IESS (Integrated European Social Statistics) Framework regulation: state of play and impact on the LFS

Transcription

1 EUROPEAN COMMISSION EUROSTAT Directorate F: Social statistics Unit F-3: Labour market statistics Doc.: Eurostat/F3/LAMAS/38/14 WORKING GROUP LABOUR MARKET STATISTICS Document for item 2.1 of the agenda IESS (Integrated European Social Statistics) Framework regulation: state of play and impact on the LFS December 2014 BECH Building Room Quetelet Luxembourg

2 1. Introduction The objectives of this document are: First to inform LAMAS about the state of play of the discussions on the modernisation of social statistics programme by the European Statistical System Committee (ESSC) and on the IESS (Integrated European Social Statistics) Framework Regulation by the Directors of Social Statistics (DSS). Second, to look at the consequences for the review of the LFS, focusing on issues which are not discussed in other LAMAS documents, but will be part of the IESS regulation or its implementing regulations. Four issues will be tackled: the list of standardised variables, requirements in terms of quality reporting, use of administrative data as well as compulsory transmission of monthly unemployment data. The financing of ad-hoc modules is outside the scope of this document and will only be tackled in June State of play of the discussions at ESSC/DSS levels 2.1 Discussion on the modernisation of social statistics at ESSC level The ESSC already discussed two times about the modernisation of social statistics: The first time, in November 2013, the ESSC was consulted (and unanimously agreed) on a very specific issue of the modernisation strategy: the introduction of the infra-annual rotational pattern as a compulsory element of the LFS; In November 2014, the ESSC discussed the overall modernisation strategy for social statistics under which three framework regulations are envisaged. The first framework regulation (IESS - Integrated European Social Statistics) would provide a unified legal basis for the existing social statistics based on micro-data collections while enabling the future integration and improvements foreseen under the modernisation programme. The second framework regulation would cover population statistics including population and housing censuses, and the third framework regulation would cover the (mainly) administrativelybased statistics and accounts. In November 2014, the ESSC generally supported the modernisation programme for social statistics, stressing the importance of the actions on the integration of the data collections, on quality, and on the wider use of administrative data as well as innovative data sources such as Big Data. It was noted that the issue of statistics on income, consumption and wealth, and their distribution, was of high importance and should be thoroughly examined. The ESSC members appreciated the approach of closer links to macroeconomic statistics, while maintaining the social perspective. They supported the principle of framework regulations for social statistics but there were concerns about the scope of the first framework regulation, IESS (Integrated European Social Statistics), and the use of delegated acts. Concerns were raised around the inclusion of additional variables in the Labour Force Survey (LFS), and it was agreed that the Directors of Social Statistics (DSS) should continue work with a view to finding a consensus on the open issues. Finally, the inclusion of the Household Budget Survey (HBS) and the Time Use Survey (TUS) in IESS is a strategic issue which will be discussed at the seminar to be organised back-to-back with the ESSC meeting in February

3 2.2 Discussion on the IESS Framework Regulation by the DSS At DSS level, discussion is currently focusing on the IESS Framework Regulation (FR). Since the last LAMAS meeting in June 2014, two discussions took place, one in the DSS meeting at the end of September 2014 and one in the DSS Board composed of a subset of the directors - on 1 and 2 December The new legal architecture for social micro-data collections will comprise three layers: the IESS Framework Regulation, delegated acts on elements common to several data collections (including a multiannual planning of social surveys), and implementing acts on each data collection. A fourth layer, encompassing manuals and guidelines will remain outside the legal framework. In this architecture, the level of detail will increase from the first to the last layer: the FR will include topics and detailed topics covered by each data collection; the list of modules will be in a Delegated Act while implementing measures will include the list of variables. Explanatory notes and model questionnaires will not be regulated. In terms of scope, the current version of the IESS FR covers nine European social data collections for which micro-data is transmitted to Eurostat. 1 The IESS FR will ensure a better harmonisation between all data collections in terms of quality requirements and assessment, structure of data collections (especially for the ad-hoc modules), planning, names of modules and variables, funding, and structure of data and metadata. 3. Consequences for the review of the LFS 3.1 General overview The IESS Framework Regulation will comprise a core text and annexes. The annexes will include the main elements which impact on the costs for Member States. Developing the core text is a DSS responsibility and LAMAS will be regularly informed. An oral presentation of the outcome of the discussion in the DSS Board (held on 1/2 December 2014) will be provided in the LAMAS meeting. The IESS FR comprises 6 annexes on topics/detailed topics to be covered, precision requirements, sample characteristics, periodicity, provision deadlines and macro-data (transmission on monthly unemployment data). The elements covered by the FR annexes as regards the LFS are appended to this document (Annex 1). Elements on precision requirements and on sample characteristics have already been agreed by LAMAS. For topics, periodicity and provision deadlines, some discussions still need to take place on the list of topics/detailed topics/modules and variables, on the periodicity of the system of AHMs (regular/supplementary modules) and on timeliness of structural and AHM variables. This will be tackled respectively under agenda items 2.2 (EU LFS new content: variables & modules), 2.3 (Future system of LFS AHMs) and 2.4 (Sub-sampling architecture for EU LFS). LAMAS will also discuss under section 3.2 of this document the consequences for the LFS of four elements included in the core of the IESS FR, i.e. standardised variables, quality reporting, use of administrative data and compulsory transmission of monthly unemployment data. It is to be noted that discussions on standardised variables and on monthly unemployment will be split between this 1 Labour Force Survey (LFS), Statistics on Income and Living Conditions (SILC), Information and Communication Technology (ICT) survey, Adult Education Survey (AES), European Health Interview Survey (EHIS), European Health and Social Integration Survey (EHSIS), Household Budget Survey (HBS), Time Use Survey (TUS) and tourism survey 2

