About the statistics
Name and topic
Name: Employment, register-based
Topic: Labour market and earnings
Division for Labour Market and Wage Statistics
Definitions of the main concepts and variables
Employed persons are persons between 15-74 years of age who performed work for pay or profit for at least one hour in the reference week, or who were temporarily absent from work because of illness, holidays etc. Conscripts are classified as employed persons. Persons engaged by government measures to promote employment are also included if they receive wages. This follows recommendations from the International Labour Organization, ILO.
The age is given in complete years at the end of the year.
If employed persons have more than one employment in the reference week, one is determined as the most important. Information on individuals’ work- and business related variables comes from the main employment.
The features residence, gender and age are collected from the National Registry.
Information about residence is collected at the end of the reference week. For the period 2000-2014 this implies the 3rd week of November. From 2015 onwards residence is collected at the end of the given statistical month.
Information about age is collected at the end of the year for the period 2000-2005. In 2006 the minimum age to be counted as employed was lowered from 16 to 15 years, in accordance with international recommendations. Also, the definition of age was changed from age at the end of the year to age at the end of the reference week. From 2015 onwards age is collected at the 16th of November.
From 2015 onwards
Working hours are calculated using the number of hours per week that constitute a full-time position, the percentage of the position and the number of paid hours.
Contractual percentage of full-time equivalent
Contractual percentage of full-time equivalent is what you have agreed to work according to your contract of employment. The employer shall not consider additional work, overtime or different types of absence from work or if the hours have been paid or not. The information on contractual percentage of full-time equivalent is based on what is reported to the a-ordningen.
For persons that are paid by the hour without contractual working hours per week, e.g. on-call temporary workers, the employer can report 0 as contractual percentage of full-time equivalent. Statistics Norway will then calculate contractual percentage of full-time equivalent from what is reported as paid hours and number of hours that corresponds to a 100 per cent position in the moth of reporting.
Contractual working hours per week
By combining information regarding contractual percentage of full-time equivalent and number of hours per week in a full position, contractual working hours per week is calculated for each employment (job) and wage earner (person).
The number of hours in a full position is the number of working hours that makes a full position in a similar employment. Unpaid lunchbreaks are withdrawn, but it is not adjusted for potential additional work, overtime or different types of absence from work
Full time is when the contractual percentage of full-time equivalent equals 100 or more. Part-time is when the contractual percentage of full-time equivalent is less than 100.
In a-ordningen occupational code is also delivered for employees in the central government. Previously, the position percentage for this group had to be calculated using statistical methods based on other sources of data.
Working hours is the scheduled working hours per week divided into three intervals: 1-19 hours per week, 20-29 hours, and 30 hours per week and above. For 90 per cent of the employed persons, information is obtained from the Ee-register. For those classified as employed persons on the basis of information from The Register of End of the Year Certificates (as one of the sources), the size of the annual income is used to calculate weekly hours of work. The working hours for self-employed persons are the average working hours per week, calculated on the basis of entrepreneurial income and information from the Labour Force Survey, LFS.
For employed persons in the Ee-register, the occupational code is supplied by the employer, or encoded from the occupational title. For other employed persons, the occupational distribution is estimated using statistical methods and other sources of data.
The information about the municipality of work and industry is obtained from the Central Register of Business Establishments and Enterprises (CRE) and uses information on the enterprise where people work. For self-employed who are not related to either an enterprise or an establishment, information on residency, education and the independent enterprise of family members are used to decide the necessary variables. The work place for seafarers and employees in the National defence is set equal to the municipality of residence.
Information about the sector is based on the institutional sector classification and is retrieved from the CRE.
The industrial classification is in accordance with the revised Standard Industrial Classification (NOS D 383), which is based on the EU-standard of NACE Rev. 2.
The occupational classification is in accordance with Standard Classification of Occupations (STYRK 2008), which is based on ISCO 08 (COM).
The educational classification is in accordance with the Norwegian Standard Classification of Education Revised 2000 (NOS C 617).
Breakdown by municipality is in accordance with the list of municipalities as of January 1st in the production year (which is the following year when the data was gathered).
