Contact infoLabour and Income, Social Statistics.
+45 3917 3423
Get as PDF
Data is collected from a sample of companies and organisations as well as the entire public sector, covering the middle month of the quarter.
Data is validated by using fixed boundaries, both at individual and company level. Manual corrections are also made if required. Only companies that are present in both quarters are included in the calculations.
In the calculation of the most detailed sub-indices, data for the private sector are weighted to the target population and the individual employment types are weighted with the hours worked.
For the sector enterprises and organisations the data is collected quarterly, partly on the basis of a sample of approximately 5000 enterprises from the private sector with 10 employees or more, partly on information from all public undertakings not included in general government. Statistics Denmark collects the information for the private enterprises in collaboration with the Danish EmployersConfederation, the Danish EmployersAssociation for the Financial Sector and the Danish Pharmaceutical Association.
For the public undertakings, Statistics Denmark collects data from the Agency for Econmics, DSB (Danish state railways) and KRL (payroll data from local and regional authorities). The sample of private enterprises is stratified and drawn from ESR (the statistical business register) by size groups 10-19, 20-49, 50-99, and 100+ full-time employees. All enterprises with 100+ employees are included in the sample.
The following applies for the sector general government: First and foremost, data stem from the major public payroll transfer systems (such as SLS (the governments payroll system) and KRL) supplemented by a small number of private payroll processing services. E.g., payroll information from the Lutheran Church of Denmark is reported via a private payroll processing service.
Frequency of data collection
For the Private sector data is collected quarterly, for the middle month in the quarter.
For the Public sector data is collected monthly, although only the data for the middle month of the quarter are used to calculate the wage index.
For the sector enterprises and organisations, the following applies: The collected data consists only of sampling from the IT systems of reporting enterprises and organisations for payroll administration of their employees on an individual level. For the main part, this takes place as system-to-system reporting, where the e.g. the payroll systems report data for their customers (enterprises and organisations) via extensive bulk submission directly to Statistics Denmark. Enterprises and organisations with proprietary payroll systems report their data either by uploading it via a web application or by submitting an encrypted file via e-mail to a separate e-mail address. The Danish Employers Confederation and the Danish Employers Association for the Financial Sector collect data from their own members and report it to Statistics Denmark via a special system-to-system solution. As provided by the Act on Statistics Denmark, reporting of this information is mandatory, and failure to do so will be reported to the police and is punishable by a fine.
The following applies for the sector general government: Statistics Denmark receives the main part of the data material as extracts from the major public payroll transfer systems via system-to-system solutions. For government employees, Statistics Denmark receives data from the Agency for Modernisation, Danish Defence, and Silkeborg Data. For local and regional authority employees, data is provided mainly by KRL. In addition, data is reported from some relatively small private payroll processing services, also via system-to-system solutions.
There are two main validation processes in the production of the Implicit index of average earnings.
The first process is performed on the individual records, and is identical across sectors. Data is validated in relation to a lower and upper limit of the earnings per hour. The limits are revised in the first quarter each year by extrapolating the limits from last year with the increase in the average earnings per hour from the same period. Records outside the limits are removed from the data. This first validation process is necessary as it removes extreme observations that otherwise would potentially distort the results.
The next validation process is performed on the level of enterprises. And here the process is different depending on data belongs to enterprises and organizations or public administration and service.
For the sector enterprises and organizations the following applies: First places of work that are not present in both the last and the current quarter are removed. Then enterprises with a change in the average earnings per hour outside the acceptance interval -10 and 10 percent are removed. The same applies for enterprises where the change in the number of employees is outside the acceptance interval of -25 and 25 percent. Large enterprises that fall outside these limits, are investigated more thoroughly, i.e. by contacting the enterprises and asking them to retransmit.
For the sector public administration and service the following applies: Organizations or enterprises with a change in the average earnings per hour outside the acceptance interval of -5 and 5 percent are either removed or investigated more thoroughly.
In addition to the two validation processes above, the following applies to all sectors: If large enterprises or organizations are removed from the data, it is investigated what the removal means for the results of especially the economic industry in which the enterprise belongs. Small economic industries, or industries with one or a few very large enterprises, are often more sensitive to removal of enterprises. In these cases, the enterprises are often contacted in order to receive new and correct data, or a method to correct the current data.
Calculation of elementary aggregates (EA) Elementary aggregates in the implicit index of average earnings also constitute the most detailed level of publication, which is the 36-groupings according to Dansk branchekode (DB07). The elementary aggregates are calculated as the change in the average earnings per hours worked for all employees in the same sector and economic industry from the previous to the current quarter. The wages in the middle month of a quarter is used as a proxy for the quarter.
Data regarding enterprises and organizations are based on sample survey and is therefore weighted. The weighting is performed by splitting the data into several strata based on economic industries and the following size grous: 10-19, 20-49, 50-99 and 100+. The weights are taken from the most current Business Register and constitute the number of full time employees in each strata.
For the sector public administration and service no weighting is performed, as the data here is based on a census. The elementary aggregates (which are the same as the most detailed level of publication) are then calculated as the change in the average earnings per hours worked from the previous to the current quarter.
For all sectors index values are calculated by chaining the change from the previous to the current quarter onto the index value of the previous quarter.
Calculation of aggregated indices that are published Sub-indices refer to the aggregated indices, that are normally published. In the implicit index of average earnings the most detailed sub-indices are also the same as the elementary aggregates (4 sectors and 36 economic industries). There are no total index across all sectors, so the most aggregate level are the sector totals.
For the sector enterprises and organizations the following applies: The sub-indices and total indices are calculated by weighting the elementary aggregates with the number of full-time employees from the Business Register from the previous year. These weights are updated with the calculation of the first quarter of the year.
For the sector public administration and service the sub-indices are calculated by summarizing the earnings and hours from each elementary aggregate, and then calculating the change in the average earnings per hour in each sub-index from the previous to the current quarter, which means that the sub- and total-indices are basically calculated in the same way as the elementary aggregates.
No corrections are made besides what is described under data validation and data treatment.