4 agenda item (where issues linked to the IESS FR will be tackled) and agenda items 3.1 (that will focus the detailed guidelines for 4 labour market variables to be included in a methodological manual) and 3.5 (where the progress report of the Task Force on monthly unemployment will be presented) respectively. 3.2 Specific elements: standardised variables, quality reporting, administrative data and compulsory transmission of monthly unemployment data A. Standardised variables Eurostat is currently working on a list of 28 variables to be standardised with the aim to issue a detailed methodological manual. These variables are collected from more than one data collection. A subset of these 28 variables are variables to be included in all social micro-data collections and for which some more precise elements will be included in the legal basis. The current draft of the IESS FR includes an Article on 'Elements common to several data collections' including the list and description of common variables. 16 (possibly 19) variables are concerned: Sex Date of birth / Age Country of birth Country of birth of mother Country of birth of father Country of main citizenship De facto marital status Household composition / Intra-household relationships Country of residence Region of residence Degree of urbanisation Status of activity (self-declared) Status in employment Occupation in employment Economic activity of the local unit where employed (Economic sector in employment) Educational attainment (highest level of education successfully completed) (Possibly) Net monthly income of the household (Possibly) Self-declared health status (Possibly) Limitation in activities because of health problems (GALI) Compared to the current list of core variables, 'legal marital status' is dropped, 'country of birth of mother' and 'country of birth of father' are added. The 3 variables 'net monthly household income', 'self-declared health status' and 'limitation in activities because of health problems (GALI)' will be part of the common variables only if the decision is taken in 2015 to include these variables in the LFS. For each of these core/common variables, the following information will be included in the legal basis: The label and corresponding codes of the variable; A short description of the variable. 3

5 As an example, for status of activity (self-declared) and depending on the result of the discussion and decision of LAMAS under agenda item 3.1, the legal basis could include the following label/codes: Total Employed / at work Main Activity Status Unemployed / seeking work Student, pupil (in formal education) Housewife/-husband Retired (incl. early retirement) Permanently unable to work Other (including conscripts) Not stated As well as the following short description: The main activity status represents a person s self-perception. It refers to socially relevant statuses like being in employment, unemployed, retired, a housewife/-husband etc. A person can have more than one of these statuses at the same time but the variable only counts the main status, i. e. the status about which the person thinks describes her-/himself best. For this purpose, the respondent is asked to choose from a predefined list of statuses. In terms of timetable, the final list and description of all variables included in all micro-data collections and with a more specific legal requirement will be available at the end of Each sectoral Working Group is responsible for developing the variables in its field of competence and will be regularly informed about agreement reached in the other Working Group meetings. DSS will always have the final word in case of disagreement on some decisions taken by other WGs. B. Requirements in terms of quality reporting: Current situation: o In the current LFS Council Regulation N 577/98 (Art. 3 Representativeness of the sample, paragraph 6), the requirement in terms of quality reporting for the EU-LFS is very general. It is described as follows: Member States shall provide the Commission (Eurostat) with whatever information is required concerning the organisation and methodology of the survey, and in particular, they shall indicate the criteria adopted for the design and size of the sample. o In practice, three types of quality reports are transmitted to Eurostat: Quarterly accuracy reports transmitted 16 weeks after the end of the quarter; Annual quality reports sent mid/end August of year T+1; Quality reports for the ad-hoc modules sent by the end of March T+1. On the basis of the information in both quarterly and annual reports, Eurostat issues an annual consolidated quality report around mid-november of year T+1. The report 4

