Reference metadata describe statistical concepts and methodologies used for the collection and generation of data. They provide information on data quality and, since they are strongly content-oriented, assist users in interpreting the data. Reference metadata, unlike structural metadata, can be decoupled from the data.
Directorate Business Statistics, Unit "Structural Business Statistics"
1.3. Contact name
Confidential because of GDPR
1.4. Contact person function
Confidential because of GDPR
1.5. Contact mail address
Guglgasse 13, 1110 Wien, Austria
1.6. Contact email address
Confidential because of GDPR
1.7. Contact phone number
Confidential because of GDPR
1.8. Contact fax number
Confidential because of GDPR
2.1. Metadata last certified
25 August 2025
2.2. Metadata last posted
25 August 2025
2.3. Metadata last update
25 August 2025
3.1. Data description
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).
3.2. Classification system
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 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).
3.3. Coverage - sector
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.
3.4. Statistical concepts and definitions
SBS constitutes an important and integrated part of the new European Business Statistics Regulation (EU) 2019/2152
Statistical units are based on the Council Regulation (EEC) No 696/93 and the operational rules based on this regulation, too.
3.5.1. Treatment of complex enterprise
Data treatment
Sample frame based on enterprises
No
Surveying all legal units belonging to a complex enterprise
No
Surveying all legal units within the scope of SBS belonging to a complex enterprise
No
Surveying only representative units belonging to the complex enterprise
Yes (see below)
Surveying of legal units within the scope of SBS belonging to a complex enterprise exceeding the threshold for the statistical survey, model based estimation for all other legal units belonging to complex enterprises
Yes
Comment:
For selected large units internal flows are collected in the annual SBS survey. Some of these large units provide a matrix of internal flows between national legal units.
In fact, a full set of SBS data for each legal unit is available serving as basis for manual and automatic consolidation.
3.5.2. Consolidation
Consolidation method
Consolidation carried out by the NSI
yes
Consolidation carried out by responding enterprise/legal unit(s)
No
Other methods, please specify
No
Comments:
For selected large units internal flows are collected in the annual SBS survey.
3.6. Statistical population
The statistical population includes all active market enterprises of Sections B-N, P-R and divisions S95, S96 of NACE Rev. 2 in the Statistical Business Register (SBR). Active means that the enterprise has either turnover and/or employment at any time during the reference period. No cut-off threshold is applied.
3.7. Reference area
Reference area is Austria.
Registered branches of foreign enterprises are included in SBS if an activity relevant for value added in Austria is carried out or employees are reported to the Austrian Social Security Authorities.
According to definitions, activities abroad of Austrian enterprises are included if the activity abroad is carried out by a non-autonomous local unit of the enterprise. Autonomous branches of Austrian enterprises in foreign countries should be excluded, the actual exclusion depends on data availability.
With the support of the LCU and National Accounts the largest branches (in Austria or abroad) are discussed in order to be depicted correctly in the different domains of business statistics. In some cases, the delimitation is very difficult and a certain bias cannot be eliminated for the full target population.
3.8. Coverage - Time
Length of time series: 1995, 1997-2017, 2018-2020, 2021-2023
Most important breaks in time series:
2021 (implementation of EBS requirements, extended coverage by activities and size classes)
2018 (switch from the legal unit to the statistical enterprise)
2008 (switch from NACE Rev. 1.1 to NACE Rev. 2)
2002 (change of national methods)
1997 (change of national methods)
see also sub-concept 15.2
3.9. Base period
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.
Year 2023
Data refer to calendar or fiscal year. If a fiscal year differs from the calendar year, the fiscal year that has more months in the calendar year in question is used; special adjustments are made for exceptional (large) units. For 2023, 8,894 legal units had a different fiscal year, counting for about 16% of persons employed.
6.1. Institutional Mandate - legal acts and other agreements
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
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.
6.2. Institutional Mandate - data sharing
Not applicable.
7.1. Confidentiality - policy
According to Article 19 of the Federal Statistics Act data shall be published in a manner that prevents any connection being made with statements concerning specific or identifiable data subjects unless the data subject concerned has no interest warranting protection in his/her data remaining confidential. In the event such connections cannot be ruled out, publication is subject to the explicit prior written consent of the data subject involved.
The SBS Confidentiality charter is the main reference for publication of SBS results on European level.