The sector classification is in accordance with the Classification of Institutional Sector.
Municipality, County and the whole country.
Frequency and timeliness
Annual per 4th quarter.
Source data are permanently stored.
Background and purpose
The main purpose is to describe the employment on a detailed regional level.
The register based employment statistics is a continuation of the employee statistics, which was first published in 1983.
Users and applications
Public administration and local administrations in particular, employers' and employees' organisations, media, individual companies and researchers.
Equal treatment of users
Coherence with other statistics
From 2015 onwards
The register based employment statistics was until the end of 2014 built on various public registers, the most important was the The Register of Employers and Employees, (Ee-register) produced by The Norwegian Labour and Welfare Organization (NAV). From 2015 onwards a new joint reporting solution called a-ordningen gathers the reporting from the employers to the Ee-register and some reports to the Tax Administration and Statistics Norway. More information about a-ordningen under “Production”.
Before 2015 the total number of employed persons was determined by the Labour Force Survey (LFS), collected and distributed by employees and self-employed. From 2015 onwards only the number of self-employed is determined by the LFS. The number of employees is determined solely from the a-register, which excludes the sample uncertainty that follows with the LFS-figures. The register based figures from 2015 onwards will therefore differ from the LFS-figures, while they are identical at the national level in the period between 2000 and 2014.
4th quarter 2000 – 4th quarter 2014
The figures on employed persons divided into employees and self-employed persons at the national level will be identical to the LFS figures. Due to different sources of data, differences will occur when the figures are distributed by various characteristics such as gender, age, industry, municipality of work and working hours.
Before the 4th quarter 2000
Prior to the register based employment statistics (the first publication was in 2001 including numbers from 2000), an employee statistics by municipality of residence and work was published using only the Ee-register. These statistics covered approximately 80 per cent of all employed persons. (For more information on these statistics seeEmployee statistics by municipality of residence and Employee statistics by municipality of work)
Comparison with other statistics on employed persons
More detailed figures on employment in health and social services are also published based on the register based employment statistics, but the definition of the population is expanded to include people who are non-residents as well. Second jobs are included too. Statistics Norway also publishes figures for the employment among immigrants in addition to one special statistics which covers employed persons on short-term stay. These statistics also have a-ordningen as their main data source. For more information about these statistics, go to the individual statistics’ “About the statistics”.
Regarding the measurement of employed persons in the national accounts and other economic statistics, other international recommendations apply than those for the labour market statistics. Instead of using residents in Norway as the population, the national accounts cover all employment in domestic enterprises, i.e. enterprises that are expected to produce goods in Norway for at least 12 months. This issue was discussed in further detail in the last publication of the quarterly national accounts.
The Statistical Act $3-2 (Administrative Registers)
The population of the register based employment statistics is persons between 15 and 74 years old, with one or more working hours per week or who is temporary absent in the reference week. The exception is persons living in Norway but working abroad.
From 2000-2014 the reference week was the 3rd week of November. From 2015 onwards the reference week is the week containing the 16th of November, which usually is the 3rd week of November.
Data sources and sampling
From 2015 onwards
From 2015 onwards a new joint reporting solution called a-ordningen gathers the reporting from the employers to the Ee-register, produced by the Norwegian Labour and Welfare Administration, as well as some reports to the Tax Administration and Statistics Norway. A-ordningen is a coordinated digital collection of data on employment, income and tax deductions to the Tax Administration, the Norwegian Labour and Welfare Organization and Statistics Norway. This new joint collection provides Statistics Norway with information on wages and employed persons directly from a-meldingen, which is the electronic message containing all the information collected, rather than the multiple sources used until 2014. You can find more information about the new reporting at www.altinn.no/en/a-ordningen.
A-ordningen has replaced the Register of Employers and Employees and the Register of End of the Year Certificates (Register of Wage Sums). In addition to a-ordningen other registers provide additional information, the most important is the Register for Personal Tax Payers administered by the Directorate of Taxes, the Register of Conscripts from the Armed Forces Personnel and National Service Centre and The Central Coordinating Register for Legal Entities.
