving later in life. Households may remove superannuation assets directly (for example, by retiring early), or by consuming more than would otherwise be the case once income and wealth have increased and liquidity constraints have eased. For those households, a partial effect of the new arrangements may be to postpone consumption (and temporarily increase current saving), rather than to increase longer-term saving for retirement.
Given households' high propensity in the past to switch between superannuation saving and other forms of saving, sustainable increases in saving are only likely to occur if households can be persuaded of the need to provide for their own retirement and if institutional structures are put in place that encourage or force them to do so.