Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Wealth on retirement: latest estimates for Australia. Paper presented to the Ninth Annual Colloquium of Superannuation Researchers, University of New South Wales, 9-10 July 2001

Download PDFDownload PDF

National Centre for Social and Economic Modelling

• University of Canberra •



Simon Kelly

Paper Presented to the Ninth Annual Colloquium of Superannuation Researchers University of New South Wales

9-10 July 2001

National Centre for Social and Economic Modelling • University of Canberra •

The National Centre for Social and Economic Modelling was established on 1 January 1993, and supports its activities through research grants, commissioned research and longer term contracts for model maintenance and development with the federal departments of Family and Community Services, Health and Aged Care, and Education, Training and Youth Affairs. NATSEM aims to be a key contributor to social and economic policy debate and analysis by developing models of the highest

quality, undertaking independent and impartial research, and supplying valued consultancy services. Policy changes often have to be made without sufficient information about either the current environment or the consequences of change. NATSEM specialises in analysing data and producing models so that decision makers have the best possible quantitative information on which to base their decisions. NATSEM has an international reputation as a centre of excellence for analysing microdata and constructing microsimulation models.

Such data and models commence with the records of real (but unidentifiable) Australians. Analysis typically begins by looking at either the characteristics or the impact of a policy change on an individual household, building up to the bigger picture by looking at many individual cases through the use of large datasets. It must be emphasised that NATSEM does not have views on policy: all opinions are the authors’ own and are not necessarily shared by NATSEM or its core funders.

Director: Ann Harding © NATSEM, University of Canberra 2001 National Centre for Social and Economic Modelling University of Canberra ACT 2601 Australia

170 Haydon Drive Bruce ACT 2617

Phone + 61 2 6201 2750 Fax + 61 2 6201 2751 Email Website

NATSEM research findings are generally based on estimated characteristics of the population. Such estimates are usually derived from the application of microsimulation modelling techniques to microdata based on sample surveys. These estimates may be different from the actual characteristics of the population because of sampling and nonsampling errors in the microdata and because of the assumptions underlying the modelling techniques. The microdata do not contain any information that enables identification of the individuals or families to which they refer.

For additional information contact: Simon Kelly NATSEM University of Canberra CANBERRA ACT 2601

Ph : 61-2-6201-2784 Fax : 61-2-6201-2751 E-mail:

This paper is available at:

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 1

Wealth on Retirement -

Latest Estimates for Australia

Simon Kelly

National Centre for Social and Economic Modelling

University of Canberra, Australia

1. Introduction

For most people, the option to retire from the workforce and then receive an income from the government for the remainder of their life is a very attractive one. In Australia, the Age Pension

provides this option for people who meet minimum age and residency requirements and pass assets

and income tests. The catch is that the maximum pension is set at 25% of average weekly earnings

and most people would require supplementary income to the Age Pension to be able to live

comfortably. In 1998, the minimum age for women to be eligible was 61 and for men it was 65 (the

pension age for women is being increased from 60 to 65 over a 20 year period to align with that for


The value of assets held by a person as they approach their 61st or 65th birthday will have a major

impact on the decision about whether to permanently retire or not. Will the pension plus the

income produced from assets allow the person to live comfortably? A better understanding of the

assets owned by people about to make this decision would allow us answer this question.

This paper looks at the estimated wealth of people in the year preceding possible entitlement to the

Age Pension, i.e. men aged 64 and women aged 60. This About to Retire group is compared with

the remainder of the Australian population in regards to general characteristics and wealth. The

physical and financial assets of these two groups are estimated and compared. While there is very

little direct information available on people’s assets, a picture of wealth can be derived from a

variety of sources. Some assets, such as the value of the family home and the outstanding mortgage

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 2

are available directly from ABS survey data, other assets can be estimated through the income that

the assets produce, while still others can be estimated by combining information from more than

one source. All of these methods are used in this paper.

Some of the methods used in this paper to estimate wealth have been used before and simply

substitute the appropriate numbers for that year to produce new estimates, while other estimates use

either improved techniques or new sources to improve the imputation process.

After describing what is meant by wealth and the techniques used to model it, an updated profile of

wealth in Australia is presented. After looking at the general population, the About to Retire group

is compared with the Australian population at large. Non-wealth differences between this group

and the remaining population are discussed , and then comparisons of the forms of assets held and

the value of these assets are presented.


Wealth in this paper is used in the sense of “Net worth”, net worth being defined as the total value of assets held by the household minus total debt (Podder and Kakwani, 1973). Wealth is regarded as

the sum of the following assets less the following liabilities:


- Interest bearing deposits (savings and deposit accounts)

- Dividend paying investments (Equities, royalties)

- Owner-occupied housing

- Investment rental properties

- Net equity in businesses

- Superannuation


- Home mortgages

- Rental investment mortgages

This is clearly not a complete list of assets and liabilities. However, it does provide good coverage

of the more significant items owned by most people. Notable items excluded from the above list

are the value of consumer goods (including cars, antiques and artworks), cash holdings, zero-

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 3

interest accounts, zero-dividend shares, life insurance and credit card debts. These items have not

been included because distributional data is not readily available on these items or the valuations

are transitory in nature. The net effect of the exclusions is likely to result in an underestimate of the

wealth of the very rich and an overestimate for the very poor. This is because personal loans and

credit card debts are likely to be small in comparison to the values of the consumer goods and other

excluded assets for the very rich while the opposite is most likely the case for the very poor.

