ECONOMIC RESOURCES

HOUSEHOLD DEBT

This article is also available as a PDF - Download the PDF here

INTRODUCTION

The onset of the global 'credit crunch' in mid 2007 and the subsequent financial turmoil appears to have discouraged households from taking on new debt. As an example, the number of owner occupied housing commitments (including refinancing) fell by 22% from the December 2007 quarter to the December 2008 quarter (from 200,000 to 156,000).

However, these developments follow an unprecedented period of growth in household debt in many developed countries over the past 30 years. (Endnote 1) In Australia, this was particularly the case from the early 1990s, as more competition among lenders increased the supply and reduced the cost of credit. (Endnote 1) On the demand side, the 1990s saw a decline in inflation leading to lower interest rates. This, combined with low unemployment and increasing income and wealth, meant that households were able to afford higher levels of debt than before.

Household debt is not in itself a cause for concern, as incurring a debt presents opportunities as well as risks. However, the rapid increase in household debt relative to the growth of income and wealth attracted a lot of attention. While some of the increase was used to finance consumption, most was used to buy assets. One concern about this high level of debt is that in the event of an economic downturn some households may have trouble servicing their debt. Households with high levels of debt that wish to reduce their level of gearing may be affected by falling asset prices.

HOW MUCH HAS DEBT INCREASED?

Based on information from the Reserve Bank of Australia, over the last 18 years the total amount of debt owed by Australian households rose almost six-fold. At September 1990 the level of household debt was almost $190 billion, increasing to around $1.1 trillion by September 2008 in real terms (i.e. adjusted to remove the effect of inflation).

Most debt was incurred to buy houses. Between 1990 and 2008, debt for investor housing increased from 11% to 27% of all household debt. Debt for owner occupier housing was consistently the largest component, ranging from 56% to 67% of debt (59% in September 2008). Other personal debt (for example credit card debt) halved as a proportion of all debt.

Relative to assets

Comparing levels of debt with assets provides context on how they have changed over time. Between September 1990 and September 2008, the ratio of total household debt to assets held by households rose from 9% to 19%. In other words, debt grew twice as fast as the total value of assets held by households. The sharp increase in the debt to asset ratio from December 2007 to September 2008 was due to a decline in the value of household assets.

Among the different types of debt, housing debt as a proportion of housing assets rose from 11% to 29%, which means overall, households have come to own a relatively smaller proportion of their houses. On the other hand, the total amount of equity households hold in their houses increased by 62%, from an average of $185,000 to $299,000 per household. Borrowing for owner-occupation and investment both contributed to the rise in housing debt. In contrast, the ratio of other personal debt to assets was around 2% in both periods.

WHO OWES MONEY?

Based on information from the Survey of Income and Housing, in 2005-06 there were 5.7 million households with some debt, that is, loans outstanding for houses, cars, investment, study or other purposes, or money owing on credit cards. The proportion of households with debt (72%) was similar to that in 2003-04 (73%).

In 2005-06, almost two-thirds (65%) of all households had non-property debt, including 55% with credit card debt. Almost one-third (32%) of households owed money for owner occupied property, and 12% had debt for rental properties and/or other property (including households living in rental properties who had borrowed money to buy or build a home somewhere else).



Age

Linked with changes in employment and income, people are generally more likely to owe money from young adulthood to middle age, and less likely to have debt during retirement and older age. Non-property debt was fairly common (between 69% and 77%) among younger and middle aged households (where the reference person was aged 15-24 years to 45-54 years) and then declined in older age groups. Owner occupied property debt increased steadily with age, peaking for households where the reference person was aged 35-44 years (51%) before declining in older age groups. Other property debt (such as debt used to buy a rental property) had a slight peak at 16-17% among households where the reference person was aged 35-44 and 45-54 years.

Income

In 2005-06, the prevalence of all types of debt increased with income, as the more income and assets a household has, the easier it is for that household to borrow money. The proportion of low income households (those in the lowest income quintile) with non-property debt (40%) was around half that of high income households (those in the highest income quintile). Owner occupier debt was less common among low income households (12%) than among households in the other income groups (ranging from 26% to 45%). Other property debt was concentrated among high income households, where 24% had this type of debt.

Households in the top two income quintiles owed almost two-thirds (64%) of all debt. These households accounted for 62% of owner occupier debt and almost three-quarters (73%) of rental property debt.



HOW MUCH DEBT?

From 2003-04 to 2005-06, the median amount of debt among indebted households rose from $37,700 to $50,500 (in 2005-06 prices), a real increase of 34%. While the average (mean) amount of debt was substantially higher than the median at both points in time ($101,000 and $128,000, respectively), the real increase in average debt was slightly lower (27%) over the period.



