APPENDIX 3 COMPARISON OF WEALTH BETWEEN SIH AND THE AUSTRALIAN SYSTEM OF NATIONAL ACCOUNTS
INTRODUCTION
This publication contains estimates of the wealth of Australian households compiled from data collected in the 2003-04 Survey of Income and Housing (SIH). The Australian System of National Accounts (ASNA) also provides estimates of the wealth of Australian households. This appendix compares wealth data from the two data sets and describes and quantifies some of the major scope, definitional and methodological differences between the two.
The concepts of wealth used in the two data sets have much in common, although there are also substantial differences. There are also differences in the definition of the household sector covered by each data set, and in the methodologies and data sources used to compile the estimates.
As the SIH and ASNA estimates of household wealth have been developed for different purposes using different data sources, they have different strengths and weaknesses. The SIH data are collected from individual households via a single collection and can be used to analyse the distribution of wealth across the population and to compare levels of wealth between various population subgroups. The ASNA data are collected from many sources, most of which do not provide information for different population subgroups within the household sector. However, the ASNA data provide a comprehensive picture of the household sector as a whole, presented within a national accounting framework. The data therefore show how the household sector relates to the corporate and government sectors in Australia and to the rest of the world. Details are available in Australian National Accounts: Concepts, Sources and Methods (cat. no. 5216.0).
Sources of error
Both data sets are subject to error. As discussed in the Explanatory Notes of this publication, the SIH data are subject to both sampling error and non-sampling error, including misreporting by respondents. Some wealth items collected in the survey require respondents to make estimates, such as the value of their owner occupied dwelling, for which they may have no precise information available.
The ASNA is compiled from a wide range of sources, with widely ranging reliability. Some estimates are based on sample surveys of households or businesses and some are based on collections that capture all the relevant transactions in the economy. Some are based on data collections that are only conducted occasionally, with estimates interpolated and extrapolated over time. In some cases, there may be a reliable estimate for an aggregate for the whole economy, but less reliable indicators or assumptions are used to disaggregate the economy wide estimate into individual sector estimates.
Data used in comparisons
Elsewhere in this publication, the SIH estimates of net worth are expressed in terms of mean value per household. To facilitate comparison with ASNA estimates, in this appendix they are presented as aggregate values for Australia. Multiplying the mean value per household for all households by the estimated number of in scope Australian households (7,735,800) provides the aggregate Australian value.
The SIH was conducted throughout 2003-04 and respondents were asked to report the value of their assets and liabilities at the time that they were surveyed. Therefore the wealth estimates are assumed to relate to the average level of household net worth during that year.
The ASNA estimates of net worth are those underlying the household balance sheet presented as table 51 in Australian System of National Accounts, 2004-05 (cat. no. 5204.0), with the memorandum item for consumer durables being taken from table 16 of that publication. Balance sheet data in those tables are presented with respect to 30 June of each year, and therefore the estimates for 2002-03 and 2003-04 are averaged to improve comparability with the SIH estimates. In the detailed comparisons provided later in this appendix, more detailed component data used to compile the ASNA balance sheet aggregates are sometimes presented to maximise comparability with SIH items.
In 2003-04 the aggregate SIH value of household net worth was $3,618 billion and the aggregate ASNA value of household sector net worth was $3,239 billion.