2. Techniques Used to Model Wealth

As noted in the previous section, information on wealth holdings is limited. It is often necessary to use some indirect method of obtaining a valuation on an asset. Information on assets is usually

found in one of three forms. Firstly, direct information on asset values or liabilities may be

available. This is the ideal situation. Secondly, an income stream related to ownership of a

particular asset may be available. The income stream can then be used to estimate the value of the

asset. The third, and the least informative, is an indication of ownership of a particular asset. All

of these forms were used to estimate wealth for this paper.

In this paper, current data from the 1997-98 ABS Survey of Income and Housing Costs

confidentialised unit record file (SIHC CURF) is used as the basis for calculating wealth. This is

the latest available unit record file and contains detailed information on people aged 15 and over

living in private dwellings. Dependent children were removed for this analysis. The SIHC data

was collected between July 1997 and June 1998. The prime reason for selecting this source is that

direct information on housing values and mortgages is available and for most Australians the equity

in the family home is the dominant form of wealth. The file also contains significant detail on

income sources and income streams, such as interest bearing deposits, dividend-based assets and

rental properties. This income information, when combined with other external information,

allows the value of the underlying assets to be calculated. Finally, the presence of business income

and the presence of superannuation income on the SIHC can be combined with other external data

to estimate the worth of other assets.

In the next few paragraphs the techniques used to impute the various assets are discussed.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 4


The 1998 SIHC does not contain current bank or any other investment balances. However, it does contain current weekly income from interest. The SIHC also contains flags indicating the sources

of interest income. However, the flags relate to investment income received in the last financial

year rather than current income. These flags are not used in this assessment of current asset values.

The value of the underlying asset can be imputed using income capitalisation. This technique uses

the known interest income and an estimate of the interest rate received to calculate the value of the

asset. For example, if someone receives $100 in interest income and the interest rate is 5.0% then

the value of the asset can be estimated as 100/0.05 = $2,000. Estimating asset values in this way

has been done by a number of researchers including Dilnot (1990), King and Bækgaard (1996) and

Wood (1999). For this technique to capture the full value of the asset the following must be met:

- All assets must produce income i.e. there cannot be any assets receiving zero interest

- All income must be reported in the survey

- The actual interest rate paid on each asset must be known.

In reality none of these requirements are fully met, but reasonably reliable estimates can be made

concerning each of them. The third point, the actual interest rate received by an individual, is the

area of most uncertainty. Most researchers take a weighted average interest rate and apply that rate

to each person with interest income. In King and Bækgaard, the calculation is enhanced through the

use of a distribution of rates rather than just one weighted average. The distribution is based on the

range of interest rates available at that time and the proportion of funds in each type of investment.

King and Bækgaard use the distribution to assign an interest rate to each individual according to the

probabilities of holding a particular type of investment and the spread of available interest rates for

that type of asset. Only one interest rate is assigned to each person.

This paper builds on the work of King and Bækgaard by combining the distribution of interest rates

with the possibility of assigning more than one rate to each person. The calculation assumes that

each person receiving interest income has some money in a low interest “at call” account. This is

probably the account into which earned income is paid and from which day-to-day living expenses

are made. If a person has more cash than required for this day-to-day living then it is assumed that

the excess is invested in an account or form that pays higher interest. For the purposes of the

capitalisation calculations, it is assumed a cash balance of up to $5000 will be at the low interest

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 5

rate, while any amount over $5000 will be at a higher “investment” rate. Following the

methodology outlined in King and Bækgaard, a distribution of interest rates is used for selecting

both the low and high rates.

Table 1 1997-98 Interest Rates

1997/98 Summary




Share of














Up to $5000 Over $5000

Bank Deposits 177.7 81.4% 0.10% 0.30% 2.70% 4.90%

Non-Bank Deposits 40.5 18.6% 0.60% 0.80% 3.95% 5.44%

Total 218.3

Source: RBA Tables B2 Financial Assets, F3 Retail Interest

Table 1 shows the total household funds deposited in banks, non-bank financial institutions and

other financial institutions and the range of interest rates that were available in 1997-98. The RBA

estimates the value of these deposits at $218.3b while the income capitalisation technique discussed

above produces an aggregate of $177.9b. Given differences in definition of “households” (RBA

includes unincorporated businesses) and the exclusion of deposits of dependent children and zero

interest assets the estimate looks reasonable.


