921_om_not-searchable
/en/arbeid-og-lonn/statistikker/arbkost/arkiv
921_om
statistikk
2010-02-17T10:00:00.000Z
Labour market and earnings
en
false

Labour costs2008

Content

About the statistics

Definitions

Name and topic

Name: Labour costs
Topic: Labour market and earnings

Next release

Responsible division

Division for Structural Business Statistics

Definitions of the main concepts and variables

Total labour costs is the employers' total cost of having employees. This includes both direct and indirect costs deducted subsidies from the Government.

Direct cost is salaries and wages and is defined as all cash compensation to employees. Payments for days not worked is included.

Payment for days not worked is a subset of direct costs and includes wages during leaves, extra days off.

Indirect costs includes payments in kind, health and safety measures, social costs, employers' social contribution, other taxes and training costs.

Payments in kind represent compensation, services and benefits that are not in cash.

Other costs includes Safety and health costs and Recruitment costs

Training costs is the employers' costs for employees' participation in external courses, educational spending, apprentices  and other costs in connection with training employees.

Social costs employers costs for mandatory and supplementary pension plans and insurances. Employers social contribution is not included.

Employers' social contribution is the mandatory cost of participation in the national social security scheme. It is paid as a percentage of wages, vacation pay and contributions to pension plans.

The Number of full-time equivalent employees is calculated based on the number of full-time and part-time employees and the number of hours each employee worked during the year. The statistic includes all employees regardless of tasks.

Paid hours consist of all hours the employer paid for, regardless wheter the hours where worked or not. It includes overtime and hollidays. Hours worked, but not paid is not included.

Hours acctually worked is calculated as paid hours deducted all paid leave such as hollidays, sickness, public hollidays and other paid leave.    

Standard classifications

The industries are classified according to the official standard industrial classification SN 2007, which is the Norwegian version of the European NACE Rev. 2.

For 2008 the statistics are published according to both SN 2002 and SN 2007. For earlier editions, SN 2002 is used.

Administrative information

Regional level

National level

Frequency and timeliness

Every fourth year, about 18 months after the counting year

International reporting

Eurostat

Microdata

Files with labour costs data and estimation programs are stored.

Background

Background and purpose

The purpose of the statistics is to provide an overview of the business' total costs of having an employee. The statistics was established in 1998 with statistics for the year 1996.

Users and applications

Major users are Eurostat, the Technical Reporting Committee on the Income Settlement, business and industry.

Equal treatment of users

Not relevant

Coherence with other statistics

The statistics are to a certain degree comparable with the development in National accounts and structural business statistics. 

Legal authority

Statistics Act Sections 2-1, 2-2 and 2-3.

EEA reference

Council Regulation (EC) no. 530/1999 of 9 March 1999 Concerning Structural Statistics for Wages and Labour Costs and Commissions Regulation (EC) nr. 1737/2005 of 21. October 2005 and Commissions Regulation (EC) nr. 698/2006 of 5. May 2006

Production

Population

The population covers all enterprises in Statistics Norway's Central Register of Establishments and Enterprises in major industries except for the Public administration and Defence and Agriculture, forestry and fishing.

 

Until 2012 the statistics was based on a sample survey covering 3500 enterprises with 10 or more employees. For the years 1996, 2000 and 2004 the population only covered the Private sector.

 

From 2016 the statistics is based on data from a-meldingen. All employers who have paid wages, cash benefits or payments in kind must submit a a-meldingen report each month. The data includes enterprises with 10 or more employees.

Data sources and sampling

Until 2012 the information about wage and salary cost were gathered from the End of the Year Certificate Register and Employers' social contributions from the tax accounts of the Directorate of taxes. Information about other costs was gathered from an electronic questionnaire sent out to all enterprises included in the sample. The questionnaire included questions about salaries in kind, safety and health, training costs, subsidies from the government and taxes for each enterprise. In addition, the questionaire contained questions regarding number of full time and part time employees and working hours.

 

The sample consisted on enterprises randomly drawn from the population. Enterprises with less than 10 employees were excluded from the sample. The sample included both a representative sample and a complete count. The complete count included all enterprises with a sufficient number of employees while the sample was stratified, random sample of small and medium-sized enterprises.