A3.1 SIH ESTIMATES OF HOUSEHOLD NET WORTH |
|  |
| | Mean value per household | Aggregate value | |
| | $'000 | $b | |
| |
| Assets | | | |
| Value of accounts held with financial institutions | 21.1 | 163 | |
| Value of shares (excl. own incorporated business) | 18.2 | 141 | |
| Value of trusts | 9.2 | 71 | |
| Value of debentures and bonds | 0.9 | 7 | |
| Value of own incorporated business (net of liabilities) | 22.8 | 177 | |
| Superannuation | 63.5 | 491 | |
| Value of owner occupied dwelling | 249.0 | 1 926 | |
| Value of other property | 70.8 | 548 | |
| Value of own unincorporated business (net of liabilities) | 15.6 | 121 | |
| Value of contents of dwelling | 47.4 | 366 | |
| Value of vehicles | 17.2 | 132 | |
| Value of assets nec | 0.6 | 5 | |
| Total assets | 537.1 | 4 155 | |
| Liabilities | | | |
| Principal outstanding on loans for owner occupied dwelling | 40.0 | 309 | |
| Principal outstanding on other property loans | 19.9 | 154 | |
| Debt outstanding on study loans | 1.2 | 9 | |
| Amount owing on credit cards | 1.9 | 14 | |
| Principal outstanding on loans for vehicle purchases (excl. business loans) | 2.7 | 21 | |
| Principal outstanding on investment loans (excl. business and rental property loans) | 2.4 | 18 | |
| Principal outstanding on loans for other purposes (excl. business and rental property loans) | 1.5 | 11 | |
| Total liabilities | 69.4 | 537 | |
| Net worth of households | 467.6 | 3 618 | |
| |
A3.2 ASNA HOUSEHOLD BALANCE SHEET |
| |
| | At 30 June 2003 | At 30 June 2004 | Average 2003-04 | |
| | $b | $b | $b | |
| |
| Assets | | | | |
| Machinery and equipment | 81 | 83 | 82 | |
| Non-dwelling construction | 71 | 79 | 75 | |
| Livestock - fixed assets | 15 | 15 | 15 | |
| Dwellings | 849 | 944 | 896 | |
| Intangible fixed assets | 3 | 3 | 3 | |
| Inventories | 21 | 22 | 22 | |
| Land | 1 335 | 1 612 | 1 474 | |
| Currency and deposits | 318 | 351 | 335 | |
| Securities other than shares | 14 | 14 | 14 | |
| Loans and placements | 15 | 17 | 16 | |
| Shares and other equity | 233 | 265 | 249 | |
| Insurance technical reserves | 585 | 693 | 639 | |
| Unfunded superannuation claims | 139 | 142 | 145 | |
| Other accounts receivable | 23 | 20 | 22 | |
| Total assets | 3 703 | 4 260 | 3 981 | |
| Liabilities | | | | |
| Securities other than shares | 10 | 13 | 12 | |
| Loans and placements | 669 | 782 | 726 | |
| Other accounts payable | 7 | 3 | 5 | |
| Total liabilities | 687 | 798 | 742 | |
| Net worth of household sector | 3 016 | 3 462 | 3 239 | |
| Memorandum item from National Balance Sheet: Consumer durables | 187 | 196 | 192 | |
| |
DIFFERENCES IN SCOPE
The SIH estimate of household net worth includes as assets the value of household contents and motor vehicles used for private purposes, whereas the ASNA estimate excludes them. However, the ASNA includes the value of consumer durables as a memorandum item outside the scope of its aggregate estimate of household net worth. The ASNA concept of consumer durables is essentially the same as the combined values of household contents and vehicles in the SIH, so the value of the ASNA memorandum item ($192b) has been added to the ASNA aggregate value of net worth of the household sector to improve comparability with the SIH aggregate.
The ASNA measure of net worth also includes some components which are not included in the scope of the SIH measure. Components which can be identified and deducted from the ASNA aggregate to improve comparability with the SIH aggregate include the technical reserves of general insurance corporations ($31b) and unfunded superannuation claims ($141b).
The technical reserves of general insurance corporations represent policy holders’ net equity in, or claims on, the reserves of general insurance corporations. This equates to prepayments of premiums and reserves held to cover outstanding claims. These funds are included in the ASNA value of net worth because they are collectively owned by the household sector. However, the value of the technical reserves is not attributed to individual households, since households only access the funds when they lodge a claim or receive a similar insurance payout. The technical reserves of general insurance corporations are therefore excluded from the household statistics collected in the SIH. The other insurance technical reserves included in the ASNA equate to the value of superannuation assets of households and the value of whole of life (termination) insurance and savings/endowment policies. Households normally receive annual statements providing the value of these policies, and they are therefore included in the scope of the SIH.