Again the 1998 SIHC does not contain actual asset values but does contain income information on shareholdings and other investments. Current weekly income from dividends, royalties and other

investments are available on a personal basis. Using the range of monthly average dividend returns

obtained during 1997-98 (see Table 2) and assuming the bond rate is an appropriate rate to attribute

to royalties and other investments, income capitalisation is used to estimate the value of shares,

royalty paying investments and other investments. The RBA estimates the value of these shares and

other investments at $136.9b while the income capitalisation aggregate is $119.5b.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 6

Table 2 1997-98 Various Rates of Return

1997/98 Summary













Shares 3.34% 3.98% 111.0

Other (use the Bond rate) 3.44% 6.74% 8.5

Total 136.9 119.5

Sources: interest rates - RBA Tables F5 Share Markets, F2 Bonds

assets - RBA Tables B2 Financial Assets and Author’s calculations

The distribution of share ownership based on 1998 SIHC records differs quite markedly from ASX

survey data for the same period. Based on SIHC data, 9.0% of adults had dividend income, while

an ASX survey in May 1997 found 20.4% of adults directly owned shares (ASX, 1997). In

addition, the SIHC imputations suggest an average share portfolio size of around $96,800 while the

ASX survey average was $33,300. It appears the SIHC is biased towards capturing those with

significant stock holdings rather than the full range of shareowners. One reason for this bias relates

to the method used to record dividend income in the SIHC. Those with little or no weekly dividend

income are not captured. The minimum dividend income that can be recorded is $1 per week. This

equates to a share portfolio of up to around $1500. Portfolios which are less than this size are

unlikely to be identified in the SIHC data. Secondly some new investors may not have recorded

their dividend income. During the ABS survey period, it is estimated 559,000 entered the share

market with the partial float of Telstra (ASX, 1998) and another 730,000 first time investors entered

in June 1998 when the insurance company AMP listed (ASX, 1999). It seems unlikely that much

of the dividend income from the two companies will have been recorded in SIHC file. In summary,

the proportion of Australians with shares varied markedly during the year and this variation impacts

directly on the proportion reporting share income. The inclusion of these first time investors with

their small portfolios would probably have reduced the SIHC calculated average portfolio value to

much the same level as the ASX average.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 7


The 1998 SIHC contains the estimated current sale price of the dwelling in which an income unit lives. An income unit is either one person or a group of people who share income (typically a

family). This estimate is assigned as the owner-occupied housing value to the individual if the

income unit only has one adult or is evenly divided between the head and spouse if they exist. If

two income units (say the parents and an unemployed son) live in the same house, then the value of

the house is only allocated to parents i.e. only the people who are a head and a spouse can be

owners. The averages by state before and after the allocation to individuals are shown in Table 3.

Table 3 1997-98 Value of Housing by Owner (with or without a mortgage) ($)

NSW Vic Qld SA WA Tas


NT Australia

Total Value

Income Units 231,641 158,192 169,109 127,436 182,020 116,619 177,343 183,002 $905.5b

Persons 136,881 93,725 99,773 76,152 107,703 69,469 106,656 108,362 $905.5b

Source: 1997-98 ABS Survey of Income and Housing Costs

The total amount owing on the dwelling is derived in the same way. The file contains an income

unit field, which contains “amount currently owing on all loans”. Again, this field is assigned as the

mortgage value to the individual if the income unit only has one adult or if a head and spouse exist

then the mortgage is divided between them. The reason the variable “all loans” is used rather than

some other variable is that it captures all loans for which the house was used as security. It thus

captures the increasing use of home equity for non-housing purposes. Table 4 presents the average

mortgages by state for homeowners with mortgages.

Table 4 1997-98 Value of Mortgages by Owner (with a mortgage) ($)

NSW Vic Qld SA WA Tas


NT Australia

Total Value

Income Units

(With a mortgage) 84,814 64,691 76,839 59,707 75,289 49,910 84,589 73,735



(With a mortgage) 47,810 37,605 43,202 34,178 43,140 28,809 49,554 42,086


Source: 1997-98 ABS Survey of Income and Housing Costs

The final question in regard to the home values is: Are estimates obtained from the owners

accurate? Yates (1991) compared self-reported valuations with professional valuations and other

sales data and concluded the estimates were likely to be within three percent on average. However,

the variability was large with only 33 percent able to estimate the value within $5000 and 60

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 8

percent within $10,000. Bækgaard (1998) concludes that variation is random and no adjustments

can be applied to improve the estimates. I follow this practice and have made no adjustment to the



Information from the 1998 SIHC is used in conjunction with the 1997 ABS Survey of Rental Investors (RIN) to ascertain rental property values and rental investment loans. Current weekly

income from rent (both non-residential and residential) and a flag indicating profit, loss or breaking

even are found on the SIHC file. The RIN file records values, loans and net income for both the

reference person and their spouse. Based on the net rental income and the rental flag recorded on

the SIHC file, the probable rental property equity can be estimated from the RIN file. Then using

the net rental income and the imputed equity, an estimate can be made of the rental property value

and mortgage. For example if a person is recorded as a reference person and has a net rental

income of $250 and $349 per week then, using Table 5, there is a 32.5% chance the person’s equity

in the rental property is between $150-200,000; a 25.7% chance it is between $200-250,000; 19.1%

chance it is between $250-300,000; and a 22.8% chance it is between $300-350,000. A rental

property equity is imputed for this person on the basis of these probabilities.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 9

Table 5 1997 Distribution of Equity by Net Rental Income (%)

REF Equity in Rental Property

Net Rental


(per week)






