Annual Enterprise Survey

In addition to its use in the national accounts, the AES is also a data source for other Stats NZ existing and upcoming outputs, including:

• industry benchmarking (see Business Performance Benchmarker),
• the Tatauranga Umanga Māori publication (see Tatauranga Umanga Maori),
• Longitudinal Research of Business Dynamics project (see Longitudinal business database),
• Regional Gross Domestic Product (GDP) (see Regional Gross Domestic Product),
• Tourism Satellite Account (see Tourism Satellite Account),
• Non-profit Institutions Satellite Account (see Non-profit Institutions Satellite Account),
• Household distribution (see National Accounts – Distribution of Household Income, Consumption, and Saving ).

The AES is also used by external customers including for:

• requests from government departments, such as the Ministry of Business, Innovation and Employment,
• requests for data/information by non-standard turnover bands.

Limitations of using the AES data

The AES data is to be used with caution below industry design level (New Zealand Standard Industrial Output Classification (NZSIOC) level 4). The survey is not designed to support analysis below this level.

The AES is designed to measure industry levels for a given year. Incremental improvements in measurement, sample design, classification, and data collection may influence the inter-period movements, particularly over longer time periods. We work to minimise the impact of these changes and to present a consistent time series in the published tables.

Interpreting the data

When using the AES data, it is important to be aware of these design issues that may affect results:

1. The AES data is collected from businesses with balance dates between 1 October in one year and 30 September the following year. For example, the AES 2020 financial year covers balance dates from 1 October 2019 to 30 September 2020.

2. The Annual Enterprise Survey measures economic activity by industry. How companies structure themselves can affect how their data is captured and reported in the AES. Large corporates often set up separate entities to manage different divisions of their business. These divisions are classified based on their predominant activity. For example, a manufacturing business (in division C) could have their administration (head office) classified to management and related consulting services (in division M), and their asset-owning to financial asset investors (in division K). This would mean that the manufacturing unit will not have these support activities recorded in the manufacturing industry. If a business is divided into different divisions, this can mean that the AES results will include inter-company flows between divisions (gross flows). This can inflate the total of some variables including interest paid, interest received, dividends paid and, dividends received. Ratios, such as return on equity, should also be interpreted with caution.

3. The AES time series can be affected by the restructuring of companies. For example, if divisions within a company are restructured or amalgamated, the following could happen:

• consolidating the units would remove the gross flows and leave net flows,
• the reverse may also occur, when restructuring results in previously recorded net flows being represented in a gross form,
• after determining the industrial classification of the resulting unit/s by predominant activity – activity in the other industries would disappear,
• value added would remain the same in all options.

4. The 'all industries' table sums divisional tables and therefore includes gross flows. The ‘all industries’ table still excludes some activity out of scope of AES, for example, residential property operators.

5. Due to rounding of values and imputation approaches (eg: balancing of financial position totals), individual figures in our outputs may not sum to the stated total(s).

6. Inland Revenue's IR10 data is the main administrative data source for the AES and is the primary data source for a number of industries, particularly agriculture. IR10 data does not provide direct estimates of resale, purchases for resale, additions and disposals of fixed assets, so AES uses modelling to calculate these. The modelling of IR10 data for additions and disposals of fixed assets is considered of lower quality and have been suppressed from the 'all industries' table and all agricultural industries tables.

7. Stats NZ has a legal obligation to protect survey respondents' privacy and industry-sensitive information. Confidentiality rules are applied to all data released – to protect the information supplied by an individual company. Once all confidential financial items are identified, suppression of further items is done to complete the protection of the confidential value.

8. The AES includes only economically significant enterprises, understating New Zealand activity.

9. The AES excludes some industries; therefore, the all industries table understates New Zealand business activity. Exclusions include:

• superannuation funds (K633000),
• residential property operators (L671100),
• foreign government representation (O755200),
• religious services (S954000),
• private households employing staff and undifferentiated goods and service-producing activities of households for own use (S96100-300),
• non-market local and central government.