6 is complemented by a summary describing the main characteristics of the national surveys and disseminates selected information of the specific country quality reports using the ESQRS structure. Proposal in the context of the IESS Framework Regulation (FR) The IESS FR specifies that Member States shall provide quality reports and that the Commission shall be empowered to adopt implementing acts setting out the modalities, content and frequency of the quality reports. On this basis, it is proposed to: Move from three to two types of quality reports: (1) quarterly accuracy reports and (2) annual quality reports including quality reports for the AHMs. In order to comply with the IESS FR while keeping some flexibility (and to be able for example to focus on a specific issue each year), the following approach is proposed for discussion with the LAMAS members: The implementing measure of the IESS FR dedicated to the LFS would only define the main quality requirements and components for the quality reporting (quarterly accuracy report and the annual quality report) (see first proposal in annex 2) Detailed quality reporting will to a large extent remain in accordance with the current practice. In terms of timetable, quarterly accuracy reports would be transmitted within two weeks of the quarterly data transmission. Annual quality reports, including the quality report of the AHMs, would be transmitted by the end of August T+1. C. Use of administrative data: The current LFS Council Regulation N 577/1998 (article 2.3) states that: The variables used to determine labour status and underemployment must be obtained by interviewing the person concerned, or, if this is not possible, another member of the household. Other information may be obtained from alternative sources, including administrative records, provided that the data obtained are of equivalent quality. This implies that administrative data is not allowed for these two parts of the LFS. In line with the work of the Task Force on the harmonisation of the measurement of employment and unemployment, it is proposed in the context of the implementing measure of the IESS FR to keep this principle but to reformulate it in terms of detailed topics as follows: The variables for the detailed topics "employment status" and "looking for work" shall be obtained by interviewing the person concerned, or, if this is not possible, another member of the household. In addition, in line with the work of the Task Force on the measurement of absences and working time, Eurostat is proposing to extend the requirement to use interview data to the detailed topic 'working hours. D. Compulsory transmission of monthly unemployment data In the current situation, Eurostat releases data on the Monthly Unemployment Rate ( MUR ) around 30 days after each reference month. This news release has become one of the flagship publications 5

7 of Eurostat and MUR data are now used in the context of the Macro Imbalance Procedure (MIP) scoreboard. However, data inputs are still collected from Member States on a Gentlemen s agreement basis and the methodology to compile MUR estimates is not subject to common quality requirements. A dedicated Task Force ( TF-MUR ) has been mandated by the LAMAS to define such quality requirements and will report to the LAMAS meeting of June 2015 (final report). An interim report is presented under item 3.5 of this agenda. The future legal basis for MUR transmissions should have two layers. The first layer, to be part of the IESS Framework Regulation, should only describe the monthly input to be transmitted by the Member States and the corresponding deadline. It should mention that only macro-data (aggregates) shall be transmitted with respect to a specific reference month. Such data transmissions shall be regarded as standard input for the compilation of monthly unemployment statistics, but not necessarily as the final estimates released by Eurostat. As a matter of fact, these data may have to be further processed in case the data transmitted by a country do not meet some quality requirements to be agreed on by LAMAS on the basis of TF-MUR proposals. The second layer would consist of a dedicated implementing regulation specifying in particular: a) the nature, the length and the level of detail of the time series to be transmitted; and b) the abovementioned quality requirements. In the IESS Framework Regulation, the text would then read as follows: Member States shall transmit to Eurostat monthly aggregated results, for the compilation of monthly unemployment statistics, within the deadline set in Annex V. The nature, the length and level of detail of the time series to be transmitted together with the quality requirements shall be described in implementing acts adopted in accordance with the examination procedure referred to in Article 15(2). LAMAS members are asked to: Take note of the state of play of the discussions on the modernisation of social statistics by ESSC and DSS Express its views on the proposals in terms of standardised variables, quality reporting, use of administrative data and compulsory transmission of monthly unemployment data 6