7.2. Confidentiality - data treatment
Aggregates for Austria are flagged confidential, if data refer to less than 3 statistical units (primary confidentiality). Secondary confidentiality pattern is calculated by an algorithm making it impossible to recalculate confidential data by subtraction from row and column aggregates. No dominance criterion is applied, but the right of the enterprises themselves to protect their individual data has to be respected. So to speak, if a company proves dominance in a specific aggregate, the cells of the table might be closed, even if there are more than 3 contributing units. For Austria, currenty no anterprise has applied for such an exception.
The confidentiality pattern for SBS is consistent with the confidentiality pattern for Business Demography.
For calculation of an efficient secondary confidentiality pattern, a heuristic algorithm, developed by the Center for Methodology of Statistics Austria is applied.
7.2.1. Confidentiality processing
Data treatment
Confidentiality rules applied
yes
Threshold of number of enterprises (Number)
<3
Number of enterprises non confidential, if number of employments is confidential
yes
Dominance criteria applied
no
If dominance criteria applied specify the threshold (Number)
Secondary confidentiality applied
yes
Comment
8.1. Release calendar
Dates of the release of SBS results in the first publication media (=Website of Statistics Austria) is preannounced in the release calendar of Statistics Austria, which is publicly accessible.
Final results are released by end of June (t+18 months), preliminary results released by end of October (t+10 months).
Results are disseminated to all users at the same time; standard tables are for free access; more detailed tables in the Statistics Austria dissemination database STATcube are with costs. Specific data is provided to everyone with a specific request. Feasibility of the request and communication of possible expenses are done in advance.
Annual.
10.1. Dissemination format - News release
No news releases.
10.2. Dissemination format - Publications
SBS data is disseminated on the Website of Statistics Austria. No paper publication. Website is available in German and English.
10.3. Dissemination format - online database
SBS data is disseminated in the online database STATcube on an annual basis, in German and English. STATcube allows the processing of a large amount of data, generating tables that can be downloaded in different formats according to individual needs. Access is possible by: - subscription: it offers access to larger data volumes, to data or detailed classifications locked in the Guest Access (depending on the matters), as well as the possibility to store query tables - registration without subscription: it offers access to free data only and possibility to store query tables.
10.4. Dissemination format - microdata access
Since July 2022, in the Austrian Micro Data Center micro-data access is possible for researchers under certain conditions. The rules for the micro data access are laid down in the national Statistical Law.
10.5. Dissemination format - other
Data are sent to Eurostat to be released as national data and to be used in the European aggregates.
10.6. Documentation on methodology
Metadata about methodology and quality is available in all publication media. The most detailed documentation on methodology and quality is the so called "Standard Documentation", which is released on Statistics Austria's website. The Standard Documentation is available in German; a summary is available in English as well.
Statistics Austria is committed to ensuring the highest quality with respect to the compilation of statistical information. In accordance with the Federal Statistics Act (Article 24), Statistics Austria has to use statistical methods and processes in compliance with internationally recognised scientific principles and standards, conduct ongoing analyses of the statistics with a view to quality improvements and ensure that statistics are as up-to-date as possible. This commitment to quality is also specified in Statistics Austria’s mission statement. In adopting the European Statistics Code of Practice, Statistics Austria has committed to adhering to principles and standards for the production and dissemination of high-quality statistics. As part of the European Statistical System (ESS), Statistics Austria recognises the Quality Declaration of the ESS. The Statistics Council has set up a Quality Assurance Committee, a primary focus of which is the ongoing examination of potential quality improvements
Statistics Austria uses standard-documentation to provide users with information about the underlying concepts, definitions, methods used and the quality of the statistics (meta-information) in a standardised format.
11.2. Quality management - assessment
In cooperation with the Statistic Committee’s Quality Assurance Committee, feedback meetings concerning the quality of the different statistical products are held regularly within the framework of Statistics Austria’s quality management programme. In addition, internal revisions and quality audits are carried out. The content and objectives concern critical examination of different aspects with particular consideration of the methods, ressourses and processes used; identification of quality improvement potential; development of recommendations for improvement measures; and improvement of the standard documentation relating to the statistics in question, with special attention to the views of users and external experts. A high quality is attributed to SBS data in Austria, nevertheless the following issues are raised in the course of the last internal quality audit (in 2021):
clear description of consequences of the switch to the statistical enterprise, further availability on data for legal units (already implemented)
detailed requests for data on regional level (implemented with the reference year 2021; more information are available at local unit level)
consistency with BD (implemented with the reference year 2021)
With reference year 2021 the open issues were implemented.
