Comparability

Comparability - geographical

When the register based unemployment statistics depends on a number of national administrative registers it is not a good statistics for international comparisons. For international comparisons of unemployment you ought to use the unemployment figures from the labour force surveys in the different countries instead.

Comparability over time

A few main series go back to 1910, see the publication Labour Market 1996:28 (Statistical News). However, there have been several discontinuities of the data series, e.g. the statistics until 1973 only covered the population insured against unemployment, but since then the statistics now cover all persons who are unemployed.

From 1979 up to and including 2007, the statistics are strictly comparable. However, changes to the unemployment insurance funds have an effect on the data series at the level of unemployment insurance funds. To this is added that since 1 September 2002, all unemployment insurance funds have been able to decide whether they want to be interdisciplinary or professionally defined.

The statistics on the registered number of net unemployed have been compiled by Statistics Denmark since 1979. To avoid a great number of discontinuities of the data series, the statistics have roughly speaking remained unchanged since 1979. However, the unemployment statistics have been adjusted as from the publication of unemployment figures for January 2008. The underlying primary data have been edited and changes have been made to the concept of unemployment, which imply that the Danish concept of net unemployment is now closer to the definition of unemployment applied by the International Labour Office, ILO, see item 3.1. All changes have been made back to April 2000.

As from the publication of unemployment statistics for January 2010, attention is now also focused on the gross unemployed population, defined as the sum of the registered (net-) unemployed population and persons in activation programmes and who are, at the same time, considered to be available for work. The figures on gross unemployment are calculated from January 2007 and onwards.

As from the publication of unemployment statistics for July 2017 the data source concerning the recipients of unemployment benefit was changed from RAM to FLEUR. One consequence of this change is that the reference period of the monthly unemployment statistics changed from unemployment benefit periods (4 or 5 whole weeks) to real calendar months.

Coherence - cross domain

Statistics Denmark's Labour Force Survey conducted on the basis of samples and interviews contains quarterly unemployment figures, which are compiled in accordance with common EU rules laid down by the ILO. The above surveys are conducted by each EU member state in accordance with EU regulations and enable comparisons of unemployment figures among the member states. In Denmark, the sample survey consists of about 18,000 persons each quarter. The reference period in the labour force surveys is the interview-week implying that the surveys are restricted to cover only those persons who were fully unemployed during the interview-week (unlike the register-based unemployment statistics). Furthermore, the Labour Force Surveys set out precise criteria/questions to which the persons must provide an affirmative reply in order to be categorized as unemployed in accordance with the ILO definition. The Labour Force Surveys also cover a group of unemployed persons, who are excluded from the register-based unemployment statistics, as they do not receive social assistance benefits, initial assistance, unemployment benefits or a similar benefit. The difference between the two unemployment concepts has previously (in 1990's) implied that the number of unemployed persons in the Labour Force Survey has been lower than the corresponding level in the register-based unemployment statistics. However after the year 2000 the level for unemployment in each of the two statistics has been closer to each other, which must partly be attributed to the tightenings made among the respondents for the register based unemployment statistics, with respect to when the recipients of cash benefits can be regarded as unemployed in accordance with the ILO definition. As from 2007, the sample for the Labour Force Surveys has been doubled, which has implied that the level of unemployment according to the Labour Force Surveys is above the level in the register-based unemployment statistics, while developments, i.e. falls and increases from 2007 to 2014, are largely identical. As from the fall 2013 Statistics Denmark has published (together with the monthly register based unemployment statistics) monthly unemployment estimates based on the labour force survey.

Coherence - internal

No internal consistency problems in the register based unemployment statistics.