$-1 Zero





















$500k $500k+

< $-350 23.1 7.8 9.5 39.8 15.0 4.9

$-350 - $-251 55.6 44.4

$-250 - $-151 5.3 8.0 3.2 53.3 11.6 7.4 3.3 3.4 4.5

$-150 - $-50 2.9 11.6 2.0 48.3 12.9 13.1 3.8 2.8 1.0 1.5

$-50 - $ 49 3.7 7.5 1.9 31.6 24.7 14.4 6.3 3.1 2.6 3.2 1.0

$50 - $149 12.4 28.1 28.8 12.6 3.7 5.9 6.0 2.5

$150 - $249 14.2 18.6 14.5 12.3 7.6 20.6 12.2

$250 - $349 32.5 25.7 19.1 22.8

$350 plus 21.3 7.0 6.8 46.8 18.0

SPOUSE Equity in Rental Property

Net Rental


(per week)






$-1 Zero





















$500k $500k+

< $-350 14.8 85.2

$-350 - $-251 68.6 31.4

$-250 - $-151 24.4 8.8 38.4 20.2 8.1

$-150 - $-50 3.5 6.4 4.2 39.8 26.4 10.4 2.4 4.9 2.0

$-50 - $ 49 2.3 5.1 1.6 31.5 27.0 15.3 7.5 2.9 3.2 2.7 0.8

$50 - $149 1.4 3.1 6.5 20.4 18.9 15.5 10.1 6.9 12.1 5.1

$150 - $249 34.6 13.0 8.4 11.8 32.1

$250 - $349 32.0 27.3 40.6

$350 plus 18.3 81.7

Source: Author’s calculations based on ABS Survey of Rental Investors, 1997

Using the equity imputed through the previous step and the rental income on the SIHC, a mortgage

amount can be imputed (Table 6). The rental property value can then be calculated.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 10

Table 6 Average Loans for Various Amounts of Equity and Net Rental Income ($)

REF Equity in Rental Property

Net Rental


(per week)






$-1 Zero





















$500k $500k+

< $-350 300,332 120,000 300,000 137,106 239,533 175,000

$-350 - $-251 205,430 142,000

$-250 - $-151 180,840 117,148 92,813 117,469 120,398 72,650 153,970 215,000 154,333

$-150 - $-50 135,507 120,365 145,844 104,375 99,200 70,552 133,983 74,079 303,023 290,171

$-50 - $ 49 233,222 108,628 105,569 68,073 48,602 43,217 55,831 73,530 69,094 84,289 93,266

$50 - $149 46,962 18,993 30,410 26,809 74,596 45,000 39,273 32,609

$150 - $249 79,300 0 0 0 30,880 36,647 15,876

$250 - $349 0 0 0 0

$350 plus 0 0 0 0 7,173

SPOUSE Equity in Rental Property

Net Rental


(per week)






$-1 Zero





















$500k $500k+

< $-350 250,000 85,457

$-350 - $-251 201,000 71,000

$-250 - $-151 234,852 85,045 122,425 82,060 297,000

$-150 - $-50 325,000 92,805 118,830 93,242 96,916 118,213 107,037 110,111 100,000

$-50 - $ 49 239,526 108,396 114,346 77,005 44,939 31,247 53,170 42,240 48,794 71,615 9,134

$50 - $149 213,000 211,829 112,674 31,666 1,938 49,169 114,271 80,248 21,444 152,884

$150 - $249 0 2,666 0 85,319 5,781

$250 - $349 0 0 0

$350 plus 240,000 0

Source: Author’s calculations based on ABS Survey of Rental Investors, 1997


A flag indicating a profit, loss or nil income from your own business is found on the 1998 SIHC file. Current weekly income from a person’s own business is also available. Both of these are used

to impute the value of the business assets. The method used to calculate business assets is described

in detail in a number of previous NATSEM publications (for example, see NATSEM Technical

Paper No.14) and this same method was used in this research. The methodology uses unpublished

working estimates provided by the ABS on the gross capital stock for unincorporated enterprises.

Using this data and the number of people who operate unincorporated businesses from the SIHC,

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 11

average capitalisations and net profit rates are calculated for nine industries. Individual net profit

rates and net capitalisation are then calculated.


NATSEM has developed a dynamic microsimulation model that models the accumulation of superannuation assets at the individual level. A full description of the microsimulation model is

published in King, Bækgaard and Robinson (1999) while the superannuation modelling is discussed

in detail in Kelly, Percival and Harding (2001). People aged 75 and older are excluded. The

probability of people having superannuation is then calculated, based on sex, age and labour force

status. If a person is found to have superannuation coverage then a value is imputed for the

superannuation asset. The value is again a function of sex, age and labour force status.

3. Wealth - the General Population

Combining the assets imputed in Section 2, wealth is calculated. The aggregate asset totals are compared with 1986 asset values in Table 7. The 1986 values are from NATSEM Technical Paper

No. 14. These 1986 figures were calculated using similar imputation techniques to that described in

the previous section and have been converted to 1998 dollars. The changes in value represent

changes in real 1998 dollars and the annual percentage changes are real compound growth rates.

