Stats NZ has a new website.

For new releases go to

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Institutional Sector Accounts: 1999–2008
Embargoed until 10:45am  –  16 December 2010
Technical notes

Institutional sector classification

This report classifies institutional units (such as enterprises and households) by broad economic sectors using the New Zealand Standard Institutional Sector Classification (NZISC).

Subsectors and groups within some sectors recognise differing functions and the degree of government control. In this release the producer enterprises sector is further broken down into four different subsectors, namely:

  • private corporate producers
  • private unincorporated producers
  • central government producers (for example, state-owned enterprises)
  • local government producers (for example, council-controlled organisations).

The general government sector is broken down into two subsectors:

  • central government
  • local government.

It should also be noted that the private non-profit institutions serving households sector differs from the coverage of the published non-profit institutions satellite account, in that it does not, for example, include non-profit organisations that serve business.

The unit considered when applying the NZISC for the collection and publication of statistics is referred to as the ‘institutional unit’. The most common institutional unit is the enterprise unit. This is the smallest business unit having a separate legal existence. It includes companies, partnerships, sole proprietorships, incorporated societies, government departments, and trusts. The non-financial corporations sector, for example, includes limited liability companies and unincorporated businesses, as well as state-owned enterprises and (local government) council-controlled organisations. The other principal unit is the household. These units include resident individuals and families in their role as final consumers and as owners of factors of production. Households, in their capacity as owner-occupiers of dwellings, are included in the household sector.

Some legal entities need to be separated into two or more institutional units in order to compile meaningful statistics. This is particularly so where the significant activities of a single legal entity span two or more institutional sectors. For instance, individuals are legal entities classified to the household sector because of their roles as final consumers. However, some individuals are also proprietors of farms or small businesses and as such, for statistical purposes, form separate enterprise units classified to the producer enterprises sector.

Although government departments are considered to be separate legal entities, some combine both entrepreneurial and governmental functions. Similarly, large territorial local authorities blend a variety of trading and administrative functions under one corporate heading. In the NZISC, separate enterprise units are formed to identify separate substantial trading and financial functions.

It should also be noted that in the NZSNA, transactions with the rest of the world are presented from New Zealand's perspective, whereas in the NZISA, the sector is presented from the perspective of the rest of the world. This means that in the NZISA transactions that are negative in the published national accounts appear as positive in the NZISA. For example, net lending has positive values in all years in the NZISA rest of the world sector account as the rest of the world has been a net lender to New Zealand. However, in the National Accounts: Year ended March 2010 release it is negative in these years, reflecting that New Zealand has been a net borrower.

General method

The institutional sector accounts are compiled by transaction. Each transaction is allocated to sectors separately, and then full sector accounts are compiled.

The principal data sources for most transactions are the same business surveys as used in the national accounts. These business surveys, such as the Annual Enterprise Survey, collect information on financial flows as well as the productive activity that is already published in the annual national accounts. Business surveys are also designed to collect data on an industry by sector basis.
For income and outlay, and capital account items there are also significant flows relating to the household sector. Therefore, additional data sources are incorporated into the method to supplement the business survey data. For example, household interest flows are estimated using Reserve Bank data.

Large financial flows are reconciled as far as possible at the enterprise level. For example, large dividends reported as being paid by New Zealand resident enterprises in the balance of payments are checked against dividends paid recorded in business surveys. Also, large inter-company flows are cross-checked. Where the data sources are inconsistent, other sources including annual reports are consulted, and the source data is then adjusted. By this process of reconciliation the gap between national totals for transactions (such as interest paid and interest received) are brought close together, and then a final adjustment is made to match the flows exactly.

In cases where conflicting information on interest flows are reported by the debtor and creditor, and both flows appear to be reported correctly, the debtor numbers have been used.

In general, data has been recorded gross. Even between institutional units within the same sub-sector, receipts are not netted off payments, and vice versa. It has also meant that where inter-company financial flows are recorded in unconsolidated form, level shifts in some series can occur that are entirely due to the inter-company flows.

Standard treatments

Saving as a residual

Since saving is estimated as the residual between total income and total outlays it is highly sensitive to any errors in estimating income and outlay components. Income and expenditure estimates (including interest and dividend flows) are now reconciled within the institutional sector framework.

Primary income and secondary income

Income is subdivided into primary and secondary. Primary income is directly linked to the production process: compensation of employees, gross operating surplus, entrepreneurial income, net taxes on production and property income (rent on natural resources, interest, dividends, and earnings attributed to insurance policyholders). Secondary income covers redistribution in the form of current transfers, including current taxes on wealth and income, social security and pension transfers.

Owner-occupied dwellings

Ownership of owner-occupied dwellings is considered to be a market activity undertaken by households. All expenditure associated with the purchase, alteration, and maintenance of owner-occupied dwellings is classified as gross fixed capital formation or intermediate consumption and is therefore excluded from final consumption expenditure. Included, however, is the payment of imputed rent by owner-occupiers. This measures an income flow back to households valued at market rates.

Consumption of fixed capital (depreciation) is recorded separately in the household capital account. A consequence of this treatment is that the saving residual in the household income and outlay account is net of the depreciation on owner-occupied dwellings.

