Bulletin – May 2004 Residential Property Investors in Australia^[1]

Over the past decade, there has been a substantial increase in investor activity in the residential property market. As a result, borrowing for investment purposes now accounts for a much larger share of total mortgage borrowing than it did a decade ago.

Despite the increased prominence of investor activity in the housing market, until recently detailed data on investors have been sparse. Although some data were available from income tax records, ABS household surveys, ABS Censuses and the 2001 HILDA Survey, these data only provided a partial picture of investor households and their economic characteristics. Recently, the 2002 Household, Income and Labour Dynamics in Australia (HILDA) Survey has become available. This survey, compiled by the Melbourne Institute of Applied Economic and Social Research, provides the first comprehensive data set on the demographics, debt and assets of residential property investors in Australia.

Property Investors in the HILDA Survey

The 2002 HILDA Survey contains a number of questions that help to identify households owning investment properties. For the purposes of this article, we identify as property investors those households that reported owning a residential property that was not their primary residence and reported receiving rental income. According to this definition, 10.3 per cent of all households owned investment properties.

However, this figure of 10.3 per cent of households is likely to be a lower bound on the number of property investors. A further 6.5 per cent of households reported owning a residential property that was not their primary residence but did not report receiving rental income. While most of these responses may represent holiday homes or instances where a family member or friend was occupying the property and not paying rent, some proportion presumably represents cases where an investment property was vacant and not yielding income or where the respondent did not answer the question about rental income correctly. In addition, the survey also included some households (around 3 per cent) that reported receiving rental income but did not report owning a second property. Although this group would include households earning income from lodgers or from letting non-residential property, it may also include an additional group of investors in residential property.

Overall, these figures confirm the general view that investment property ownership in Australia is widespread, and is considerably higher than that in other countries, such as the US, Canada or the United Kingdom.^[2]

Demographic Characteristics

Not surprisingly, investor households differ from non-investor households in a number of aspects. The survey revealed that investor households tended to be somewhat older than the rest of the population (Graph 1). The median age of the reference person in investor households was around four years older than for non-investors, largely reflecting a greater concentration of households in the 35 to 64 age groups, especially those aged between 45 and 54. The bunching of investors in the working-age segment of the population would be consistent with tax considerations being one of the motivations for owning investment property.

Investor households across all age groups tended to have higher incomes than their non-investor counterparts (Graph 2). In line with this, 71 per cent of investors were in the top two household income quintiles and the propensity to own an investment property increased at higher incomes, with households in the highest quintile being almost three times more likely to own an investment property than households in the middle quintile (Graph 3). Associated with their higher income, investor households were much more likely to be working than their non-investor counterparts, and especially to be self-employed.

Not surprisingly, investor households were also wealthier than non-investor households, partly because higher wealth reflects property ownership (Graph 2). More than half of all investor households were in the quintile with the highest net wealth, while only 2 per cent were in the least wealthy quintile (Graph 3).

A conventional view of property ownership is that households tend to purchase their own home before buying an investment property. The HILDA Survey provides some support for this view, with 85 per cent of investor households also owning their own home, considerably more than the 68 per cent of home owners in the general population. Still, at 15 per cent, a surprisingly large share of investors were households who rented, but owned a residential property which they did not normally reside in. The latter group tended to be younger than those investors who also owned their own home: the median age of renter investors was 39 years, versus 49 years for owner-occupier investors.

Investor Property Gearing

The HILDA Survey also provided details on assets and debt associated with residential property ownership. The median value of investment property assets reported was $250,000, the same as the median value for all owner-occupied homes (Table 1). For owner-occupier investors, the typical value of the own home was higher, at $310,000, in line with the finding that investors tended to be wealthier than non-investors.

Previous snapshots of Australian investors have suggested that many investment properties are geared. The HILDA Survey confirmed this, with 55 per cent of those households owning an investment property also owing debt on that property. However, this figure may understate the true proportion of geared investment properties. In particular, it is considerably lower than would be implied by ATO data, since 74 per cent of rental income earners deducted interest payments on their tax returns in 2000/01. This discrepancy is puzzling, but might in part be explained by the different definitions of investors used in the two sources.^[3]

The proportion of investor households with one or more investor property loans was somewhat higher than the proportion of all owner-occupier households with debt on their own home (55 versus 50 per cent). In addition, the median size of an investor property loan ($120,000) was larger than that of a typical owner-occupier home loan ($90,000). In fact, the median gearing (the debt-to-asset ratio) on investment properties on which debt was outstanding was, at 52 per cent, markedly higher than the 40 per cent for owner-occupied homes with debt outstanding. This difference would be consistent with the different tax treatment of interest payments on investment, as opposed to owner-occupied, property.

Renter investors were much more likely to have debt on their investment property (85 per cent) than investor households who owned their own home (50 per cent). Moreover, geared investment properties owned by renters typically had somewhat higher debt-to-asset ratios than those owned by owner-occupiers (58 versus 51 per cent).

The median size of investor loans increased with income, but so did the value of investor properties (Graph 4). As a result, the gearing of investor properties did not change much across income quintiles. In addition, investment properties on which debt was outstanding had a somewhat higher median value than those which did not have debt outstanding, possibly because gearing allows investors to purchase higher-valued properties.

The likelihood of having debt on investment properties and the level of gearing were clearly associated with age (Table 2). Young investor households were much more likely to gear their investment properties than older investors: more than two-thirds of property investors in the 25–44 year age group used gearing. In addition, among those investors with debt outstanding on investment properties, the typical gearing ratio was higher for younger investors. Young investors were also more likely than older investors to report receiving a negative net income on their property investment: indeed nearly 50 per cent of investors in the 25–44 year age group reported negative or zero net income from letting their property. Interestingly, 26 per cent of residential property investors in this age group were renters. Taken together, these observations, and the fairly sharp contrast between young investors and those at or near retirement age would be consistent with both life-cycle and tax considerations playing a role in decisions about the purchase and financing of residential property as an investment.

Conclusion

Given the sharp growth in borrowing for residential property investment in recent years, it is noteworthy that there has been relatively little hard data on this important aspect of household behaviour. The 2002 HILDA Survey has partly filled the gap in this area. It shows that ownership of investment property in Australia is quite widespread and is positively correlated with both income and wealth. In addition, a significant proportion of property investment is by households or individuals that rent the home in which they are living. Finally, the survey shows that borrowing is an important source of finance for such investment. Indeed, investors are more likely to have loans on their properties than are owner-occupiers, and investment properties are typically more highly geared than owner-occupied properties.

Footnotes

This article was prepared by Natalie Parlett and Anthony Rossiter of Economic Group. For previous Reserve Bank articles on residential property investors see ‘Recent Developments in Housing: Prices, Finance and Investor Attitudes’, RBA Bulletin, July 2002, ‘Innovations in the Provision of Finance for Investor Housing’, RBA Bulletin, December 2002, ‘Household Debt: What the Data Show’, RBA Bulletin, March 2003, ‘Submission to the Productivity Commission Inquiry on First Home Ownership’, RBA Occasional Paper No 16, ‘Tax Data on Households' Property Investment Exposures’, Financial Stability Review, Box A, March 2004. [1]

For more details on data available for investment property ownership in other countries, see ‘Submission to the Productivity Commission Inquiry on First Home Ownership’, RBA Occasional Paper No 16. [2]

For example, the tax data refer to individuals while the data from HILDA refer to households. Moreover, the tax data also include rental income earned from letting a room in an owner-occupied property. [3]