  

The purpose of the sampling was to find a representative sample for the survey and to avoid burdening all enterprises with questionnaires. An effort had been made to ensure that small enterprises were also least burdened by questionnaires. The sample included about 3 500 enterprises

 

From 2016 the statistics is based on data from a-meldingen, Trading Statements (NO), the Register of Annual Company Reports in Brønnøysund and other sources. The main source is a-melding, which contains information about wages, cash benefits, payment in kind, employers' national insurance contributions, full-time equivalent and hours paid and worked. The Trading Statements and the Register of Annual Company Reports include information about employers’ costs for mandatory and supplementary pension plans. The statistics also use data from LFS, CVTS and apprenticeship scheme.

Collection of data, editing and estimations

Until 2012 information was gathered from an electronic questionnaire sent out to all enterprises included in the sample.

 

The collected data should represent the average level of labour costs in the industries included in the survey and the data was therefore weighted. The population was stratified by industry and size as measured by number of employees. The largest enterprises in each industry were all included. For enterprises with fewer employees a sample was drawn, ranging from 5 to 30 per cent of the population.

 

The weighting was based on the inverse probability of inclusion in the sample such that the weighted number of employees in the sample was equal to the number of employees in the population. The weighting was based on enterprises with sufficient data quality such that any biases introduce by variably data quality was adjusted for.

 

Average costs were calculated by divided the weighted total costs by the weighted number of full-time equivalent employees.

 

From 2016 the statistics is based on data from registers. To ensure coherence with the previous years, the cost components from register contains the same information as the one from the questionnaire. The variables for costs, full-time equivalent and hours are computed as a total of the year. Average costs are calculated by dividing total costs by total full-time equivalent of the year.

 

For 2012 and 2016 indirect cost is lower than the cost components due to subsidies reducted from the total indirect cost.

  

The data is controlled and revised through several processes, both automatically and manually.

 

Automatic controls correct mathematical mistakes and limit or reduces data that are not within a reasonable range. Some data points are imputed where the enterprise has missing values or values that are obviously erroneous.

 

Important units with large deviations are revised manually and compared with other sources, mainly the annual accounts of the enterprises in question.

Seasonal adjustment

Not relevant

Confidentiality

Not relevant

Comparability over time and space

The labour costs statistics were produced for the first time in 1996, and statistics are comparable through time for some section. From 1996 through 2008 was according to NACE Rev.1 From 2008 statistics has been published according to the new industry standard, NACE Rev.2 (SN2007). From 2012 the public sector is included in the sections.

Accuracy and reliability

Sources of error and uncertainty

Measurement errors occur when the provider of the data report incorrect answers due to forgetfulness, misunderstanding of questions etc. Processing errors are errors from coding or errors that occur during revisions.

  

The statistics are based on data from several registers. Errors can be introduced by variable or lagging update frequency. E.g., enterprises that merged in December are in some registers counted as one enterprise while other registers maintain the original number. Incorrect industry codes and/or employment data in Statistics Norway's Register of Establishments and Enterprises may result in the establishments being placed in the wrong industry.

  

Until 2012 the non-response rate was between 3-10 per cent of the sample. The main reason was that the enterprise no longer had employees due to cessation of activities, bankruptcy, mergers or acquisition by new owner. Moreover, certain enterprises consistently failed to respond while others responded so belatedly that they were not included from the survey. Partial non-response occurred when enterprises did not respond to certain questions that were hard to quantify, e.g. social costs and different kinds of leave from work.

 

All sample-based surveys will be burdened with a certain uncertainty. Generally, the results are less certain the fewer the observations they are based on. Uncertainty also depends on costs dispersion and rate of coverage for the various variables in the population from which the sample is drawn. Groups that are based on relatively few observations will easily be affected by so-called extreme observations, or observations that deviate markedly from the group average.

 

From 2016 the statistics is based on data from a-meldingen. Here also, observations can be discarded due to missing information and/or incorrect values for the variables. Extreme observations are carefully considered on a case-by-case basis for inclusion in the statistical basis.

Revision

Not relevant