12.1. Relevance - User Needs
Internal/national: National and Regional Accounts, Statistical Business Register, Social Statistics, Environmental Statistics, Science and Technology Statistics, Austrian Economic Chamber of Commerce, Austrian Institute of Economic Research and other research institutes, Universities, Ministries, press and media, Central Bank, enterprises, students etc.
External/international: Eurostat (European Commission), ECB, OECD etc.
Different user needs (National SBS Regulation, National Accounts, other user needs) define the requirements for SBS at national level. Beside the discussions at European level, key users are consulted in the course of so called technical subcommittees. A technical subcommittee for enterprise statistics, which is responsible for SBS, comprises representatives from different user groups. There is at least one meeting in the course of the year. In the case of revisions and conceptual changes of particular statistical products ad hoc meetings take place. The changes in the national SBS concept with the reference year 2021 were discussed with the national users in detail. To fulfil national user needs SBS data published at national level is different (more detailed) from data sent to Eurostat.
12.2. Relevance - User Satisfaction
A survey related to the user’s satisfaction is organized by Statistics Austria for key statistical domains. However, the information derived from this survey, which was carried out for the last time in 2024/beginning 2025, does not provide specific information for SBS, because various business statistics for industry, construction, trade and services are covered in total. Nevertheless results can be used for evaluating the fulfilment of user needs in general. Users have attributed an adequate satisfaction with results; more results on regional level are desired. In general, more technichal problems with the Website of Statistics Austria are mentioned than problems with the content of the information requested.
12.3. Completeness
National data for the reference 2023 are complete.
13.1. Accuracy - overall
The most significant errors are:
measurement errors (distinction market-non market, wrong NACE code, fiscal year differs from the calendar year)
modelling errors (model based estimation, delineation of statistical enterprises and consolidation)
Non-response and data processing errors can occur, but these errors are not significant. Coverage errors were significant till the reference year 2020. With the inclusion of very small enterprises and alignment of coverage to BD scope coverage errors are negligible as well.
Preliminary results are not biased; the period from the first release of results to the final release is 8 months.
13.2. Sampling error
Calculation of a sampling error is not possible. The national SBS strategy (exhaustive primary survey of all units exceeding legally defined thresholds and model based estimation using SBR and administrative data on micro level for all units below reporting thresholds) does provide a micro data set for all units in the coverage. Therefore there is no sampling error - or the error is zero.
13.3. Non-sampling error
Non-sampling error is applicable, but it is negligible or rather small.
Unit non-response:
SBS survey is compulsory (National SBS Regulation) – therefore bias resulted from unit non-response is very small. Statistics Austria gives notice of units refusing to respond to administrative authorities. For treatment of unit non response, the same estimation model as for legal units below thresholds is applied. As large units are checked manually in the course of the data editing process, impact on results is negligible.
The weighted unit response rate for legal units was transmitted to Eurostat according to SBS requirements.
Sample size: 36,397 legal units, 5.7% of the total population of legal units were sampled.
Unweighted response rate in number of legal units: 97.3%; weighted response rate 98.7% (unit response is weighted in terms of employees and self employed persons).
The bias resulting from non-response is very low. Large units are checked manually in the course of the data editing process.
Bias from model based estimation: for model based estimation a bias (mainly for detailed variables) is very small, because estimation parameters are calculated mainly with legal units in the same economic activity above thresholds and applied to enterprises below thresholds. The bias is minimized, because only those units above thresholds are taken for calculation of model parameters, which are most similar to units below thresholds (=smallest 70% of units above thresholds measured in terms of turnover). Regarding structural differences between economic activities a bias is possible, but it has to be noted that for most economic activities (about 80% of NACE Rev. 2 classes covered in SBS) the absolute frequency for estimation at the most detailed NACE level was sufficient. If a strong bias is observed in the course of model based estimation expert rating (=corrections in the estimation model) is applied for selected economic activities.
