unemployment statistics better and cheaper

Lower costs, more respondents, better international comparability and web surveys. Today Statistics Netherlands publishes the first outcomes of the revision of labour force statistics. This revision consists of two parts. Definitions have been adjusted to comply with internationally agreed ones, and data collection has been improved: Statistics Netherlands is the first national statistics office in Europe to use the internet for its Labour Force Survey. This means more respondents will be able to participate in the survey, and it responds to their demands to complete the questionnaire when and where they want. In addition, it reduces the costs of the survey considerably.

The change in definition results in a lower number of unemployed, the new data collection method in a higher estimate of unemployment. The net result of these two adjustments is a relatively small difference: for 2013, the figure after revision was 647 thousand, 9 thousand fewer than before the revision.

Revision based on new definitions and statistics

In July, Statistics Netherlands announced that from 2015 it will use the definition of the International Labour Organisation as its main unemployment indicator. This means that, from 2015, the monthly unemployment press release will be based on the ILO definition. This is part of a broader strategy to comply with international definitions as much as possible in order to increase the international comparability of statistics. As unemployment is just one indicator for the situation on the labour market, Statistics Netherlands will make more and more use of supplementary indicators to describe developments on the various dimensions of the labour market adequately and consistently. In its recently published quarterly labour market review, for example, it used unemployment rates in combination with figures on unutilised labour supply and on jobs, vacancies and labour market dynamics to describe how the labour market showed signs of recovery in the third quarter.

Better international comparability

According to the ILO definition, everyone with a paid job of at least 1 hour a week is included in the employed labour force.  Pupils and students with part-time jobs, for example, all belong to the employed labour force under this definition. The national definition of the labour force comprises only people who work or want to work for at least 12 hours a week. According to this national definition, the Dutch employed labour force comprised nearly 7.3 million people in 2013, compared with 8.3 million according to the ILO definition. In addition to the hours threshold, there are other differences between the definitions, such as the period in which jobseekers have looked for work, and how soon they can start work (see table 1).  In 2013, according to the ILO definition 600 thousand people were unemployed in the Netherlands, 56 thousand fewer than according to the national definition.


Labour Force Survey

The Labour Force Survey (LFS) is the main data source for labour force statistics. Every month a representative sample of 30 thousand people take part in this survey, answering questions on their labour market position: whether or not they have a paid job, and if so for how many hours a week. Is this the number of hours they want to work per week? And if they do not have work, do they want to work? Are they looking for work and can they start a job in the short term?

Every month Statistics Netherlands uses advanced statistical methods to process the answers in order to compile reliable unemployment estimates.  It uses a time series method that is unique in Europe to publish monthly statistics on the basis of the survey. Although the figures on the unemployed and employed labour force are the basis for unemployment statistics, to get a more complete picture of labour market, supplementary indicators on various dimensions of the labour market are compiled.  For example, the number of changes between labour market positions, duration of unemployment and length of service are also measured. Weekly working hours, both actual and desired, and job characteristics such as working hours, occupation, and sector of employment are also included in the survey.

First in Europe to use web surveys

In recent years, the survey used to collect information about the labour force has been updated and made more efficient by allowing respondents to complete the questionnaire online. The Netherlands is the first country in Europe to use the internet for this purpose. At the same time, the questionnaire has been adapted slightly to make it suitable for online use. These adjustments have resulted in a higher estimate of the number of unemployed according to the ILO definition: +47 thousand. The combination of the switch to the ILO definition     (-56 thousand) and the updating of the survey method thus gives a net revision of the unemployment rate of -9 thousand persons, resulting in 647 thousand for 2013. At -80 thousand, the adjustment of the employed labour force as a result of the changes in survey method is relatively small compared with the total of nearly 1 million extra jobs of less than 12 hours a week that are now included under the ILO definition. The unemployment rate for 2013 drops from 8.3 percent to 7.3 percent (see table 2).

Revised figures for 2003-2014

This press release details the changes in labour force figures for 2013 as a result of the revision. For the monthly unemployment statistics, Statistics Netherlands will incorporated revision for the first time in the figure for January 2015, to be published on 26 February 2015. To prevent a break in series, the revised figures for 2003-2014 will be published on Statistics Netherlands’ website. The unemployment press releases published in December 2014 and January 2015 will be based on pre-revision figures. From February 2015, the supplementary labour market indicators based on the Labour Force Survey will also be adjusted in line with the ILO definition. Figures according to the national definition will remain available in the next few years.