Comparability
Contact info
Labour and Income, Social statisticsJarl Quitzau and Uwe Pedersen.
+45 39 17 35 94, +45 39 17 34 24
Get as PDF
Especially the extraordinary payout of 2 weeks worth of wages as part of a COVID 19-relief package affects the income levels in 2021. COVID-19 also makes indicators on relative poverty - designed for measuring social exclusion difficult to interpret. Due to the high inflation rate of 7.7 percent in 2022, the income statistics 2022 is affected due to special payouts of governmental inflation aid.
Income statistics are quite consistent over time if you use the coherent time series on our website. The statistics were last revised in 2013. Data dating back to 1987 has been revised. A change in the vacation law has caused differences between national accounts and income statistics in the years 2020-2021. Eurostat and OECD are the recommended sources for making international comparisons.
Comparability - geographical
Reliably comparing income for different countries is very difficult. Insurance-, pension- and welfare systems differ from country to country. In addition, the price level and the level of the welfare services make it hard to interpret differences in the disposable income when estimating economic living conditions. Denmark, the other Nordic countries, Holland and Slovenia are among the only countries that base income statistics on complete registers. In other countries, the statistics are survey-based and thus affected by statistical uncertainty.
Comparability over time
Especially for 2019 and 2020, as a result of the situation surrounding COVID-19, the Government launched a number of aid packages in the spring of 2020 and in the spring of 2021 with the aim of supporting the Danish business community. One of the measures referred to a general postponement of the general deadlines for personally owned companies with regard to reports of company profits etc. to SKAT for the income year 2019 and 2020. In this connection, the deadlines were moved from June 2020 to 1 September 2020 and from June 2021 to September 2021. However, Statistics Denmark received the first delivery of the Tax Agency's data for the final calculation after the 8 months in August, i.e. before the reporting deadline of 1 September for personally owned companies. At that time, the data base was thus extremely deficient, as a large number of self-employed persons were simply not included. As a result of this, the Income Statistics for the census years 2019 and 2020 were extraordinarily postponed until November 2020 and 2021, as it had to be done on the basis of a later delivery of data for the final equation, which was only received in the month of November.
In 2020 and 2021 as well COVID-19 related benefits has a large imapct on the timeseries. The additional payout of "holiday funds" equal to 2 weeks worth of wages is included in the wage income. Due to the high inflation rate of 7.7 percent in 2022, the income statistics 2022 is affected due to special payouts of governmental inflation aid.
COVID-19 makes the interpretation of the indicators on relative poverty difficult. Normally they are used as indicators for social exclusion due to economic reasons. However COVID-19 lockdowns meant that economy might not have been the primary reason for social exclusion in low-income households in 2020 and 2021. Furthermore there is an temporary effect caused by the extraordinary payouts of holiday funds in 2020 and 2021 as well as the extraordinary payouts of inflation aid in 2022. See the attached Annex ) details the effect of COVID-19 on poverty estimates (Danish only). 1994 tax reform: Some tax-free transfers were increased and made subject for taxation. These transfers includes disability pensions, old.age pensions and cash benefits. This increases the pre-tax income for the recipients but has little effect on the disposable income. The same reform introduced the Labor Market Contributions (”Arbejdsmarkedsbidrag”).
History: From 1905 to 2016 Statistics Denmark has published an annual publication on income. The income statistics are for the first time published in The Statbank in the year 2002. Income statistics for the period 1983 - 2001 are published in the ”Tax-oriented Income”-publication. The information on income was before 1983 published in the so-called, ”Tax Return Survey (Selvangivelsesundersøgelsen)”. It was based on a sample of income tax returns.
The ”Tax-Related Income”-statistics are based on The Income Statistics Register, which has been constructed since 1970. Comparing data on income over time was not possible until 1976 when the design of The Income Statistics Register was changed.
Income statistics for families and households were published annually in an extended newsletter, ”Statistiske Efterretninger”, from 1984-2005 and in the annual income publication.
Coherence - cross domain
As most income related statistics share data sources, the data is often comparable.
Statistics on certain transfers are available in The Statbank. They are mostly based on data from municipal social registers. There will not be an exact match between these statistics and the income statistics, but for the large majority of individuals the data should match. Income and Wage statistics are often mixed up by users of the statistitics. The primary unit in Wage Statistics is the standardised hourly wage at a selected time for a given job position. Wage statistics are not directly comparable to the wages in the annual income statistic as employees may have more than one job, work part-time, change job or be partially unemployed during a year.
EU-SILC is a survey carried out in more than 30 European countries and is the primary source for income statistics in many European countries. In SILC the income definitions are harmonised and - with the above mentioned reservations - somewhat comparable. The primary difference between the income concepts in this statistic and the Danish SILC income statistic is that imputed rent of housing and interest on mortgages is not included in the SILC-incomes. Furthermore SILC uses a different household definition than the families used in this statistic.
Finally it's worth noting that SILC is a survey and thus is subject to statistical uncertainty.
Danish Paper on difference between the national registers and SILC.
Coherence - internal
There are no inconsistencies.