Financial Stability Review – September 2005 Box A: Rates of Indebted Home Ownership
The rapid growth in housing debt over recent years is largely accounted for by an increase in average loan size. For example, since 1996 the average new owner-occupier housing loan has more than doubled in size from $99,000 to $215,000, and the ratio of the average new loan to average household income has increased by 50 per cent (Graph A1). These outcomes largely reflect the fall in nominal interest rates that was associated with Australia's return to low inflation in the early 1990s, and a relaxation of banks' lending criteria.
The growth in housing debt is, however, also partly explained by an increase in the number of households with housing debt. Although comprehensive time-series data are not available, the number of households with owner-occupier housing debt appears to have increased by over 40 per cent since 1996, considerably faster than growth in the total number of households. With the share of households that are owner-occupiers remaining relatively stable over this period, the result has been an increase in the share of owner-occupier households with housing debt.
The main sources of data on home ownership in Australia are the Census, the Household Expenditure Survey (HES), the Survey of Income and Housing (SIH) and the Household, Income and Labour Dynamics in Australia (HILDA) Survey. The various surveys show similar and broadly steady rates of overall home ownership, with just over 70 per cent of households owning their home either with or without debt (Graph A2). On the share of owner-occupier households with housing debt, there is greater variation among the surveys, but all show some pick-up in this share since the mid 1990s. For example, over the decade to 2004, the HES suggests that the share of households with debt secured on their home has increased from less than 30 per cent to 36 per cent. The current share is broadly comparable to that in the early 1980s although housing debt levels were much lower relative to income at that time.
The fact that the latest Census estimate (for 2001) of the share of owner-occupier households with housing debt is below estimates from the other surveys is partly explained by the relatively narrow nature of the Census questions on housing tenure. In particular, the Census asks households if their home is ‘being purchased’, whereas the other surveys effectively ask households if they have ‘mortgages or other loans secured over the property’. To the extent that households increasingly have debt secured on their dwelling for purposes other than purchasing the dwelling, such as home-equity loans used to finance renovations or other spending, the Census will tend to underestimate the proportion of owner-occupiers with housing-secured debt.
The upward trend in the share of owner-occupier households with housing debt reflects a number of factors. One is that households now have larger debts relative to their income than was the case previously, and therefore the average time taken to pay off the debt is likely to have increased. This is particularly the case where households draw down home-equity loans, or refinance and take on a larger loan when the value of their property rises. A second reason is that there has been an increase in the share of households owning investment properties, with investors – who are typically (but not universally) owner-occupiers – often having some debt secured on their primary residence.
In addition to owner-occupiers with housing debt, a number of renter households also have housing debt, typically on investment properties. While there is little information on the evolution of the importance of these households, the HILDA Survey suggests that around 7 per cent of renter households had property debt in 2002.
These figures exclude owner-occupier households that only had housing debt secured over investment or other residential properties.