Unfunded superannuation claims reflect the liability of some governments to pay superannuation benefits to their employees for which they have not set aside funds. The benefits are often in the form of a pension that is determined on the basis of the employee's length of service and final salary at retirement, and they are funded from general government revenue at the time of payment. The value of this item is imputed on an actuarial basis for administrative purposes, and is an estimate of the amount of money that is required to pay the unfunded superannuation benefits to which the employees are entitled. As estimates of this liability are not available on an individual basis, they are excluded from the household statistics included in the SIH.
The net worth of non-profit institutions serving households (NPISHs), which include charities and religious organisations, are included in the ASNA definition of the household sector but are not in scope of the SIH. Only limited information is available about their net worth, and bank deposit and borrowing estimates are the only scope adjustments that can be made to the ASNA aggregates included here. The potential influence of other NPISH assets and liabilities is noted in some of the item comparisons that follow, although it is not expected that their impact would be large.
The net result of adjusting for the four scope differences described above is to raise the ASNA based estimate of net worth from $3,239 billion to $3,253 billion. This is 10% below the SIH estimate of $3,618 billion.
There are other scope differences which cannot be as easily quantified. In particular, the SIH estimate is confined to the net worth of households excluding households in very remote and Indigenous Communities. It therefore omits the net worth of people who live in non-private dwellings such as hotels, boarding houses and institutions and those who live in very remote and Indigenous Communities. In total, about 2 per cent of the Australian population are out of scope of the SIH. The net worth of these people are included in the scope of the ASNA estimates.
COMPARISON OF ITEMS
The following paragraphs compare the items published in this publication with corresponding items published in the ASNA, or source data available for those items. The categorisations of data in the two data sets differ substantially, limiting the detailed comparisons that can be made.
For example, the SIH only collects the value of unincorporated businesses on a net basis, rather than collecting the value of business assets and liabilities by type. The ASNA, on the other hand, often does not distinguish between assets and liabilities relating to households' business activities and their non-business activities. Therefore there is difficulty in making precise comparisons between any asset or liability items that are likely to relate to both business and non-business activity.
Accounts held with financial institutions
The SIH item 'accounts held with financial institutions' is compared to the deposits component of the ASNA item currency and deposits. That component is also known as balances with authorised deposit-taking institutions excluding deposits belonging to NPISHs (NPISHs are outside the scope of the SIH estimates). The SIH item is only about half the value of the ASNA component, as shown in the following table:
There are three major factors which limit the comparability of the SIH and ASNA values for this item.
First, the ASNA component includes deposits in cash management trusts, estimated to be worth about $22 billion. In many cases, SIH respondents would be expected to report such deposits in the value of trusts data item, but the SIH does not differentiate between cash management trusts and other types of trusts. The total value of trusts in SIH is $71 billion. Adjusting ASNA for this difference in item classification reduces the difference in the two sources to $129 billion.
Second, the ASNA component includes deposits belonging to unincorporated businesses, whereas the SIH item excludes them. In the SIH they are one of the many unidentifiable components of the value of unincorporated business (net of liabilities). In SIH there are 1.3 million respondents reporting that they own an unincorporated business (a little lower than the 1.5 million people reporting such businesses for taxation purposes). While it is not known how much these businesses may hold in either current or investment accounts, the amounts would be significant, perhaps running to tens of billions of dollars.
Third, the scope restriction in SIH (excluding 2% of the total population, and a higher proportion of the adult population) will contribute to the difference. Many of these people will be older persons living in non-private dwellings (about 7% of persons aged 65 and over). As can be seen from table 20 in this publication, households with reference person aged 65 and over have the highest deposit balances of any of the age groups. They account for about 40% of all deposits reported even though these households account for only 13% of the people living in private dwellings.
