Earnings

Name: Earnings
Topic: Labour market and earnings

Division for Labour Market and Wage Statistics

National, County and Municipality.

Annually with November as the reference month.

The statistics are published in February the following year.

Council Regulation (EC) No 530/1999 on statistics on the structure of earnings, as amended by Regulation (EC) No 1893/2006 Reporting shall take place within 18 months of the end of the census year.

Basic files for statistics on earnings are saved.

The purpose of the statistics is to provide an overview of the level of earnings and change for all jobs, regardless of business affiliation and working hours. The statistics were established in 1997, based on sample surveys in the private sector and register information in the public sector.

There are two breaks in the time series in the wage statistics, in 2008 and 2015.

1997-2008: Wage statistics were based on the old Standard Industrial Classification (SN2002) and are available as closed series in StatBank Norway.

2008-2015: As of 2009, a new Norwegian Standard for Industrial Classification was adopted (SN2007). The figures for 2008 were recalculated according to the new classification. For 2015, the figures were based on information from the a-ordningen, but to obtain comparability with 2014, the figures were prepared based on a sample of enterprises from the a-ordningen with the same sampling model utilized up until 2014. This time series is closed figures can be found in StatBank Norway.

Up to and including 2014, data were obtained electronically from a sample of enterprises in the private sector, and full counts from registers in the public sector. As of 2015, the sample survey in the private sector and the public registers were replaced by the a-ordningen.

2015-: As of 2015, the statistics are based on a-ordningen and new time series in the StatBank Norway. There is also a breach in the time series related to occupational information, the new series is based on a new occupational standard in accordance with the Standard for occupational classification-08 (STYRK-08). Employers report employees according to the previous occupational standard STYRK-98, after which Statistics Norway converts to the current standard.

For more information about the breach in the time series under «Production».

Major users are the Technical Reporting Committee on the Income Settlements, research and official studies institutes, employee and employer organisations, Eurostat, the media, trade and the business sector and private persons. The statistics are used in Statistics Norway’s Labour Accounts.

No external users have access to the statistics and analyses before they are published and accessible simultaneously for all users on ssb.no at 8 am. Prior to this, a minimum of three months' advance notice is given in the Statistics Release Calendar. This is one of Statistics Norway’s key principles for ensuring that all users are treated equally.

The annual wage statistics are closely linked to the quarterly statistics «Number of employments and earnings». Both are based on data from the a-ordningen.

Basic earnings in the annual and quarterly statistics on earnings are directly comparable, but the level and development in monthly earnings are not. The reason being that variable additional allowances and bonuses are calculated as an average over the year in the annual statistics on earnings, while the average in the quarterly statistics on earnings only calculated for the current quarter.

Register-based employment statistics also utilize a-ordningen as a source for statistics on the number of employed. These statistics also include self-employed but are limited to residents aged 15-74.

The national accounts prepare figures on accrued annual earnings from the a-ordningen. This concept is relatively closely linked to annual earnings in the annual statistics on earnings but has some differences. In contrast to the statistics on earnings, the national accounts also include foreign workers in foreign shipping. The population in the national accounts is all employees in the companies that are included in the production delimitation for Norway, i.e. all employees who contribute to Norwegian value creation.

«The Official Statistics and Statistics Norway Act § 10, cf. The employer's reporting of employment and income conditions Act, etc. (the a-opplysnings Act) § 3.»

EEA reference Council Regulation (EC) No 530/1999 concerning statistics on the structure of earnings, as amended by Regulation (EC) No 1893/2006

Commission regulations that also regulate the statistics:

• Commission Regulation (EC) No 1916/2000 implementing Regulation (EC) No 530/1999 concerning structural statistics on earnings, as amended by Regulations (EC) No 1738/2005, (EC) No 1022/2009, (EU) No 317/2013 and (EC) No 973/2007
• Commission Regulation 1726/1999 implementing Regulation (EC) No 530/1999 on statistics on the structure of earnings, as amended by Regulations (EC) No 1737/2005 and (EC) No 973/2007
• Commission Regulation 698/2006 on the implementation of Regulation (EC) No 530/1999 concerning structural statistics on earnings, as amended by Regulations (EC) No 1022/2009 and (EC) No 317/2013

Council Regulation (EC) No 530/1999 of 9 March 1999 concerning structural statistics on earnings and on labour costs (HTML).

Statistics Norway’s statistics on earnings are compiled in accordance with the European regulations for measuring earnings, in addition to covering national needs related to, among other things, wage negotiations.

The reporting unit in the wage statistics is enterprises. Each enterprise comprises one or more local units (sub-units) which are categorised by sector and Industry (for more information see Standard Classifications). The unit of analysis in the statistics is employments, mainly referred to as jobs, by local unit. An employee may have several jobs in different enterprises or local units, but jobs within the same local unit are aggregated together into one. The number of jobs will thus exceed the number of employees in the population and the statistics.

