CENTRAAL BUREAU VOOR DE STATISTIEK (CBS, STATISTICS NETHERLANDS)

EBD (Enterprise statistics Den Haag)

Postbus 24500

2490 HA Den Haag

The Netherlands

Structural business statistics (SBS) describes the structure, conduct and performance of economic activities, down to the most detailed activity level (several hundred economic sectors). SBS covers all activities of the business economy with the exception of agricultural activities, public administration and (largely) non-market services such as education and health. Main characteristics (variables) of the SBS data category:

• Business demographic variables (e.g. Number of active enterprises)

• "Output related" variables (e.g. Net turnover, Value added)

• "Input related" variables: labour input (e.g. Number of employees and self-employed persons, Hours worked by employees); goods and services input (e.g. Purchases of goods and services); capital input (e.g. Gross investments)

Business services statistics (BS) collection contains harmonised statistics on business services. From 2008 onwards BS become part of the regular mandatory annual data collection of SBS. The BS’s data requirement includes variable “Turnover” broken down by products and by type of residence of client.

The annual regional statistics collection includes three characteristics due by NUTS-2 country region and detailed on NACE Rev 2division level (2-digits).

Statistical Classification of Economic Activities in the European Community (NACE):  NACE Rev.2 is used from 2008 onwards. Key data were double reported in NACE Rev.1.1 and NACE Rev.2 only for 2008. From 2002 to 2007 NACE Rev. 1.1 was used and until 2001 NACE Rev.1

The regional breakdown of the EU Member States is based on the Nomenclature of Territorial Units for Statistics (NUTS). 

The product breakdown is based on the Classification of Products by Activity (CPA) as stated in the Regulation establishing CPA 2008 and its amending  Commission Regulation (EU) No 1209/2014  (from reference year 2015 onwards).

Starting reference year 2021 onwards SBS cover the economic activities of market producers within the NACE Rev. 2 Sections B to N, P to R and Divisions S95 and S96.
Until 2007 the SBS coverage was limited to Sections C to K of NACE Rev.1.1 and from the reference year 2008 to 2020 data was available for Sections B to N and Division S95 of NACE Rev.2.
From 2013, as the first reference year, to 2020 information is published on NACE codes K6411, K6419 and K65 and its breakdown.

From 2008 reference year data collection BS covers NACE Rev 2 codes: J62, N78, J582, J631, M731, M691, M692, M702, M712, M732, M7111, and M7112.

SBS constitutes an important and integrated part of the new European Business Statistics Regulation N° 2152/2019

Data requirements, simplifications and technical definitions are definied in Commission Implementing Regulation (EU) 2020/1197

Enterprise

Local units for regional SBS

- Sections B to N, P to R and Divisions S95 and S96 are fully covered (market producers)

- All size classes are covered

- Market producers are selected using sector codes in the business register

- The frame for SBS statistics is the business register, all data is provided on the enterprise level.

The Netherlands, including the requested information on NUTS 1+2 level.

Data for the Caribbean Netherlands (special municipalities that are located in the Caribbean sea) are not included

1997-present

Not applicable.

• Number of enterprises and number of local units are expressed in units.

• Monetary data are expressed in millions of €.

• Employment variables are expressed in units.

• Per head values are expressed in thousands of € per head. 

Ratios are expressed in percentages.

Starting with reference year 2021 two new regulations currently form the legal basis of SBS:

• Regulation (EU) 2019/2152 repealing 10 legal acts in the field of business statistics (EBS Regulation), and
• Commission Implementing Regulation (EU) 2020/1197 laying down technical specifications and arrangements (general implementing act) pursuant to Regulation (EU) 2019/2152

Year 1995 was the first year for the implementation of the Council Regulation No 58/97 (SBS Regulation).