Coverage errors: imperfection in SBR and consequently in SBS was reduced with the reference year 2021 (inclusion of units with less than 10,000 EUR turnover), so the imperfection is almost negligible. Nevertheless coverage errors due to wrong NACE classification are possible: legal units were informed about their NACE code by classification notifications; wrong NACE codes are changed in the SBR. In the course of the primary survey a wrong NACE code was observed for only about 1% of legal units. It has to be noted that the quality of NACE is better for enterprises covered in statistical surveys, so for the small units a higher share of wrong NACE classification is possible, but the share is not known. The distinction between market/non-market is according to ESA 2010; especially in the acitvities covered for the first time some bias can't be excluded. But the bias is very small, as there is an annual update of non market activities in SBR by National Accounts.
14.1. Timeliness
Data collection deadline: 9 months after the end of the reference year National dissemination of preliminary results: 10 months after the end of the reference year Post-collection phase: 15 months after the end of the reference year
Manual consolidation: 15 months after the end of the reference year
Model based estimation: 16 months after the end of the reference year
Automatic consolidation: 17 months after the end of the reference year National dissemination: 18 months after the end of the reference year
The timetable for producing and disseminating SBS data is aligned to data availability in legal units and administrative sources, which is laid down in EU and national regulations.
14.2. Punctuality
Transmission deadline to Eurostat was respected for the reference year 2023.
15.1. Comparability - geographical
For the entire national territory the same concepts and definitions are applied. In general definitions of EBS Regulation are applied but different accounting schemes (IFRS not obligatory in Austria) may have an impact on comparability.
15.2. Comparability - over time
The length of comparable time series and the reasons for differences are provided in sub-concept 15.2.1.
15.2.1. Time series
Time series
First reference year available (calendar year)
1995
Calendar year(s) of break in time series
1995, 1997, 2002, 2008, 2018, 2021
Reason(s) for the break(s)
1995: census for all economic market activities
1997: stratified random sample with free grossing up; yearly rotation of sampled units, reduced coverage
2002: new statistical law, better access to administrative sources and increasing coverage of SBR; new strategy for compiling SBS results (cut-off sample and model based estimation in combination with use of administrative sources for enterprises below thresholds)
2008: introduction of NACE Rev.2 and SBS-Recast-Regulation
2018: application of the statistical enterprise
2021: implementation of EBS requirements, extended coverage
Length of comparable time series (from calendar year to calendar year)
1995 1997 – 2001 2002 – 2007 2008 - 2017
2018 - 2020
from 2021
Comment
Important information about relevant developments for the reference year 2022 can be found in chapter 19.
15.3. Coherence - cross domain
SBS - SBR: SBR and SBS is coherent for active enterprises.
SBS - BD: Data are fully coherent from reference year 2021 onwards.
SBS - Inward FATS: Beside information about foreign control from the Austrian Central Bank, SBS serves as basic data source for compilation of Inward FATS. So coherence of both statistical products is 100%.
SBS - PRODCOM: PRODCOM is based on KAU. SBS is based on enterprises, which influences coherence. Above this, the two statistical domains are compiled at different points of time, small deviations (e.g. due to retrospective reorganisation) have to be taken into account as well.
SBS - Value added of National Accounts: SBS is the main data source of National Accounts. For the calculation of output, intermediate consumption and gross value added in NA some adjustments of SBS data are necessary to meet the requirements of ESA 2010. These are, for example: definition differences between NA and SBS based on financial statements, therefore output and intermediate consumption in NA are adjusted; estimation of hidden economy, addition of tips, insurance premiums reported as operating expenses in SBS are reduced to the service charge concept, adjustments for non-calender business year, FISIM and Market Making Services are added to intermediate consumption, changes in inventories are adjusted for holding gains and losses a.s.o
SBS - STS: STS data can directly be compared with data from the annual SBS for NACE B-F. As these two statistics are carried out at different points of time, small deviations (e.g. due to retrospective reorganisation) have to be taken into account. Additionally different interpretations (e.g. employees: in SBS yearly average, in STS monthly average resp. number of employees at a special date.) have to be taken into account. Moreover it has to be taken into account that the reference period in SBS is not always the calendar year due to different possible financial years of legal units. From the reference year 2018 the application of the statistical enterprise has to be considered as well. Evolution of turnover and persons employed from STS in trade and services: checked on a yearly basis. Generally speaking, both data sources show a very good coherence. Some restrictions are caused by the following reasons: different strategy. In STS a stratified random sample with grossing up to the population is used for all economic activities except section I of NACE Rev. 2; VAT advance return is used for turnover for most of enterprises in STS (except selected enterprises in trade); definition differences to turnover in company accounts; different reporting periods; financial year is not the calendar year; retrospective reorganisation of enterprises; for STS in trade and services only indicators are published, no gross data. The developments of most economic activities show coherence in general. From the reference year 2018 the application of the statistical enterprise have to be taken into account as well.