The registers mentioned above are still being used to assure the quality of the data, to consolidate the consistency between different data sources as well as choosing the right main job and to classify people as employed. A few additional registers are also being used for these purposes, where the ARENA-register produced by the Norwegian Labour and Welfare Organization is one example. It provides data on unemployed and people engaged in labour market projects.
4th quarter 2000 – 4th quarter 2014
Labour market data is based on several registers. The most important ones are the Ee-register, The Register of End of the Year Certificates (Register of Wage Sums), The Register for Personal Tax Payers, The Register of Unemployed, the register of conscripts and civil service employees collected from respectively the Armed Forces Personnel and National Service Centre and The Norwegian State Administration Database and The Central Co-ordinating Register for Legal Entities.
The Register of Employers and Employees is the main source for data on employed persons, while The Register of Wage Sums gives additional information about persons who do not have to report to Ee-register. Both registers are job-registers. The tax form register is the main source of data on self-employed persons.
The Register of Unemployed holds data on unemployed persons and persons engaged in labour market measures. The Business Register holds information on the workplaces. Several registers give additional information: register of conscripts, registers of employees in central and local government, register of sick leave etc. The Business Register and the central Register of Business Establishments and Enterprises (CRE) provide information about the companies where the employed people work. To assure the quality of the data, additional data from other sources are used: Statistics on unemployment and labour market measures are based on data from the administration system of NAV, ARENA, registers on employed persons in central government and municipals, wage statistics for employed persons in the private sector, the register of sick leave etc.
The definition of employment is thus based on a variety of sources and Statistics Norway has established a system to jointly utilise these. The systems include modules for consolidating the consistency between various data sources, for selection of the most important job and for classifying people as employed.
Collection of data, editing and estimations
A-ordningen: Statistics Norway receive monthly data from the Tax Administration Shared Services Agency (Etatens fellesforvaltning, EFF), a management entity of the Tax Administration, which administers a-ordningen on behalf of the three owner agencies; the Tax Administration, the Norwegian Labour and Welfare Organization and Statistics Norway.
The Register of Employers and Employees: Statistics Norway received weekly files on changes that occurred which NAV reported to Ee-register, in addition to an annual total report. From 2015 onwards, the Ee-register is replaced by a-ordningen.
The Register of End of the Year Certificates: A provisional extraction from the register was made in March, which formed the basis of the figures in the statistics. From 2015 onwards, the register is replaced by a-ordningen.
The Register for Personal Tax Payers: All information about the tax reporting and settlement is available electronically. Statistics Norway obtains an annual sample of the data from the Tax Directorate.
The register of civil workers and conscripts: Statistics Norway received quarterly files from Armed Forces Personnel and National Service Centre and The Norwegian State Administration Database at Dillingøy until the end of 2014. From 2015 onwards, these files are received monthly.
The Central Coordinating Register for Legal Entities: Variables like municipality of work and industry are collected from these registers.
For the two most important registers for the production from 2015 onwards, the control and revision is done in the following way:
Controls are carried out at several stages:
- Rules of business in the reception at The Tax Administration Shared Services Agency (EFF)
- In the production system for wages and employment at Statistics Norway
Rules of business (controls) in the reception at EFF:
A number of rules of business (controls) are being run to detect errors/omissions in the submitted information, after a-meldingen is received by EFF. Small to medium sized legal entities with a declaration obligation receives a feedback in a short amount of time (less than a minute). The larger legal entities have to wait a little longer for the feedback. The feedbacks from EFF include all uncovered discrepancies, where in the submitted information the discrepancies take place and what rule of business has been broken.
Rules of business and errors that can occur are documented at the homepages of a-ordningen: https://www.altinn.no/en/a-ordningen/
Controls in the production structure of Statistics Norway:
A number of checks and automatic actions are taking place during the production, the purpose is to ensure the quality of the data for statistical purposes. We distinguish between three types of controls:
a) Automatic checks and actions
b) Reports (monitoring)
c) Manual checks
In addition to this, controls and checks are being carried out throughout the year, regardless of publication.