The Council Regulation No 58/97 has been amended three times: by Council Regulation No 410/98, Commission Regulation No 1614/2002 and European Parliament and Council Regulation No 2056/2002. As a new amendment of the basic Regulation it was decided to recast the Regulation No 58/97 in order to obtain a new "clean" legal text. In 2008 the European Parliament and Council adopted Regulation No 295/2008 and the provisions of this Regulation were applicable from the reference year 2008 to reference year 2020. Regulation No 295/2008 was amended by Commission Regulation (EU) No 446/2014.

Not applicable.

Statistics Netherlands guarantees the privacy of respondents (persons, enterprises and other), the confidential nature of the information provided and the sole use for statistical purposes.

This is captured in law:Act of 20 November 2003, enacting a law governing Statistics Netherlands (Statistics Netherlands Act)

Section 37 of this act:
1. The data received by the director general in connection with the performance of his duties to implement this act shall be used solely for statistical purposes.
2. The data referred to in the first subsection shall not be provided to any persons other than those charged with carrying out the duties of the CBS.
1.3. The data referred to in the first subsection shall only be published in such a way that no recognisable data can be derived from them about an individual person, household, company or institution, unless, in the case of data relating to a company or institution, there are good reasons to assume that the company or institution concerned will not have any objections to the publication.

SBS is also subject to the General Data Protection Regulation (GDPR) as enterprises also comprise sole proprietorship.

Primary confidentiality according to P% rule. This approach takes into account that a contributor to a publication group can use own data to make an estimation of competitors values.

Furthermore, the weight of secondary confidentiality cells are minimised.

The confidentiality treatment is performed using the Tau-Argus package available within the NSI.

Measures to reduce the number of confidential cells:

- Use the P% rule.

- Apply the same confidentiality pattern to strongly correlated indicators.

- The procedure of confidentiality treatment is only applied to (potential) output cells.

The SBS indicators are derived from several datasets which are available online, each with its own release policy. The release policy is articulated in the explanation accompanying the datasets.

The release calendar is part of the explanation paragraph of the on-line tables:

Links to online databases:

Number of enterprises:

Employment Statistics

https://opendata.cbs.nl/statline/#/CBS/nl/dataset/83583NED/table?ts=1523624981763

The policy of Statistics Netherlands is that data which is provided to a third party should also be available at the on-line database.

Annual.

There are no regular or Ad Hoc news releases on SBS.

A selection of SBS results is published in cooperation with the Ministry of Economic affairs:

Annual publication of SME's 2023 (Dutch language)

Jaarbericht 2023 (staatvanhetmkb.nl)

Figures on business demography are published on a quarterly basis, the other indicators on an annual basis.

Links to online databases:

Number of enterprises (quarterly):

https://opendata.cbs.nl/statline/#/CBS/nl/dataset/81589NED/table?ts=1523624088065

Business Demography

https://opendata.cbs.nl/#/CBS/nl/dataset/85201NED/table?dl=91A7B

Production Statistics

https://opendata.cbs.nl/statline/#/CBS/nl/dataset/81156ned/table?ts=1523624176265

Investment Statistics

https://opendata.cbs.nl/statline/#/CBS/nl/dataset/81351ned/table?ts=1615305346448

https://opendata.cbs.nl/statline/#/CBS/nl/dataset/81352ned/table?ts=1615305442297

Employment Statistics

https://opendata.cbs.nl/statline/#/CBS/nl/dataset/83583NED/table?ts=1523624981763

Access to microdata for research purposes is only possible for specific organisations under strict conditions:

https://www.cbs.nl/en-gb/our-services/customised-services-microdata/microdata-conducting-your-own-research/applying-for-access-to-microdata

The data are sent to Eurostat to be used in European aggregates and to be released also as national data.

