Accuracy and reliability

Overall accuracy

As the methods used for collection and production of the indices differ in all the countries included in the international labour cost indices, it is only possible to make any conclusions on the accuracy of the Danish index.

The overall accuracy in the case of the Danish index is mainly affected by two factors. The first relates to the general uncertainty of the index being based on a sample survey. The second factor is the completeness of the transmitted data, which often suffer from errors that can be traced back to the enterprises payroll systems and how data are registered there. Another factor that might cause inaccuracy is the enterprises transmission of data to the survey on other labour costs. The challenge here is that the information mainly is collected manually through their accounting system, and does not always entail all the information requested in the survey. Nevertheless, this last factor is considered to be of less influence on the overall accuracy then the two first ones, as the other labour costs only constitute around 5 percent of the total labour costs in Denmark. Furthermore, as the sample survey of other labour costs is stratified in a way that includes all big enterprises, the labour cost index is still considered to be quite accurate, especially at the most aggregate levels.

Sampling error

There has not been produced any estimate of the sampling error of the Danish labour cost index. At the same time it is not mandatory for the EU member countries to estimate the sampling error when producing the indices.

Non-sampling error

Other uncertainties are primarily linked to mistakes in enterprises reporting of components for the wages of their employees. This could for example concern components such as irregular payments, e.g. like bonuses and delayed regulations of wages. When these are not correctly registered in the payroll system of the enterprise, they can have unintentional consequences for the calculations of wages. As such, the problem is not not caused by non-response, but rather errors present in the enterprises payroll systems, which can be quite difficult for the enterprises to detect. At the same time it is not always possible to detect the errors during the coarse search for errors before the production process, as it is not necessarily given that there should be payments of this kind in a given period.

From 4th quarter 2019 onwards the figures on Danish labour cost are affected by a new legislation concerning payments for days not worked, imposed on enterprises from 1st September 2019 onwards due to The Danish Holiday Act 2020, entering into force from 1st September 2020. During the transitions period 1st September 2019 to 31th August 2020 the Danish government has decided the earned holiday payments during that period will be frozen until the employee finally leave the labour market for retirement. The employers (enterprises) are instructed not to pay out these funds with the wages and salaries but to withhold and later on transfer them to a special pension fund. These funds on holiday payments are not part of the usual quarterly data reports by the enterprises, but represents a substantial part of their total labour costs. As a consequence it has been necessary to estimate these missing holiday payments as 12.5 percent of the basic wage, to compensate for the lacking data. However as they are estimated, they represents an additional source of uncertainty in measuring the development in the Danish labour costs.

Another source of uncertainty is due to the COVID-19 crisis. The annual growth rates in EU-member states, UK and the USA are affected in different size due to differences in political decisions by their respective governments regarding non-standard schemes, such as direct payments to employees to compensate lower wages due to short-time work or direct subsidies to enterprises for compensating labour costs during the periodic shut-down. Moreover the shut-downs itself in some cases might have caused challenges in the process of collecting data by the national statistical institutions within some countries. As a consequence it is likely that revisions for some countries will be at a larger scale than usual in the forthcoming months due to the outbreak of COVID-19.

Quality management

Statistics Denmark follows the recommendations on organisation and management of quality given in the Code of Practice for European Statistics (CoP) and the implementation guidelines given in the Quality Assurance Framework of the European Statistical System (QAF). A Working Group on Quality and a central quality assurance function have been established to continuously carry through control of products and processes.

Quality assurance

Statistics Denmark follows the principles in the Code of Practice for European Statistics (CoP) and uses the Quality Assurance Framework of the European Statistical System (QAF) for the implementation of the principles. This involves continuous decentralized and central control of products and processes based on documentation following international standards. The central quality assurance function reports to the Working Group on Quality. Reports include suggestions for improvement that are assessed, decided and subsequently implemented.

Quality assessment

It is not possible to assess the quality on the indices included in the international labour costs, as they are all to some degree based on different collection and production methods. For information on quality of the different indices, it can be referred to the homepages of Eurostat and BLS on the subject, or the individual countries national statistical institutions.

Data revision - policy

Statistics Denmark revises published figures in accordance with the Revision Policy for Statistics Denmark. The common procedures and principles of the Revision Policy are for some statistics supplemented by a specific revision practice.

Data revision practice

Revisions of historical figures for EU member countries occur on a more or less comprehensive scale continuously from quarter to quarter. For more information about this it can be referred to Eurostat.