Checks that aim to uncover defects and discrepancies that should be corrected are mostly aimed at the employments. Several controls have uncovered errors and missing values concerning the number of hours in a full-time position and the percentage position, combined with other features like whether a job is a main job or second job, the contractual working hours and the payment arrangements, such as:
- The number of hours worked are missing
- The number of hours that qualifies as full-time is probably reported wrongly
- The number of hours that qualifies as full-time deviates from the reported contractual working hours
- The number of hours that qualifies as full-time are extremely high or low
- The position percentage is missing
Many registered employments are removed (not considered active) during the production because wage of the employment relation has not been registered in the reference month. This could apply to seasonal workers who have not performed any work in the reference month (and therefore not been paid), and where a date of termination of employment (wrongly) has not been reported, or in cases where individuals have mistakenly been reported with an active employment (e.g. temporary staff who has not been working during the given period).
The Register for Personal Tax Payers
For the Register for Personal Tax Payers, a number of controls by machines are being run. Firstly the sum of partial records should match the main records from the Tax reporting. Further on, outliers are investigated to decide whether they are logical in relation to the rest of the Tax reporting. Internal consistency controls are also carried out.
Until the end of 2014, when the Ee-register and the Register of end of the Year Certificates (LTO-register) were the main data sources, control and revision were performed in the following way:
The Norwegian Labour and Welfare Organization conducts an annual control of the Ee-register. A sample of employers who reports manually receives lists containing all individuals they have registered as actively working. Errors are reported to the Ee-register. Statistics Norway checks that multi-enterprise establishments have separate numbers for each enterprise, and that the employed persons are linked to the proper enterprise. Occupational codes are also being controlled. If errors are discovered, the employer is notified via The Norwegian Labour and Welfare Organization to enable corrections to be made in the register. This is important if the information about industry and municipality of work is to be correct.
In addition, the automated controls and corrections performed at Statistics Norway are based on considerations about what is reasonable when it comes to total hours of work in the employment arrangement. For example, there must be registered earned income from the LTO-register for a person and the individual cannot be registered as unemployed in the ARENA-register.
The control of The Register of End of the Year Certificates is carried out in several steps. Employers do the first control before the information is sent to the Tax Administration. Then the Tax Administration executes controls both locally at the municipal treasurers’ and centrally at the Directorate of Taxes. The information is also controlled by Statistics Norway, where control and assurance of quality of the institutional sector and industry is the most comprehensive.
The main rule is that 1s and 2s have been replaced by 0 or 3 in order to meet privacy requirement.
Comparability over time and space
The register based employment statistics was first published in association with the Population and Housing Census 2001, which also included comparable figures for the 4th quarter of 2000.
A new data source
From 2015 onwards the register based employment statistics will be based on a new data source for employees. Until the end of 2014, the main data source was the Employees register, produced by the Norwegian Labour and Welfare Organization. In 2015 this reporting was coordinated with the reporting of earnings and personnel data to the Tax Administration and Statistics Norway. This common reporting system is called a-ordningen. A-ordningen provides a better data basis compared to the Ee-register, as it is more accurate on the individual level in addition to covering a greater number of employments. Furthermore, it provides more frequent statistics for wage earners and will supply highly detailed figures on the development in the labour market. These include numbers on the municipal level by age and industry.
Change in the time series
With the transition to the new data source, we will obtain more precise figures on the number of employed persons. In the new base, the number of employed is 60,000 lower than in the old data base, which used the number of employed in the country from the Labour Force Surveys. This is a modest difference of 2.4 per cent compared to the total number of employed, but as a result of the change in database the number of employed in the 4th quarter of 2014 compared to the 4th quarter of 2015 does not express an actual change. When looking at changes in the number of employed at the national level between 2014 and 2015, one should therefore use figures on the employed in Norway from the LFS. It is also possible to use the quarterly national accounts, which provides numbers for employment by industry. Those who work in Norway without being a part of the Norwegian population are also included in the quarterly national accounts.
Overall, Statistics Norway assesses the quality of the register based employment statistics to have improved from 2015 onwards. Because of some challenges in the data in the new register, we will not publish tables with a distribution on number of hours worked. Efforts are being made to improve the quality of reporting, while Statistics Norway tests new methods to adjust for deficiencies in this reporting. Tables including hours worked will thus be published at a later time.