With real growth of 4.5% pa in net wealth and an annual inflation rate average of 4.0% over this

period, the commonly quoted growth rate for wealth of 10% per annum appears to be perhaps to be

slightly high. This research suggests the growth rate has been 8.5%. The significant growth in

shares and other investments reflects the growing popularity of shares as an investment vehicle and

the greater diversify of investment opportunities available in 1998.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 12

Table 7 Estimated Wealth, 1986 and 1998 (in 1998 dollars)



Share of Total

(% of total)



1986 1998 1986 1998

Interest Bearing Deposits 106 178 10.4 10.4 4.6%

Shares and Other Investments 40 120 3.8 7.0 10.0%

Housing (net) 547 745 53.2 43.5 2.7%

Rental Properties (net) 71 130 7.0 7.6 5.4%

Business (net) 103 222 10.1 13.0 6.9%

Superannuation 159 316 15.6 18.5 6.2%

Wealth (net) 1,026 1,711 100.0 100.0 4.5%

Sources: 1986-NATSEM Tech Paper No.14, 1998-Author’s calculations

A breakdown of wealth by asset type and net value is presented in Table 8. The table is divided

into wealth percentiles. Those with the lowest net wealth are in the lowest percentiles and the

wealthiest Australians are in the highest percentiles. The richest ten percent is further broken to give

more insight into this group. The average person in Australia has assets worth $127,000. However

there is a considerable range. The bottom 10 percent of the population have a net worth of less than

zero (-$1,000) while the top 10 percent have over $550,000.

The prevalence and value of certain assets also changes significantly as the total wealth increases.

The least wealthy half of the population have $1,000 or less in shares and other investments, rental

property investments or their own business. The wealthiest 50% have significant amounts in each

of these assets. The major asset for the least wealthy is their superannuation; it represents 90% of

the wealth of the poorest 20 percent (see Table 9).

As wealth increases, home ownership becomes a factor. As equity in the home increases so does

the wealth percentile. By the 40th percentile, equity in the family home has increased to represent

61% of the wealth and the significance of superannuation has dropped to be 28% of the total. As

the climb up the wealth ladder passes the 50th percentile, a number of other forms of wealth start to

become significant. Interest bearing deposits, equity in rental properties and business assets are all

reducing the importance of the family home and superannuation. By the time we reach the top 10

percent we are looking at people with not only significant equity in the family home, but also with

over $70,000 in deposits, $70,000 in superannuation, $76,000 in shares and other investments,

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 13

$70,000 in equity in their rental properties, and net business assets worth almost $100,000. The

wealth is evenly across these assets (except for the family home which is worth twice the others at

30% of the total).

Table 8 Personal Wealth by type of Asset and Wealth Decile 1997-98 ($)


Number of

People in this






Shares and



















1-10 1,340,339 0 0 0 -2 0 1 -1

11-20 1,342,013 1 0 0 0 0 6 7

21-30 1,341,271 2 0 2 0 0 12 17

31-40 1,341,056 3 0 13 0 0 17 35

41-50 1,341,314 4 1 35 1 1 16 58

51-60 1,341,522 6 1 53 1 3 19 83

61-70 1,341,484 10 2 70 3 6 23 114

71-80 1,341,613 13 3 98 7 12 28 162

81-90 1,341,200 22 7 114 17 44 42 247

91-100 1,341,801 71 76 168 70 98 70 553

91-95 670,400 26 16 124 33 123 49 371

96-99 536,977 83 59 188 82 83 90 585

100 134,376 247 441 304 204 38 96 1,330

Overall 13,413,566 13 9 56 10 17 24 127

Source: Author’s calculations

* Australians aged 15 years and over

Overall, the family home is the major component of wealth. On average it represents 44 percent of

a person’s wealth (see Table 9) but for middle Australians (41st-70th percentiles) it represents more

than 60% of their wealth. As expected, the housing contribution to wealth in the lower and higher

percentiles is not as strong as it is in the middle. In the lower percentiles, people do not have the

deposit required to purchase a home or the value of their home is negated by the outstanding

mortgage. In the higher percentiles the equity in the home is only one of a range of assets

contributing to the person’s wealth.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 14

Table 9 Assets as a Proportion of Personal Wealth 1997-98






Shares and

















11-20 8 1 2 -1 90

21-30 12 1 14 73

31-40 9 1 39 1 50

41-50 8 2 60 1 2 28

51-60 7 1 64 1 4 23

61-70 9 1 62 3 5 20

71-80 8 2 61 4 8 17

81-90 9 3 46 7 18 17

91-100 13 14 30 8 13 19

91-95 7 4 33 9 33 13

96-99 14 10 32 14 14 15

100 19 33 23 15 3 7

Overall 10 7 44 8 13 19

Source: Author’s calculations

Note: Rows may not add to 100% due to rounding

Superannuation is the next biggest asset for most people and appears to be the least concentrated.