The three factors limiting the comparability between the aggregates are not considered likely to account for all of the difference between them. The ASNA estimate of the balances for all sectors with authorised deposit-taking institutions is considered reliable because it is based on regulatory information obtained from financial corporations, and the household sector is the sector owning the largest share of those deposits. It may also be that the household sector information in ASNA includes some balances that relate to small incorporated businesses. However, the ASNA estimate is considered to be relatively accurate, and it is concluded that SIH respondents have substantially under-reported this item.
Shares, including own incorporated business
The SIH asked households who owned their own incorporated businesses to report the value of the businesses net of liabilities. In principle, this should equate to the share value of those incorporated businesses. The SIH also collected the value of other shares held by households. The values of these two items were $177 billion and $141 billion respectively, summing to $318 billion.
The corresponding ASNA value is the sum of listed and unlisted shares owned by households, estimated at $242 billion. The values of listed and unlisted shares are both calculated as residuals in the ASNA, that is, the total value of each type of shares owned by all sectors is estimated and then the value of shares owned by sectors other than households are subtracted to derive the value of shares owned by the household sector.
It could be assumed that listed companies would rarely be identified by SIH respondents as 'own incorporated business'. It could therefore be expected that the SIH value of shares (excluding own incorporated business) would equate to the ASNA value of listed shares plus some proportion of the ASNA value of unlisted shares. However, at $141 billion, the SIH estimate is below the ASNA value of $155 billion. This may partly reflect the ownership of some shares by unincorporated businesses or by trusts and therefore their exclusion from this SIH item but captured elsewhere in SIH and wealth measures.
There appears to be much less congruity between the SIH value of incorporated own business ($177 billion) and the ASNA value of unlisted shares ($87 billion). The difference in the measurement of investment in trusts is again likely to explain some of the difference between the two sources of data. First, the largest components of unlisted shares in the ASNA aggregate are equity in unlisted property and trusts. In many cases, SIH respondents would be expected to report such equity in the value of trusts data item, further increasing the gap between the two measures. But the total value of trusts in SIH is relatively small ($71 billion). It is therefore likely that some SIH respondents have incorrectly identified businesses operated under a trust structure as being operated as incorporated businesses and reported the equity as share assets in SIH. However, this is unlikely to fully explain the gap between the SIH value of own incorporated business and the ASNA value of unlisted shares. The ASNA data sources are recognised as being of relatively poor quality and may have led to underestimation of this item in the ASNA. SIH respondents may also have valued their incorporated business incorrectly.
Superannuation
Superannuation is the most important form of household financial asset recorded in the SIH, and survey respondents reported a balance of $491 billion held in superannuation funds. The corresponding ASNA items are the technical reserves of pension funds ($569 billion) and part of the $38 billion of the technical reserves of life insurance corporations. The ASNA estimate is considered reasonably reliable and therefore it appears that at least part of the difference between the two estimates reflects under-reporting in the SIH. Under-reporting will inevitably occur for households that are unaware of superannuation assets that they hold, and the Australian Taxation Office has reported unclaimed superannuation accounts of about $8 billion.
Some SIH respondents may also have mistakenly included the assets of their self-managed superannuation funds as trust assets. The SIH estimate of the value of trusts is $71 billion.
Property assets
For many households, their dwelling is their main asset. The SIH collects data about the value of their dwelling from owner occupier households, which account for about 70% of all households. The SIH also collects data about the value of other property owned by households, if that property is not considered by the respondent to be part of the assets of an unincorporated business.
The ASNA data set includes the value of dwellings and the value of land separately. The estimate of dwelling value is compiled by estimating the total number of dwellings and applying unit values to those dwellings, excluding the value of land. A proportion of the total value is then assigned to the household sector. The value of land is separately estimated, using information from state and territory valuer generals, with separate totals for rural, commercial and residential usage. The household sector is assigned 92% of total land value.
