RDP 2006-08: A Survey of Housing Equity Withdrawal and Injection in Australia 8. Conclusion

The survey results provide a wide range of information relating to housing equity flows. In addition to being the first survey of its kind in Australia, the comprehensive approach extends the more narrowly focused surveys conducted internationally on this topic. This survey captured flows of both housing equity withdrawal and injection by all households including flows associated with deceased estates, non-transaction-related debt repayments, and non-debt-financed renovations. Another innovation is information gathered on the features of each household's mortgage, to help gauge the importance of new financial products to housing equity flows.

The results of the survey suggest that any aggregate series for net housing equity withdrawal or injection masks large aggregate withdrawals and injections by households. Over 2004, 30 per cent of households made net equity injections, while 12 per cent made net equity withdrawals. The values injected were, however, typically much less than those withdrawn.

The most common methods of withdrawing or injecting housing equity were through altering debt levels on already-owned property holdings. Households that were net withdrawers over 2004 tended to favour methods such as refinancing and increasing loan size, or drawing down home-equity loans. Net injections were most commonly made through regular principal repayments. In addition, a number of households injected equity into already-owned properties through renovations.

Though fewer in number, withdrawals and injections of housing equity associated with property transactions were typically significantly larger in value, accounting for the bulk of the value of housing equity flows. In turn, the most important property transactors by value were those changing the number of properties owned.

The survey data show a significant life-cycle influence on housing equity flows, particularly among property transactors. Over 2004, the bulk of equity withdrawal was undertaken by older households, while younger households typically injected through deposits for property purchase or mortgage repayments. To our knowledge this intuitive result – evident both in bivariate and logit analysis – has not previously been demonstrated empirically. Age aside, there were few differences in the characteristics of households that injected without transacting and those that withdrew without transacting, although access to flexible mortgage features appeared to play some role in explaining household behaviour.

The use of equity withdrawn tended to vary with the method by which it was accessed. Withdrawals associated with property transactions were used significantly more for accumulation of non-property assets than consumption, a preference less evident for non-transaction-based withdrawals. Overall, around two-thirds of equity withdrawn by net withdrawer households in 2004 was mainly invested in other assets or used to pay down other loans. In contrast, only a relatively small proportion of equity withdrawn was mainly used to fund consumption in that year.

These results have some potentially important aggregate implications. Swings in housing equity withdrawal are likely to be heavily influenced by turnover in the property market, given the importance of such transactions to gross equity flows and the observation that the typical property transaction results in net equity withdrawal. This effect is likely to be amplified following a period of sustained house price growth, and is consistent with the large increase in aggregate housing equity withdrawal in Australia between 2001 and 2003, along with its subsequent decline. Secondly, the survey results also suggest that a significant number of households have used refinancing opportunities over recent years to increase the size of their debts, for purposes including consumption and renovation. Thirdly, only a relatively small portion of overall equity withdrawn from the housing stock in 2004 was used for consumption.