EU labour force survey – development and history
This Eurostat article portrays briefly the history of the European Union labour force survey (EU-LFS). It gives information on the most important developments of the EU-LFS, describes the cornerstones of today's EU-LFS "output harmonisation" and "continuous survey", and gives an overview of main stages in the creation of the EU-LFS.
This article is part of a set of online articles on the EU-LFS.
Development of the EU-LFS
In Europe, the labour force survey is a long-standing survey, going back in some countries to the 50s or 60s. Ever since the EU-LFS has grown in several directions:
- The number of participating countries has increased. The original number of 6 EU Member States has grown to presently 34 participating countries, of which the 28 Member States, three EFTA countries (Iceland, Norway and Switzerland), and three candidate countries, i.e. Montenegro, the former Yugoslav Republic of Macedonia and Turkey.
- It has become more harmonised across countries. The national surveys were originally quite far apart from each other. This harmonisation was achieved mostly through European legislation and output harmonisation approach. The definition of internationally agreed standards prepared by the International Conferences of Labour Statisticians convened by the ILO in 1982 and 1987 were also important milestones towards a high degree of comparability across countries.
- The survey has gained steadily in importance. In the last decades the European labour market has been transformed by strong structural changes which have taken place. Women have been increasingly joining the labour force, mass unemployment emerged, the weight of the various sectors of the economy has evolved, new forms of work (part-time work, work at home...) arose. Over time, the EU-LFS has proved to be the only source of information in these areas to provide truly comparable data in the sense of being independent of the national administrative and legislative framework. The EU-LFS is now universally recognised as an indispensable tool for monitoring labour market developments and for taking the appropriate policy measures.
Milestones in the EU-LFS
|1960||The first EU-LFS was organised in the six original Member States in 1960 by Eurostat.|
|1983-1991||Annual surveys on the basis of a revised set of concepts designed to guarantee an improved degree of comparability between the member states. The concepts and definitions used were those adopted by the 13th International Conference of Labour Statisticians of 1982.|
|1992-1997||The Regulation 3711/1991 leads to further improvement of the quality of the EU-LFS.|
|1995||EU-LFS covers 15 Member States.|
|1998-2000|| Regulation 577/1998 set the regularly framework and the corner pillars for today's EU-LFS by introducing a continuous survey and adapting the output harmonisation approach.|
Ad-hoc modules were also introduced.
|2000||EU-LFS covers 27 Member States.|
|2005||Quarterly, continuous survey in 27 Member States (see section 'continuous survey' below).|
|2006||'Structural variables' introduced (Regulation 430/2005). These variables need to be surveyed only annually instead of quarterly.|
|2007||Quarterly, continuous survey in 28 Member States (see section 'continuous survey' below).|
More information on the EU-LFS regulation can be found EU-LFS main features and legal basis.
Labour force surveys are run in many countries around the world (a compilation of those surveys can be found in the International Labour Organisation (ILO) LABORSTA internet, volume 3, 'household sources and methods'. In spite of all being named labour force survey, they can actually be very different surveys, just having some common features: they are household surveys and they are mostly targeted to collecting data on the labour situation of respondents. However they may vary in many aspects: the frequency of the survey may not be the same (ranging from monthly to multi-annual depending on the country), the definitions used may not be identical (although ILO issues guidelines that countries are encouraged to adopt), the list of variables, the survey may or may not be used as a vehicle to collect information unrelated to the labour market, sampling design and data collection mode (face-to-face interview, telephone interview, etc.), IT technologies used, etc. In Europe, the labour force survey is a long-standing survey, going back to the 50s or 60s in many countries. These national surveys were originally developed independently by countries. The first attempt on the way to an EU-LFS was made in 1960 to collect comparable data on employment and unemployment from the six original Member States of the then European Community by means of a labour force survey. In the following decades and until the early 1990s the comparability of countries' data relied mainly on Member States' agreements to use international statistical definitions. The concepts and definitions used were those adopted by the 13th International Conference of Labour Statisticians of 1982.
Use of EU legislation for harmonisation
In the early 1990s EU legislation was first used to further assure the comparability of the EU-LFS. EU regulations in the field of statistics legally bind the Member States to fulfil the standards set out in the regulations. They stipulate the rules and guidelines to assure the comparability of the results by regulating the survey designs, the survey characteristics, and methods. Council Regulation 3711/1991, introduced in 1992, aimed to improve the quality of the data and their reliability at national and regional level by specifying the contents and reliability criteria for the EU-LFS survey as well as the frequency of the survey and the conduct of the survey. Nevertheless in the late nineties, in a context of different, established national LFS surveys across Europe, there was a further need to set up an EU-LFS that, while ensuring continuity with the established national surveys, delivered further enhanced comparable and harmonised results. This eventually led to the adoption of Council Regulation 577/1998 of 9 March 1998. This regulation has set the regulatory framework and the pillars for the today's EU-LFS by changing the survey to a continuous, quarterly sample surveyand adapting an output harmonisation approach. This approach is important and it has consequences visible in the current conduct of the EU-LFS.