Finance service charge and future developments

The finance service charge is earned by banks and other enterprises classified as financial intermediaries.

Conceptually the interest received by these intermediaries, in the income and outlay account, should be net of the finance service charge earned. For all sectors to balance, however, this requires the corresponding payments to be also recorded net. This in turn requires the payment of the finance service charge to be allocated by sector. Currently in the NZSNA the payment of the finance service charge is not allocated, rather, it is recorded as being paid by a nominal industry, which has a corresponding negative operating surplus.

Rather than include the nominal industry as a separate institutional sector with negative saving, the finance service charge received has been subtracted from the operating surplus of the financial corporations sector. It also avoids double-counting the relevant interest flows. Moreover, because gross flows of interest received and paid are recorded in the income and outlay accounts, the non-allocation of the finance service charge, though affecting operating surplus for each sector, has no effect on sectoral saving or net lending.

Work is underway in the national accounts to replace the finance service charge with the 1993 SNA concept – financial intermediation services indirectly measured (FISIM). When allocating the cost as FISIM, the GDP total will change (increase) by the value of the amount classified as final consumption expenditure of households, government, and private non-profit institutions serving households, as well as the value allocated to exports, net of that estimated as imported.

Pension and social security schemes

According to the 1993 SNA, income for life insurance and superannuation and pension schemes is imputed. Employer contributions to these schemes are included as part of an individual’s income. Since the accumulated pension and superannuation funds are regarded as part of household assets, interest earned by the funds is included in household income. To avoid double-counting this income, actual pension payments are treated as a rundown in assets, or dissaving. Internationally, there may be differences in where the line is drawn between funded social security schemes (not classified as part of private saving) and funded pension schemes (usually for state employees).

While every effort is made to exclude capital gains and losses from the earnings of pension and life funds, to the extent that any remain, then household income will be over or understated compared with the conceptually correct measure.

In addition, income earned by New Zealand residents’ investments in overseas pension funds is not currently measured.

Insurance premiums and pension fund contributions

Only the service or administration charge component of insurance premiums and pension fund contributions paid by households is treated as final consumption expenditure. The balance is treated as a transfer payment classified as secondary income payable.

Actual final consumption

Actual final consumption of households measures the goods and services acquired by households, whether purchased by them directly or by government or non-profit institutions on their behalf.

Capital gains and losses

As noted, capital gains and losses associated with holding or trading capital and financial assets are, in the SNA, recorded in reconciliation accounts. Therefore, these gains/losses are excluded from the concept of saving. Similarly, foreign exchange movements are eliminated.

The concept of saving

Over recent years, interest in accurately estimating the saving of New Zealanders and identifying in which part of the economy it occurs has grown substantially. Interest in this question predates the global financial crisis, but has intensified in the face of the crisis. This is reflected in recent government efforts such as the Tax Working Group, the Capital Market Working Group, and the Savings Working Group. A level of national saving lower than the level of investment necessarily results in a higher current account deficit and a higher net level of international debt. In contrast, higher national saving means that more domestic funds are available to invest in productive activity.

National saving (for the whole economy) is published as part of the Annual National Accounts. Conceptually, national saving is the sum of each sector’s saving, determined from the income and outlay accounts of each sector.

This release uses the term ‘saving’ in its conventional economic sense employed in the national accounts. Under this definition, saving is the portion of national income accruing to a country (or sector thereof), that is, not consumed and is therefore available to finance investment and/or fund future consumption.

Some alternative savings estimates include balance sheet items, for example, the market value of a sector’s stock of assets less the market value of its stock of liabilities. Wealth measures that adjust for the change in the value of assets are called estimates of net worth. Wealth or net worth can therefore be viewed as the accumulated stock of savings. Wealth estimates are outside of the current scope of the institutional sector accounts.

An example of how household sector saving as measured in this release differs from the change in net worth is illustrated as follows:

  • Increases in the value of the owner-occupied housing stock will be included in measures of household net worth but are not included in household saving as measured in the NZISA.
  • However, increases in mortgage interest payments related to increases in housing values do affect (reduce) household saving as measured in the NZISA.

It should be noted that measuring saving as a flow measure (income not spent) or savings as a stock measure (change in net worth) are not competing methodologies. Instead the key objective should be to reconcile them in a full set of income, capital, financial, and balance sheet accounts.

At the sector level it is possible to compile saving estimates using either macroeconomic data, as in the institutional sector accounts, or using microeconomic data. While, for example, the Household Economic Survey is not designed to produce household saving statistics, and Statistics NZ cautions against doing so, it has been used for this purpose because of its comprehensiveness and richness of data on respondent characteristics. There are a number of possible conceptual or definitional differences that may exist between the 1993 SNA-based household income and outlay account and survey-based household income/expenditure series.

For a more extensive discussion of saving measures, refer to Selected Issues in the Measurement of New Zealand’s Saving(s) (Bascand, Cope, & Ramsay, 2006), available on the Statistics NZ website,


More information

For more information, including terms and definitions, follow this link from the 'Technical notes' section of this release on the Statistics NZ website.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+