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.
Employments: Number of employees and self-employed persons - based on individual PAYE tax returns collected by the Swedish Tax Agency (swedish: arbetsgivardeklarationer på individnivå). Main purpose is to show changes in the level of employment in Sweden.
3.2. Classification system
NACE Rev. 2.
3.3. Coverage - sector
The STS for industry covers Nace Rev. 2 Section B to E, in construction Nace Rev.2 Section F, in retail trade Nace Rev. 2 Section G (excluded G473) and in other services NACE Rev.2 Section H to J and L, M (excluded M701, M72 and M75) and N. In all annexes values are for enterprises with 1 or more employees.
3.4. Statistical concepts and definitions
The definition of employment includes all persons employed by the public and private sectors. Included are persons working full time and part time, persons on leave (sickness, abroad, maternity, training, military service and so on), persons on strike and employed on short-term contracts. The number of employees and self-employed persons is approximated by the number of employees. Persons on temporary leave are counted for if they have recieved any payment from the employer. Changes in personel are calculated from the PAYE tax returns using information on wether a person has recieved payment from a unique employer or not.
3.5. Statistical unit
Legal unit is used as the reporting unit in both private and public sector. The observation unit used is Kind-of-activity unit (KAU).
3.6. Statistical population
All registered legal units with at least one employee who pay salary to their employees are included in the statistical population. Buisness units have to contribute to Swedish production. The register covers legal units. A distribution model is used to distribute number of employments to different KAU.
3.7. Reference area
No regions of Sweden are excluded. Activities outside the geografical coverage are not included in the data.
3.8. Coverage - Time
Time series is available from 2000Q1. Due to a change of the data source there is a time series break 2020Q1.
3.9. Base period
Base year is 2021. Indices are reported to Eurostat.
Number of employees.
The data are transmitted to Eurostat in indices and are published on national level in absolute values.
The register includes monthly data. However the indices reported to Eurostat refer to quarterly numbers based on an average of the three months in the reference quarter.
6.1. Institutional Mandate - legal acts and other agreements
Since of the 1st of January 2019 all employers are obligated to report individual PAYE tax returns to the Swedish Tax Agency. These data is later transmitted to Statistics Sweden. Statistics on employment is classified as official statistics in Sweden according to the Official Statistics Act of 2001 (SFS 2001:99) and to the EBS regulation.
6.2. Institutional Mandate - data sharing
Data are not sent to other international organisations.
7.1. Confidentiality - policy
Policies are available to eliminate the risk of disclosure of private information.
7.2. Confidentiality - data treatment
No single unit can be identified.
8.1. Release calendar
Monthly data is released every quarter. Dates for the releses are determined one year in advance.
8.2. Release calendar access
A release calendar with information on release dates of statistics on national level is available on Statistics Sweden’s web site Publishing calendar (scb.se) .
8.3. Release policy - user access
The monthly data is released every quarter simultaneously on the website and in Swedish statistical database. Employment data are transmitted to Eurostat as soon as they are released in Sweden (approximately 2 months after the end of the reference period).
The monthly statistics are published quarterly on Statistics Sweden's website (see Employments) and within a week the data is transmitted to Eurostat.
10.1. Dissemination format - News release
The data are released approximately 2 months after the end of the reference quarter in the Statistical news.
10.2. Dissemination format - Publications
Not applicable.
10.3. Dissemination format - online database
The monthly data are released approximately 2 months after the end of the third month in the reference quarter. Data are published in the Swedish statistical database (SSD) AM0211. The data are published at 2-digit NACE level, regional data on the NUTS 3 level, according to the size classes; all on a monthly basis.
10.4. Dissemination format - microdata access
Micro-data access is granted to researchers. The micro-data is always deidentified.
10.5. Dissemination format - other
Data transmission to Eurostat every quarter.
10.6. Documentation on methodology
Updates of documentation on the website once a year.
10.7. Quality management - documentation
The documentation updates on the website once a year. The updated version can be found on Statistics Sweden's website (see Employments).
11.1. Quality assurance
Quality declarations are published annually together with data for the first month of the reference year.
11.2. Quality management - assessment
Validation of data is conducted on regulary basis. Deviation and extreme values are brought to attention when compared to previous values and data sources.
12.1. Relevance - User Needs
The primary users:
Eurostat.
National Account Departement within Statistics Sweden includes the data in quartarly national account processing.
Ministry of finance, Central bank of Sweden, Swedish Public Employment Service and Swedish Association of Local Authorities and Regions use the data as a backdrop to decision making.
The Swedish National Institute of Economic Research uses the data for economic evaluations and forecasts.
Other users:
Enterprises, students and researchers request data for analysis.