Cross domain coherence is checked in the course of data analysis on macro level (data production) and in the course of meta data reporting on national level.
15.4. Coherence - internal
Internal coherence is 100%
Restricted from publication
17.1. Data revision - policy
Preliminary data at t+10 are revised with final data at t+18 months.
17.2. Data revision - practice
Normally the only revision is the revision of the preliminary data with the final data.
The data revision (preliminary data versus published final data) for 2023 is -2.9% for turnover and -0.5% for persons employed. So the concept and models for calculation of preliminary data work quite satisfactory.
For calculation of preliminary results already available data from the primary survey, administrative sources, STS in industry and construction and an estimation model are applied.
For reasons of data quality of preliminary results additional plausibility checks are carried out for larger enterprises and enterprises with major deviations from the previous year. Birth and death of enterprises and NACE changes are also checked.
18.1. Source data
Type of Source:
Data source is a statistical survey for legal units combined with administrative sources.
Cut off sample: in the national regulation activity-based survey thresholds and the flexible application of the thresholds are written down. In general, the thresholds can be adjusted every year, provided that number of sampled units are not more than 38,000 and quality criteria for coverage are met. Legal units exceeding the thresholds are exhaustive in the sample. All units below the thresholds are calculated with information from SBR, administrative sources and model-based estimation on micro level, so for each unit a complete micro dataset is available.
Stratification criteria: not applicable
Selection schemes: activity based; employment and/or turnover thresholds based on structure of different economic activities; a certain coverage is necessary for application of model-based estimation
Threshold values for the reference year 2023 see annex "Thresholds for the Structural Business Statistics Survey 2023 to delimit the survey mass"
The effective sample size (sampled units without unit non-response) for the reference year 2023 was 36,397 legal units (5.7% of the population), covering 62.8% of persons employed and 85.9% of total turnover.
Administrative sources:
Social security data: Number of self-employed persons and employees; basic data source and for plausibility checks, model-based estimation
Wage tax data: Wages and salaries, part time employment, social security contributions; plausibility checks, imputation, model-based estimation
Turnover and Income tax data: Turnover and purchases of goods and services, selected items of stocks; plausibility checks, imputation, model-based estimation
Financial Statements
Data from Supervisory authorities for insurance services and credit institutions
STS for industries and construction and data from imports of goods and services are used; basic data source and for plausibility checks, model-based estimation
Administrative data are available as micro data for the entire population covered by the data sources; for tax data: 1:1; 1:n in exceptional cases; n:1 in the case of group taxation of enterprise groups; for social security data 1:1, 1:n for some cases
Frequency of updating: the frequency to which the used administrative data sources are updated is satisfactory. Administrative data are subject to several revisions with increasing degree of completeness.
Frame:
The frame from which the sample is drawn is the SBR of Statistics Austria for legal units and enterprises. The frequency of updating the SBR is as follows:
Register of companies: daily
Tax Register: monthly or quarterly (income tax)
Social security register: weekly
Membership file of the Austrian Federal Economic Chamber: weekly
Information obtained from economic surveys and classification notification: ad hoc
The sample is updated every year, with the possibility of slight updates in the course of the survey.
The classification of the principal activity is based on the value added. If this information is not available, turnover or employment is used as approximation. Secondary activities are determined in the same way as principal activities. Ancillary activities are classified with the principal activity. A top-down approach is applied for classification of units.
The principal activity can change whenever new information is available but in general the principal activity is held stable in the SBR within a calendar year with the exception of the reorganization of an observation unit. Within a calendar year it should not be changed for reasons of continuity and coherence with short-term statistics. Units which are part of the cut-off survey for SBS are directly asked for their principal activity. The SBR is updated with activities changed in the course of surveys. Additionally, legal units are informed about their activity classification in the course of classification notifications.
Delineation of statistical enterprises and definition of ancillary activities is aligned to proposals of Eurostat. Some national particularities are considered. Enterprises of about 80 EG (Enterprises Groups) are profiled and consolidated manually – for the other EG automatic methods are applied. Every change of a legal units can have consequences for the enterprises. The update of statistical enterprises in SBR is done on a monthly basis.