The gap between the median and average amount of debt indicates that debt, like income and wealth, is not distributed evenly and that, while a considerable number of households have relatively low levels of debt, there are some households with very high levels. In 2005-06, 39% of households had debt less than or equal to $20,000, and 4% had debt greater than $500,000.

The greatest real increases in the median amount of debt from 2003-04 to 2005-06 applied to loans for other property, loans for investment purposes (to buy shares, for example) and loans for other purposes (which all increased by around 40%). The median amount of owner occupier debt increased from $106,000 to $131,000.

INDEBTED HOUSEHOLDS: COMPONENTS OF DEBT - 2005-06

	Proportion with type of debt	Median amount owing on type of debt among households with type of debt
Type of debt	%	$'000
Owner-occupied property	44.9	130.8
Rental property	9.8	202.0
Other property	6.8	138.0
Investment	4.1	79.6
Credit card	76.6	2.3
Vehicle purchase	23.0	14.0
Other purposes	13.8	7.0
Student	16.3	9.0
	'000	$'000
Total households with debt	5 723.4	50.5
Source: ABS 2005–06 Survey of Income and Housing

Some home loans have features that enable the borrower to draw on their asset for non-housing purposes. This can help households pay for expenses such as medical bills, to buy a new car or help with living costs during a transitional phase such as unemployment. In 2005-06, almost one-quarter (24%) of households with a mortgage on their home had money owing on it for non-housing purposes (excluding business or investment purposes). The median amount owing among these households was $23,100. The proportion of lower income households (those in the second, third and fourth equivalised disposable household income deciles) with debt for other purposes on their mortgage was similar (27%) as was the median amount owing ($20,400).

Credit card debt and student loans

Among households with credit card debt, the median amount owing on credit cards did not increase by much from 2003-04 to 2005-06 ($2,200 to $2,300). This was also the case among those with student loans ($8,500 to $9,000).

In 2005-06, over half (58%) of all lower income households had one or more credit cards. This was the most common type of debt among these households, with 43% reporting some credit card debt. While the median amount owing for lower income households with a credit card debt was just over $1,700, 18% (180,000 households) reported a debt of $5,000 or more, including 60,800 households who owed $10,000 or more.

DEBT, ASSETS AND INCOME OVER THE LIFE COURSE

In 2005-06, the median amount owed by indebted Australian households was $50,500, the median value of their assets was $488,000 and their median disposable household income per year was $56,900. Relationships among debt, assets and income vary over the life course and are influenced by home buying patterns and other investment strategies. As an example, people borrowing money for investment purposes would have higher levels of debt than would otherwise be the case. However, the asset holdings of such households would be higher.

During the early stages of the life course, the level of debt and value of assets tend to be relatively low, reflecting low home buying rates. In 2005-06, the median level of debt among young lone person households (reference person aged under 35 years) ($23,100) was lower than the median for all households. The median value of their assets ($121,000) was also lower as was their median disposable income per year ($35,200).

For the main home buying groups, such as couples with young children (eldest child aged under five years), the level of debt tends to be high. Among life-stage groups, couples with young children had the highest median amount of debt ($167,000). While their median level of debt was much higher than the median for all households, the median value of their assets ($501,000) and median disposable income ($63,200) were not far above the corresponding figures for all households.

Households with low income and low assets are not as readily able to borrow as households that are better off financially. (Endnote 4) Among lone parent households, the median level of debt was relatively low ($23,400) as was the median value of assets ($234,000) and income ($39,100).

INDEBTED HOUSEHOLDS: SELECTED LIFE COURSE GROUPS - 2005-06

		Median debt	Median assets	Median net worth	Median disposable household income per year	Median ratio of debt as a proportion of assets	Median ratio of debt as a proportion of annual disposable household income	Total house- holds with debt	Proportion of households with debt among the life course group
Selected life course groups		$'000	$'000	$'000	$'000	%	%	'000	%
Lone person aged under 35		23.1	120.5	67.0	35.2	36.2	78.6	281.3	76.2
Couple only, reference person aged under 35		90.0	338.0	156.0	66.1	35.9	138.4	388.4	91.7
Couple with dependent children only
	Eldest child aged under 5	167.0	501.2	306.6	63.2	31.9	246.3	399.7	93.0
	Eldest child aged 5–14	148.1	560.5	377.6	65.8	25.7	217.2	782.6	91.1
	Eldest child aged 15–24	122.1	723.5	569.7	79.5	18.5	169.6	426.7	90.9
One parent with dependent children		23.4	233.5	139.1	39.1	19.3	60.6	364.2	67.6
Couple with
	Dependent and non-dependent children only	121.9	682.7	569.9	96.8	17.7	119.3	234.6	88.7
	Non-dependent children only	52.9	694.0	601.0	82.0	8.3	67.8	379.5	84.5
Couple only, reference person aged 55–64		19.0	807.5	743.8	51.1	3.2	32.5	368.3	72.7
Couple only, reference person aged 65 and over		1.6	590.0	582.5	30.3	0.7	3.9	296.5	43.7
Lone person aged 65 and over		0.9	374.0	362.2	16.7	0.8	4.4	218.9	29.4
Other		33.0	431.0	314.1	51.2	15.1	66.3	1 582.6	72.2
All households		50.5	488.0	359.8	56.9	16.1	89.8	5 723.4	72.2
Source: ABS 2005–06 Survey of Income and Housing