Methodology (in Dutch)

Number of Enterprises:

https://www.cbs.nl/nl-nl/onze-diensten/methoden/onderzoeksomschrijvingen/korte-onderzoeksbeschrijvingen/populatie-van-actieve-bedrijven-in-nederland

https://www.cbs.nl/nl-nl/onze-diensten/methoden/onderzoeksomschrijvingen/korte-onderzoeksbeschrijvingen/bedrijvendemografie

Production Statistics

https://www.cbs.nl/nl-nl/onze-diensten/methoden/onderzoeksomschrijvingen/korte-onderzoeksbeschrijvingen/productiestatistiek

Investment Statistics

https://www.cbs.nl/nl-nl/onze-diensten/methoden/onderzoeksomschrijvingen/korte-onderzoeksbeschrijvingen/investeringen

Employment Statistics

http://www.cbs.nl/nl-NL/menu/themas/arbeid-sociale-zekerheid/methoden/dataverzameling/korte-onderzoeksbeschrijvingen/statistiek-werkgelegenheid-lonen-banen-art.htm

Quality declaration:

https://www.cbs.nl/en-gb/about-us/organisation/quality

Statistics Netherlands follows the Code of Practice for European Statistics

More information on quality assurance is available on the website:

https://www.cbs.nl/en-gb/about-us/organisation/quality

Statistics Netherlands uses administrative data as much as possible as it provides almost data completeness.

National accounts use the structural information and Year-to-Year growth rates to calculate the Gross Domestic product.

The Centre for Policy Related Statistics: Under certain conditions, Statistics Netherlands can make microdata (anonymous data at the level of individual persons and businesses) available for statistical research. The data can, for example, be used to measure the performance of specific categories of enterprises.

Ministry of Economic Affairs and Climate Policy: SBS data is used for policy purposes. Special interest in data on SME's (small and medium sized enterprises).

Ministry of Health, Welfare and Sport: SBS data on NACE Q (86-88) and R (93).

Eurostat, ECB, and other international organisations for various policy purposes.

Feedback of the most important user National Accounts is received on a structural basis. Data quality in general is perceived as good. A quality issue of SBS statistics is the limited sample size for SME enterprises. The accuracy of the results on the most detailed level is therefore also limited.

Feedback from Eurostat (e.g. the compliance score but also the results from the validation process) is used to improve statistics.

Most datasets are 100% complete or almost complete.

Statistics Netherlands uses administrative sources as the main source for Number of enterprises and employment indicators. For some sections (NACE Q and R) corporate (for legal persons) and individual (for natural persons) tax data is also used for financial variables. NACE Q also uses administrative data from the ministry of Health, Welfare and Sport. Administrative data is furthermore used as auxiliary information in the weighing process of most financial variables and most NACE groups. The sampling errors which are associated with traditional survey sampling are therefore reduced, especially on the higher levels of NACE aggregation.

Remaining sources for inaccuracies:

- Sampling errors (not all variables are covered by administrative data)

- The use of administrative data could potentially result in a small bias in some cases (e.g. different definitions of the source compared to SBS definitions).

- Measurement errors are possible especially for multinational enterprises. As a countermeasure Statistics Netherlands has installed a large case unit for the largest enterprise groups.

The preliminary results can be slightly biased. For preliminary data the estimates rely more heavily on previous year data, and extrapolations from administrative data. For NACE Q larger biases can occur as VAT data is not used in this sector because of many VAT exemptions. 

The first release of the data is 10 months after the reporting period (preliminary data deliveries)

The second and final release is 17/18 months after the reporting period.

Sample surveys are used for the calculation of financial variables. The sampling error is usually small but especially in smaller NACE groups or NACE x size class groups the sampling error can be large.

Sampling errors are calculated following the guidelines provided by Eurostat.

Non-sampling errors could potentially arise in SBS:

- Coverage errors: The completeness of the Business Register in Statistics Netherlands is considered to be very good. Most SBS indicators are based on statistics that use the business register. Lack of completeness is therefore not considered to be a big problem. Effect negligible.

- Data collection and access errors: In general the results of mistakes in data collection are considered to be small and acceptable. Most difficulties arise with multinational enterprises as they can have complex business structures within and outside the borders of the Netherlands. The effect is small.

However, there is an exemption into section Q and R. The tax corporate and individual tax data is incomplete due to tax exempt status of enterprises, reporting after data collection period, and incomplete linking with Business Register. The effect is still unknown and more research is needed.