Two reporting tools are applied for data collection:
eQuest (80.9% of legal units)
Balance of accounts transmitted by WebService (implemented with reference year 2021) (19,1 of legal units)
Postal questionnaires are allowed in exceptional cases (only if the technical preconditions in the reporting unit are not given).
In electronic questionnaires some built-in consistency checks are applied.
5.7% of the population have received a request to report for reference year 2023. Respondents are contacted by postal letter or by mail. Access codes for the electronic reporting media are enclosed, if units didn’t have an individual access data from previous years. Non-response is monitored in the SBR. The following actions are taken to speed up or increase the response:
reminder for the closing date of the survey
one or two e-mail reminder after the closing date, if required combined with at least one postal reminder (certified mail with delivery confirmation)
reminder by telephone (mainly for very important units)
finally, Statistics Austria gives notice of legal units refusing to respond to administrative authorities.
Administrative data:
Administrative units are linked to reporting units in the SBR; administrative data relevant for SBS is extracted from SBR and/or different databases and depicted in the data editing tool and provided as data files for model-based estimation.
18.4. Data validation
First checks are carried out in the electronic reporting tools, but most of the checks on micro level are done in the course of data editing: mainly intra-dataset checks but also intra domain or intra source checks with information from administrative sources or other statistical domains.
Data editing is done by using a special IT tool, where predefined plausibility checks are carried out resulting in real errors, possible errors and information points. First of all, data are checked on the level of legal units. If all legal units of a statistical enterprise are checked, data consolidation is carried out for large units on manual basis; data for smaller enterprises are consolidated automatically. All in all, there are more than 100 checks concerning the structure of units, comparisons with administrative sources and the SBR, checks within a category of variables and checks between categories of variables, but also checks between statistical units. Real errors have to be corrected, possible errors and information points have to be checked by our data editing staff and possibly revised or confirmed as correct.
For statistical enterprises, which are consolidated in the same tool, more or less the same validation rules are applied. Consolidation is considered in the validation rules and some special rules are applied. Validation rules of Eurostat are applied, if possible, on the level of statistical enterprises.
The next stage of data validation is done on macro level (aggregated data, ratios, comparison with results of other statistics, etc.). An important step in the course of data validation is checking the impact and quality of model-based estimation on overall results by economic activity. Data visualisation is applied as well. If there are some questionable developments of macro level, the micro level is checked again, first for legal units and then for enterprises.
The final checks are done prior to the data transmission to Eurostat, but many of the data checks are already implemented in the data editing tool.
18.5. Data compilation
Legal units:
The data compilation process distinguishes between primary survey and model-based estimation. After data editing process of survey data is finished, model-based estimation is carried out, for which survey data is beside SBR and administrative sources an important source of information for calculating model parameters. For reasons of manual consolidation model-based estimation for legal units belonging to large enterprises to be consolidated is carried out earlier. Unit non-response is imputed in the course of model-based estimation. In addition to model-based estimation the following methods have been used for unit and item non-response in the course of the data editing process:
Information from administrative sources
Financial statements
Imputation based on information from the previous year
Use of other statistical information (e.g. STS)
Mean imputation
Automatic corrections in the data editing tool (predefined corrections of selected variables under certain conditions to save time and resources for data editing)
Grossing up is not applied. Instead of grossing up a "mass imputation" for all enterprises below thresholds values and unit non-response is done by model-based estimation. Model-based estimation is carried out at micro level. SBS results are compiled by aggregating the survey data and estimated data to overall results by economic activity, size classes and regions.
Enterprises:
Data for enterprises of the about 80 largest enterprise groups were consolidated manually with the use of consolidated accounts, financial statements of legal units, SBS reports of legal units and reported data about internal flows between national legal units.
Internal flows for non-additive variables were identified with the help of data sources mentioned above and consolidated, where necessary; additive variables were summed up. Flows between legal units of the same enterprise were consolidated, flows between legal units of different enterprises in the same group were considered as market values.
Data for statistical enterprises of smaller groups were consolidated by using automatic consolidation methods. They were implemented with the reference year 2022.
18.6. Adjustment
If the accounting period of an enterprise differs from the calendar year the units are asked to indicate the exact reference period provided. In general this accounting period should be reported, in which more months are in the reference period.
