Household saving
In the life cycle model of saving, individuals will try to smooth consumption over their life times. As incomes vary over a lifetime, this implies that the saving rate should also be expected to vary. Typically, saving rates will be low in early adulthood when earnings are low, and then increase with higher earnings, becoming negative in retirement as people “dis-save”, drawing down on previous accumulations. This would lead to a hump pattern in lifetime saving rates.
Ideally, data is needed for each age cohort through time in order to follow the actual lifetime saving patterns. A proxy for this is the cross sectional pattern of saving rates at a point in time. The raw saving rates by age groups from the Household Economic Survey are shown in the second column of Table 7.3. The unadjusted rates rise as predicted by the life cycle model of saving. However, they remain higher than predicted in retirement. If an adjustment is made for health and superannuation, then the pattern is markedly changed.
New Zealand has a pay-as-you-go system of taxes and transfers to meet superannuation payments to retirees. In contrast, the tax contributions paid by the working age population could be regarded as forced saving, and the pension payments received by the retirees as decumulation of previous pension contributions.
If the tax paid is treated as forced saving, then the saving rate of the working age population would increase by about six percentage points. If the pension is treated as a decumulation of assets by retirees, their saving rates will become negative.
As the bulk of health expenditure is incurred by the elderly, total consumption by retirees is understated in the HES. If public health expenditure is included as part of the consumption by the retired age groups, then they become substantial dis-savers.
| Age group |
Unadjusted saving rate |
Adjustment for superannuation |
Adjustment for health |
Total adjusted saving rate |
|---|---|---|---|---|
| 19-24 | 7.6 | 6.0 | 4.0 | 17.6 |
| 25-29 | 7.9 | 6.0 | 4.0 | 17.9 |
| 30-39 | 8.3 | 6.0 | 4.0 | 18.3 |
| 40-49 | 12.7 | 6.0 | 4.0 | 22.7 |
| 50-59 | 19.4 | 6.0 | 4.0 | 29.4 |
| 60-64 | 12.6 | 0.0 | 0.0 | 12.6 |
| 65-74 | 15.3 | -56.3 | -44.0 | -85.0 |
| 75+ | 28.8 | -37.8 | -108.0 | -117.0 |
Source: Coleman (1999)
The effect of both these adjustments (for superannuation and health expenses) is shown in the last column of Table 7.3 (details of the adjustments are given in Coleman, 1999). Two features of these adjusted saving rates are noteworthy. In the first place the absolute level of saving is much higher than the HES or Household Income and Outlay Account data would suggest. Secondly, the life cycle model of saving receives much stronger support, with the distinctly humped pattern of saving rates now strongly evident.