The population is furthermore reduced to encompass jobs where cash compensation has changed hands in the reference month, and then only cash compensation that fall within the framework described under definitions.

The statistics do not include remunerations in kind, insurance, non-deductible expense allowances or the like. Furthermore, the cash amount is limited to the amount paid for the job / position to which they belong. It follows from this that cash benefits such as severance pay, board fees, fees for other councils / committees or operating allowances are also not included in earnings. Overtime pay is included in the statistics as a separate term but is not included in the monthly earnings. Beyond this, there is no restriction related to characteristics of the employee (working hours, main position / assisted position, resident / non-resident, age, occupation, temporary or similar). Self-employed are not included.

The delimitation of which jobs are included in the statistics is as follows:

1. Jobs where a person is registered as employed during the reference week in November, which is the reference period, are included (considered employed).
2. There must have been a transaction of at least a fixed earnings or hourly wage in cash for the job in the reference month.

The reference week is the week that contains the 16th, which is generally the 3rd week in the reference month.

Practical challenges in the population delimitation are largely related to determining what kind of cash payment the employee has received for the job in the reference month, as well as assessing the earnings against information about the employment, i.e. whether the employee has been employed, performed and received payment.

As of 2015, the data basis for the statistics is based on information from the a-ordningen (for more information, see Skatteetaten). The unemployment insurance scheme is a coordinated reporting of earnings and employment information to the Tax Administration, the Norwegian Labour and Welfare Administration and Statistics Norway regulated by the Unemployment Insurance Act. Data are obtained through digital machine to machine solutions.

All employers who have paid cash benefits or benefits in kind must submit an a-melding monthly. The data thus include all employees in all enterprises in the industries covered by the earnings statistics.

Up to and including 2014, data were obtained electronically from a sample survey of enterprises in the private sector, and from registers in the public sector. Information was obtained on, among other things, the agreed earnings, bonus, Variable additional allowances and working hours for the individual employee in the company.

Data collection

The statistics are entirely based on information from different registers. Data from a-ordningen is obtained monthly from the Tax Administration, which administers the collection on behalf of the three owner agencies: The Tax Agency, Social and labour services (NAV) and Statistics Norway. Variables such as workplace location, industry and sector are obtained from the Register of Legal Entities and Statistics Norway's Business and Enterprise Register (VoF). Person-related identities are obtained from the Central Population Register (DSF).

Editing

Checks are carried out in several stages where to major stages are, in when data is reported and received in the tax administration and later when data are processed I Statistics Norway.

Business rules (controls) in the reception at the Tax Administration:

After the a-melding has been received by the Tax Administration, several business rules (checks) are run to uncover errors and deficiencies in the submitted information. Small and medium-sized information providers receive feedback quickly (less than a minute), while large information providers must wait a little longer. The feedback from the Tax Administration contains all discrepancies that have been discovered and are reported back with references to where reporting is incorrect. All the business rules and error reports are documented on the a-ordningen's web-pages (Skatteetaten).

Controls in Statistics Norway's production system:

After receiving data from the Tax Administration Statistics Norway applies several automated controls and measures are carried out, the purpose of which is to ensure the data quality for statistical purposes. We distinguish between three types of controls: a) Automatic controls and measures b) Reports (monitoring) c) Manual controls in addition to this, inspections will be carried out continuously throughout the year, regardless of publication.

Checks aimed at uncovering errors and omissions that should be corrected are largely aimed at information about the employment. Many inspections have been made that show errors and omissions in the number of full-time hours and the percentage of positions, combined with other characteristics such as type of employment and working hours, including some typical faults such as:

• Full-time hours are missing
• Number of hours full time probably incorrectly reported
• The number of full-time hours deviates from the reported working time
• Number of hours full time extremely high or low
• Position percentage is missing

In the production plan, many registered employments are removed (not considered active) because there is no registered earnings for the employment relationship in the reference month. This may apply to seasonal workers who have not performed work in the reference month (and therefore have not been paid), where an incorrect end date for termination of the employment is missing, and cases where persons have been incorrectly reported with active employment which also could be persons on temporary staff who have not performed work during the period.

Control and editing of the wage statistics takes place in several stages, where most of the operations are automated, both with regards to the control itself and the correction that follows. All key variables are checked. Furthermore, the earnings data is checked, both at micro and macro level.

The A-report is based on a principle where all transactions monetary or otherwise are reported the month they are executed. This implies that reported earnings in some cases will deviate from the agreed monthly earnings for an employee. For example, this can be done by post-payment of earnings settlement. In such cases, Statistics Norway will calculate the agreed monthly earnings by comparing information from several months in context.

Obvious errors and omissions in payroll reporting trigger imputation of values ​​using statistical methods. The extent of this is about 5 percent of the total number of employment relationships. The imputation utilizes information from periods before and after the reference period, if the corresponding employment relationship is comparable.