- Unit non-response: The weighted response rates are considered high (around 90% depending on the variable and NACE group). The strategy for minimising non-response is mainly focussed on one or several telephone reminders. Effect is  negligible.

- Item non-response for the key variables. For Total turnover and Total investments, the weighted response is around 90%. Number of enterprises and number of employed persons are calculated using administrative data (non-response not applicable). Effect negligible.

- Editing, coding and imputation errors: Imputation errors for the financial variables are minimised by using T-1 data and VAT information on the microdata level and in some cases using information from the company financial statement. Editing errors are minimised by following a different approach for small enterprises (automatic editing) and large enterprises (manual editing). Effect negligible.

- Modelling errors: The grossing procedure uses auxiliary information (VAT data and information on persons employed for the whole population). This reduces bias resulting from selectivity in response. Effect small.

-Modelling errors for Nace Q:
Flag D (definition differs) is applied in NACE Q because administrative data used does not align with SBS variables:
- Purchases of goods and services for resale: included in this SBS-variable are the fee costs for independent medical specialists from the annual reporting data of health care institutions data and the purchased services of costs of sales from the tax data.

- Value of output: see different definition for ‘Purchases of goods and services for resale’

- Expenses on services provided through agency workers: included in SBS-variable are all personnel costs for personnel not on payroll (all data sources).
Flag E (Estimated) is applied for NACE 86.22 because of attempt to adjust for incorrect and partly missing relations of tax data to the business register.

-Modelling errors for Nace R:
Flag U (unreliable) is applied for Purchases of goods and services for resale because available data sources do not align with this SBS variable.
Flag E (estimated) is applied for almost all NACE R categories for Expenses on services provided through agency workers because of adjusting for incorrect and partly missing relations of tax data to the business register.

Depending on the indicator, figures are published 3 to 16 months after the end of the reporting period.

The data is in general delivered before or at the target date.

The same concepts apply for the different regions of the Netherlands

Break in series are usually small but sometimes non-negligible and due to changes in the business register or due to redesign.

Minor breaks in series have occurred in 2000, 2006, 2009, 2013 and 2021 for different reasons and depending on the group of indicators.

The SBS statstics use (the market producers of) the business register as a population frame (employment variables excluded):

- The number of enterprises in SBS equals the number of enterprises in the business register (correcting for administrative renumbering). Turnover, is not available in the business register.

- The 'size class for persons employed' in the business register is a 'coordinated value' used by most statistics for sampling and raising data/calculating totals. It is also used as a size class proxy for SBS. The business registers uses previous year (T-1) data from the Policy administration (administrative data on wages, benefits and labour relations) to calculate the persons employed and corresponding size class. SBS uses this size class information in calculating size class data. The SBS indicator on persons employed however is based on the reporting year (T) information from the same Policy register. The size class of SBS and the actual number of employed can be inconsistent due to timelag and an incomplete match between business register and the Policy administration.

Prodcom within statistics Netherlands is based on a separate questionnaire from SBS. The Large Case unit compares production from SBS and Prodcom on the Enterprise level (correcting for different concepts: Ownership principle versus flow of goods). Coherence level is considered to be high.

Value added of national accounts is not directly calculated from SBS. Both SBS and national accounts use the same underlying base statistics in its calculations (production statistics, investments, wage statistics etc). Several differences between SBS and national accounts can be detected: valuation against basic prices (national accounts) versus factor costs (SBS), undeclared work etc. However the production statistics used by both is the main source for national accounts in calculating 'Production' and 'Use' and the resulting Value Added.

STS uses SBS Turnover data as weights. In both STS and SBS VAT data is used in the estimation process. The STS approach is more focused on correcting for non-real mutations (in the business register or observations).

Labour statistics (LFS, LCI and JVS) and SBS differ in the way employee benefits are calculated. SBS for wages/employee benefits is based on profit and loss information. Labour statistics are primarily based on the Policy Administration. Health and welfare financial statistics within Statistics Netherlands (NACE Q) and SBS differ in the way population is determined. Consolidation of enterprises in annual reporting to ministry of Health, Welfare and Sport and the use of the register for Professions in Individual Health Care are causes for difference in population.