1969 Recommendation 4 of the 1969 Technical Committee on Statistics states "That the Department of Statistics integrate as far as possible and as soon as possible, all statistics of economic activity whether for national accounts or other purposes, by means of nation-wide enterprise enquiries rather than the existing individual surveys." At the time the recommendation was made the typical economic census was largely run as an independent exercise, each one designed to meet its own particular requirements and each one largely unrelated to any other. The Technical Committee characterised the existing statistical surveys as unsuitable since they suffered from overlaps, gaps and inconsistent boundaries and valuations, and overall they lacked integration. The system of statistical surveys was as a whole falling well short of meeting the increasing demands for comprehensive and timely statistics.

1974 An inter-departmental committee was created to 'advise on current and anticipated future statistical needs in both Government and the private sector'. The recommendations produced, incorporating recommendations by the 1969 Technical Committee, were accepted by the Government and these then provided the basis for the Department's development of business statistics. The report stressed the need for standard statistical classifications, and the integration of both coverage and concepts in the existing system of statistical surveys and censuses. Consequently, the development of integrated economic censuses was recommended and the development of an integrated census of manufacturing was given first stage priority.

1975 This year saw the introduction of the first integrated manufacturing census, which replaced the old Industrial Production Census. The Central Register of Enterprises (known as the Business Register) provided a register of manufacturing enterprises, used via addressograph plates, to structure the census population.

1976 On 16 December 1976 the Minister of Statistics approved in principle a programme of economic censuses. In seeking this approval, the Department said, "The department's future work programme provides for the development of an integrated series of enterprise based economic censuses". Census of Manufacturing conducted.

1977 Census of Manufacturing conducted.

1978 Census of Distribution conducted.

1979 The following economic censuses were conducted: Manufacturing; Building and Construction; and Mining and Quarrying.

1980 The following economic censuses were conducted: Transport and Communications; Agricultural Contracting; and Forestry and Logging.

1981 The following economic censuses were conducted: Fishing and Miscellaneous Services.

1982 Census of Manufacturing was conducted.

1983 The following economic censuses were conducted: Distribution, and Financial and Insurance Services.

1984 The following economic censuses were conducted: Manufacturing, Mining and Quarrying, Forestry and Logging, and Fishing.

1985 The following economic censuses were conducted: Building and Construction, Transport and Communications, Agricultural Contracting and Hunting.

1986 The AES was introduced for the 1985/86 survey year. This was intended to overcome the weaknesses of the economic censuses, particularly the inability to accurately conduct cross-sector analysis. The survey covered all parts of the economy except for agriculture production, central government administration & defence, water works & supply, local government administration, electricity generation & distribution, communications, insurance, sanitary & cleaning services, all social & related community services, and recreational & cultural services.

1987 The AES population was updated and extended to cover additional industry groups. A separate Economy Wide Census of all industries, focusing on the sales and purchases of goods and services, was also conducted in 1987.

1988 The Communications industry was surveyed in the AES for the first time.

1989 The Electricity industry was surveyed in the AES for the first time.

1990 Beginning in 1990, the survey population was re-selected from the Business Directory, on an annual basis.

1992 New imputation methodology introduced. The Census of Distribution was conducted in conjunction with the AES (by extending the AES population for that industry to full coverage).

1993 The Census of Distribution was conducted in conjunction with the AES. The following industries were surveyed within the AES for the first time:

• Hunting and Trapping; Other Welfare Institutions; and Live Entertainment and Cultural Services nec,
• More detailed Wholesale and Retail Trade industries introduced,
• More detailed Health and Welfare industries introduced,
• Fishing was split into two industries: Ocean Fishing and Other Fishing. Prior to 1993 Fishing was spread across eight industries: Trawling, Dredging, Squid, Potting, Hand, Shellfish, Farming, and Other,
• The superannuation industry was dropped from the AES coverage due to quality concerns.

1994 The AES population was expanded to include units introduced to the Business Directory and Inland Revenue FIRST reconciliation project. This event mainly impacted on the Agricultural Contracting industry where total income jumped by about 10 percent ($83 million).

1995 The Census of Manufacturing was conducted in conjunction with the AES. The AES population definition changed to exclude compulsory Goods and Service Tax (GST) registrations that fall below the threshold of economic significance.

1996 The Census of the Rest of the Economy was conducted in conjunction with the AES:

The Business Directory was renamed to Business Frame.