A household approaching retirement age may borrow against the equity in its home, to buy shares for example. Households at the pre-retirement stage, such as couples with non-dependent children only, had a relatively high median level of debt ($52,900) but the median value of their assets ($694,000) and income ($82,000) were also high. A small proportion (7%) of households in this life-stage group had rental property debt and/or loans for investment purposes.

Among older households, such as couples where the reference person is aged 65 years and over, most debt has been paid off and the value of assets is generally high while income is low. In 2005-06, the median level of debt for this group was very low ($1,600) while the median value of their assets was $590,000 and median income was $30,300 per year.

DEBT SERVICING

The extent to which repaying their debt places a financial burden on households, in particular households with relatively low income, is also of interest. In 2005-06, there were 526,000 lower income households with a mortgage for owner occupied property, representing 23% of households in this income group. The median proportion of gross household income spent on mortgage repayments among lower income households decreased slightly in the ten years to 2005-06 (from 27% to 23%). However, there was a slight increase in the two years following 2003-04 from 20% to 23%, which may have been related to higher interest rates and house prices. Among these lower income households in 2005-06, 20% reported that they were paying between 30% and 50% of their gross household income on mortgage repayments and one-tenth were paying more than 50%.



HOUSING LOAN ARREARS

Another measure of the extent to which households are having difficulties servicing their debt is whether they are behind on their mortgage repayments. In September 2008 the Reserve Bank estimated that there were around 17,000 borrowers who were more than 90 days behind on their mortgage repayments, up from 15,000 borrowers in March 2008. (Endnote 5) These and other arrears data cover a period of higher interest rates; when later data are available they may show some impact from the recent falls in interest rates. The share of housing loans that are in arrears has increased. From January 2001 to June 2008, the proportion of securitised housing loans more than 90 days in arrears increased from 0.15% to 0.35%.

There has been a general increase in arrears rates since 2003, in part reflecting the easing of credit standards over the past decade. This meant that many borrowers were able to obtain a housing loan who previously may not have been eligible, and many others have been able to borrow larger amounts. (Endnote 6) From September 2003 to June 2008, the 90-day arrears rate for housing loans on banks' balance sheets (which account for around 75% of all outstanding housing loans) increased from 0.18% to 0.41%. However, it is currently no higher than it was in the mid 1990s, and low by international standards. (Endnote 7) The arrears rate for prime securitised housing loans increased from 0.21% to 0.57% from January 2001 to June 2008. (Endnote 5) For securitised prime low-doc loans, the arrears rate was 1.2% in June 2008, more than double that for securitised prime full-doc loans. The arrears rate for non-conforming loans was much higher at 8.5%.

In New South Wales

The overall arrears rate for prime securitised loans was higher in New South Wales than the other states and territories, and increased to a greater extent between April 2003 and July 2008 (from 0.18% to 0.84%). Higher arrears rates in New South Wales have been related to relatively weak economic conditions and housing markets in areas of the state. As an example, the increase has been greatest in western Sydney (up from 0.25% in March 2004 to 1.28% in July 2008) where house prices have been under downward pressure and where a disproportionately large number of borrowers took out investment housing loans around the peak of the house price cycle. (Endnote 5 and 6) The 2004–2006 increase in arrears rates in New South Wales resulted in a sharp increase in the number of court applications for property repossession as a proportion of the dwelling stock, from 0.10% in 2003 to 0.22% in 2006 (steady at 0.22% in 2007). (Endnote 5 and 6)

LOOKING AHEAD

Although Survey of Income and Housing data suggest that during 2005-06 most households with high levels of debt were well placed to service it by way of high income and assets, more recently financial conditions for many households have changed. The banks' standard variable interest rate went up by 205 basis points between July 2006 and August 2008, and since then has fallen sharply in the wake of the global financial crisis. (Endnote 8) More recent data, such as housing loan arrears rates, indicate that some households have been facing a more difficult financial situation. Most recently, the level of wealth has fallen in many households, affected by substantial declines in share prices and the value of housing assets falling slightly. (Endnote 5) However, households with mortgages may also be starting to benefit from lower interest rates. Housing comprises nearly 60% of household assets (and the majority of household debt) in Australia.