The introduction of award-based superannuation in 1986 and the Superannuation Guarantee Charge

(SGC) in 1992 appear to have helped all people to make some savings for their retirement. With

contributions being made for almost every employee under SGC and these funds not being available

until retirement, people who formerly would have had no assets at all are now saving 5-9% of their

salary each year. As noted above, for the least wealthy, this forced saving is very significant and

often represents almost all of their wealth.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 15

Table 10 Proportion of Personal Wealth by type of Asset and Wealth Decile 1997-98



of Pop’n in

this Group





Shares and



















1-10 10

11-20 10 0 0 0 0 0 3 1

21-30 10 2 0 0 0 0 5 1

31-40 10 2 0 2 0 0 7 3

41-50 10 3 1 6 1 1 7 5

51-60 10 4 1 10 1 2 8 7

61-70 10 8 2 13 4 3 10 9

71-80 10 10 3 18 7 7 12 13

81-90 10 17 7 21 18 27 18 19

91-100 10 54 85 30 72 60 30 43

91-95 5 10 9 11 17 37 10 15

96-99 4 25 27 14 34 20 15 18

100 1 19 50 6 21 2 4 10

Source: Author’s calculations

Note: Columns may not sum correctly due to rounding

While the distribution of superannuation is across the entire adult population, the distribution of

wealth is most certainly not. The wealthiest 10 percent have 43% of the total wealth and the top 50

percent have 90% of the wealth. Viewed from the other direction, the bottom half of the population

have only 10% of the wealth, the least wealthy 20 percent have only one percent and the poorest 10

percent has no wealth at all.

The distribution of shares and other investments is the most concentrated, with 85% being owned

by the top 10 percent. In fact half of the wealth in this form is held by the richest one percent of

Australians. Rental properties and businesses are also highly concentrated in the hands of the rich.

4. Wealth - On Retirement

The decision to retire permanently from the workforce is one that is not lightly taken. A number of factors need to be considered. A person’s financial situation is clearly one of these factors. In this

section, the assets of those people who are considering permanent retirement are compared with the

general population. The ‘About to Retire’ group is considered to be those about to meet the

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 16

minimum age requirement for the Age Pension. The minimum age for eligibility to the Age

Pension in 1998 was 61 years for women and 65 years for men. To actually qualify for the Age

Pension they would also have to meet residency requirements and have income and assets below

certain thresholds. Assuming people will consider permanent retirement in the year leading up to

becoming eligible (if they are not already retired), the About to Retire group is composed of women

aged 60 and men aged 64 at the time they were interviewed.


In 1998 the About to Retire group had 151,441 members. This represents 1.1 percent of the non-dependent child population aged 15 and over. Some characteristics of this group, other than

wealth, which will be considered later, are significantly different from the remainder of the

population. The first one is gender. In the About to Retire group females make up some 54.9

percent while for the remainder the share is 50.7 percent (Table 11). This can be attributed to two

facts - the earlier retirement age of women and the lower mortality rates. The higher life

expectancy of women suggests that age increases the proportion of women should increase. In this

group the women are not the same age as the men but in fact four years younger, it is not surprising

then to find that the proportion of women is higher than for the total population.

The marital status of the About to Retire group shows a bias towards two groups - married people

and those who are separated, widowed or divorced. The bias towards marriage is particularly

strong for men. Not surprisingly, given the age of the members, those that have never married are


Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 17

Table 11 Characteristics of those about to become Eligible for the Age Pension, 1998 Males Females

Aged 64 Remainder Aged 60 Remainder

Numbers 68,280 6,537,892 83,161 6,724,233

Percentage of total 45.1 49.3 54.9 50.7

Marital status

Married or defacto 71.8 65.4 66.7 63.4

Separated, widowed or divorced 22.4 9.0 31.7 19.1

Never married 5.8 25.6 1.7 18.5

Family Type

Couple with dependent children 9.4 31.5 5.0 30.7

Couple without dependent children 62.4 33.2 61.7 32.2

One parent 2.0 1.1 1.2 6.6

One person 26.2 34.2 32.1 30.5

Labour force status

Employed full-time 39.5 64.4 10.3 32.0

Employed part-time 6.5 6.1 18.3 19.1

Unemployed 5.2 6.3 2.3 4.0

NILF 48.8 23.2 69.1 44.9


Not applicable 54.0 29.5 71.5 48.9

Managers and administrators 12.3 8.3 4.5 2.4

Professionals 3.7 11.5 4.4 11.1

Associate professionals 5.4 8.8 1.9 5.1

Tradespersons 5.3 15.0 1.8

Advanced clerical and service 0.8 3.5 4.7

Clerical and Sales 8.1 6.3 5.5 14.4

Production and transport 5.0 9.7 1.8 1.1

Elementary clerical 5.2 3.2 2.8 6.4

Labourers 1.1 7.0 4.2 4.1

Source: 1997-98 ABS Survey of Income and Housing Costs

The impact of the higher ages of those in the reference group can be seen in the proportion of

people who are members of couples without dependent children (62.4% for males and 61.7 % for

females against population averages of 33.2% and 32.2% respectively). A surprising statistic is that

men living alone (26.2%) was below the average for men (34.2%) while women were slightly

above the average at 32.1% compared with 30.5% for the remainder.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 18

Examination of the labour force status of the About to Retire group shows that 48.8% of the men

and 69.1% of the women are already not in the labour force while the population averages are

23.2% and 44.9%. This suggests the About to Retire nomenclature may be inaccurate, as it appears

a large proportion of the population have already retired. For men the NILF rate is twice the rate

for the remainder while for women it is 50% more than the average. This is a little surprising as the

government, in general, will not contribute to the living costs of the person if they have withdrawn

from the labour force. With some exceptions such as partners of pensioners or carer’s (both of

whom receive their own government pension), people who are not in the labour force must be living

off their partner’s income, deriving income from their assets or liquidating some of their assets.