1998 The AES sample was based on the Australian and New Zealand Standard Industrial Classification (ANZSIC) classification, replacing the New Zealand Standard Industrial Classification (NZSIC). For the first time the private sector participants in the Sewerage and Drainage Services and Water Supply industries were included in the AES sample. In addition, a number of industries were brought into the AES for the first-time using data from IR10s' and existing surveys of the government sector:

• From IR10: Horticulture and Fruit Growing; Livestock and Cropping Farming, Dairy Cattle farming, Other Farming and Commercial Property Operators and Developers,
• From existing government surveys: Central Government Administration, Defence, Public Order and Safety Services and Local Government Administration,
• IR10 data was used for all individual and partnership business types (instead of direct surveying),
• Output editing processes were reviewed and updated,
• Other changes were made as part of the AES re-design.

1999 The following changes were made to the AES:

• Creative Arts, Sports and Services to Sports, and Interest Groups n.e.c. were surveyed for the first time,
• Sample re-selected,
• Questionnaires re-designed,
• Survey re-designed as a two-component survey,
• New computer processing systems (Sprocet & Sybase) introduced,
• Output variables introduced. This is the level of variable that the AES is published at, such as total income and the individual components that make up total income e.g. sales of goods and services not further processed,
• Commercial property estimates were retrospectively included when first available in 2005. See ‘2005’ for more details.

2001 Early pilot survey of Commercial Property Operators conducted within the AES framework.

Commercial property estimates were retrospectively included when first available in 2005. See ‘2005’ for more details.

2003 A trading account summary was introduced into the retail trade industry questionnaire to simplify the recording of sales, cost of goods sold and stock values for respondents. Full time equivalents were replaced by rolling mean employment.

Commercial property estimates were retrospectively included when first available in 2005. See ‘2005’ for more details.

2004 Introduction of scanning and image recognition to the survey process. Further pilot testing of Commercial Property Operators conducted with intention to publish a re-casted time series for this industry in the AES 2005.

Commercial property estimates were retrospectively included when first available in 2005. See ‘2005’ for more details.

2005 Information for Commercial Property Operators was sourced from direct survey for the first time in 2004 and was introduced in the 2005 publication. Time series of the Commercial Property industry was backdated from 1999 to 2003 using the 2004 survey data as a benchmark.

Results retrospectively changed to ANZSIC06 classification. See ‘2007’ for more details.

2006 Sample boost to help with the revisions due to the introduction of the new version of Australian and New Zealand Standard Industrial Classification (ANZSIC06).

Results retrospectively changed to ANZSIC06 classification. See ‘2007’ for more details.

2007 Sample re-optimised in response to the classification change in versions of Australian and New Zealand Standard Industrial Classification (ANZSIC96 to ANZSIC06).

Survey designed and published on an ANZSIC06 basis. 2005 and 2006 revised to reflect ANZSIC06 classification version.

2009 The AES was redesigned. Changes in the 2009 year included:

• Sample re-optimised,
• Questionnaires re-designed,
• Survey reverted to a one-component survey,
• Increased use of administrative data (IR10),
• Change to the range of the random number line used for selecting businesses to survey.

2010 Further improvements were made including:

• Additional use of administrative data (IR10),
• A more efficient sampling strategy,
• Enhancements to editing and imputation processes,
• A timeliness gain (the AES was previously published in the first week of October, in AES 2010 this was progressed two weeks to the middle of September).

2011 Improvements made and significant events in the AES 2011 included:

• A further increase in the use of administrative data (IR10) to replace sampled units,
• Started using Charities Services data and modelling to replace not-for-profit sampled units,
• The release date moved to the end of August, two weeks earlier than the AES 2010,
• 2010/2011 Canterbury and Christchurch earthquakes had significant influences on the ongoing economic output of businesses located in Christchurch,
• Additional tables were released at level 4 of the New Zealand Standard Industrial Output Classification (NZSIOC),
• Results retrospectively changed to account for new tax law. See ‘2012’ for more details.