Nowadays the EU-LFS does not rely only on specific legislation for this survey. There are other initiatives across domains to ensure data quality. For instance, the European Statistics Code of Practice clearly states in Principle 14 that European Statistics are consistent internally, over time and comparable between regions and countries.
More information on the EU-LFS regulation can be found EU-LFS main features and legal basis.
Council Regulation (EC) No 577/98 also saw the introduction of ad-hoc modules in the labour force survey. Each year, a set of additional variables covering a specific topic of the labour market is collected. More information can be found EU-LFS ad hoc modules
Output harmonisation approach
Broadly speaking one can differentiate between two harmonisation strategies for European surveys: input and output harmonisation. "With input harmonisation – which is also referred to as method harmonisation or harmonisation of data sources – all participating countries use precisely the same survey procedures in an idea case. Country-specific particularities are only permissible where they are indispensable – in other words for example in the language used on the questionnaire. With output harmonisation – also known as product harmonisation – in contrast, only the goal is determined – in other words the value to be surveyed. The selection of suitable survey methods is left to the participating countries themselves. In general terms, output harmonisation only sets an international concept by which the circumstances are defined that are to be surveyed. The countries are then left with the task of working out suitable national concepts and measurement procedures with which the international concept can be portrayed." (Roland Günther, p.4 Working Paper 19 – The change from Input Harmonisation to Ex-post Harmonisation in national Samples of the European Community Household Panel - Implication on Data Quality, Wiesbaden, Statistisches Bundesamt, 2003)
The EU-LFS uses output harmonisation. This means that the EU issues standards for the output of the LFS in the Member States, but those standards do not impose a single way of designing or conducting the survey. In practice, the EU standards regard the list of variables and categories, the minimum sampling precision, the observation period (i.e. a reference week and the survey must collect data referring to every week in the year), the periodicity of the survey, and reporting obligations of data and quality reports to Eurostat. All these standards were established in the EU-LFS Regulation 577/1998.
This harmonisation approach has strong consequences in several aspects.
- Firstly, in the distribution of the tasks between Eurostat and the NSIs of the Member States, in particular: the national statistical institutes of the Member States are responsible for designing national questionnaires, drawing the sample, conducting interviews and sending results to the Commission (Eurostat) in accordance with a common coding scheme established by Commission Regulation 2008/377. Member States transmit the LFS data to Eurostat (at microdata level with the corresponding weighting coefficients) as well as quality reports. Eurostat is in charge of monitoring the implementation of EU legislation on the EU-LFS, providing assistance to national statistical institutes, promoting harmonised concepts and methods, and disseminating comparable national and European labour market statistics.
- Secondly because of the output harmonisation, countries do not necessarily use the same questionnaire, the same sample design, the same sampling unit, the same panel rotation structure for the sample, etc.
- Thirdly, countries can (and do) add questions of national interest to the questions necessary for the EU-LFS. They can also have national ad-hoc modules, in addition to the EU ones.
In general, the use of output harmonisation is transparent to users and it has no visible effect on the results, but it changes significantly the way Eurostat works with the National Statistical Offices of the Member States.
Historically, many labour force surveys in Europe were conducted only once a year in spring. For the EU-LFS 'spring' corresponded to quarter 2 for all countries participating in the survey except France and Austria for which quarter 1 was used. Spring was chosen as a 'normal' and 'representative' period of the year because the number of employed persons was not affected by seasonal effects in winter, summer, Christmas, etc. However this practice did not allow monitoring short-term trends the labour market, nor estimating yearly averages. With the move from an annual survey to a quarterly survey, it became important to collect data covering all the weeks in the year. This is a continuous survey. Some countries first introduced a continuous annual survey (meaning the reference weeks were uniformly distributed throughout the reference spring quarter) and then switched to a quarterly collection, whereas the others moved directly to a continuous quarterly survey. Based on the Regulation 577/98 most participating countries changed their labour force surveys into continuous surveys in the period 1998 to 2004, although some did it earlier and others later:
|1998||IE, PT, RO, SK|
|1999||BE, ES, CY, SE|
|2000||EE, NL, PL, FI|
|2002||SI, LT, LV|
|2003||BG, FR, LU, HU, IS|
|2004||IT, MT, AT|
Since 2014 all countries conduct the LFS as a continuous survey.
Given this gradual changeover, Eurostat provides EU results on the basis of spring results until 2004. Starting from 2005, the EU results are calculated every quarter.