Skip to content

Comparability

Contact info

Labour and Income, Social Statistics.
Jesper Moltrup-Nielsen
+45 3917 3423

jmn@dst.dk

Get as PDF

Implicit index of average earnings

The standardised index of average earnings is comparable since first quarter 2005 but for some sectors, comparable wage indices also exist further back in time. The implicit index of average earnings is based on the same data as the standardised index of average earning, but there are significant differences in methodology that allow the two wage indices to be used only partially for comparison.

Internationally, the implicit index of average earnings can be compared to the labor cost index collected and published by Eurostat for all EU countries.

Comparability - geographical

The implicit index of average earnings is best comparable with the labour cost index, which is an index that measures the development in labour costs, and which is collected and published by Eurostat for all the EU countries. The main difference is that the implicit index of average earnings do not cover other labour costs such as contributions to social funds, education costs, etc. In addition the labour cost index also includes irregular payments, such as bonuses and payments for vacations.

Comparability over time

To explain the comparability over time, it is necessary to split the description into two parts, one for the sector enterprises and organizations and one for public administration and service.

Enterprises and organizations The implicit index of average earnings for the sector enterprises and organizations has since the 1st quarter of 2013 replaced the index of average earnings for the private sector which was published for the last time in the 4th quarter of 2013. The difference has to do with the classification of sectors, which was changed in 2013 (see the section on Sectors). This means that there is a small difference regarding which enterprises and organizations that are included in the sector enterprises and organizations in the implicit index of average earnings and the index of average earnings for the private sector which was finished in 2013. Other than that the two index series are fully comparable. In the two Statbank tables ILON12 and ILON15, these two index series are chained, so that there despite a break in 2013 exists a longer time series going back to 2005.

In addition to the change in sector classification, the largest break in the index relates to a changes in the classification of economic industries. For the sub-industries the series are not fully comparable between the two classifications. Despite this the total (for the whole sector) is fully comparable. This means that the industry total in the Statbank tabel ILON12 and ILON15 are comparable with the industry total in ILON2X and ILON3X that are available all the way back to 1994.

For ILON12 and ILON15 there was a break in the series in the 3rd quarter of 2008, where a new classification of economic industry was introduced. The classification hence changed from Dansk Branchekode 2003 (DB03) to Dansk Branchekode 2007 (DB07).

The index of average earnings in the private sector was first made in the 1st quarter of 1994 and can be found in the Statbank tables ILON2X and ILON3X. Until the 1st quarter of 1996 figures for nine economic industries were published. From the 1st quarter of 1996 the index was expanded to 27 economic industries. The index was at that point published according to the classification of economic industries in DB93 where the most detailed level of publication was the 27-groupings in that classification. The index was based on all employees, including both white- and blue-collar workers, as well as apprentices and young people under the age 18, that were emplyed in enterprises with at least 10 employees.

From the 2nd quarter of 1998 the effect of special holidays (feriefridage) and children holidays (børnefridage) is ad hoc taken account of in the calculations of the index.

The most detailed description of the old index series are found in the earlier Statistiske Efterretninger. These are unfortunately not available in English.

The index of average earnings for the private sector replaced the hourly earnings index for industrial workers and the monthly earnings index for the white collar workers in the industry. These indices were discontinued in 1997 (ILON1 and ILON6). Since then there has been radical changes in methodology, data collection methods and population, which makes these early indices incomparable with the later indices.

Public administration and service The implicit index of average earnings for the sector public administration and service (ILON22, ILON25, ILON32, ILON35, ILON42 and ILON52) supersede the index of average earnings for the public sector (ILON4, ILON5, ILON6, ILON7 and ILON8), which were discontinued in the 4th quarter of 2013. There are large consistencies between the two series of indices. as the difference mainly refers to the change in sector classification (explained in more detail in the section Sector). The main change is that public enterprises (such as DSB and some municipal owned utilities) now belongs to the sector enterprises and organizations.

Figures following this new classification of sectors have been revised back to 2013, which means that there is a (small) break in the index series. The index development beyond 2013 is therefore identical across the two index series.

In addition to the change in sector classification, the largest breaks are caused by revisions in the classification of economic industry. At the level of sub-industries the series are thus not fully comparable. Despite this the industry total (for the whole sector) is fully comparable between the indices. Furthermore, the reference year has been changed every few years.

The index of average earnings for the public sector was introduced in 1998 and showed the development going back to the 1st quarter of 1995. The first Statbank tables for the period 1996-2008 are the following ILON4X, ILON5X, ILON7X and ILON8X. The quality of the indices have continuously been improved. From 1999 the effect of the sixth week of vacation, or the special holidays (de særlige feriedage) has been taken into account in the calculations.

The most detailed description of the old index series are found in the earlier Statistiske Efterretninger. These are unfortunately not available in English.

Coherence - cross domain

There is a fairly high degree of coherence between the standardised index of average earnings and the implicit index of average earnings, but as for method and delimitation of population, there are still major differences between the two indices. The implicit index of average earnings is a so-called index of average values, where the development of earnings is generated based on a summary earnings average for all employees in the same industry, independently of the individual characteristics of employees. Consequently, changes in the composition of staff in a given industry can affect the measured development of earnings. The standardised index of average earnings is designed as a price index, where employees are divided into groups according to work function and age, and where the development of earnings in the separate groups is weighted together with fixed weights. In this way, the measured development of earnings cannot be influenced to the same extent by changes in the composition of the labour market. Further, there is a difference of population between the two indices, as adolescents (persons under the age of 18) and apprentices are included in the implicit index of average earnings but not in the standardised index of average earnings. The differences are described in more detail in the method report for the standardised index of average earnings. The coherence has to do with the fact that both of the indices use the same data basis (with population differences, however) and the fact that the same earnings and hour concept is used for calculation of the development of earnings.

Coherence - internal

There is internal coherence between content, names and definitions of variables across sectors in the compilation of the implicit index of average earnings. Even though the reported information is relatively coherent across the sectors, certain differences exist. Some components are reported on a more detailed level in one sector compared to the others.