2012 Improvements made and significant events in the AES 2012 included:

• Use of more administrative data for industries other than agriculture, forestry, and fishing division (ANZSIC06 division A). IR10 data did not provide direct estimates of additions and disposals of fixed assets, so modelling was used to calculate these. -Additions and disposals of fixed assets were suppressed from the "all industries" table, all agricultural industries, and the accommodation industry tables. -The increased use of administrative data in 2012 also caused a discontinuity in shareholders' funds/owners' equity in the repairs and maintenance industry, and the accommodation industry,
• 2010/2011 Canterbury and Christchurch earthquakes continued to have significant influences on the ongoing economic output of businesses located in Christchurch,
• A new tax law was introduced on the 1st of April 2011 which affected depreciation of buildings. The effect of the law change was primarily seen in the rental, hiring, and real estate services industry. It was the main contributor to the decrease in depreciation across all industries in the 2012 financial year.

2013 Improvements made and significant events in the AES 2013 included:

• A further increase in the use of the Charities Services data.
• Sample boost for improved estimates of Tatauranga umanga Māori (see Tatauranga Umanga Maori).
• The IR10 form was re-designed to improve the quality of data collected. The IR10 form changes resulted in measurement discontinuities of some variables such as shareholders' funds/owners' equity, current liabilities and non-operating income. The discontinuities were more significant in industries which used IR10 data as their predominant source.
• A new processing and analysis platform was used for the first time (MEP). 2011 and 2012 data were also included in the new platform retrospectively.
• Classification changes to the agriculture industry as well as the removal of non-market government data were made retrospectively to align with changes made in 2015. See '2015' for more details.

2014 Improvements made and significant events in the AES 2014 included:

• In 2014, the AES started used the Statistical Classification for Institutional Sectors (SCIS). Because of this change, family trusts were classified under the household sector and are excluded from AES scope. Before 2014 they were classified under the non-financial business sector and included in the financial assets investing industry in the AES. As a result, over 10,000 family trusts included in the scope of the AES in 2013 were excluded in 2014. The impacts of the change was primarily seen in the financial position of the financial asset investors industry (NZSIOC KK112) for the 2014 financial year, particularly in the decrease in fixed asset holdings measured by the AES.
• Classification changes to the agriculture industry as well as the removal of non-market government data were made retrospectively to align with changes made in 2015. See ‘2015’ for more details.
• Results were revised with AES 2018 to re-introduce all family trusts previously excluded. See ‘2018’ for more details.

2015 Improvements made and significant events in the AES 2015 included:

• Non-market central and local government units were excluded from AES and removed from AES results for all data published since the AES 2013. This is because coverage of non-market government data was incomplete within the AES and the information is available from other sources. For data on non-market government activity, customers should refer to the Government finance statistics. Market government units are still included in AES scope (e.g. For-profit and state owned enterprises).
• To continue to reduce respondent load, in the 2015 financial year the AES sourced more data from the inland Revenue (IR10).
• A large number of units were re-classified in the agriculture industry in the AES 2015, as a result of information received from Agricultural Production Statistics. The units affected were mainly in the horticulture industry. Most of these re-classified units remain in the wider agriculture industry. The largest changes, with the most industry, were back-dated to 2013.
• In the AES 2015 additional data was collected for financial assets and liabilities. This was to complete the alignment of the AES collection to the Statistical Classification of Financial Assets and liabilities (SCFAL).
• Results were revised with AES 2018 to include trusts previously excluded. See ‘2018’ for more details.

2016 Improvements made and significant events in the AES 2016 included:

• 2016 Kaikoura earthquakes had significant influence on the economic output of businesses located in Kaikoura and affected areas.
• Results were revised with AES 2018 to include trusts previously excluded. See '2018' for more details.

2017 Improvements made and significant events in the AES 2017 included:

• The AES 2017 was released at the end of June, two months earlier than the AES 2016.
• More data was made freely available through the Stats NZ website. Examples are: the Business Performance Benchmarker release now has benchmarks available for over 200 industries; and more years’ data were published in the csv files.
• There was further increase in the use of administrative data (IR10) in the AES 2017 to reduce respondent burden. As a result, 11 more industries at the survey design level moved from sample collection to full coverage, including: agriculture, forestry, and fishing support services; primary metal and metal product manufacturing; water supply; motor vehicle and motor vehicle parts wholesaling; other goods wholesaling; specialised food retailing; recreational, clothing, footwear, and personal accessory retailing; non-store retailing and retail commission based buying and/or selling; postal and courier; warehousing and storage services; and telecommunications services.
• New reporting standards for registered charities in New Zealand came into effect on 1 April 2015. This impacted on data quality of variables related to charities activities such as government funding in industries with significant non-for-profit units (such as division P Education and Training and division R Arts and Recreation Services).
• The Business Register (superseding the Business Frame) was used for the AES population sample for the first time.
• Some trusts were retrospectively included when first available in 2018. See ‘2018’ for more details.