Data source from the 4th quarter of 2000 – 4th quarter of 2014:
Revised figures ofr the 4th quarter of 2003 were published together with figures for the 4th quarter of 2004. This is partly due to improvements in the production method and to some extent due to improvements in data reported to the Ee-register. Both old and revised figures for 2003 are available in the StatBank.
Improvements in the production procedures affect employed persons in smaller jobs, where the information about the actual workplace has been insufficient. This was especially true for the central government in Oslo. The change in the figures from the 4th quarter of 2003 to 2004 should not contain errors because the new program is run on the figures from 2003.
The reporing to the Ee-register by the municipalities has previously had some deficiencies and measures were taken at the end of 2004 to improve the quality. The improvements mainly led to a more correct distribution of employees in the municipal activities. By the 4th quarter of 2005, most of the improvements had been included in the data source. The improvements were added to the data of the previous years back to the 4th quarter of 2003. In some municipalities the improvements may cause large changes in the numbers over time. Especially in industries where county municipality is immense, like health and social services, education and public administration.
In 2006 the minimum age to be counted as employed was lowered from 16 to 15 years, in accordance with international recommendations. Also, the definition of age was changed from age at the end of the year to age at the end of the reference week. This resulted in 10 000 more employed in the 4th quarter of 2005. Comparable figures according to both new and old definition of age are available in the StatBank for 2005.
Occupation (based on Standard Occupational Classification, STYRK 1988) was included in 2003, with the exception of municipal sector and the publicly owned health enterprises which were not included until 2007. In 2011 a new Standard Occupational Classification (STYRK 2008) was adopted.
In 2009 a new Standard Industrial Classification, NACE Rev 2, were adopted. The new classification replaced NACE Rev 1.1. Comparable figures for 2008 according to both new and old classifications are available in the StatBank.
Sources of error and uncertainty
From 2015 onwards
In the reporting of the percentage of the position, which is used to calculate contractual/usual working hours, deficiencies have occurred. To calculate the position percentage for some of those with deficiencies the number of paid hours are used as a base (mainly applies to employees who get paid by the hour). Efforts are being made to improve the quality of reporting, while Statistics Norway tests new methods to adjust for deficiencies in this reporting.
Self-employed are identified using information from the Register for Personal Tax Payers. Due to the long production process, information about business activities is from the previous year. Thus, a person may wrongly be defined as employed even if the activity ended the previous year.
Before 2015 the total number of employed persons was determined by the Labour Force Survey (LFS), collected and distributed by employees and self-employed. From 2015 onwards only the number of self-employed is determined by the LFS. The number of employees is determined solely from the a-register, which mitigates the sample uncertainty that follows with the LFS-figures. On the other hand, the level number of employees will be a bit low.
For individuals defined as employed based solely on information from the End of the Year Certificates Register (about 10 per cent of the employees), the job is not dated. Additional information on about half of these employees is collected from other registers to help date the jobs. Wages are used to determine if the person is employed for those where we do not have additional information. For the remaining persons, an uncertainty is attached to whether the person actually was employed in the reference period.
The Register of Personal Tax Payers is used to identify self-employed persons. Due to the long production process, information about business activities is from the previous year. Thus, a person may wrongly be defined as employed even if the activity ended the previous year.
Some individuals are defined as employed and employees based solely on information from the Register of End of the Year Certificates, which also attaches the job to an enterprise. A routine for identifying the correct establishment has been created for situations when the enterprise has several establishments. Whether the job is connected to the proper establishment is somewhat uncertain, thus if follows that the information on workplace and industry is uncertain to some extent.
The quality of the Ee-register is sometimes faulty. For larger enterprises with multiple establishments this may cause employees to be registered at the wrong establishment. This can have a noticeable effect on a regional level, such as municipality, both when the error occurs and when the errors are corrected.
The base registers do not provide reliable information about all employed people at the reference time. The national level on employed and self-employed from the LFS-survey is therefore used to determine the same numbers in the register based employment statistics. This implies that the uncertainty in the sampling selection also applies to the register numbers for employment at the national level.