RDP 9004: Saving and Investment in the 1980s 5. Conclusions

It is widely believed that during the 1980s Australia has developed a serious problem of under-saving. This belief appears to be based on movements in the conventionally quoted household saving ratio, which has been in steady decline since the mid 1970s. This paper has argued that the impression given by the household saving ratio is misleading, for three reasons. First, the distinction between the household and corporate sectors is to some extent artificial, and much of the decline in household saving has been offset by increased saving in the corporate sector. Secondly, household saving is usually quoted in net of depreciation terms, but the depreciation estimates are unreliable and highly sensitive to assumptions used in their construction. Thirdly, the official statistics on saving do not take into account the effects of inflation on the real value of nominal balances. Failure to allow for this effect generally results in an overstatement of private saving and an understatement of public saving, and this measurement bias has produced a spurious peak in recorded private saving in the mid 1970s.

Broadening the focus to consider gross saving of the private sector as a whole, and adjusting for the inflation transfer, it becomes apparent that private saving rates have been quite stable throughout the past three decades. Thus at least on the simplistic criterion of comparison with past standards, Australia does not seem to have developed any problem of under-saving during the 1980s. A slightly less comforting picture emerges however when international comparisons are made; these show that although Australia's private saving rates have generally been around the levels typical of other OECD countries (with the exception of Japan), Australia has tended to fall toward the lower end of that range. On this basis a weak case could be made for the under-saving hypothesis, but it would remain true that this is greatly overstated by the decline in the household saving ratio.

In reviewing explanations for the behaviour of private saving and consumption, a number of reasonably solid conclusions have emerged. The short-term movements in the saving ratio appear to be qualitatively well explained by simple principles of consumption smoothing, causing the saving ratio to move inversely with income. However, both the time series and cross sectional data suggest that consumption follows income much more closely over longer periods than is predicted by standard versions of the life-cycle theory. It is likely that the explanation for this must include a combination of factors including rule of thumb behaviour by households, borrowing constraints interacting with uncertain future income, and disincentives to saving for retirement arising from the system of publicly funded pensions. Other factors thought to be important, such as movements in interest rates, and changes in the value of wealth, appear to have little or no effect on short-run saving behaviour on the basis of the evidence reviewed here.

Aggregate investment by the private sector has shown a certain amount of cyclical variation in the 1980s but on average has been as strong as in previous decades and has been consistently higher as a ratio to GDP than the OECD average. The investment ratio has also been on an upward trend through the decade. The most important determining factor in the short run appears to be profitability, but as a long-run proposition at least part of the the strong investment performance is probably due to Australia's relatively high population growth. The extent to which population growth can explain the persistent gap between investment and saving is an important unresolved issue.

Distortions in the tax and benefit system form the main additional source of explanation for Australia's persistent private sector deficits. Two main features seem biased against saving; the tax treatment of nominal interest, and the age pension benefit system. The latter provides a strong incentive to match consumption to income in the years around the retiring age. The tendency for participation rates in those age groups to decline suggests that this is being achieved at least partly by earning less, rather than by spending more, a tendency which could well become a major issue as it interacts with the projected ageing of the population structure during the next forty years. The tax treatment of nominal interest represents a distortion on both the saving and investment sides tending to increase the attractiveness of investment, and to reduce the incentive to save when inflation is high. On the investment side, this effect is reinforced by the asymmetric treatment of interest costs and capital gains.

The net effect of these and other distortions on saving and investment is not known, partly because it depends on complex interactions with similar distortions in other countries. What is clear however is that there are significant non-neutralities in Australia's tax system, and the paper has argued that their likely net effect is in the direction of encouraging the net importing of capital. One important reason for this is that the distortions arise in part from interactions of the tax system with inflation, and will therefore tend to be more serious in countries with relatively high inflation rates.

A final issue of policy relevance which should be highlighted concerns the public sector contribution to national saving and investment levels. A major goal of fiscal policy in the second half of the 1980s has been to reduce public sector deficits, a goal which has been dramatically achieved with the shift into surplus by 1988/89. Without wishing to minimise the importance of this achievement, it is often overlooked that fiscal balance has been restored at much lower levels of public investment than were typical prior to the mid 1970s. The cumulative effect of reduced public capital formation is a neglected area of study in Australia. Presumably there is some degree of subslitutability between public and private investment spending, and indeed this may have been a contributing factor in the expansion of business investment ratios through the 1980s. However, to the extent that these expenditures are not close substitutes, the decline in the relative provision of public sector capital may have a detrimental long-term effect on aggregate productivity.