The following information needs to be regarded for interpretation of results of SBS 2023:
The reference year 2023 was characterised by smoother increases to the reference year 2022 compared to the previous year (2021 to 2022). Nevertheless, prices had an impact for many economic activities as well. Especially Section D is affected by changes in prices – so in 2023 there was a considerable higher decrease in purchases than in turnover – so there was an increase of value added by 24%. In sections I and M the rise of value added was caused by the same development.
The decrease in wholesale trade was the result of the rising prices in 2022 as well.
In Austria, tax authorities have introduced a simplified system for tax declarations for very small businesses – in SBR and SBS units, which have changed this option were considered with the reference year 2023 for the first time. This is the reason for the increase of enterprises in sections J, M, N, P, Q, R und S of NACE. Values are not affected by this measure.
Due to considerable increases in wages and salaries gross operating surplus decreased in some economic activities (mainly in sections C and G).
In Austria, an annual check of units regarding market – non-market is carried out. If units show a tendency to non-market for some time, the attribution of the units in question is changed. For reference year 2023 section H is affected by such a measure.
Turnover in K66 was influenced by the change from a gross to a net depiction of turnover.
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).
25 August 2025
SBS constitutes an important and integrated part of the new European Business Statistics Regulation (EU) 2019/2152
Statistical units are based on the Council Regulation (EEC) No 696/93 and the operational rules based on this regulation, too.
The statistical population includes all active market enterprises of Sections B-N, P-R and divisions S95, S96 of NACE Rev. 2 in the Statistical Business Register (SBR). Active means that the enterprise has either turnover and/or employment at any time during the reference period. No cut-off threshold is applied.
Reference area is Austria.
Registered branches of foreign enterprises are included in SBS if an activity relevant for value added in Austria is carried out or employees are reported to the Austrian Social Security Authorities.
According to definitions, activities abroad of Austrian enterprises are included if the activity abroad is carried out by a non-autonomous local unit of the enterprise. Autonomous branches of Austrian enterprises in foreign countries should be excluded, the actual exclusion depends on data availability.
With the support of the LCU and National Accounts the largest branches (in Austria or abroad) are discussed in order to be depicted correctly in the different domains of business statistics. In some cases, the delimitation is very difficult and a certain bias cannot be eliminated for the full target population.
Year 2023
Data refer to calendar or fiscal year. If a fiscal year differs from the calendar year, the fiscal year that has more months in the calendar year in question is used; special adjustments are made for exceptional (large) units. For 2023, 8,894 legal units had a different fiscal year, counting for about 16% of persons employed.
The most significant errors are:
measurement errors (distinction market-non market, wrong NACE code, fiscal year differs from the calendar year)
modelling errors (model based estimation, delineation of statistical enterprises and consolidation)
Non-response and data processing errors can occur, but these errors are not significant. Coverage errors were significant till the reference year 2020. With the inclusion of very small enterprises and alignment of coverage to BD scope coverage errors are negligible as well.
Preliminary results are not biased; the period from the first release of results to the final release is 8 months.
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.
Legal units:
The data compilation process distinguishes between primary survey and model-based estimation. After data editing process of survey data is finished, model-based estimation is carried out, for which survey data is beside SBR and administrative sources an important source of information for calculating model parameters. For reasons of manual consolidation model-based estimation for legal units belonging to large enterprises to be consolidated is carried out earlier. Unit non-response is imputed in the course of model-based estimation. In addition to model-based estimation the following methods have been used for unit and item non-response in the course of the data editing process:
Information from administrative sources
Financial statements
Imputation based on information from the previous year
Use of other statistical information (e.g. STS)
Mean imputation
Automatic corrections in the data editing tool (predefined corrections of selected variables under certain conditions to save time and resources for data editing)
Grossing up is not applied. Instead of grossing up a "mass imputation" for all enterprises below thresholds values and unit non-response is done by model-based estimation. Model-based estimation is carried out at micro level. SBS results are compiled by aggregating the survey data and estimated data to overall results by economic activity, size classes and regions.
Enterprises:
Data for enterprises of the about 80 largest enterprise groups were consolidated manually with the use of consolidated accounts, financial statements of legal units, SBS reports of legal units and reported data about internal flows between national legal units.