12.2. Relevance - User Satisfaction
There isno user satisfaction report available.
12.3. Completeness
The STS and SCB requirements are fulfilled.
13.1. Accuracy - overall
The accuracy is good since the data cover the whole population of employers.
13.2. Sampling error
Not applicable.
13.3. Non-sampling error
Errors that could occur are related to accessing the administrative data. However, possibility that it happens is low.
14.1. Timeliness
Nationally, the publications of monthly data are released approximately 2 months after the end of the third month. Dates for the quarterly releases of the monthly data are determined one year in advance. Data is not revised. Shortly after the data is published nationally, it is transmitted to Eurostat.
14.2. Punctuality
The statistics always published on schedule. All deadlines for publishing are met.
15.1. Comparability - geographical
Not applicable.
15.2. Comparability - over time
There has been changes to NACE classifications over time. These have affected the comparability to some extent, over the period. The time series are available from 2000. In addition, in 2020Q1 there is a time series break caused by change in the data source.
15.3. Coherence - cross domain
Continual comparisons are made with the data from the Labour Force Surveys and Population by Labour market status. For more information about these surveys, please read more on Statistics Sweden's website.
15.4. Coherence - internal
Since the statistics is registerbased the summarized sub-aggregates are consistent with the totals. Some sub-aggregates are however market due to discosure.
Since the data is collected by the Swedish Tax Agency the cost associated with the collection of data and burden on respondents can not be calculated by Statistics Sweden. However, since Statistics Sweden can use this source in production the cost and burden are lower in comparison to parallell collection of data by Statistics Sweden.
17.1. Data revision - policy
Revisions are made for the STS every fifth year due to rebasing. Nonscheduled revisions may occur due to errors if changes in published data are considered large.
17.2. Data revision - practice
Users are informed in advance of methodological and others changes which may lead to revisions. Big revisions are announced when data is released. However, big revisions are unusuall. Smaller revisions are not announced.
18.1. Source data
Administrative data source - Employers declarations on individual level (PAYE) collected by the Swedish Tax Agency.
18.2. Frequency of data collection
Monthly.
18.3. Data collection
The monthly administrative data is transmitted to Statistics Sweden from Swedish Tax Agency.
18.4. Data validation
Comparison is done with previous years data, previous months data and, other data sources. The data is aldo validated before it is sent to Eurostat. The numbers are compared with the published data in Sweden. It is also compared with the data sent to Eurostat in the previous quarter and even compared with data from the same quarter previous year.
18.5. Data compilation
Less than one percent of the objects are imputed. Non response are on a very low level.
18.6. Adjustment
None at present.
No further comments.
Employments: Number of employees and self-employed persons - based on individual PAYE tax returns collected by the Swedish Tax Agency (swedish: arbetsgivardeklarationer på individnivå). Main purpose is to show changes in the level of employment in Sweden.
15 June 2025
The definition of employment includes all persons employed by the public and private sectors. Included are persons working full time and part time, persons on leave (sickness, abroad, maternity, training, military service and so on), persons on strike and employed on short-term contracts. The number of employees and self-employed persons is approximated by the number of employees. Persons on temporary leave are counted for if they have recieved any payment from the employer. Changes in personel are calculated from the PAYE tax returns using information on wether a person has recieved payment from a unique employer or not.
Legal unit is used as the reporting unit in both private and public sector. The observation unit used is Kind-of-activity unit (KAU).
All registered legal units with at least one employee who pay salary to their employees are included in the statistical population. Buisness units have to contribute to Swedish production. The register covers legal units. A distribution model is used to distribute number of employments to different KAU.
No regions of Sweden are excluded. Activities outside the geografical coverage are not included in the data.
The register includes monthly data. However the indices reported to Eurostat refer to quarterly numbers based on an average of the three months in the reference quarter.
The accuracy is good since the data cover the whole population of employers.
Number of employees.
The data are transmitted to Eurostat in indices and are published on national level in absolute values.
Less than one percent of the objects are imputed. Non response are on a very low level.
Administrative data source - Employers declarations on individual level (PAYE) collected by the Swedish Tax Agency.
The monthly statistics are published quarterly on Statistics Sweden's website (see Employments) and within a week the data is transmitted to Eurostat.
Nationally, the publications of monthly data are released approximately 2 months after the end of the third month. Dates for the quarterly releases of the monthly data are determined one year in advance. Data is not revised. Shortly after the data is published nationally, it is transmitted to Eurostat.
Not applicable.
There has been changes to NACE classifications over time. These have affected the comparability to some extent, over the period. The time series are available from 2000. In addition, in 2020Q1 there is a time series break caused by change in the data source.