The two aggregate estimates appear very similar, with the SIH value at $2,474 billion and the ASNA value at $2,445 billion. But the SIH estimate should be somewhat lower than the ASNA estimate, since the latter includes substantial amounts for farm land and other property, which would mostly be included in the net value of unincorporated businesses when reported to the SIH. The unexpected similarity of the aggregate estimates reflects substantial differences in the average value of residential land and dwellings underlying these estimates. For its June 2004 estimate, the ASNA data incorporated an implicit average value of residential land and dwellings of $282,000 per dwelling, whereas the average value of owner occupied dwellings reported in the 2003-04 SIH was $355,000 per dwelling. This difference suggests a $0.5 billion difference in aggregate valuation levels for residential land and dwellings. The ASNA methodology for this item is currently under review.
Own unincorporated business
The SIH asked respondents to provide a single net estimate of the value of their own unincorporated business. Therefore no information is available about the asset and liability composition of that part of household wealth. Due to the different data sources and methodology used in compiling the ASNA estimates, the ASNA includes information about the individual assets and liabilities, but does not identify which of those relate to unincorporated business activities and which are used for other purposes. However, some forms of assets and liabilities are most likely to relate to business activities and those are included in the table below.
The net value of the selected ASNA assets and liabilities was $157 billion, which is broadly comparable to the SIH net value of own unincorporated business of $120 billion. As discussed above, the valuation for farm land held by unincorporated businesses and some property other than residential property, are included in the ASNA estimates for household sector 'total land and construction' whereas the net worth in these activities will be captured in the unincorporated business assets item in SIH. No further conclusions have been reached about these items.
Contents of dwelling
The SIH value of contents of dwelling was $366 billion. The most closely related ASNA value is the non-vehicles component of the value of consumer durables, which is included as a memorandum item on the national balance sheet. This component equalled $92 billion, but the ASNA value of consumer durables excludes clothing, personal effects such as watches and jewellery, some recreational goods, books and electronic recordings, or jewellery, artworks and antiques that are held as a store of value. In total, these are likely to be worth more than the consumer durables that have been included in the ASNA item.
Part of the difference between the SIH and ASNA values will also reflect the different valuation bases underlying the estimates. In the SIH, respondent households were asked to provide the value of their dwelling contents insurance cover, when available. Otherwise, respondent households were asked to estimate the value of their dwelling contents. The ASNA estimates the aggregate value of the goods held by householders by using a personal inventory model methodology, which tracks purchases of new items by households and assigns them reducing values over their assumed life time. Therefore the ASNA methodology can also be expected to result in substantially lower estimates than those obtained by using dwelling contents insurance cover because the former attempts to estimate actual value while the latter is normally based on a 'new for old' valuation basis. The SIH estimate will also be influenced by any tendency for households to over-insure or under-insure the contents of their dwellings, while the ASNA estimate is dependent on the validity of the assumptions underlying the perpetual inventory methodology.
Vehicles
The SIH value of vehicles was $132 billion, while the motor vehicles component of the ASNA memorandum item consumer durables was $100 billion.
The valuation bases underlying the estimates of vehicles from the two sources are more comparable than for dwelling contents described above. In the SIH, survey repondents are asked the value of their vehicles. In the ASNA, a perpetual inventory method, as described in the previous section, is used. In principle, the two approaches reflect the same valuation basis. However, while the SIH estimate excludes motor vehicles that are solely used within businesses owned by households, it does include vehicles that are used for both business and personal use. These are likely to have been excluded from the ASNA estimates. It is not known how much of the difference between the two estimates is attributable to the different treatment of vehicles used for business.
Property loans
The property loans data as published from the SIH includes only a proportion of the principal outstanding on the loans where the loans are used both to finance the purchase or construction of, or alteration or addition to an owner occupied dwelling and to finance other activity. Loans for housing in the ASNA relate to loans originally for the purpose of housing. The SIH estimate has therefore been adjusted to obtain better comparability with the ASNA estimate, using relevant data collected in the SIH.
The SIH estimate (of $470 billion) is $87 billion, or 16%, below the ASNA estimate. The SIH estimate is broader than the ASNA estimate in so far as it includes loans for property that is not residential property, but is narrower in so far as it does not include loans for dwellings regarded by SIH respondents as the assets of their unincorporated businesses. However, the net effect of these two scope differences would seem unlikely to be the main reason for the 16% difference between the two estimates.