Internal flows for non-additive variables were identified with the help of data sources mentioned above and consolidated, where necessary; additive variables were summed up. Flows between legal units of the same enterprise were consolidated, flows between legal units of different enterprises in the same group were considered as market values.
Data for statistical enterprises of smaller groups were consolidated by using automatic consolidation methods. They were implemented with the reference year 2022.
Type of Source:
Data source is a statistical survey for legal units combined with administrative sources.
Cut off sample: in the national regulation activity-based survey thresholds and the flexible application of the thresholds are written down. In general, the thresholds can be adjusted every year, provided that number of sampled units are not more than 38,000 and quality criteria for coverage are met. Legal units exceeding the thresholds are exhaustive in the sample. All units below the thresholds are calculated with information from SBR, administrative sources and model-based estimation on micro level, so for each unit a complete micro dataset is available.
Stratification criteria: not applicable
Selection schemes: activity based; employment and/or turnover thresholds based on structure of different economic activities; a certain coverage is necessary for application of model-based estimation
Threshold values for the reference year 2023 see annex "Thresholds for the Structural Business Statistics Survey 2023 to delimit the survey mass"
The effective sample size (sampled units without unit non-response) for the reference year 2023 was 36,397 legal units (5.7% of the population), covering 62.8% of persons employed and 85.9% of total turnover.
Administrative sources:
Social security data: Number of self-employed persons and employees; basic data source and for plausibility checks, model-based estimation
Wage tax data: Wages and salaries, part time employment, social security contributions; plausibility checks, imputation, model-based estimation
Turnover and Income tax data: Turnover and purchases of goods and services, selected items of stocks; plausibility checks, imputation, model-based estimation
Financial Statements
Data from Supervisory authorities for insurance services and credit institutions
STS for industries and construction and data from imports of goods and services are used; basic data source and for plausibility checks, model-based estimation
Administrative data are available as micro data for the entire population covered by the data sources; for tax data: 1:1; 1:n in exceptional cases; n:1 in the case of group taxation of enterprise groups; for social security data 1:1, 1:n for some cases
Frequency of updating: the frequency to which the used administrative data sources are updated is satisfactory. Administrative data are subject to several revisions with increasing degree of completeness.
Frame:
The frame from which the sample is drawn is the SBR of Statistics Austria for legal units and enterprises. The frequency of updating the SBR is as follows:
Register of companies: daily
Tax Register: monthly or quarterly (income tax)
Social security register: weekly
Membership file of the Austrian Federal Economic Chamber: weekly
Information obtained from economic surveys and classification notification: ad hoc
The sample is updated every year, with the possibility of slight updates in the course of the survey.
The classification of the principal activity is based on the value added. If this information is not available, turnover or employment is used as approximation. Secondary activities are determined in the same way as principal activities. Ancillary activities are classified with the principal activity. A top-down approach is applied for classification of units.
The principal activity can change whenever new information is available but in general the principal activity is held stable in the SBR within a calendar year with the exception of the reorganization of an observation unit. Within a calendar year it should not be changed for reasons of continuity and coherence with short-term statistics. Units which are part of the cut-off survey for SBS are directly asked for their principal activity. The SBR is updated with activities changed in the course of surveys. Additionally, legal units are informed about their activity classification in the course of classification notifications.
Delineation of statistical enterprises and definition of ancillary activities is aligned to proposals of Eurostat. Some national particularities are considered. Enterprises of about 80 EG (Enterprises Groups) are profiled and consolidated manually – for the other EG automatic methods are applied. Every change of a legal units can have consequences for the enterprises. The update of statistical enterprises in SBR is done on a monthly basis.
Data collection deadline: 9 months after the end of the reference year National dissemination of preliminary results: 10 months after the end of the reference year Post-collection phase: 15 months after the end of the reference year
Manual consolidation: 15 months after the end of the reference year
Model based estimation: 16 months after the end of the reference year
Automatic consolidation: 17 months after the end of the reference year National dissemination: 18 months after the end of the reference year
The timetable for producing and disseminating SBS data is aligned to data availability in legal units and administrative sources, which is laid down in EU and national regulations.
For the entire national territory the same concepts and definitions are applied. In general definitions of EBS Regulation are applied but different accounting schemes (IFRS not obligatory in Austria) may have an impact on comparability.
The length of comparable time series and the reasons for differences are provided in sub-concept 15.2.1.