The proportion of the About to Retire group in various occupations also reflects to some extent their

age. Male occupations that are physically hard work are under-represented (for example,

tradespersons, production and transport, or labourers) while white-collar positions are over-represented (managers and administrators, clerical and sales, or elementary clerical). For women,

these trends are not as noticeable. It is also noticeable that the traditionally higher income

occupations of professional and associate professional are under-represented in the About to Retire

group. One possible explanation for this is that those with the higher incomes have already retired.

Unfortunately we do not have the former occupation of those currently not working.


Examination of the assets owned by the About to Retire group is presented in Table 12. Home ownership is higher than the remainder of the adult population while the number with mortgages is

considerably lower. These results are not unexpected, as traditionally Australians have wanted to

purchase their own home and have then used extra money to reduce their mortgage. Historically,

the mortgage is often paid out when large lump sum payments, such as redundancy payments and

job termination payments, are received. The large proportion of About to Retire group that are

NILF may already have received such a payment and paid off their house.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 19

Table 12 Assets of those about to become Eligible for Age Pension Males Females

Aged 64 Remainder Aged 60 Remainder

Own Home

Yes 82.6 60.5 77.7 63.7

No 17.4 39.5 22.3 36.3


Yes 4.5 29.0 15.1 28.4

No 95.5 71.0 84.9 71.6

Rental Properties

Yes 2.2 8.7 6.9 7.0

No 97.8 91.3 93.1 93.0

Shares and Other Investments

Yes 12.2 9.8 8.0 8.0

No 87.8 90.2 92.0 92.0

Business Assets

Yes 16.8 12.3 13.3 7.1

No 83.2 87.7 86.7 92.9


Yes 79.5 82.5 71.3 73.2

No 20.5 17.5 26.8 26.8

Source: Author’s calculations

Rental property investments do not appear to be as popular with men in the About to Retire group as

the remainder of the population. Only 2.2% of the men had rental properties against an average of

8.7% for the remainder of the male population. One of the major attractions of investment in rental

properties is negative gearing where a property is financed in such a way as to produce a loss in

terms of income taxation while (hopefully) returning a capital gain which is taxed at a later date and

at a reduced rate. Since the attraction of reducing income taxation is proportion to tax paid, the

under-representation of rental property assets in the About to Retire group may relate to the large

proportion with a labour force status of NILF. Without a high income, the attraction of reducing tax

is reduced and the attraction of rental properties may not be as strong.

The reason women are investing in rental properties may relate to another feature of this form of

asset. Real estate as a physical asset is perceived to be more secure and less volatile than some

other financial assets, for example, shares. This security combined with a known, reasonably

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 20

reliable income stream may be an attraction for women. This attraction would not vary with age

and would explain why women in the About to Retire group behaved in line with the population.

The proportions owning their own business and having shares and other investments are as expected

for people in this age group.

The proportion having superannuation was a little lower for the About to Retire group than for the

remainder of the population. This reflects the lack of exposure to compulsory employer

contributions for those in the About to Retire group who may have left the workforce before the

introduction of these initiatives.


It is reasonable to assume that wealth increases with age until retirement when it starts to decrease

due to the decrease in income. This effect can be seen in Figure 1. As the age of those in the About

to Retire group is 60-64, we would expect their average wealth to be higher than the overall mean.

This is the case, with the average of the group being $216,000 while the population average is

$127,000. The average for the group is home equity of $95,000, rental property equity of $11,000,

shares and other investments of $17,000, net business assets of $28,000 and superannuation valued

at $40,000.

Figure 1 Wealth by Age Group, 1998








15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 50-59 60-64 65-69 70-74 75+

Age Group

Net Wealth ($'000s)

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 21

However, there are still large numbers of the group who are not wealthy. There are 5,413 people

(3.6%) from the About to Retire group in the lowest wealth decile where the average wealth is less

than zero. Figure 2 shows the proportion of the group found in each of the wealth deciles. As

expected there is over-representation in the higher wealth deciles.

Figure 2 Wealth Deciles of Those About to Retire, 1998







10 20 30 40 50 60 70 80 90 100


Proportion (%)

About to Retire Remainder of Population

Another interesting feature of this group is that while they have more assets than the average

person, the form in which the assets are held is similar to the remainder of the population (Figure 3).

However, there is an under-representation of rental property assets, already discussed and slightly

higher amounts in interest bearing deposits and shares and other investments. This may indicate a

change to assets that can readily be liquidated if required.