Other loans
The SIH estimate of loans other than for property or unincorporated business purposes is $73 billion. The corresponding component of the ASNA item loans and placements is about 60% higher, at $118 billion. It is not clear whether the large difference mainly reflects under-reporting in the SIH, or differences in scope between the SIH and ASNA components included here (eg. loans for the purpose of funding an unincorporated business may be captured in ASNA data for household sector loans but be reported in SIH indistinguishably as part of unincorporated business net assets), or difficulties in dividing reliable aggregate financial data into sector specific components in the ASNA. Some loans, such as those associated with vehicle finance leases, can be difficult to allocate between business purposes and personal purposes, for both reporting in SIH and ASNA compilation.
SUMMARY
The SIH 2003-04 estimate of total household net worth is $3,618 billion. The corresponding ASNA estimate is 10% lower, at $3,239 billion. There are some scope differences between the two estimates. The ASNA total can be adjusted by adding the ASNA memorandum item for consumer durables and deducting the ASNA components for the technical reserves of general insurance corporations and unfunded superannuation claims. However, this only narrows the gap slightly, raising the ASNA estimate to $3,253 billion, $365 billion lower than the SIH estimate. No adjustment has been made to reflect the exclusion by SIH of the 2% of the population who do not live in private dwellings or who live in very remote and Indigenous Communities, suggesting that the gap between the two aggregate estimates is somewhat higher than $0.4 billion. Only limited adjustments have been made for the inclusion by the ASNA of the net worth of charities, religious organisations and other non-profit institutions serving households (NPISHs).
The sources of data used in the two data sets provide somewhat different decompositions of the aggregate amounts, and detailed item level comparisons between the data sets are difficult. It is therefore only possible to draw broad conclusions about the differences in aggregate wealth provided by the two data sets.
Dwellings are the most important form of asset held by households, and as discussed above the SIH valuation for residential property is probably about $0.5 billion higher than the ASNA valuation. The ASNA methodology used for this aggregate is currently under review.
There is a large difference between the contents of dwelling item ($366 billion) from the SIH compared to the ASNA memorandum item consumer durables ($92 billion). The difference reflects the exclusion of a number of items including clothing, books, electronic recordings and other personal items from the ASNA measure and the inclusion of 'old for new' valuations in the SIH 9as measured from contents insurance). The SIH estimate for motor vehicles ($132 billion) is also somewhat higher than the ASNA value ($100 billion), which in part at least reflects the inclusion of vehicles used for both business and private purposes in the SIH estimate.
The SIH estimate for accounts held with financial institutions is $163 billion, which was only about half the ASNA estimate of balances with authorised deposit-taking institutions, excluding NPISH deposits ($314 billion). The ASNA estimate includes the value of balances in cash management trusts and the value of accounts held by unincorporated businesses, which SIH would have recorded elsewhere, while SIH excludes the deposits of people living in non-private dwellings. However, these differences are unlikely to account for all difference between the two aggregates, and it is likely that this item was under-reported in the SIH.
At $491 billion, the SIH estimate for balance of accounts with superannuation funds compares with the ASNA estimate of $569 billion for the item of pension funds' technical reserves. The ASNA also includes an unquantified superannuation amount within the $38 billion of technical reserves of life insurance corporations. While some of the SIH superannuation assets may have been reported as trust investments in the case of self-managed funds, it is likely that the SIH item was somewhat under-reported.
The $76 billion gap between the SIH and ASNA items for own incorporated business and for other shares is in part due to classification differences but also likely to reflect inconsistencies between households' valuation of their incorporated businesses and the ASNA methodology which uses sources of lower quality.
It is difficult to draw conclusions about the comparability of the other components of the SIH and ASNA aggregates. Overall it can be concluded that SIH respondents have under-reported some items and that for some items the ASNA data sources and methodology provide estimates that are below the corresponding valuations provided by households.
Print Page
Print All