2018 Improvements made and significant events in the AES 2018 included:

• An increase in the number of trusts in the financial assets investing industry (NZSIOC KK112). There was a change in the classification of these units under the Statistical Classification for Institutional Sectors (SCIS) which was incorporated into the AES population in 2014 and reversed in 2018. As a result, around 10,000 family trust units were included. The revision was taken back to 2014 when the classification was released, which creates a consistent time series. The most significant impact of the revisions can be seen in the financial position, particularly in the fixed tangible assets.
• Improved modelling to utilise the reported variables in the Charities Services data. This improves data quality of variables related to charities activities that were impacted in 2017, such as government funding.

2019 Improvements made and significant events in the AES 2019 included:

Change to the range of the random number line used for selecting businesses to survey.

• Introduction of new accounting standard

International Financial Reporting Standard 16 (IFRS 16) is a significant change to the way leases are treated in financial accounts. This standard was effective from 1 January 2019. Some businesses adopted the standard earlier and responded using this new accounting treatment for leases in AES 2019. Most businesses started reporting on the new basis in AES 2020.

See NZ IFRS 16 for further information including exceptions from using the new standard.

IFRS 16 has the potential to cause significant changes to some of the data collected by the survey. All other things being equal, this change to the accounting of leases is expected to cause: assets, equity and liabilities to increase; interest expense and depreciation expense to increase; and expenses paid to other businesses (purchases) to decrease.

Analysis of the final output data has shown only minor impacts due to IFRS 16 at the industry design level and higher output levels in the 2019 financial year.

2020 Improvements made and significant events in the AES 2020 included:

Change to the range of the random number line used for selecting businesses to survey.

• Impact of COVID-19

The COVID-19 pandemic had significant impact on NZ business activity, particularly from 24th March until 14th May 2020, when COVID-19 alert levels 3 and 4 were in place. Auckland was under alert level 3 again on the 12th August until 23rd September.

As the 2020 Annual Enterprise Survey results cover enterprises with balance dates between 1 October 2019 to 30 September 2020, the impact of COVID-19 was varied.

'Government funding, grants and subsidies' include government funded COVID-19 wage subsidies in the 2020 Annual Enterprise Survey. Wage subsidies passed on to employees are captured under salaries and wages paid. Significantly more COVID-19 wage subsidies are expected in the 2021 Annual Enterprise Survey results.

• New accounting standard for leases

On 1 January 2019 most businesses changed the way they account for leased (‘right-of-use’) assets under the International Financial Reporting Standard 16 (IFRS 16).

See NZ IFRS 16 for further information.

The change to the accounting of leases caused:

• assets and liabilities to increase
• depreciation and interest expenses to increase
• other operating expenses to decrease.

2021 Improvements made and significant events in the AES 2021 included:

• Impact of COVID-19

The COVID-19 pandemic had significant impact on NZ business activity over 2020 and 2021 years, particularly when COVID-19 alert levels 3 and 4 were in place.

As the 2021 Annual Enterprise Survey results cover enterprises with balance dates between 1 October 2020 to 30 September 2021, the impact of COVID-19 was varied.

'Government funding, grants and subsidies' include government funded COVID-19 wage subsidies in the 2021 Annual Enterprise Survey. Wage subsidies passed on to employees are captured under salaries and wages paid. Significantly more COVID-19 wage subsidies are in the 2021 Annual Enterprise Survey results.

Internal

The AES data forms the basis of Stats NZ's national accounts statistics including:

• Annual GDP,
• Financial flows and balance sheets,
• Household saving estimation.

External

The AES provides a wealth of information to businesses, economic researchers, forecasters, entrepreneurs and government agencies. Areas of interest include the financial performance across industries and the more detailed information for industries less well covered by more timely statistics.

4. Annual