OTHER INFORMATION

Data sources and definitions

The main sources of data used in this article are the ABS 2003-04 and 2005-06 Surveys of Income and Housing (SIH). Reserve Bank of Australia (RBA) tables are also used.

'Households', as used in ABS surveys, refer to people who usually live in private dwellings in Australia. 'Household sector', as used by the RBA, comprises all people, including those living in non-private dwellings such as aged care residential facilities. About 2% of the Australian population are outside the scope of the SIH. These people would be included in the RBA data. The household sector excludes unincorporated businesses and non-profit institutions serving households for levels of debt, but includes their financial assets for levels of assets.

'Debt or liabilities' in the SIH includes: principal outstanding on loans for owner-occupied dwellings, rental and other property (including non-residential property), cars, investments and other purposes; debt outstanding on study loans; and amounts owing on credit cards. As used by the RBA, debt refers to loans made for owner-occupier and/or investor housing (i.e. residential dwellings) and other personal debt (personal loans and credit card debt).

'Assets' are owned by the members of a household, and provide economic benefits. In the SIH, assets include money held in bank accounts, the family home, other property and land, cars and home contents, amounts accumulated in superannuation funds, shares, trusts, debentures and bonds. The term 'assets' as used by the RBA includes dwellings (such as houses and apartments), consumer durables (such as cars and furnishings) and financial assets (such as shares). For more information see the RBA Bulletin Notes to Tables <http://www.rba.gov.au>.

'Median' refers to the level of a measure (such as income) which divides the units in a group (such as households) into two equal parts, one half having, for example, incomes above the median and the other half having incomes below the median.

Income definitions

'Gross household income' refers to the sum of income from all sources (wages and salaries; profit/loss from own unincorporated business; investment income; government pensions and allowances; private cash transfers) before income tax or the Medicare levy are deducted. Deducting these produces disposable household income.

'Equivalised disposable household income' is disposable household income adjusted to take account of the economies of scale obtained when more than one person lives in a household and shares housing costs and other expenses. For a lone person household it is equal to disposable household income.

'Quintiles' are the groupings that result from ranking all households or people in the population in ascending order according to some characteristic such as their household income and then dividing the population into five equal groups, each comprising 20% of the estimated population. 'Deciles' are ten equal groups, each comprising 10% of the estimated population.

For more information see Household Income and Income Distribution, 2005–06 (ABS cat. no. 6523.0).

Negative gearing

Taking on debt for investment purposes, for example to buy shares or a rental property, can offer tax benefits to the borrower. Negative gearing applies when the costs of earning income (such as interest on a loan) are greater than the income earned from the investment. This difference is a loss for taxation purposes, and so reduces the amount of tax borrowers pay, which makes it cheaper to borrow.

The annual tax deductions allowed for rental properties provide another perspective on increased investment in rental properties. In 1995-96, 1.1 million Australians declared net rental income on their tax returns. Of these, 68% claimed rental interest deductions, totalling $6.4 billion. Over half (56%) had a taxable loss after interest and other deductions (that is, they had negatively geared rental properties). (Endnote 2) By 2005-06, 1.6 million individuals declared net rental income on their tax returns. Most (79%) claimed rental interest deductions, which amounted to $13.8 billion in initial taxation deduction claims. Just over two-thirds (67%) had negatively geared rental properties. (Endnote 3) While some of the increase in interest deductions can be explained by higher interest rates, most is related to an increase in the number of investors and in the amount being borrowed for investment. Over that same decade gross rental income reported for tax purposes increased six-fold to just over $19 billion.

Housing loan definitions

'Prime loans' are provided to borrowers who meet the standard lending criteria of mainstream lenders.

'Full-doc loans' are provided to borrowers who are able to provide full documentation as proof of their income, assets and debt.

'Low-doc loans' require less documentation to prove the borrower's income, assets and debt.

'Non-conforming loans' are provided to borrowers who do not meet the standard lending criteria of mainstream lenders, such as those with impaired or incomplete credit histories. These currently make up less than 1% of outstanding housing loans.