8 Annex 1: Elements covered by the annexes of the IESS FR as regards the LFS Topics to be covered For all domains Topic Detailed topic Technical items Person and household characteristics Labour market participation and perspectives Technical items Person characteristics Household characteristics, including monthly income (t.b.c.) Self-perceived main activity status Elementary job characteristics Education and training Elementary education attainment Health Elementary health status (t.b.c.) Domain Topic Detailed topic Labour market Labour market participation and perspectives Job tenure and previous work experience Working conditions including working hours and working time arrangements Education and training Health Care needs Socio-economic background Employment status Job characteristics Occupation Supervisory responsibilities Characteristics of the workplace Looking for work Second job Looking for other job Reconciliation between work and family life Young people on the labour market Labour market situation of migrants Transition into retirement Start of job Way job found Previous work experience Working hours Working time arrangements Work organisation Educational attainment Participation in education and training (4 weeks) Participation in education and training (12 months) Health and safety at work Care needs 7

9 Precision requirements 1. Precision requirements for all data collections are expressed in standard errors and are defined as continuous functions of the actual estimates and of the size of the statistical population in a country or in a NUTS 2 region. 2. The estimated standard error of a particular estimate SE (p ) shall not be bigger than the following amount: 1. p (1 p ) f(n) 3. The function f(n) shall have the form of f(n)=a N+b 4. The following values for parameters N, a and b shall be used. p N a b Labour market domain: 3 precision requirements Estimated (national) quarterly unemployment-topopulation ratio Estimated (national) quarterly employment-topopulation ratio Estimated quarterly unemployment-to-population ratio in each NUTS II region Country population aged residing in private households, in million persons and rounded to 3 decimal digits Country population aged residing in private households, in million persons and rounded to 3 decimal digits Population aged in the NUTS II region residing in private households, in million persons and rounded to 3 decimal digits See footnote 5. Should countries have negative f(n) value with the parameters expressed above, they will be exempted from the corresponding requirement. For the estimated ratio unemployment to population in each NUTS II region, the function f(n) is defined as follows: 1300, if N r, million inhabitants f(n r,15 74 ) = { N r,15 74, if N r,15 74 < million inhabitants 8

10 Sample characteristics 1. Sample characteristics of the Labour Market domain shall be: (a) (b) The national sample for the reference quarter (aggregation of consecutive reference weeks) shall be distributed uniformly in the 13 weeks of the quarter. The sample for the reference quarter (in each NUTS II region) shall be distributed in the 3 months proportionally to the number of weeks in each month The sample shall have an infra-annual rotation scheme. There must be a minimum sample overlap of 20% between the same quarters in consecutive years and of 50% between consecutive quarters. Periodicity 1. For the Labour Market domain, the data collections shall consist of quarterly, annual and biennial information. Data for regular modules shall be collected every eight years and for supplementary modules every four years (to be confirmed after LAMAS meeting of December 2014). Data provision and transmission deadlines 1. For the Labour Market domain, the Member States shall provide Eurostat with: (1) Validated micro-data with the results of the data collection without direct identifiers, according to the following two-step procedure 2 : During the first three years of implementation of this regulation: o o For the quarterly data: transmission within ten weeks of the end of the reference period For the annual variables and for regular and supplementary modules: transmission by 31 March of year Y+1 At the occasion of the fourth year of implementation and onwards, the transmission shall be as follows: o For the quarterly data: transmission of data for quarters 1, 2 and 3 respectively by 29 May, 29 August and 29 November of the same year and of data for quarter 4 by 28 February of the next year o For the annual variables and for regular and supplementary modules: transmission by 15 March of year Y+1 In years when those deadlines fall on a Saturday or Sunday, the effective deadline shall be the next Monday. (2) Monthly aggregated results for the compilation of monthly unemployment statistics forwarded at 25 days after the reference month. 2 To be confirmed after the LAMAS meeting of December

11 Macrodata 1. Member States shall transmit to Eurostat monthly aggregated results for the compilation of monthly unemployment statistics within the deadline set in Annex V. 2. The nature, the length and level of detail of the time series to be transmitted together with the quality requirements shall be described in implementing acts adopted in accordance with the examination procedure referred to in Article 15(2). 10

12 Annex 2: Proposed main requirements, components and transmission deadlines for quality reports to be transmitted by Member States The quality reporting on the organisation of a labour force sample survey in the Community is composed of a quarterly accuracy report and an annual quality report. The quarterly accuracy report shall contain basic information on the accuracy and reliability of the survey and describe changes in the basic concept and definitions that affect the comparability over time. Member States shall forward the quarterly accuracy report within two weeks after the defined quarterly data transmission. The criteria of annual quality report orientate itself onto the Standard for Quality Reports Structure of the European Statistical System. The content of the annual quality report shall be whatever information is required about the accuracy, timeliness and punctuality, accessibility, comparability and coherence of the labour force sample survey and describe the sample design and methods used conducting the survey. Member States shall forward the annual quality report by the end of August T+1. 11