As of the release of November 2020, a new method for calculating working hours has been implemented, the method has been applied in all years of the statistics from November 2016. This new method adjusts the agreed working hours for jobs where information on agreed working hours are missing or are deemed insufficient information or faulty. In total, about 6 per cent of the jobs receive a change in the agreed working hours.

Not relevant

The Act on Official Statistics and Statistics Norway and Statistics Norway's rules for confidentiality are complied with. Confidentiality is ensured in the following way:

1. If there are fewer than three companies behind the figures are supressed
2. If one enterprise accounts for more than 90 per cent of all jobs or more than 90 per cent of the sum of all monthly earnings in total figures are supressed, relates to all tables distributed by industry, sector and region
3. If two enterprises account for more than 95 per cent of all jobs or more than 95 per cent of the sum of all monthly earnings figures are supressed, relates to all tables distributed by industry, sector and region
4. In addition: figures are supressed for groups encompassing less than100 jobs

The last point ensures both privacy, quality and helps to reduce the costs of producing the wage statistics by avoiding that a few extreme earnings, high or low, affect the results unnecessarily.

The reference period for the statistics were changed from September to November at the time of publication in February 2021. Comparable figures are available back to November 2016. Figures for 2015 are as of September and according to the old method for calculating working hours (see information on new methods for processing working hours further down) and thus not directly comparable with the figures from and including 2016 mostly due to the difference of two months in the reference period.

Figures for September 2015 are available, see tables under «closed tables» in StatBank Norway. For more information on comparability back in time, see ‘Purpose and history’.

Figures for earnings are partly comparable with earnings stated in the statistics ‘Number of working conditions and earnings’ that come quarterly (see «Relationship with other statistics»).

New method for processing working hours

As of the publication of figures from November 2020, a new method for calculating working hours that is incompletely reported has been applied. The change has been implemented to vintages from and including November 2016. The change affects the distribution of agreed working hours, but also the calculation of earnings for those who get new agreed working hours. The method does not contribute to major systematic changes between groups, and in most cases the number of jobs will subject to only minor changes compared with previously published statistics.

Municipal and regional reform

As of the first quarter of 2020, the statistics are affected by the municipal and regional reform which was implemented on 1 January 2020. With the reform, a few enterprises in the former municipalities / counties ceased and working conditions in these were transferred to existing or newly established enterprises in the new municipalities/regions.

Structural changes in the industries have an impact on the statistics. How many and which groups of employees go in and out of the labour market, how many have changed agreed working hours and how many change jobs between the industries will affect wage developments.

Calculations

Since the statistics are based solely on register data, the presentation of statistics is largely concentrated on the presentation of frequencies, averages, and measures of distribution such as median and quartiles among others. Otherwise see full-time equivalents and checks and error correction above. Nevertheless, there will be errors in reporting or non-response that affect the results (see also sources of error and uncertainty).

Sources of error and uncertainty:

The a-ordning is the main source of these statistics. The quality of the a-ordningen is good, but there may still be errors and omissions in the data.

Measurement and processing errors

Errors / omissions can occur in that those required to provide information misinterpret and report incorrectly (measurement errors). Furthermore, errors may occur through the processing of data in Statistics Norway (processing errors). An example of the latter can be Statistics Norway's assessment of which jobs that are active or inactive. Earnings are utilized as the main criterion but can refer to payments due for an earlyer period. For the group «Freelancers / Contractors etc.» (also includes board members, elected representatives, foster parents, support contacts, people with care pay, etc.) there is, for example, only a requirement to report jobs (working conditions) in the a-ordningen when remuneration takes place therefore it can prove difficult to identify when the work is done. Therefor some jobs in this group are not included in the statistics. Only freelancers / contractors with fixed and / or hourly wage or salary are included in the earnings statistics.

There has been a gradual improvement in reporting through a-ordningen from 2015 and onwards but was regarded as good already in 2015 and an improvement over all compared with older sources.

Registry errors can occur if any errors or omissions in the register that are connected to the statistical unit have errors or omissions in identification, classification or coding. For example, a link to the Business and Enterprise Register (VoF), which has information on, among other things, industry and the sector, can lead to companies being consigned to the wrong category or assigned attributes that are incorrect. Errors or biases in registration in the National Education Database (NUDB) will also lead to errors in the statistics.

Non-response

Two basic types of non-response should be addressed, unit non-response which can refer to an employer not reporting at all or the full non-response of certain jobs and their connected transactions. The other basic form is called partial non-response and refers to missing information about some variables related to the aforementioned units. Some elements of information where partial non-response occurs may be calculated or imputed by way of utilizing information from other months or together with information from the reference period in question. Observations can also be rejected because of missing information and / or incorrect values ​​for variables. In the event of obvious errors in the reporting, the earnings can be imputed using statistical methods. The same applies to working hours, which are central to calculating monthly earnings and monthly earnings per full-time equivalent.

Not relevant