Financial Stability Review – September 2014 Box C: Households' Investment Property Exposures: Evidence from Tax and Survey Data
The characteristics and risk profile of households' investment property exposures warrant close examination given the recent strength of investor demand for housing. Investor housing loan approvals currently account for almost 40 per cent of the value of total housing loan approvals, similar to their share in the early 2000s, a period of rapid housing price inflation and strong investor demand (Graph C1). As a result, lending to households for property investment currently accounts for around 20 per cent of banks' total lending. This box reviews households' investment property exposures and resulting risk factors, using data from the Australian Taxation Office (ATO) up to 2011/12 and the 2010 Household, Income and Labour Dynamics in Australia (HILDA) survey – the latest data available from both sources.
Investor housing lending typically has attributes that differ from those of owner-occupier loans and that affect its risk profile.
- Because interest expenses on investment property are tax-deductible, investors have stronger incentives than owner-occupiers to take out interest-only loans. In Australia, around 64 per cent of loan approvals to investors are interest-only loans compared with 31 per cent to owner-occupiers. The typical interest-only period on these loans is around five years, though up to 15-year periods are also available. During this period, the loan principal is usually not being paid down, although liaison with banks suggests that some borrowers with these loans do make discretionary repayments. If the loan balance is not declining via principal repayment, it is more likely that it will exceed the property value (be in negative equity) if housing prices should fall. There is also a risk that the borrower could face difficulty servicing the higher (principal and interest) repayments after the interest-only period ends. To reduce this risk, banks assess borrowers' ability to service the higher repayments.
- Investor loans tend to have lower loan-to-valuation ratios (LVRs) at origination compared with owner-occupier loans. Part of this is likely to be driven by investors seeking to avoid the cost of lenders mortgage insurance, which is typically required for loans with an LVR greater than 80 per cent. Some institutions also have lower maximum LVRs for investor loans, partly to offset the risks from lower repayments noted above.
According to ATO data, the share of the population aged 15 years and over with an investment property grew steadily through the 1990s and early 2000s, before stabilising in the late 2000s at around 10 per cent (Graph C2). Over the same period, the share of these investments that were geared – where the investor claimed interest deductions – increased steadily before levelling off at a little over 80 per cent. Given the sharp increase in investor loan approvals over 2013 and 2014 to date, especially in New South Wales and Victoria, the share of investors is likely to have increased further of late.
While aggregate measures of property investment and gearing increased only modestly from the early 2000s to 2011/12, there were some notable changes in the distribution of investment and gearing across age groups. In particular, the share of property investors aged 60 years and over increased significantly, to account for around one-fifth of investors in 2011/12 (Graph C3). This shift in the distribution of investors towards older individuals reflects both the ageing of the population as well as an increase in the extent of investment property ownership within this age group. Individuals in this group also became more likely to have a mortgage against their property investment, with around half of these investors claiming interest deductions from their total incomes, though borrowing remained far more prevalent among younger investors, with almost all investors below the age of 40 years being geared.
ATO data also show that the incidence of property investment and the incidence of geared property investment both increase with total income (Graph C4). While the incidence of investment fell between 2003/04 and 2011/12 for people with total incomes between $15,000 and $500,000, it increased for those with very low incomes and those with very high incomes. For investors with very low incomes (below $15,000), individuals aged 60 years or older comprised a larger share in 2011/12 than in 2003/04, as an increasing number of baby boomers owning investment properties entered retirement. These investors may be more capable of servicing any investment property debt than their younger low-income counterparts. In particular, even though their total income (for tax purposes) is low, they are likely to have non-taxable sources of income (such as pension streams and drawdowns from superannuation funds) to draw on.
The HILDA survey provides further insight into the finances and total debt outstanding of geared property investors. After-tax (disposable) income can also be calculated, which is arguably a better measure than total gross income (as in the ATO data) for assessing the ability of borrowers to service their debt. Based on these data, investor households with incomes in the top 20 per cent of the income distribution owe the bulk of the investor housing debt and over a quarter of total housing debt outstanding. These households appear fairly well placed to service their debt: the highest income earners are typically using less than 30 per cent of their income to service their total property debt, and more than half are ahead of schedule on their owner-occupier mortgage repayments (Table C1).
At the other end of the income distribution, the HILDA survey suggests that households in the bottom 20 per cent account for just 2 per cent of investor housing debt. However, these investors typically have a much higher debt-servicing burden compared with their higher-income counterparts. Consistent with the tax data, the HILDA survey finds that one-quarter of these geared low-income property investors are retirees. HILDA data suggest that although the debt-servicing burden of retired low-income investor households is high, these investors typically have built up more liquid assets to draw on should they fall into difficulty servicing their property debt, than their younger low-income counterparts.
Property investors are defined as individuals who declare a net rental profit or loss on their individual tax return. The data do not distinguish between rental receipts/deductions from residential and commercial property. Property investments held in self-managed superannuation funds are not captured in these data. 
Total gross (before-tax assessable) income includes wages and salaries, net rental income, net capital gains, income derived from financial assets and certain other income items. It excludes non-taxable sources of income such as pension streams and drawdowns from superannuation funds. The average total income per taxpayer was around $55,000 in 2011/12, excluding these non-taxable sources. 
The increase in lower-income individuals aged 60 years and over was also partly driven by changes to taxation of superannuation in July 2007. Total income as recorded on the individual's tax return fell for those aged 60 years and over after the changes, because some superannuation benefits that had previously been taxed became tax-free and are therefore no longer recorded in total income. 
Debt servicing in the HILDA survey is measured as ‘usual repayments’. For an amortising loan it is likely to capture scheduled principal and interest repayments as well as any regular excess repayments made by the borrowers.