'Securitised loans' are loans that have been converted to 'residential mortgage-backed securities' and sold to institutional investors via a trust or a company known as a special purpose vehicle. These loans generally have a higher risk associated with them. They currently make up around one-quarter of outstanding housing loans, with the remainder being on banks' balance sheets.

'90-day arrears rate' is the value of housing loans 90 or more days in arrears as a proportion of the value of all housing loans.

Recent changes in household borrowing

While the global economic downturn is still unfolding, its impact on households in terms of preparedness to take on debt is evident across a range of indicators. The following graphs provide some indications of recent changes in borrowing by Australian households as well as mortgage interest rates and household wealth.

The seasonally adjusted number of housing finance commitments declined from around 65,000 per month throughout the December quarter of 2007, to an average 51,000 per month in the December quarter of 2008.

December 2008 and January 2009 have seen some recovery, with 54,000 and 56,000 commitments respectively in these months. This upswing can be attributed to an increase in first home buyer commitments.



Following the introduction of the First Home Owners Boost in mid October 2008, the number of first home buyers increased in November and increased further in December and January.

The proportion of all home finance commitments made up by first home buyers increased to 24% in November 2008, and to 26% in December 2008 and January 2009, having been less than 20% since early 2002.



The value of personal finance obtained has been in general decline since late 2007.

The average monthly amount committed in the December quarter of 2007 was $7.5 billion, falling to an average $6.3 billion in the three months to December 2008, and was the lowest level since the June quarter of 2002 ($6.2 billion).


In February 2009, the banks' standard variable interest rate charged on housing loans was 5.85%, the lowest it has been in decades.

This follows the longer period of interest rate increases from May 2002, which peaked at 9.6% in July and August 2008.


Household net worth (the value of assets less liabilities) peaked at $5,320 billion in December 2007.

In 2008, falling asset values produced a decline in the overall net worth of Australian households, to $4,780 billion in September 2008, similar to the level in September 2006 ($4,700 billion).



Over the past decade, household debt grew much faster than income. The ratio of household debt to annual disposable household income peaked at 160% in December 2007 and March 2008. The ratio decreased over the last three quarters, reaching 153% in December 2008.

Housing debt as a proportion of disposable income followed a similar pattern, and made up 87% of all debt in December 2008.


(a) Levels of household debt and debt components have been adjusted for breaks in the series (the establishment of new banks and other changes in reporting arrangements)
(b) Disposable income
Source: RBA Bulletin Statistical Table B21

Definitions

'Housing finance commitments' are loans provided to people for the purpose of buying or building owner occupied dwellings.

'Personal finance commitments' are loans given to people for personal use, such as to purchase a car, household goods or a holiday. Revolving credit, for example credit cards, is included.

'Disposable income' is sourced from the ABS national accounts (ABS cat. no. 5206.0) and is reported in annual terms. It excludes gross mixed income earned by unincorporated enterprises and is measured before the deduction of net interest payments.


ENDNOTES

1. Kent, C, Ossolinski, C & Willard, L 2007, 'The Rise of Household Indebtedness' in Kent, C, and Lawson, J (eds) The Structure and Resilience of the Financial System, Proceedings of a Conference, Reserve Bank of Australia, Sydney, viewed 24 April 2008, <http://www.rba.gov.au>.

2. Australian Taxation Office 1998, Taxation Statistics 1995–96 Detailed Tables, Australian Taxation Office, Canberra.

3. Australian Taxation Office 2008, Taxation Statistics 2005–06, Australian Taxation Office, Canberra, viewed 9 April 2008, <http://www.ato.gov.au>.

4. Headey, B, Warren, D & Wooden, M 2008, The structure and distribution of household wealth in Australia: cohort differences and retirement issues Social Policy Research Paper No. 33, Department of Families, Housing, Community Services and Indigenous Affairs, Canberra.

5. Reserve Bank of Australia 2008, Financial Stability Review September 2008, Reserve Bank of Australia, Sydney, viewed 25 September 2008, <http://www.rba.gov.au>.

6. Reserve Bank of Australia 2008, Financial Stability Review March 2008, Reserve Bank of Australia, Sydney, viewed 28 March 2008, <http://www.rba.gov.au/>.

7. Reserve Bank of Australia 2008, 'An Update on Household Finances, Address by Mr Ric Battelino, Deputy Governor, to 7th ITSA Bankruptcy Congress Sydney, 30 October 2008' in Reserve Bank Bulletin, November, viewed 20 November 2008, <http://www.rba.gov.au/>.

8. Reserve Bank of Australia, Bulletin Statistical Table F5 Indicator Lending Rates, viewed 15 January 2009, <http://www.rba.gov.au>.