Figure 3 Proportion of Wealth by Type of Asset







Home Rent Prop Deposits Shares Business Super

Type of Asset

Proportion of Total Wealth (%)

About to Retire


Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 22

5. Conclusion

In 1998, women aged 61 and men aged 65 were eligible for the Age Pension provided they met residency, income and asset test requirements. For most people, the option of retiring and

receiving a pension for the remainder of their life is an attractive one. However, the pension is set

at 25% of average weekly earnings and most would require additional supplementary income to live

comfortably. The value of assets held by the person at the time of their 61st or 65th birthday will

have a major impact on the decision of when to retire.

This paper looked at the wealth of people in the year preceding possible entitlement to the Age

Pension, i.e. men aged 64 and women aged 60. This About to Retire group was compared with the

remainder of the Australian population in regards to general characteristics and wealth.

In calculating the assets of Australians, the methods used to determine the value of interest bearing

deposits and the value of rental property equity have been enhanced. In calculating interest bearing

deposits, everyone is assumed to have some money in a day-to-day account paying low interest.

Excess funds are assumed to be invested at a higher interest rate. These assumptions have not been

included in previous estimates of this asset. The recent release of information from the ABS 1997

Survey of Rental Investors has been used to improve the imputation of rental property assets.

The personal wealth of Australians in 1998 was $1,711 billion. This represented real growth of

4.5% p.a. since 1986. The average adult Australian had assets with a net worth of $127,000.

Equity in the family home made up 44% of this while superannuation accounted for a further 19%.

For the poorest Australians, superannuation was a very significant asset often representing all of the

wealth. However for the poorest 10%, even after allowing for superannuation, they had no wealth

and were in fact $1,000 in debt. The wealthiest 10 percent had $553,000 in assets and owned 85%

of all shares and other investments, 72% of rental investment properties and 60% of business assets.

In regard to the About to Retire group, this group was found to have a higher proportion of women

than the average and a high proportion of both sexes had already left the workforce. The finding

that a large proportion (48% of men and 69% of women) were already not in the labour force was

surprising. Two types of workers in particular appeared to have left the labour force. The first was

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 23

men in physically hard occupations (like tradesmen and labourers) and the second was professionals

(both men and women) who may have had the wealth required to retire early.

People in the About to Retire group have significantly more assets than the remainder of the

population. They have assets to the value of $216,000 while the average overall is $127,000. This

high average wealth is not surprising given the ages within the group. In general wealth rises with

age and the group is considerably older the average. Despite the advantage of age, 3.4% of the

group still had no wealth and were in the lowest wealth decile.

A family home is the biggest asset of the majority of Australians and the About to Retire group.

The next biggest and most evenly spread asset was superannuation. It represented around 19% of

the average person’s assets. The spread of assets for the About to Retire group was similar to the

remainder of the population with the exception that 64 year old men were not as inclined to have

their assets in rental properties as the remainder of males (2.2% against the average of 8.7%).

Overall, the group about to retire were wealthier than the average Australian, more likely to be a

women and no longer in the labour force but otherwise they did the same things with their assets -

put them into a home and superannuation.

Wealth on Retirement - Latest Estimates for Australia

Date:6-Jul-01 Page 24


ABS (Australian Bureau of Statistics), 1999, 1997-98 Survey of Income and Housing Costs Australia

Confidentialised unit Record File Technical Paper, Catalogue No. 6541.0.30.001, ABS, Canberra.

ASX (Australian Stock Exchange), 1997, 1997 Australian Shareownership Survey, ASX Limited, Sydney.

__, 1998, 1998 Shareownership Update, ASX Limited, Sydney.

__, 1999, Australian Stock Exchange Fact Book 1999, ASX Limited, Sydney.

Bækgaard, H., 1998, ‘Simulating the Distribution of Household Wealth in Australia: New Estimates for 1986

and 1993’, Technical Paper No. 14, NATSEM, University of Canberra, June.

Dilnot, A. 1990, ‘The Distribution and Composition of Personal Sector Wealth in Australia’, Australian

Economic Review, 1st Quarter, 1990.

Kelly, S., Percival, R. and Harding, A. 2001, ‘Women and Superannuation in the 21st Century: Poverty or

Plenty?’, Paper presented to SPRC National Social Policy Conference 2001, University of NSW,


King, A. and Bækgaard, H., 1996, ‘The Dynamics of Housing Wealth’, Paper presented to the 8th National

Conference of the Australian Population Association, Adelaide, NATSEM, December.

King, A., Bækgaard, H. and Robinson M., 1999a, ‘DYNAMOD-2: An Overview’, Technical Paper No. 19,

NATSEM, University of Canberra, December.

Podder, N. and Kakwani, N. 1973, ‘Distribution of Wealth in Australia’, Review of Income and Wealth, Vol.

22, pp 75-92.

Wood, G., 1999, ‘Home-owner Residential Property Taxes and their Burden on Net Personal Wealth: An

Empirical Study for Australia’, Urban Studies, Vol. 36, No. 2, pp.239-254.

Yates, J. 1991, Australia’s Owner-Occupied Housing Wealth and it’s Impact on Income Distribution, SPRC

Reports and Proceedings No 92, May 1991, Social Policy Research Centre, University of New South

Wales, Sydney.