Tourism statistics: There is no coherence iwith SBS. Mainly because the population is based on a functional approach.

Sectorial statistics (transport, energy etc.) follow a functional approach instead of a institutional (Enterprise/NACE) approach. The data are not used for validation of SBS.

Business services of SBS are collected using the same questionnaires as the main table from SBS.

Aggregates are consistent with sub-aggregates. An exception are the regional figures. For a small number of enterprises no NUTS code may be available, leading to differences between the sum of the NUTS figures and national figures.

Coherence can be limited between different groups of indicators. The SBS statistics are produced from different sources and by different departments. Employment variables are derived from the Policy administration and the business register is only used for determining the NACE and size class of the activity. Differences between Policy administration (SBS employed persons) and other SBS indicators are possible.

Section K data is for some variables based on the business register and for other variables based on central bank data or other financial sources. 

Statistics Netherlands does not have a separate SBS unit. The data are collected from different surveys and from different statistics.

The costs associated with SBS statistics are not available.

Investment indicators are collected using surveys. The total response burden for respondents is calculated  to be 17.000 hours.

Indicators on Output and performance indicators are also collected using surveys. The total response burden for respondents is calculated  to be 81.000 hours.

Business demography: Quarterly figures. Provisional figures for current year, final figures are published in January T+2

Output and performance, purchases and employee benefits: Figures are published at T+15 months. No revisions are made.

Investments: preliminary data published at T+12 for Manufacturing, final data for all sections at T+18 months.

Employment T+9 month (preliminary), T+21 months (final).

Business demography:

Revisions should be small or zero since both estimates are based on (almost complete) administrative data.

Employment:

Revisions should be small or zero since both estimates are based on (almost complete) administrative data.

Turnover:

Preliminary data estimations are in most sections based on VAT data and SBS T-1 reporting data. For NACE Q preliminary data estimations are based on auxiliary data and SBS T-1 reporting data due to lack of VAT data (many VAT exemptions in this sector). Larger revisions are possible when observations for the reporting year are available.

Business demography: derived from the business register. All enterprises in the business register are considered to be active and used for calculating (counting) the number of enterprises.

Employment:

As from 2006 Statistics Netherlands uses a new source for the compilation of the number of jobs of employees: an integral register containing information on wages and social contributions of all employees in the Netherlands. This employee register is controlled by the Social Insurance Institute and is filled with the data of employees from the declaration of earnings which employers send to the Tax Authorities. The employee register contains all employees working for companies and institutions who are obliged to pay taxes on earnings and social contributions.

As a first step the number of employees has been determined from the employee register. Then the companies and institutions in the register are linked with the General Business Register of Statistics Netherlands through their tax number. The General Business Register contains the NACE of companies and institutions, which is used to differentiate the number of jobs by economic activity. Besides the General Business Register contains the main office of companies and institutions.

The register lacks information on the location of companies and institutions with more than one office in the Netherlands. Therefore an additional survey among these companies with more than one office is held. In this survey, companies are asked to report the number of employees on municipality level. The results of the survey are combined with the number of jobs of employees from the register and result in regional employment figures.

Investments: Sample survey, stratification variables are NACE and size class. Sampling rates vary for each stratum and are determined by Neyman allocation (Total Investments used as the variable for allocation). Total sample around 55.000

Output and performance, purchases, employee benefits:

All sections except K,Q,R: Sample survey, stratification variables are NACE and size class. Sampling rates vary for each stratum and are determined by Neyman allocation (value added used as the variable for allocation). Total sample around 75.000. For some NACE groupings a threshold of 10 or 50 persons employed is used (every 2 out of 3 years) and results are imputed using T-1 or T-2 observations. Administrative data (VAT) is used as auxiliary information to improve the estimation procedure.

Section K: Data collected by Dutch Central Bank. ca 1500 survey respondents.

Section Q: Combination of sample survey data, data from annual reporting for health care institutions (DigiMV)  collectecd by the minstry of Health, Welfare and Sport and corporate (for legal persons) and individual (for natural persons) tax data are used in the estimation process.

Section R: Combination of sample survey data, annual reports and corporate tax data are used in the estimation process.

Annual data collection for Investment indicators and Output and performance indicators (except for section K which is a combination of quarterly and annual data collection.

Quarterly data collection for employment indicators.

-For administrative data sources which are collected from the Tax authorities, the business register department links the administrative data to the business register and gives the statistical departments access to this data.

-For the administrative data which are collected from the ministry of Health, Welfare and Sport in NACE Q the statistical department links the data to the business register.

-Surveys from electronic questionnaires (or paper questionnaires by exception).

-Repeated telephone contacts are used to improve response rates. Legal enforcement is used as a last resort in a limited number of cases.

Business demography

plausibility checks

The main check is to compare the annual change in the calculated number of active units with the change in number of business register units.

Output and performance, purchases, employee benefits:

completeness checks (data integrity rules)

validity checks (internal consistency)

plausibility checks

Missing values are imputed automatically and checked manually for large enterprises. Inconsistencies are removed manually or automatically. There are a number of plausibility checks incorporated in the process. These are comparing with T-1 data and comparing with similar units.

Investments:

completeness checks (data integrity rules)

validity checks (internal consistency)

plausibility checks

Missing values are considered to be 0 values. Inconsistencies are then removed manually. Plausibility is checked using historical data.

Employment:

completeness checks (data integrity rules)

validity checks (internal consistency)

plausibility checks

The register data have gone through a general imputation-correction method whereby double records are removed and missing records or variables are imputed for example with the use of declarations from other periods. Second, through top down analysis, the units with the highest number of employees are individually controlled and when necessary corrected. The quality of the data is further checked by analysing the development of the data on employment and wages differentiated by economic activity and regions over time. When necessary corrections on the values of the variables of the reporting units are made.

Business demography:

No non-response, no methods for dealing with non-response needed.

Output and performance, purchases, employee benefits:

Imputation methods by unit non response (only single imputation is applied) are:

- Use VAT data for estimation of the Turnover (not applicaple for NACE Q becasue of VAT exemptions)

- Use unit response from T-1 multiplied by the average Year-to-Year growth rates of similar units.

- Use average or (by exception, medians) of similar units if T-1 is unavailable.

- Use the proportions between variables of similar response units.

Imputation methods by item non response (both multiple and single imputations are applied) are:

- Use the relationship between variables.

- Use the proportions between variables of responding units.

- Use of multiple imputation with Random Forest of variable ‘Expenses on services provided through agency workers’ (sections Q and R) because this item is not available in tax data.

In the grossing procedure non-respondents are excluded. The number of active units in the frame is used in the grossing procedure; VAT is used as auxiliary information in the grossing procedure.

Investments:

The methods for dealing for unit non response are:

Use T-1 data and the development of similar units (imputation of large units)

Use the average values of similar units if T-1 is unavailable (imputation of large units)

The methods for dealing for item non response are:

Missing values are considered to be 0 values

If the Total investments are unequal to the sum of the parts then a manual check and correction is applied.

Employment:

First, the register data have gone through a general imputation-correction method whereby double records are removed and missing records or variables are imputed for example with the use of declarations from other periods. Second, through top down analysis, the units with the highest number of employees are individually controlled and when necessary corrected. The quality of the data is further checked by analysing the development of the data on employment and wages differentiated by economic activity and regions over time. When necessary corrections on the values of the variables of the reporting units are made.

The reference period which represents the calendar year the most is chosen. Only for units with a large contribution a correction will be made if the length of the reference year is unequal to one year. For smaller units the fiscal year financial data is calibrated using the factor VAT data turnover / fiscal year turnover. The result is more in line with the calendar year.

No specific comments